Some Effects of Environmental Pollution on Development of Infants
Section 1.a (summary): There have been large increases in major non-communicable disorders, including neurological conditions, among American children in recent decades, beginning in the 1960’s and especially in the ‘70’s; this is according to U.S. government statistics and highly-published researchers. Mortality rates have also increased dramatically among non-Hispanic whites born in the U.S. after 1970.
(Preview of Figure 13c)
In this chart, in the left-hand section of the red line, observe the declining mortality of 30-year-old U.S. non-Hispanic whites born in the 1960's. In the right-hand section of the red line, observe the increasing mortality of the same age group of U.S. whites who were born in 1970 and later.
Increasing neurological disorders have affected males principally. Aside from diagnosed disorders, there have also been other adverse developments (especially worse performance in school) among boys and young men in the general population.
Since genetics could not change so quickly, the origins of the increasing disorders and adverse trends are almost certainly to be found in environmental exposures. Toxins present in food are a very important form of such exposure. (The full text, including references to authoritative sources, is in Section 1.a)
Section 1.b (summary): There were huge increases in numbers and quantities of chemicals after World War II, which have resulted in major increases in toxic pollutants in human environments since that time. A document of the U.S. Agency for Toxic Substances and Disease Registry (ATSDR) dealing with toxins in human milk referred to "detection of all five of these and other chemicals in breast milk, combined with the knowledge that each is able to alter neurological development.... All four studies demonstrated statistically significant associations between concentrations of these persistent chemicals in maternal fluid samples and deficits in neurological development of the children...." 85f According to experts writing in 2004, "these substances have caused contamination of human milk only during the last half century, and long-term health impacts are now being discovered."6b (The full text, including references to authoritative sources, is in Section 1.b)
Section 2, summary: Contrary to popular belief, the period after birth is a period of high vulnerability to effects of developmental toxins; also, exposures to developmental toxins are especially high during the early postnatal period.
Section 2.a (summary): Vulnerability of the brain’s development to toxins in the early period after birth:
Many authoritative sources (EPA, NIH, ATSDR, leading scientists, a council of the U.S. National Academies, WHO, etc.) recognize the early-postnatal period as being a time of vulnerability to effects of environmental toxins, because of the major development taking place after birth. Development of the brain (see preview charts below) is known to be vulnerable to effects of toxins during the period when the developmental processes are active.
Fig. 1: Brain development (preview of Figure 2)
(Section 2 summary, cont.) Vulnerability of development to environmental toxins is often greater after birth than prenatally, as indicated also by
a) over thirty studies that have found greater effects of toxins during the postnatal period than prenatally, and
b) a clear EPA statement saying that the postnatal period is the time of greatest sensitivity to a serious neurodevelopmental toxin (PBDEs), which increased dramatically in humans in the late 20th century.
(Full text, including references to authoritative sources, in Section 2.a)
Section 2.b (summary): Exposures to developmental toxins are far greater postnatally than prenatally, due to the placental barrier and to the long-term maternal accumulations of toxins that can be transferred via lactation after birth.
Section 2.b.1 (summary): According to highest authorities, developmental toxins accumulate in mothers over decades, leading to increasing concentrations of the toxins in maternal fat and then in breast milk as the mother grows older. Many studies have found ASD risk to increase with maternal age.
Section 2.b.2: Excretion in breast milk substantially reduces levels of toxins in mothers as the toxins are transferred to infants; those toxins build up rapidly in infants during a period of major sensitivity of development.
Section 2.b.3: Transfers of such toxins during gestation are so minimal that (as found in several studies).the maternal burdens of the toxins do not decline during gestation.
Section 2.b.4: Additional information about toxins' becoming concentrated in human milk, especially as compared with adult exposures and prenatal transfers.
(Details and authoritative sources in Section 2.b)
Section 2.c (summary): The vulnerability to developmental toxins of concern here is related to toxins to which most infants are exposed, at environmental background levels.
Section 2.d (summary): Several studies have shown that direct or prenatal exposures to environmental toxins may have little or no effect on children, whereas those same toxins can be very harmful as a result of their much greater exposures to infants that take place indirectly via lactation. The secondary, greater exposures could be described as "second hits." (Details and authoritative sources in Section 2.d)
Section 2.e (summary): Many studies have found autism diagnoses to be associated with histories of breastfeeding, including in dose-response relationships. (This is a brief preview of what is presented fully in Section 4)
Section 3. (summary) The State of Washington Department of Ecology considers breastfed infants to be at the very highest level of exposure to common environmental toxins, sharing that top position with only two other groups, both of which have unusually high exposures to contamination or industrial pollution. (Full text, including references to authoritative sources, in the Section 3 Introduction)
There is considerable scientific evidence indicating that these toxins, at common levels of exposure, can cause adverse effects in children. Five different developmental toxins have each been found to be typically present in human milk in concentrations exceeding established safe levels, while infant formula has been found to have very little or no detectable concentrations of those toxins. (many sources cited in the full text in Section 3)
Shown below are observed effects on 9-year-old children of fairly common developmental exposures to PCBs.
(Note that the bottom lines are not at 0)
Section 3.a (summary): PCBs, in postnatal as well as prenatal exposures, have been found in many human and animal studies to “impact normal brain development,” at levels to which humans are very often exposed; specific effects observed have included distractibility, reduced IQ, hearing loss, and poor memory and school performance; PCBs have been found to be present in human milk in doses 63 to 270 times the minimal risk level established by the U.S. Agency for Toxic Substances and Disease Registry, and essentially absent in infant formula. Exposures in human environments are still high decades after most uses were stopped. (Full text, including references to authoritative sources, at Section 3.a)
Section 3.a.2: (Summary) Lactational transfer of PCBs to infants: Exposures to PCBs after birth are likely to be far higher than before birth, especially with breastfeeding for a few months or more.
(Full text, including references to authoritative sources, at Section 3.a.2)
Section 3.a.3: (Summary) PCB exposures and neurological development: A scientist who is a member of both the U.S. National Academy of Sciences and the Institute of Medicine stated after completion of a laboratory experiment, “If human fetal and infant effects parallel rat impacts, we would predict that there would be a correlation between the PCB/PBDE levels in human breast milk — and in infant blood — with the probability of autism onset.” (Both PCBs and PBDEs are highly present in typical human milk.) Experiments with animals have been authoritatively determined to accurately indicate, or to underestimate, effects of these chemicals on humans.
Preview from Section 3.a.4
Section 3.a.4: (Summary) PCBs and learning ability: There are good reasons to see PCB exposures as having a role in the (otherwise unexplained) rapid increases in neurological impairments in children, especially among males, in recent decades.
Section 3.a.5: (Summary) There is considerable evidence indicating that postnatal exposure to PCBs is more harmful than prenatal exposure. The U.S. ATSDR, in its major document on PCBs, states that "breast milk exposures have been associated with neurodevelopmental deficits in newborn and young children."16p
Section 3.a.6: (Summary) There is good evidence indicating that autism results from impaired connectivity in the brain, and that PCBs harm formation of connections in the brain; crucial connections are being formed during the same early postnatal period when breastfed infants are ingesting large amounts of PCBs.
Other adverse outcomes of PCB exposures include effects on male and female hormones, reproduction, cancer risk, and (with especially great potential for long-term harm) disruption of the blood-brain barrier.
-- There have been high and increasing levels in breast milk of HBCD, which the EPA has designated as a “High hazard” for developmental neurotoxicity; levels have been found to be thousands of times higher in some homes than in others, with higher human exposures found to be associated with presence of typical electronic devices and carpeting. And levels of HBCD in breast milk have been found to vary according to presence in the home of furnishings that contain flame retardants.
Preview of Fig. 5.1
-- There have been ongoing high lactational exposures to PBDEs, at levels over ten times higher than a few decades ago; effects of PBDE exposure, including at human background levels, have been found to include developmental neurotoxicity, hyperactivity, learning disability, memory defects, and poor social competence.
-- The EPA says that sensitivity to this toxin is greatest postnatally.
(Full text, including references to authoritative sources, at Section 3.b)
Section 3.c (summary): Dioxins, according to a major toxicology textbook, can have “particularly devastating” effects on development, and children are most susceptible to them during development and nursing. Dioxin levels have also been associated with harmful neurological effects at ages 12 to 15, including as related to attention deficits and learning disability; and those effects were linked with elevated levels of dioxin that are very common within the general population. Breastfeeding has been found in studies to have been the main determinant of long-term levels of that toxin.
(Full text, including references to authoritative sources, at Section 3.c)
Preview of Fig. 15
Section 3.d (summary): Mercury (Hg) is a known neurodevelopmental toxin that often already exceeds the established safe level in infants at birth, and then a large part of a mothers’ long-term accumulation of that chemical is usually rapidly transferred to an infant in breast milk; “clearly documented toxic effects (of mercury) on the immature brain” are authoritatively recognized to be able to occur several months into the postnatal period.
One of the many different forms of mercury, as found in vaccines, has been found not to cause autism, whereas mercury in general has been closely associated with autism and other cognitive impairment, in many studies. According to the EPA, “There is general agreement that the nervous system continues development in post-natal life and that methylmercury can adversely affect the developmental processes.” (Methylmercury is common in human milk.)
(Full text, including references to authoritative sources, at Section 3.d)
Section 3.e: (Summary): Pesticides and lead have been found to be extensively present in human milk and undetectable in infant formula in the U.S., in recent authoritative studies. According to impeccable sources, "No amount of lead exposure is safe," and "Early postnatal exposure appears to be more effective than exposure prenatally," in reducing IQ. 11% of elderly womens' lifetime accumulations of lead were found to have been transferred to their developing infants via lactation.
Associations have been found in many studies between autism and heavy metal (lead and mercury) exposure at levels often occurring in developed countries. Many studies have also found associations closely linking ADHD with such lead exposures.
(Link to full text, with citations of authoritative sources, at Section 3.e)
Section 3.f: (Summary) These toxins are known to often have effects that are apparent in later childhood and/or adulthood, while not being noticed earlier. Various studies have failed to find adverse effects of the developmental toxins in human milk when testing children only in early childhood; surprisingly, the U.S. CDC tells about failure to find adverse effects in early childhood as if that is all that really matters, while saying nothing about the excellent evidence indicating adverse long-term effects. (Link to full text, with citations of authoritative sources, at 3.f
Section 3.g (Summary): Five toxins discussed above have been found to be present in human milk in concentrations that exceed established safe levels either by several times (in two cases), or by scores to hundreds of times (in two cases), or by an unquantified but substantial amount, in the case of lead; and the combined effects of those toxins are likely to be far more than merely additive, according to excellent evidence. Infant formula has been authoritatively found to have little or none of these toxins. Details and citations of sources are to be found in Section 3.g.
Section 3.h (Summary): Despite some reductions in pollution, emissions of developmental toxins that enter human milk continue at hazardous levels, from sources to which humans are closely exposed.
Section 4. (Summary): Studies that have found prevalence of neurological disorders to correlate with breastfeeding:
Section 4.a.1: (Summary): Studies finding autism to be higher in relation to greater duration of breastfeeding:
Four studies published in 2009 and later have found positive correlations between breastfeeding duration and autism prevalence, forming an overall dose-response relationship. One of those studies investigated data from all 50 U.S. states and 51 U.S. counties and found that "exclusive breast-feeding shows a direct epidemiological relationship to autism," and also, "the longer the duration of exclusive breast-feeding, the greater the correlation with autism." A Canadian study investigated a population of over 125,000 and found correlations of that same general nature. Another U.S. study found a 185% increase in autism among children who had been breastfed for at least six months, compared with controls, with a very high level of statistical significance. (Full text, including references to sources, at Section 4.a)
A 2015 study based on 902 ASD cases found a 47% greater risk of ASD among breastfed children than among controls. In a 2012 study, of 33 children who had been breastfed by mothers with above-average levels of PCBs, 9 had mental delay, whereas none of the 11 non-breastfed children with similar PCB exposures had mental delay.
Another 2015 study found a 3% increase in odds of autism diagnosis for each additional month of any breastfeeding and a 4% increase for each additional month of exclusive breastfeeding, as well as a 30% increase in odds of autism with six months of exclusive breastfeeding versus no breastfeeding.
Statistical significance is discussed, with regard to above studies.
In a 2011 study it was found that gestational exposure to maternal PBDEs had only relatively minor or no significant adverse effect, but exposure to those same mothers' PBDE levels via breastfeeding had effects including an 80% increase in relative risk of attention-deficit problems and a 160% increased relative risk of poor social competence (a basic trait of ASD).
(Full text, including references to sources, at Section 4.a)
Section 4.a.2: (Summary): Other studies, including very large ones, linking breastfeeding with risk of autism and cognitive deficits:
Section 4.a.2.a (Summary): The CHARGE study in California, the especially large SEED study (carried out at six different sites across the U.S.), and another important U.S. study have found that children later diagnosed with ASD or ADHD had been breastfed for average durations of 6.7, 7.3 and 5.9 months. By contrast, median durations of breastfeeding in the general U.S. population in the relevant years were found in all of the four large surveys to be about 3 months.
Section 4.a.2.b (Summary): In two separate studies, duration of breastfeeding was linked with deficits in executive function, which is seen to be integral in autism and important in ADHD.
Section 4.a.2.c (Summary): In children who had been breastfed for six months, emissions from municipal incinerators were found to be closely associated with all of the autism-related outcomes that were investigated; on the other hand, effects of those same emissions as found in children in the general population were far fewer and smaller. (Full text at Section 4.a.2.c)
Section 4.a.2.d (Summary): PCBs from breast milk have been found to greatly reduce activity levels of children. Activity is necessary for normal development of the brain, according to excellent authority.
Section 4.a.2.e (Summary): The profile of autism matches the observed effects of inactivity: major deficits in perceptual abilities, but not necessarily in intelligence.
Section 4.a.3 (Summary):
(a) Fairly common pesticide exposures have been closely linked with autism risk, in many studies;
(b) breast milk is the predominant pathway for pesticide residues to most infants, at a time of infants' high developmental vulnerability.
(a) and (b) above should be considered at the same time, when reflecting on toxins that could lead to autism and the pathways by which those toxins are known to reach infants.
In a 2007 study, autism prevalence was found to be six times as high as normal near California agricultural fields where pesticides were applied; autism prevalence varied in correlation with distance from the fields and with poundage of pesticides applied. A 2012 study found that the time of peak correlation of pesticide exposure with autism incidence was in the year after birth. Strong evidence indicates that infant exposures to pesticides are predominantly via breastfeeding.
(Full text, including references to authoritative sources, at Section 4.a.3)
Section 4.a.4 (Summary): Exposures to airborne toxins in the first year after birth have been found to be strongly associated with risk of autism. Logically implicated toxins are taken in by infants via the pathway of breastfeeding in far greater concentrations than by inhalation or other means. (Full text, including references to authoritative sources, at Section 4.a.4)
Section 4.b: (Summary): Other studies have found two substantial, parallel declines according to birth order: autism incidence, and concentrations of toxins in breast milk. (see chart and caption below) Those two declines’ taking place in parallel might not be coincidence.
Concentrations of lipophilic chemicals in human milk decline with additional births, as the mother’s long-term accumulations are excreted in the milk. It is known that some of those chemicals are neurodevelopmental toxins.
Preview of Fig. 6.a
In parallel with the above, odds of a child’s having autism also decline greatly with birth order, for unknown reasons.
But some people might see a logical connection between the decline in autism with birth order and the decline in developmental toxins ingested (see above) in relation to birth order, considering that several toxins are present in human milk in concentrations far in excess of established safe levels. (Full text, including references to sources, at Section 4.b)
Section 4.c: (Summary): Studies that have associated autism with less breastfeeding have had fairly obvious, serious weaknesses. Where genuine (but minor) autism-related benefits were associated with breastfeeding, the evidence for a favorable outcome was found to decrease with increase in breastfeeding intensity.
Fig. 7, preview
Section 5. (summary): There have been major increases in breastfeeding, which paralleled the major increases in disorders in children; increases in breastfeeding began slowly in the late 1960's, were rapid for a decade or so beginning in the early 1970's, and slowed down later, and increases in disorders followed a similar pattern. The increases of breastfeeding in relation to the 1970 rate were especially dramatic in the case of longer-term breastfeeding
In addition to major increases in what is probably the predominant pathway of developmental toxins to infants (breastfeeding -- see Section 6 just below), concentrations of some major toxins (brominated flame retardants) were also rapidly increasing in the environment and in human milk during recent decades. (Full text, including references to authoritative sources, at Section 5)
Section 6. (summary): Breastfeeding may be a unique pathway of widespread infant exposure to developmental toxins in doses exceeding established safe levels. Scientists who should know about developmental toxins apparently do not know of any other toxins to which infants or fetuses are widely exposed in doses exceeding established safe levels, aside from the five that are discussed in this article. (Full text at Section 6)
Section 7. (summary):
Section 7.a, summary: Far more young males than young females have been diagnosed in recent decades with certain neurological disabilities, including ADHD and ASD. In addition, boys and young men in general have been falling behind in education. There has been nothing but speculation as to the causes of these relatively recent adverse developments among young males. There are good reasons to see transfer of certain toxins via breastfeeding (toxins that are known to have adverse effects specifically on male cognitive development) to have been underlying these trends.
For a more complete summary, with link to the full text on this topic, go to Section 7.
Section 7.b (summary): The five specific neurodevelopmental toxins that have been found often to be substantially present in typical contemporary human milk (Section 3.g) have all been found also to have adverse effects on males far more than on females. This is very compatible with the high ratio of males to females diagnosed with ASD and ADHD or falling behind in education.
Preview of Figure 8
School test scores in relation to child mercury levels
For the complete text that is summarized above, go to Section 7.b
Section 8: (summary):
Exposures of children to pesticides at current background levels correlate well with reduced mental capacities as well as with ASD, ADHD, and other neurodevelopmental problems, in many studies. Human milk has been found to normally contain many pesticides, and infant formula in the U.S. has been found to contain essentially none.
(For a more complete summary, with link to the full text on this topic, go to Section 8.)
Section 9: (summary):
The time when toxic exposures take place is critical in determining their effects; vulnerability is greatest when development is active.
Fig. 11, preview
(Section 9 summary, cont.) Breastfeeding, transferring several different developmental toxins to infants, each in very significant amounts, is by far most prevalent during the early months after birth; also, the concentrations of most of those toxins are especially high during earlier breastfeeding. The combination of the above leads to far greater total transfers of toxins soon after birth than at any other stage of development. The feeding type that is at the base of those high early-postnatal transfers went through very major increases in the late 20th century, in the U.S. and many other countries. (For full text including citations of authoritative sources, see Section 9)
We should consider the possibility of a connection between
(a) the very large increases that took place in the late 20th century in infants' exposures to toxins during the early months after birth, as described above, and
(b) the concurrent, unexplained major growth in impairments in specific functions that are especially vulnerable to toxins during the early postnatal period.
For details and authoritative sources, see Section 9.
Evidence of the kind that is typically cited (observational) says that breastfeeding reduces risk of obesity. So we should expect to see a substantial decline in childhood obesity in the decades following the rapid increases in breastfeeding. See below what actually happened.
Preview of Fig. 11.a
Observational studies also say that diabetes is reduced by breastfeeding, but diabetes increased to epidemic proportions among children and young adults during the decades of increasing breastfeeding.
To see the complete text and chart that are summarized above, go to Section 9.
Section 9.b: (summary): In those highly vulnerable early-postnatal months, other neurological development (aside from that discussed in 9.a above) is also taking place that has not been going well during the period of major increases in breastfeeding; functions that are developing in those months include areas of attention, speech, gaze control, and many other areas.
Section 10 (summary): Essentially all of the studies that have found benefits of breastfeeding have been observational studies, according to the U.S. Surgeon General; and those studies have mainly assessed outcomes in early childhood. There are fundamental weaknesses in observational studies, as recognized by leading experts on medical evidence; and there are excellent reasons to pay far more attention to long-term outcomes.
Section 10.a (summary): Authoritatively recognized low quality of the kind of studies (observational) that have found benefits of breastfeeding; confounders.
Section 10.b (summary): The adherer effect, a well-established pattern according to which faithful followers of the established advice (such as to breastfeed) normally have better health outcomes, including when just taking placebos. That effect probably explained why the pooled estimate of effect from the best-quality observational studies indicated a 44% - 50% reduction in coronary heart disease with use of hormone replacement therapy (HRT) by postmenopausal women; this led the American Heart Association, the American College of Physicians, and the American College of Obstetricians and Gynecologists all to recommended HRT to older women as a means of warding off heart disease and osteoporosis, based on what was alleged to be "strong" and "consistent" evidence. That was until the results came in from high-quality (randomized) studies finding that, if anything, HRT led to worse health effects. For details and sources, see Section 10.b.
Section 10.c (summary): The confounder of socio-economic differences: Women of higher socio-economic status, and their children, (a) are normally of better health and (b) are much more likely to breastfeed in the countries where most studies are carried out; so observational studies in those countries typically arrive at the unjustified conclusion that breastfeeding leads to better health outcomes. Those are the studies that promoters of breastfeeding choose to cite, as opposed to the scores of studies that have found adverse effects of breastfeeding.
For details and authoritative sources, see Section 10.c.
Section 11 (summary): Promotion of breastfeeding is based mainly on observations of what appear to be short-term benefits; the evidence about longer-term effects strongly indicates worse long-term health outcomes. (For full text including citations of authoritative sources, see Section 11.)
Preview of Figure 12.5
Section 12: (summary) :Randomized trials are considered to be the gold standard of study types. Here are charts provided in a 2016 randomized study dealing with breastfeeding in relation to allergies, showing results similar to many other outcomes found in the study.
One could search in vain for any reference to the many studies negative to breastfeeding such as those also in Section 4, or even any references to "toxins," in the Policy Statement on breastfeeding of the American Academy of Pediatrics. Likewise the AAP makes no mention of the large number of other peer-reviewed studies that have found adverse effects of breastfeeding; or of anything about the public health history of recent decades showing substantial increases in the disorders that are claimed to be reduced by breastfeeding, following major increases in breastfeeding.
(Full text, including references to authoritative sources, in Section 12)
Section 13 (summary): Final remarks
Section 13.a: (summary): A reasonable question to ask:
(a) non-communicable disorders have been greatly increasing among children in recent decades -- for basically unknown reasons (Section 1),
(b) the developmental processes taking place after birth are authoritatively recognized to be vulnerable to toxins ingested postnatally (Section 2),
(c) toxins known to be typically at high levels in human milk have been found to lead to effects similar to symptoms of the increasing disorders (Section 3); and
(d) positive dose-response relationships have been found between breastfeeding and autism or ADHD, in many published studies (Section 4),
it is reasonable to ask the following question of the medical organizations that promote breastfeeding:
How has it been determined that increasing exposures to developmental toxins in human milk (Figures 7 and 7.a) have not been causing increases in disorders that outweigh the benefits of breastfeeding?
The U.S. medical associations that promote breastfeeding (pediatricians, family physicians and obstetricians and gynecologists) do not respond after being repeatedly asked the above question, with minor variations. They do not deny the presence of high levels of developmental toxins in human milk. And they do not deny the validity of the evidence indicating that those toxins, at levels such as in breast milk, can have adverse effects.
Since there is substantial peer-reviewed scientific evidence to support (a) through (d) above, it would seem to be appropriate for those physicians’ organizations to consider that evidence before they recommend feeding infants a food that has been authoritatively determined to contain several developmental toxins in concentrations far exceeding established safe levels. That kind of consideration of evidence would be especially called-for now that major, unexplained increases in child neurological disorders have followed the major increases in breastfeeding. (see Sections 1 and 5.) And a response to the above question would be even more in order considering the apparent absence of widespread infant exposure of infants to even one toxin (not to mention 5), by any pathway other than breastfeeding, in doses exceeding established safe levels (see Section 6).
If careful study had been carried out on such an important matter of public health, a written record of the analysis of the important evidence ought to be available to the public. But there appears to be no such record available from any of the organizations that promote breastfeeding, which implies that the breastfeeding recommendations may be based on something other than careful consideration of the important evidence.
Section 13.b: (summary): More reasons to question the advisability of breastfeeding:
An alternative type of infant feeding is readily available that
(a) contains less than 7% (and usually less than 1%) as much of the toxins discussed here, and
(b) was the standard feeding type for the entire U.S. generation that was born “throughout the mid-20th century,” according to the American Academy of Family Physicians. Remember that childhood disabilities and disorders, which by now have reached high levels, were reported to have first started emerging as major chronic conditions in the 1960’s, followed by more substantial increases beginning in the 1970’s and later (see Section 1). The generation that was seldom breastfed did not have the rapidly-increasing childhood health problems that were to be observed after that generation was born. (Details and authoritative sources in the full version of Section 13b)
That is something to think about, along with consideration of the increased presence of toxins in the environment and in breast milk since the mid-20th century and the considerable scientific evidence about effects of those toxins.
Section 1.a: Substantial increases in major non-communicable disorders, especially among children, beginning in the 1960’s:
As reported in a 2008 publication of the U.S. Center for National Health Statistics: “Over the past three decades in the United States, behavioral and learning disorders have emerged as major chronic conditions affecting the development of school-aged children and adolescents.”5 A 2012 study by a team of scientists from the U.S., France, U.K., Denmark and New Zealand referred to the “many… major diseases – and dysfunctions – that have increased substantially in prevalence over the last 40 years;” their primary focus, regarding possible causal factors in the environment, was on the period of early development.6 According to a document of the EPA published in 2015, "Some researchers have stated that the prevalence of certain neurodevelopmental disorders, specifically autism and ADHD, has been increasing over the last four decades." (citing five studies, and expressing no disagreement with the statements described.) The EPA continued, "Surveys of educators and pediatricians have reported a rise in the number of children seen in classrooms and exam rooms with behavioral and learning disorders." (citing three studies)25e No specific beginning of increases of mental retardation/intellectual disability has been determined, but the EPA (in a 2017 web page) points out that it has increased by more than half over the last two decades.6e
Going by the time periods related to increases in disorders as stated above, the birth years of American children at the beginning of the period of substantial increases in disorders were apparently in the 1970’s.
(Preview of Figure 13c)
Evidence published in 2017 broadens the area of concern to include all-cause mortality, when looking at outcomes in adulthood. An inquiry into origins of disease in adulthood should look into toxic exposures that could have occurred during the early developmental period, a period of high vulnerability to toxins, when exposures are recognized to be able to lead to serious health consequences in later life.6r In this chart, one can see worse outcomes for non-Hispanic white American 30-year-olds beginning in the early 2000's, which should cause researchers to look into what may have been toxic in environments of these people when they were developing in the 1970's.
There is much more on this topic later, pointing to this same time of transition to declining child health. (see Section 13.b)
The causes of the vast majority of those disorders are unknown. It is clear that their origins probably lie in environmental exposures (which include food consumed), since genetics could not change so much during such a short time span.7 It is recognized that varying individual genetic susceptibilities influence the effects of the environmental toxins.
Despite the popular belief that the prenatal period is a time of far greater vulnerability to developmental toxins, development actually continues to be highly vulnerable to toxins after birth also. And exposures to developmental toxins increase greatly after birth. Evidence to support these last sentences will follow.
Section 1.b: Substantial increases in toxic chemicals in human environments beginning after World War II:
As of the 1940's, relatively few environmental chemicals were recognized as being hazards to the growing child. It was after World War II that chemicals developed during the war were modified for use as pesticides and for other common uses and were produced in increasing numbers.6a, 6c According to data from the American Chemical Society, production of chemicals increased over 350% in the 25 years after 1947.6d Referring to specific types of chemicals that are of special concern for child development, researchers who are authors or co-authors of over 340 studies have stated, "It is now recognized that numerous endocrine-disrupting chemicals have been released into the environment in large quantities since World War II;" among other chemicals the authors listed in this category were many pesticides and PCBs, dioxins, and mercury;6c PBDEs are also recognized endocrine disruptors.
Speaking in broad terms, experts on the subject of environmental toxins pointed out in 2004 that "these substances have caused contamination of human milk only during the last half century, and long-term health impacts are now being discovered." 6b
The U.S. Agency for Toxic Substances and Disease Registry (ATSDR) issued a major document in 2004 dealing with toxins in human milk,85f in which the Agency stated, "the detection of all five of these (chemicals, including dioxins, DDE, PCBs and methylmercury) and other chemicals in breast milk, combined with the knowledge that each is able to alter neurological development, has led to epidemiological studies (in four countries).... All four studies demonstrated statistically significant associations between concentrations of these persistent chemicals in maternal fluid samples and deficits in neurological development of the children...." (emphasis added)
The ATSDR continued: "The results from the human studies are thought to implicate gestational exposure to the persistent chemicals to a greater degree than lactational exposure." (emphasis added) Aside from the very tentative nature of this suggestion that lactational exposures might be less harmful than prenatal exposures, there are other reasons not to feel comfortable about promoting an infant feeding type that is known to contain harmful chemicals:
a) As of the year of publication of the above-quoted document (2004), numerous studies had not yet been published that since then have found greater effects of postnatal/lactational exposures than of prenatal exposures to those toxins; see Section 4.a below regarding autism- and ADHD-related studies; also see http://www.breastfeeding-studies.info concerning more recent studies of other health areas, in which lactational and/or other postnatal exposures were found to be harmful, sometimes more harmful than prenatal exposures.
b) Even if lactational exposures cause less than half of the total harmful effects (the ATSDR is very uncertain about this), that is nevertheless likely to be substantial harmful exposure that could be prevented relatively easily. Much more on this will follow.
We should also consider the scientific merits of the evidence that causes people to feel that it is justifiable to expose developing infants to these chemicals as part of the pursuit of benefits of breastfeeding.
-- Essentially all of the studies that have found benefits of breastfeeding have been observational studies, according to former U.S. Surgeon General Regina Benjamin.a1a The leading authorities on medical evidence have determined that evidence from observational studies is predominantly of low quality, with only exceptional ones reaching a medium level of quality. (This is discussed in much more detail, citing expert sources, in Section 10.)
-- Evidence for benefits of breastfeeding is based mainly on observations of what appear to be short-term benefits; the evidence about longer-term effects strongly indicates worse long-term health outcomes. (This is discussed in more detail in Section 11.)
-- Randomized trials are recognized as being far superior to observational studies. Randomized trials that have dealt with breastfeeding, as well as over 50 other peer-reviewed studies investigating associations with breastfeeding, have indicated unfavorable results of breastfeeding. (For details, see Sections 12 and 13.)
It is worth carefully considering which stages of a child's development are most vulnerable to disruption by chemicals such as were referred to above, chemicals that have greatly increased in human environments since World War II. There is a widespread belief that postnatal exposures to developmental toxins are of only minor significance in comparison with prenatal exposures, but ample evidence indicates that such a notion is misguided.
Section 2.a: Vulnerability of the brain’s development to toxins after birth:
A commission of the U.S. National Research Council (of the National Academies), when discussing “specific periods in development when toxicity can permanently alter the function of a system,” states that the developing brain and certain other organs “may demonstrate particular sensitivity during the postnatal period.” (p. 43)9 Statements by the U.S. Agency for Toxic Substances and Disease Registry (ATSDR), by an international group of 24 experts, and by other academic experts also point to postnatal periods of special vulnerability to toxins;9a the ATSDR refers to the immaturity of the blood brain barrier as a reason why the developing brain is especially vulnerable to toxins during the early postnatal period.9b
A publication of the U.S. National Academy of Sciences states, "The brain develops steadily during prenatal and early postnatal periods, which are considered as the most vulnerable windows for effects of environmental exposures."12a
A publication of WHO states that “developing organs are particularly susceptible to toxic insult, given the increased rate of cell division and immaturity….”13 In a publication of the National Academies Press, the authors state that “toxic exposures at a particular time would differentially affect the structures undergoing peak development.” (p. 60)12 In the charts above and in the left-hand caption, observe evidence of the above recognized characteristics of periods of vulnerability of the brain to toxins, all during the year after birth: immaturity, rapid increases in numbers of cells, and active development, including peaks of development of individual structures.
It is of interest to observe the specific structures that are shown to be undergoing peak development during the year after birth, and whose development is therefore vulnerable to toxins during the early-postnatal period. A 2007 study in the journal Pediatrics, by 12 researchers, states that damage to the cerebellum in infancy is among the highest risk factors for developing ASD (estimated 40-fold increase);13b that risk is second only to that of having an identical twin with autism.13c And note that the structures shown in the right-hand chart to be undergoing rapid development after birth are the ones that underlie specific functions in which disabilities have greatly increased in recent decades, including the hearing and vision functions (see Figure 11); deficits in those two functions would be closely related to core traits of ASD: impaired social interaction and communication. See Section 9 for a discussion of this topic. In parallel with the increases in disabilities that could have resulted from ingestion of certain toxins after birth, there have been major increases in early-postnatal exposures of infants to developmental toxins in recent decades, which will be discussed later. (see Section 5)
Also note the EPA statement in Section 3.b about “the most sensitive outcome” of PBDE exposure being “adverse neurobehavioral effects following exposure during the postnatal period." This is especially significant in that levels of this serious developmental toxin in humans and in human milk increased tenfold and more during the late 20th century. (see Section 3.b)
The ATSDR refers to the “particularly sensitive” periods of children’s neurological development to effects of mercury, which include “the early months after birth."93b
Lead is widely recognized to be neurodevelopmentally toxic and, according to a document of the World Health Organization, "Early postnatal exposure appears to be more effective than exposure prenatally," in reducing IQ. (p. 175 of reference 98a)
In a pooled analysis of seven European birth cohorts, postnatal exposures to PCBs were found to be significantly associated with reduced growth of the child, but prenatal exposures did not have any such effect.20c Note that PCBs (to be discussed in Section 3.a) appear to be recognized principally to be neurodevelopmental toxins; it would be reasonable to expect that, since PCB exposures reduce the body's growth predominantly after birth (see just above), they would be at least as effective (probably more so) in reducing growth of the brain after birth.
Early postnatal vulnerabilities that are specifically related to ASD:
a) As indicated above, development of the hearing and vision functions are especially vulnerable to toxins during the months after birth; consider the fact that those functions are critical to communication abilities and social relations, impairments in which are basic characteristics of ASD.
b) As seen in Figure 2 above, notice that connections in those structures are the specific components that are especially rapidly developing after birth, meaning that proper formation of connections is especially vulnerable to toxins during the year after birth; in that regard note that, according to a team of authors mostly with the U.S. National Institute of Environmental Health Sciences, "PCB exposure (usually high after birth -- see Figure 4) affects neuronal connectivity.... . Altered neuronal connectivity is a common feature of many neurobehavioral conditions, including ADHD and autism spectrum disorders (ASDs)." 37w
c) The brain's prefrontal cortex also undergoes substantial development after birth (implying vulnerability to toxins at that time), and deficits in functions guided by that structure have been authoritatively linked to autism. (see Section 4.a.2.d)
d) A 2017 study found that early overgrowth in brain volume (occurring between 12 and 24 months after birth) "was linked to the emergence and severity of autistic social deficits" that were emerging at that same time.37y Initial excess growth in the developing brain, followed by "synapse elimination or pruning" of the excess, is the normal pattern of development,37x so the overgrowth was very possibly a failure of the pruning process. The precise cause of the overgrowth was not determined, but all of the toxins discussed in this article are known to interfere with normal neurological development.
Aside from the above evidence linking postnatal toxic exposures with effects on neurological development, over 30 other studies have found postnatal exposures to developmental toxins to have greater effects than prenatal exposures, in a wide range of health areas.16b
There is considerable additional evidence about postnatal neurodevelopmental vulnerability to effects of environmental toxins, especially during the early months after birth, and sometimes indicating greater effects of postnatal exposures than prenatal. But to permit progressing now to other important points, that additional evidence has been placed in Appendix F.
Section 2.b: Exposures to developmental toxins are far greater postnatally than prenatally, due to the placental barrier and also because of lactational transfers of toxins that accumulate in mothers over time.
Specific developmental toxins of concern here will be discussed in much more detail in Section 3. For now, some of them will be briefly introduced by pointing out the following:
A list of recognized neurodevelopmental toxins was provided in a 2016 scientific consensus statement; that statement (the TENDR statement) was signed by a group that includes 38 scientists and MD's, many of whom are highly-published authors of scientific studies. It provides a list of six "prime examples of toxic chemicals that can contribute to learning, behavioral, or intellectual impairment."24d Five of those -- PCBs, PBDEs, mercury, lead, and organophosphate pesticides -- will be dealt with in some detail later in this article. Recent studies have revealed that background exposures of the general population to lead, mercury, and PCBs 84a as well as PBDEs and dioxins84b all contribute to a wide variety of problems, including impairments in attention, memory, learning, social behavior, and IQ. See Section 3 later for much more about these developmental toxins.
Section 2.b.1: Concentration of toxins via lactation: An official publication of the American Academy of Pediatrics stated in 2012 that "the relatively high concentration of fat in human milk means that fat-soluble substances will, in effect, concentrate there." (Fat-soluble substances that were being discussed included PCBs, PBDEs and various pesticides.)24b According to experts on this topic (A.A. Jensen et al.), "Significantly more (10 to 20 times) of a mother's body burden of persistent organohalogens is transferred to the infant via the milk than by the transplacental route."17 (Organohalogens include dioxins, PCBs, PBDEs, and organochlorine pesticides, all of which are neurodevelopmental toxins that are normally present in human milk.)
For more about concentration of these toxins in human milk, see Section 2.b.4 to follow soon.
Accumulation of toxins in future mothers over time: Two other experts on developmental toxins reported in 2006 that “Persistent lipophilic substances... such as PCBs, accumulate in maternal adipose tissue and are passed on to the infant in breast milk, resulting in infant exposure that exceeds the mother’s own exposure by 100-fold on the basis of bodyweight.”139e The ATSDR cites findings compatible with the above but notes that concentrations in umbilical cord blood were found to be 2- to 4-fold lower than in maternal blood.139i The EPA, also, recognizes that toxins such as these are bio-accumulative18g,139m and points out that computer workers were found to have PBDE burdens that varied according to duration of that employment.139k
Figure 2.a on the right/above, from a 2012 study, provides illustration of the tendency of three developmental toxins to accumulate through the years in childbearing-aged women. (For more evidence on this topic, see Appendix O.) Parallel with that trend of accumulation of toxins in mothers with age, risk of ASD among children has been found in many studies to increase with maternal age -- 38% per 10 years in one especially large study.139n
Although the above does not constitute proof that these toxins are causing ASD, there is ample other evidence indicating that
(b) all three of those plus PBDEs and dioxins are typically ingested by breastfed infants in concentrations greatly exceeding established safe doses (see Section 3 about all of those.), and
(c) all of the above toxins have been found to have adverse effects specifically on males while having little or no adverse effects on females, which fits well with the high male-to-female ratios of those diagnosed with ASD and ADHD. (see Section 7.b)
(For information about absence of these toxins in infant formula, see Section 3.g.)
Section 2.b.2: Transfers to infants: Stages of the lactation process as related to accumulation and transfer of toxins
Part 1: Certain environmental toxins that are ingested by women before birth -- and accumulated (see Figure 2.a above right) -- are released during lactation . . .
Accumulations of toxins including PCBs, PBDEs, dioxins, mercury and others
are excreted in breast milk.*
Part 2 of lactation: The long-term build-ups that depart from adult-size bodies are then ingested by developing infants, ...
and the toxins accumulate in the infants... ->
during a time of high vulnerability of development. (see Section 2.a)
*Declines of toxins in women as the toxins are excreted during breastfeeding:
* PCBs in serum of lactating women were found in one study to decline 23% in 6 weeks,17b and several other studies have observed that the PCB body burden of a mother during breastfeeding decreases by 20% every 3 months.84c
* One major form of PBDE (BDE 209), which the EPA says has high developmental toxicity,93k and which is especially high in traffic-related pollution,93m was found in a study to decline an average of seven-fold in its concentration in human milk during breastfed infancies.56f
* Dioxins in human milk have been found to follow a downward slope during the first year after birth similar to the slope for DDEs in Figure 3 above left;75 but the specific type of dioxin that is predominant in human milk was found to decline by 50% during the first month of breastfeeding. (see 93d and Appendix F)
* In a Korean study of 157 lactating mothers, it was found that about 71% of 15-day-old infants were estimated to be at risk due to lead exposure through breastfeeding, but by 30 days the at-risk-due-to-lead percentage had declined to 56%,139q indicating a large decline in maternal lead during 15 days of breastfeeding.
The other side of the those declines of maternal levels of toxins: Several different developmental toxins are typically ingested by breastfed infants, each in possibly hazardous doses;* and those toxins accumulate, while the children are passing through their vulnerable periods of rapid development:
*(All of those toxins are typically received by breastfed infants in doses exceeding officially-established relatively safe doses -- see Section 3.g later for details and authoritative sources.)
-- PBDEs building up in breastfed children, and remaining concentrated for years: In a 2007 study of Spanish children, it was found that increases of body burdens of PBDEs since birth were over six times as high in 4-year-old children who had been breastfed as in 4-year-olds who had been formula fed; 54 also see Section 3.b.
-- Dioxins in breastfed infants: A German study found that, at 11 months of age, dioxin toxicity-equivalent concentrations in breastfed infants had become about 10 times higher than in formula-fed infants; 79 also see Section 3.c.
-- PCBs in breastfed infants: See the charts above right and Figure 4.a below, and also Section 3.a.
-- Mercury: In a study by a highly-published scientist (P. Grandjean) and his team, it was found that total mercury concentrations in infants that had been breastfed for one year were three times as high as the concentrations in infants that had not been breastfed;100 also see Section 3.d.
(For more on accumulation of developmental toxins in breastfed infants, see Appendix O.)
Note that there is apparently little or no dispute that toxins such as PCBs accumulate in women through the years except for declining with the number of children breastfed, as the toxins are transferred to infants during breastfeeding. Even a major article in the La Leche League website acknowledges that this has been authoritatively established.20b
Regarding the major increases of developmental toxins in infants during breastfeeding, compared with the declines of toxins among bottle-fed infants, as shown in Figure 4 above: This same pattern has been verified elsewhere.20a
(POPs are "persistent organic pollutants.")
The above is a summary look at the above-discussed pattern, illustrating the decline in maternal PCB levels as the toxins are transferred to the breastfed infants, until weaning. Due to the differences in body sizes, the percentage increases of PCB concentrations in the infants are much higher than the corresponding declines in the mothers.
The 2015 study from which the above charts were taken20c was by a European and Canadian team of 20 researchers, who analyzed seven European birth cohorts.
For a summary of how high the concentrations of the above-discussed toxins are in average human milk, see Section 3.g.
Section 2.b.3: Transfers of such toxins during lactation versus during gestation:
By contrast with the above evidence about declines of toxins in mothers during lactation, average concentrations of such toxins in mothers have been found not to decline during gestation, implying that relatively little from those toxins is entering the developing fetus:
-- levels of total mercury of over a hundred women were found in a study to be the same at gestational week 37 as at gestational week 12;85a
-- In a 2005 study, PCB levels in maternal blood were found to be the same at birth as during the second trimester of pregnancy, in three separate groups assessed.85h Another 2005 study determined, regarding maternal PCB levels during gestation, that appropriately-adjusted maternal PCB concentrations were similar throughout pregnancy.85k
-- The U.S. ATSDR, in support of its statement that "the amount of PCBs transferred to offspring is expected to be higher during lactation than during gestation," refers to a laboratory experiment as an example: "In female rats administered PCBs before gestation, an average of 0.003% of the administered dose was transferred to the fetus, whereas 5% was transferred to sucklings."23b That works out to lactational transfer of PCBs over 1600 times greater than gestational transfer, resulting from the same original prenatal exposure; this is a ratio that the ATSDR obviously considers to be relevant to how human lactational exposure compares with human gestational exposure to those toxins. Studies of humans have found lactational transfers of developmental toxins to be hundreds of times greater than gestational transfers -- see later, in Section 2.b.4.
Various studies have found that mothers of impaired children had received peak exposures to specific toxins during gestation, especially during the third trimester; and the assumption typically seems to have been that the harm to the developing child was done at that time, prenatally; the reality, on the other hand (in line with the example provided by the ATSDR, above), is probably that the prenatal exposure was merely when toxins were received by the mother and mainly stored in her body. The predominant exposure of the developing child (if breastfed) was actually to come only after birth, when the earlier-received-and-stored toxins were transferred to the infant by lactation. (see above and Section 2.b.1)
When observing the buildup of mercury and PCBs in breastfed infants (above charts), and when reading about the accumulations of the other toxins that also build up in infants during lactation (see Appendix O), consider the authoritatively recognized effects that these toxins might have on development; effects may include "irreversible neurological damage." Remember the TENDR statement above and see Section 3.a through 3.d for much more on the effects of these toxins. Also bear in mind that
-- the early postnatal period is recognized to be one of high vulnerability to developmental toxins (see Section 2.a above),
-- as mentioned, PCBs, PBDEs, dioxins and mercury have all been found to be contained in average human milk in doses greatly exceeding the relatively safe doses that have been established by U.S. government agencies; and infant formula has little or nothing detectable of these toxins. (See Section 3.g later for details and authoritative sources); and
A commission of the German Federal Environmental Office reported that the average daily PCB intake of an adult is 0.02 micrograms per kg of body weight, as compared with the intake of a breastfed infant (3 micrograms per kg of body weight), which is 150 times higher.18a The ATSDR refers to a study that found concentrations of PCBs in breast milk to be 13 times those in maternal whole blood; and maternal blood in turn was found to have PCB concentration two to four times higher than umbilical cord blood.(p. 392 of ref. 9b) One study, by a research team who are authors or coauthors of well over 1000 studies, estimated PBDE intake from food to be 0.9 ng/kg/day in adult females, compared with 307 ng/kg/day for nursing infants.18d Other studies have observed that nursing infants consume a daily TEQ (toxic equivalency, normally used in reference to dioxins) intake that is 50 times higher than that of adults.18b
(When reading about exposures to these toxins after birth, remember from Section 2.a the several authoritative sources that point to the early postnatal period as being one of high vulnerability to developmental toxins, such as these.)
As mentioned earlier, the U.S. ATSDR, when stating that transfers of developmental toxins are expected to be higher during breastfeeding than during gestation, illustrates that by describing a laboratory test in which rat sucklings received 1600 times as much PCB as was received via transfer to fetuses, from the same original pre-gestation dose to females.23b (For evidence indicating that animal tests accurately indicate effects in humans, see Section 3.a.3.) In human studies summarized in a publication of the National Academies Press, the comparisons that were most relevant to the above pattern yielded ratios of about 280 to 1 and 775 to 1;18 the specific comparisons in those cases were of PCB concentrations in breast milk in relation to PCB contents in umbilical cord serum and cord blood; so the actual total ingestions of PCBs by infants would obviously be determined by how long breastfeeding continued; the postnatal/prenatal ratio for total exposure could very possibly be over 1600 to 1 with extended breastfeeding.
To see evidence of likely effects of the lactational concentration of toxins that originate from background exposures, see Section 2.d a little farther down.
For information about accumulation of these and other toxins in maternal fat tissues over the years, which makes possible far larger transfers to infants than could take place otherwise, review Figure 2.a and accompanying text in Section 2.b.1, earlier.
In a 2008 American study by eleven scientists, studying a birth cohort of over 250,000, a typical fourth child’s risk of autism was found to be half as high as that of a firstborn, and the odds of being diagnosed with autism decreased from first to later children; other studies have provided general confirmation of the above relationship. 120,122 It would not be unreasonable to see a possible relationship between that progressive reduction in autism risk and the decline in accumulated stores of developmental toxins in women that occurs during breastfeeding (remember Figure 3 and accompanying text). Milk received by later-born infants has toxin levels that have been found in many studies to be far lower than milk received by first-borns.127, 127a, 127b, 128, 128a, 129, 129a 85c
Preview of Figure 6
In actuality, it indicates the doses of a few of the developmental toxins to which breastfed infants are exposed, according to birth order, which might well explain why autism declines with birth order.
Section 2.c: The vulnerability to developmental toxins of concern here is to pollutants that many or most infants are exposed to, at environmental background levels.
Quoting again from the National Scientific Council on the Developing Child: After pointing out that “developing brain architecture is disrupted by mercury” and other neurotoxins, the Council says that the dominant source of the mercury in the U.S. that underlies this developmental disruption is coal-fired power plants, emitting mercury to the air.16a (In addition to direct human exposure via inhalation, those emissions drift widely, are deposited into water bodies, and return to humans via consumption of fish.) Other known neurodevelopmental toxins in the general environment are much more concentrated in air and dust in urban and high-traffic areas,16c in indoor pollution,16d and in foods, especially foods that contain animal fats. Summarizing, toxins that are recognized to harm infant development are part of widespread background exposures of the general population. Much more on this topic of adverse effects of fairly typical background exposures to these toxins will come later in studies described in Sections 4.a.2.b, 4.a.2.c , 4.a.2.2, and 3, and in the Freire et al. study in Appendix A, and at www.traffic-pollution-autism.info.
Mild background exposures accumulating and becoming concentrated on their way to infants:
Although toxins from the general background have in many cases been linked to adverse developmental effects in children, background exposures have obviously also been unrelated to adverse effects on children in many cases. Of relevance to this, toxins from the background often become concentrated before developing children are exposed to them, which is likely to help explain why some children are affected by the background exposures and some are not; and whether that intermediate step of concentration takes place depends on what people voluntarily do or don't do.
Section 2.b.4 earlier contains information about concentration of environmental toxins that takes place via lactation, and more along these lines follows below.
Section 2.d: Some studies have shown that direct or prenatal exposures to environmental toxins may have little or no effect on children, whereas those same toxins can be very harmful as a result of their much greater exposures to the infants that take place indirectly but in concentrated form, via lactation.
A Taiwanese study (Lung et al.) reported about adverse effects of local incinerators on children. Among children in the general population living within the designated distance from the incinerators, only one area of significant effect was found; but among children who had been breastfed for at least six months, adverse effects associated with the local incinerators on measured outcomes were reported in all developmental areas that were investigated. Also, the effects (β values) were far greater, and the statistical significance figures were many times greater, with breastfed children compared with the data for the one adverse effect seen in children in the general population.119j (This study is described in more detail in Section 4.a.1)
(As preface to the following, bear in mind that testosterone is known to be important to male neurological development.16e) Also of considerable relevance is a study (Mocarelli et al.) of long-term outcomes of prenatal and early-postnatal exposures to dioxins resulting from an accident in Seveso, Italy; two decades after the prenatal exposures, when the sons’ sperm quality and hormone concentrations were examined in young adulthood, those young men who had been breastfed (and only those who had been breastfed) showed seriously adverse effects in all of the four different reproduction-related areas that were measured; by contrast, those who had been formula-fed showed no effects. Apparently the exposures that were indirect (but greatly concentrated -- see Section 2.b.4), as received via breastfeeding, made a dramatic difference in the effects of the environmental toxins on infants. (This study will be described in more detail in Section 3.c.) So the harm to neurological development that could have resulted from low testosterone when those men were infants would probably have affected only the breastfed sons.
To sum up what apparently happened in the above two studies, illustrating a pattern that is probably occurring widely:
With environmental exposures such as often occur in modern industrial societies, the direct exposures are typically not sufficiently potent by themselves to produce significant effects on developing children;
but, as a result of the subsequent, amplified, indirect exposures that take place by way of lactation (see Section 2.b.4 earlier), total combined exposures of developing infants to the original pollutants may become hazardous, exceeding the infant body's ability to tolerate the exposures without adverse effects.
What is happening might be referred to as a "double hit" of infant exposure to environmental toxins, to borrow from medical terminology such as "double hit lymphoma." In both cases, the effects of two hits are likely to be far greater than the effect of a single exposure or source of harm. The body has mechanisms for detoxification and elimination of toxins, but the capacity to thereby tolerate toxins can be exceeded.119p The role of second hits in reaching a level of harmful exposure, coming at a time of ongoing neurodevelopmental vulnerability (see Section 2.a), should be apparent. And the importance of the second hit should be especially obvious when considering that the lactational exposure is likely to be far greater than either the prenatal exposure or the exposure directly from the environment (Section 2.b.4).
Considering the above, it should only be expected that prenatal toxic exposures, which are widespread in modern environments, may often have harmful effects only when combined with the second exposures to those same toxins, received via the breastfeeding that usually follows birth. Additional evidence on this topic follows in the next subsection.
When authors of studies claim to have observed effects of "prenatal" exposures to toxins, the burden ought to be on them to explain why they feel that those effects would have taken place even without the subsequent additional doses that will usually follow after birth, via breastfeeding. There is, after all, excellent evidence indicating that such second exposures
a) do normally take place and are far greater than the original, prenatal exposures to which researchers typically point (see Sections 2.b.1 and 2.b.2), and
b) come at a time of continued vulnerability to developmental toxins (see Section 2.a)
In the cases of both of the studies described above, if the authors had not taken the unusual step of distinguishing effects on breastfed children from effects on non-breastfed children, and reporting about the far greater effects of the lactational exposures, both of those studies would probably have merely gone on record as finding effects of "prenatal" exposures to toxins. The crucial lesson would have been lost, that lesson being that the pathway for the exposures that are likely to turn out to be the harmful ones is a pathway that can easily and effectively be restricted.
Other studies, also, have provided evidence of greater effects of postnatal, lactational exposures as compared with prenatal exposures.
A 2004 prospective study (such studies allow accurate measures of exposures, compared with retrospective studies), Vreugdenhil et al.,16k investigated mental abilities of nine-year-old children in relation to prenatal exposures to toxins and also in relation to their histories of breastfeeding versus formula feeding. The formula-fed children performed better than their breastfed counterparts, and the children who were breastfed for longer periods performed worse than those breastfed for shorter periods, in scores on a test of executive function. Those relationships held consistently, when the subjects were grouped according to both higher prenatal exposures to toxins and lower prenatal exposures to toxins. For more about this study, see Section 4.a.2.b.
Another example is the Gascon et al. (2011) study described in Section 3.b. Gestational exposure to PBDEs was found to have either relatively minor or no significant adverse effect on 4-year-olds, but exposure to those same mothers' elevated PBDE levels via breastfeeding did have substantial effects, including an 80% increase in relative risk of attention-deficit problems and a 160% increased relative risk of poor social competence.57 (Bear in mind that deficient social competence is normal in people with autism.) As in the cases of other studies described above, no harm or less harm was traced to the primary exposures from the environment; the greatest (or possibly the only) harm to the developing child was linked to lactational exposure to those same toxins.
The above authors took an unusual step in reporting effects on breastfed children separately from effects on non-breastfed children, and this unusual step revealed that the actual source of the adverse effects was postnatal. What more typically occurs is that studies determine exposures from the environment before or at about time of birth, or even levels of toxins in breast milk, and the authors somehow consider that those represent only "prenatal" exposures to the developing child.
But often the distorted impression is much more the fault of reporting about the study by others, as was the case with the 2013 Roberts et al. study about effects of air pollution exposures; the authors were clearly investigating exposures that took place during the approximate time of birth, as determined by the mother's residential location at the time of birth, with no indication of the exposures' being especially prenatal;17f but reporting about the study was typically like that in Medical Daily, with its headline, "Prenatal Exposure To Air Pollution May Raise Child’s Autism Risk."17g There seems to be a general gravitation toward the notion that the prenatal period is the only significant period of vulnerability to environmental toxins, even when there is no justification for such bias.
Section 2.e: Many studies have found associations of postnatal exposures with autism prevalence:
Upcoming in Section 4 will be descriptions of many studies published since 2009 that have found direct associations between autism and a distinctly postnatal exposure (breastfeeding); among those studies are a U.S. study that found a 185% increase in autism among children who had been breastfed for at least six months, compared with controls, with a very high level of statistical significance; a Canadian study that investigated a population of over 125,000 and found 20%increased odds of autism among children who were being breastfed very shortly after birth versus children who were not, and another U.S. study that observed (based on examination of data from all 50 U.S. states and 51 U.S. counties) that "the longer the duration of exclusive breast-feeding, the greater the correlation with autism.” (Full text, including other studies with similar findings, and references to sources, at Section 4.a)
Section 3: Developmental toxins in human milk, beginning in the mid-20th century:
The American and Danish experts, P. Grandjean and A. A. Jensen, who are authors or coauthors of 481 and 127 scientific studies respectively, writing in the American Journal of Public Health and referring to certain environmental chemicals, pointed out in 2004 that "these substances have caused contamination of human milk only during the last half century, and long-term health impacts are now being discovered." The specific chemicals to which they were especially referring were PCBs, dioxins, brominated flame retardants (PBDEs), and many pesticides.22
A report by the Washington State Department of Ecology discusses “Persistent Bioaccumulative Toxins” (PBTs), which category includes almost all of the specific toxins being discussed in this article: PCBs, brominated flame retardants (including PBDEs), dioxins, methylmercury, and pesticides. 24 In their paragraph discussing high-exposure populations, they express concern about only three groups: people who consume fish from contaminated waters, those with exposures related to working in or residing near certain polluting industries, “and babies and young children who are breast-fed.” Continuing, “PBTs and metals of concern have been shown to accumulate in breast milk as a result of the mother’s exposures.” (p. 64) It is noteworthy that breastfed babies, solely because of their feeding type, are considered by the State of Washington Department of Ecology to be at the very highest level of exposure to these toxins; they share that position with only two relatively small, high-exposure groups, indicated above.
Note well that this discussion of breastfed babies as being at the very highest level of exposure is separate from the additional consideration of infants' high developmental sensitivity to toxins.
Later the report does bring up that second matter, pointing out that children are undergoing “rapid growth and development, and therefore are particularly vulnerable to exposures that disrupt the developmental process.” Special concern is expressed about “formation of vital connections between the cells” of the brain; interference with this process is said to present “a high risk of permanent neurobehavioral dysfunction.”24
Regarding the above-mentioned formation of vital connections between brain cells, and the risk of permanent neurological dysfunction if that process is interfered with, remember from Figure 2 and accompanying text the especially large amount of growth and formation of neural connections that normally take place during the first year after birth
So it should be noted that, according to the Washington Department of Ecology,
(a) there is a broad group (developing children) that is “particularly vulnerable” to disruptive toxins such as PCBs, PBDEs, dioxins, and certain metals;
(b) within that particularly vulnerable larger group, there is a smaller subgroup (the infant age group) that has greater sensitivity; and
(c) within that especially sensitive subgroup, a still smaller percentage is at the highest level of exposure: breastfed infants.
It has been authoritatively determined that typical human milk in the contemporary U.S. -- and very likely in most developed countries -- contains several developmental toxins, each in concentrations that greatly exceed established safe levels. See Section 3.g later for a summation of the exceedances, compared with the very different levels in infant formula, and see the upcoming text for details about some of the most important toxins in human milk.
PCBs are ranked by the U.S. ATSDR as being among the top 5 priority hazardous substances, along with arsenic, lead, mercury, and vinyl chloride.24e
In studies published as of 2010, PCBs had been found to be present in human milk in doses 63 to 270 times the minimal risk level established by the U.S. Agency for Toxic Substances and Disease Registry;.23 a 2013 publication of the UN Environmental Programme also pointed out that PCB exposures of breastfed infants in relation to established safe exposures were in the range just stated.23f Both of the above reports were published more than 30 years after most intentional uses of PCBs had been phased out; PCBs are classified as “persistent” for good reason, and they are also substantially present in current vehicle emissions and other current sources (see text below Figure 13 in Appendix E). From 1949 to the late 1970's, PCBs were widely used in building materials, in high concentrations, and are still being emitted in those buildings.23g A 2009 review of 45 studies found that some types of PCBs have decreased in the environment since 1979 but others have not decreased at all.23c See more on this topic in Appendix E.
By comparison with the major transfer of PCBs by breastfeeding: in a study by scientists with the U.S. National Institute of Environmental Health Services, who examined 104 samples of infant formula, no detectable PCBs were found in all but one sample.30
As indicated earlier in Figure 4 and in the study from which it came,27a with corroborating evidence in this chart from the Danish Health and Medicines Authority.39 and in many other studies,29 exposures of breastfed infants to PCBs have been found to normally be far higher after birth than before birth, especially with nursing for a few months or more.23b
These exposures would be likely to affect the highly-vulnerable neurological development that takes place after birth (Section 2.a).
As indication that these exposures are not merely passing occurrences of infancy, note that in a 2014 study, children at 45 months of age who had been breastfed for 6 to 12 months were found to still have over 9 times the PCB concentrations compared with non-breastfed children; there were even greater differences if the children had been breastfed for longer.30a This was fully compatible with other studies,30c including one cited in an ATSDR report.30b
For additional information about concentration of toxins in developing infants via breastfeeding, see Appendix G,
Section 3.a.3: PCB exposures and neurological development:
As published in the journal, Proceedings of the National Academy of Sciences, biological evidence of a link between toxins in breast milk and autism was provided in a 2007 study (Kenet et al.) carried out in the laboratory of and with guidance from M.M. Merzenich, a member of both the U.S. National Academy of Sciences and the Institute of Medicine.31 According to Dr. Merzenich, “PCB intoxication was achieved by feeding the rat mother just enough of the poison to match the levels of PCB recorded in nursing human mothers in high-PCB-exposure areas of the U.S.,” which he indicated to be "all over the country." 32 Relating the results, Dr. Merzenich stated, “normal, progressive development (of the infant rats’ brains) was almost completely blocked,” about half of the rats had a “dramatically altered organization of the representations of sound frequency,” and this was “of special interest because the same bizarre abnormalcies have been recorded in autistic individuals.”
According to a summary of the above study published by the National Scientific Council on the Developing Child (which works in close collaboration with the Center on the Developing Child at Harvard University),33 “the key mechanism that the brain uses for learning new skills in all animal species and humans” was found to be impaired.34
Dr. Merzenich also pointed out that, as a result of these high-background-level PCB exposures, "the brain remained into adulthood in a very primitive underdeveloped state." And he stated that, in addition to PCBs, we should also be concerned about PBDEs, which are “close cousins” of PCBs and “very structurally similar” to the specific type of PCBs used in the experiment;35 PBDEs, like PCBs, accumulate especially in fat and breast milk. He concluded, “If human fetal and infant effects parallel rat impacts, we would predict that there would be a correlation between the PCB/PBDE levels in human breast milk — and in infant blood — with the probability of autism onset.”
Animal tests accurately indicate or under-estimate effects of chemicals on developing humans:
There is good reason to believe that effects on human infants would parallel the effects observed in animals.36 According to a consensus statement signed by 57 scientists, researchers and health professionals, including many who are highly published experts, "The concordance between human and animal neurotoxicity assessment is remarkable as demonstrated for lead, mercury and PCBs."36a But there is also good basis to believe that animal studies under-estimate the effects on humans; according to a 2002 article (Stein et al.) by a research team that included three MD's, "animal studies commonly underestimate human vulnerability to neurotoxicants because of the obvious difficulty testing uniquely human cognitive, language, and behavioral functions within animal models. In the case of lead, mercury, and PCBs, animal studies underestimated the levels of exposure that cause effects in humans by 100- to 10,000-fold.... " Continuing, "Current testing protocols also underestimate toxic threats by exposing subjects to only one chemical at a time, although children are exposed to complex mixtures of chemicals throughout development. It is now well established that such multiple chemical exposures can be far more damaging, or cause damage at lower levels of exposure, than single exposures generally addressed in animal models." 84a
Considering the link suggested by Dr. Merzenich between autism and exposures to PCBs (typically received in high doses via breastfeeding), note that many studies have found breastfeeding duration to be associated with autism prevalence, including in dose-response relationships. Those will be described in Section 4.
In relation to the dramatically altered representations of sound that were found to result in PCB-exposed rats in the above study, it is relevant that
-- over 77% of children with autism were reported in a survey to have hypersensitivity to particular sounds, very often causing them to try to run away from the sound;32a
-- in a study published in 2016 (Bennetto et al.), it was found that children with ASD had greatly reduced perceptions of sound in frequencies that are important to understanding speech, and the severity of autism symptoms was in proportion to the degree of hearing impairment.37c
Review of Fig. 4
The reader should consider how auditory impairment in those with autism would be related to the impairment of communication that is a core characteristic of autism, including the way hearing affects learning to speak. Then note that it was lactational exposure to PCBs at levels as found in many human environments that caused distorted representations of sound in brains of one of our fellow mammals, in a controlled laboratory experiment as discussed by Dr. Merzenich. And remember from above that animal studies typically underestimate effects of toxins on humans.
There is additional evidence of a relationship between background PCB exposure and hearing deficits (and therefore very possibly with autism): a 2014 study found an inverse association between postnatal PCB concentrations and hearing ability measured at 45 months.37d Measures of prenatal exposure to PCBs were not associated with hearing loss.
Section 3.a.4: PCBs and learning ability, with especially great effects on males:
The reader should think about adverse effects on learning ability that could occur when representations of sound are distorted as a result of developmental PCB exposure, such as discussed above. Also see the Colciago et al. study about the strong negative effect on learning ability that has been found distinctly in male animals as an apparent result of PCB exposure (see below, with more at www.male-development.info).
Male-female PCB effects
Bear in mind that it is unknown why males vastly outnumber females among those diagnosed with ASD and ADHD. Consider whether the explanation for that could be related to
a) the far greater adverse neurological effects of PCBs on males than on females, as found under controlled laboratory conditions (see above), combined with
Also bear in mind that there is apparently only one pathway via which infants are exposed to PCBs in concentrations recognized to exceed safe levels: breastfeeding. This is not disputed by any of the many appropriate doctors and scientists questioned on this matter -- see Section 6. Note that activity on this pathway increased greatly during the years leading up to the major increases in diagnoses of ASD and ADHD (Section 5); this might not be pure coincidence.
The reader should also consider the significance of the above in relation to the 730,000 American youth, in the 15-to-17 year age group alone, who were classified as having learning disability in 2001, a 63% increase over that same category in 1987. (see Fig. 11) (The child population increased about 6% during those same years.) Then note (in Figure 7) that breastfeeding rates more than doubled between the birth years of the children born at the beginning of that increase in disability and the birth years of the children in whom that impairment was to become so much higher. And remember the effects of high-background-level PCB exposures on hearing and brain development as found in the Kenet/Merzenich laboratory study. (Section 3.a.2) and the effect of PCBs on male learning ability as found in the Colciago et al. study.37L
Section 3.a.5: Effects of postnatal versus prenatal exposures to PCBs: Among studies that have investigated effects of developmental exposure to PCBs, a thorough search by the authors of a review article (published 2009) found eight studies that fulfilled the authors’ requirements; six of those studies measured PCB levels in cord plasma and two measured PCBs in breast milk at two weeks after birth. None of the studies that took measurements in cord plasma (which were considered to reflect prenatal exposure) found significant adverse mental effects of PCB exposure in later testing; on the other hand, the studies that measured PCB levels in breast milk did find adverse mental effects that were either fully statistically significant (in two cases) or of borderline significance (one case).38 The authors pointed out that previous studies, also, had found (a) associations between child PCB concentrations and hearing impairment at 8-9 and at 12 years of age, and (b) no associations between prenatal PCB exposures and hearing loss.
The U.S. ATSDR, in its major document on PCBs, states that "breast milk exposures have been associated with neurodevelopmental deficits in newborn and young children."16p
Other studies, also, have found greater effects of postnatal exposures to PCBs, often far greater, compared with prenatal effects. See Figure 14.a and accompanying text for data from two of those studies. Descriptions and citations of some of the other studies with similar findings have been placed in Appendix B at www.pollution-effects.info/appendixBandC.htm in order to allow moving on now to the next topics.
Section 3.a.6: Other serious effects of PCBs, especially in the long term:
According to excellent authority, PCB exposure affects neuronal connectivity; 37w and, according to the director of the NIH's National Institute of Mental Health (Thomas Insel), "Previous studies have suggested that autism is a developmental disorder resulting from abnormal connections in the brain."13a For substantial evidence of active formation of connections taking place in the human brain during the very period when a typical infant is taking in large amounts of PCBs via lactation, see Figure 2, Figure 4, and "Early postnatal vulnerabilities..." )
According to WHO, "Experimental studies have shown PCBs to interact with the estrogen/androgen systems and, in doing so, to have long-lasting or irreversible effects in altering gender-specific behaviour." (p. 173 of reference 98a) At a time when birth rates in most developed countries are substantially below replacement levels, despite greatly increased spending on assisted reproductive technologies, and when viability of some nations' social security systems is therefore considered to be in question, widespread exposures of developing infants to substances that alter gender-specific behavior should be of concern.
Human exposures to PCBs have also been closely linked with certain cancers.98b And, according to a study by seven U.S. researchers, "several studies have ... demonstrated adverse reproductive, developmental, immunologic, and neurologic effects (citing 6 studies as evidence)...The influence of PCBs on cancer risk is well established in animal studies."98c It is surprising that scientists have sometimes measured effects of infant exposures to PCBs as indicated by certain cognitive tests a few years after birth and found no significant cognitive harm, and then the researchers have often concluded that the PCB exposures do not have harmful effects. Such a conclusion is grossly premature, considering the evidence of slowly-emerging types of damage that would usually become detectable only many years later, such as the above-mentioned reproductive and cancer-related effects and the harmful effects on the blood-brain barrier (see below); also bear in mind the many kinds of developmental and neurological disorders (autism, ADHD, schizophrenia, etc.) that would very often not be detected in tests of cognitive abilities at early ages.
Disruption of the blood brain barrier is a result of PCB exposure that may have the greatest long-term effect. In a 2009 study (Seelbach et al.98c), mice were administered PCBs in doses (150 μmol/kg) that result in a plasma PCB level that is comparable to the levels found in some human populations that have received brief exposures to PCBs.98d Shown in this chart is the following:
Disruption of the blood brain barrier by three types of PCBs administered in the above-mentioned study:
The right-hand three bars represent measures of permeability of the blood brain barriers of the PCB-dosed test animals, compared with permeability in the control animals that received only the vehicle (represented by bar on far left).
Considering all the harmful chemicals that have become prevalent in the environments of modern industrial societies in recent decades, and considering how many of those chemicals have been found in human blood streams,98e it should be apparent that optimum shielding of the brain (by means of a healthy blood-brain barrier) from those chemicals is of critical importance.
According to a 2009 review, "Accumulating experimental and clinical evidence indicates that BBB (blood-brain barrier) dysfunctions are associated with a number of serious CNS (central nervous system) diseases ... such as multiple sclerosis, stroke, brain tumors, epilepsy or Alzheimer's disease." 98g Autism, also, is a CNS disease. And there is good evidence indicating that the blood brain barrier is disrupted among those with autism -- see Appendix R.
Review of Fig. 4
a) autism and other neurological disorders are linked with disrupted blood-brain barriers (see above), and also that
b) PCBs have been found to disrupt the blood brain barrier (see above),
is there reason for serious concern about long-term effects of PCBs on developing brains?
The first few years of relatively protected early childhood would be before occupational, craft-related, and long-term household exposures to neurological toxins, and also before normal exposure to traffic pollution, which is a significant source of neurological toxins with well-substantiated links to neurological harm.98f Existence of a well-formed versus a compromised blood-brain-barrier could make an immense difference in outcomes in later years. This is one additional reason why failure to find evidence of harm from exposures to PCBs in early childhood is meaningless.
See Section 3.f for much more about long-term effects of PCBs and other toxins in human milk.
Of the two principal types of brominated flame retardants, PBDEs are more thoroughly studied, but HBCDs are also recognized to have substantial adverse effects. Typical levels of PBDEs in human milk have been found in at least three studies to exceed the EPA's relatively-safe reference dose (RfD), including in a study that found median levels of one type of PBDE to be over three times the RfD.37h In a study of breast milk exposures in Europe, the authors used a Hazard Index (HI) that they derived from reference doses determined by the European Food Safety Authority based on animal testing; according to this Hazard Index, using their "moderately-conservative" determination method, any value above 1 was considered to be unacceptable; the HI was determined to be 11 assuming mean PBDE levels measured in some European studies, and 5.4 assuming PBDE levels measured in another European study.37i (Bear in mind that PBDE levels in North America have repeatedly been found to be many times higher than in Europe.66b)
Given this data about exceedances of satety thresholds by average levels of PBDEs in breast milk, it is relevant to note that, in another U.S. study, PBDE levels in the top 5% of the population were found to be 10 to 100 times the median levels.60
A U.S. study published in 2017 states that "flame retardants (in preschool children) were found to be linked with both more negative and less positive social behavior," and "this represents substantial risk for difficulty with academics, social relationships, and mental health."66c Considering how central social behavior, as well as other aspects of mental health, are to ASD, it should be apparent how important exposures to flame retardants are to discussion of that disorder.
3.b.1: For information about HBCD flame retardants, to which the EPA has assigned its “High hazard” designation for developmental neurotoxicity, go to Appendix X.
3.b.2: More on adverse effects of PBDEs: According to the U.S. Agency for Toxic Substances and Disease Registry (ATSDR), “results from human studies are suggestive of an effect of PBDEs on neurodevelopment in children, including impaired cognitive development (comprehension, memory), impaired motor skills, increased impulsivity, and decreased attention.”42 Whereas the ATSDR indicates probability above, nothing but certainty is indicated in a statement by the EPA regarding PBDEs' “adverse neurobehavioral effects following exposure during the postnatal period." And the EPA clearly sees those effects as being serious; they refer to neurobehavioral effects of PBDE exposures as being a “critical endpoint of concern.”43 Given the ongoing, high infant exposures to PBDEs via breastfeeding (see Figure 5.1 and accompanying text, below), there is special significance in the EPA's statement about particularly great sensitivity to adverse effects of PBDEs during the postnatal period.
An international team of 15 researchers published a study in 2016 in which they assessed the consequences of exposures to many different types of endocrine-disrupting chemicals in the European Union.43a The authors chose to focus on "only those (15) endocrine disrupting chemicals with the highest probability of causation." The team determined that the toxicological evidence for "IQ Loss and Intellectual Disability" as outcomes of exposure to PBDEs was strong, with a 70-100% probability of causation. These were not determinations that these authors made lightly; only one other toxin (organophosphate pesticides) was considered to have a probability of causation of adverse effects as high, or strength of human evidence as high, as the authors attributed to PBDE toxicity.
When reading the above, bear in mind that those determinations were based on PBDE exposures in Europe, and that PBDE exposures in North America have been found to be 10 to 100 times as high as in Europe.66b
A study of effects of PBDE exposures in the U.S., by a team of nine authors and published in 2016, estimated that 43,000 cases of intellectual disability per year were due to PBDE exposures; those disabilities plus millions of IQ points lost due to PBDEs were estimated to cost the U.S. $266 billion per year.43f Substantial other evidence is compatible with the above.66a
Published criticism of the above 2016 studies of PBDE effects have been surprisingly minimal, considering how unusual it is for scientists (large teams of scientists, at that) to make such unqualified statements about serious causal effects of specific toxins to which the public is widely exposed. As of July, 2018, there have apparently been only two critiques published, one by a sole author and one by a pair of authors, the lead author of which was acknowledged to be a consultant for industry. (PBDEs, as used commercially for decades, were industrial products.)43g
A 2008 study found strong inverse correlations between concentrations of PBDEs and sperm concentration and testis size in contemporary Japanese males, with high levels of statistical significance. The authors noted that PBDE levels in Japan are comparable to those found in European countries.43e (Remember that levels in the U.S. are many times higher than in both of those regions.) See Section 3.a.6 earlier about concerns regarding the role of such toxins in causing unwanted decreases in birth rates, and bear in mind the important role of testosterone in male mental development, including during the crucial early postnatal period. (see Section 7)
For further evidence that widespread background exposures of infants to PBDEs are strong enough to have adverse effects on children (especially boys), see www.male-development.info, specifically Figure 5 and accompanying text.
According to early data, concentrations of PBDEs in infant formula were found to be at levels less than 3% as high as in average U.S. breast milk.55 But according to more recent data that applies to the specific type of PBDE (BDE 209) that has been by far the most widespread type in recent years, that chemical has been found to be one eight-hundredth to one nine-thousandth as high in infant formula as in human milk.56 Comparison with average adult exposure is also extreme: As mentioned, a U.S. study estimated PBDE intake from breast milk to be over 300 times higher than adult females' intakes of PBDEs.18d
3.b.3: Adverse effects of brominated flame retardants in general, especially related to ADHD, have been observed.56a See also two of the studies at "Effects of exposures..." below for other evidence about associations of these toxins with ADHD-related behavior.
PBDE levels were doubling in humans approximately every three to five years during the later decades leading up to the 2000's.114d According to a 2007 report to the Maine legislature by an official of the Maine CDC, ”Several studies reported that levels of PBDEs in human tissues are now higher than PCBs, including in the U.S.;” in two out of 52 individuals in one study, PBDE levels were found to be 100 times greater than PCBs.114d
To put in perspective the apparent fact that PBDE levels are now typically even higher than PCB levels, and sometimes far higher, remember the data presented earlier (in a 2010 report from the Oregon Department of Environmental Quality) stating that PCBs have been found to be present in human milk “in doses 63 to 270 times the minimal risk level established by the U.S. Agency for Toxic Substances and Disease Registry.”23 Also remember (from Section 3.a) that Dr. Merzenich considered PBDEs to be approximate equivalents of PCBs, regarding risk to development.
In the chart below, showing increases in PBDEs in Canadian breast milk over two decades, notice the 100-fold increase that took place in the median levels during that period.
Figure 5.2: PBDEs in Canadian breast milk 1982-2002
Also notice the wide variations within the population, with the highest level measured in 2001-2002 being over 40 times the median level, and over 4,000 times the median level of 20 years earlier.
An EPA web page on PBDEs,50 accessed in January of 2016, pointed out that “despite the United States having phased out the manufacture and import of (two types of PBDEs) in 2004, their component congeners are being detected in humans and the environment. Some reports indicate that levels are increasing.” The EPA mentions nothing about any reports of declining levels of PBDEs; but it does refer to the chemical’s persistence (including in relation to continued use of furnishings and electronics containing them), and PBDEs' presence in imported products. In addition, PBDEs are also components of combustion emissions, including from power plants and vehicles,44 especially diesel-powered vehicles,45 46 and those emissions are continuing.
For other studies providing additional evidence regarding the ongoing high human exposures to PBDEs, the extremes of exposures of breastfed children to PBDEs, and effects of PBDEs in reducing levels of the neurodevelopmentally-important testosterone, see Appendix C.
Levels of PBDEs in the U.S. have been found to be especially high largely due to a California law which, for decades before 2015, required use of flame retardants in cushioning, electronics, and some other products. But even the comparatively low exposures outside the U.S. appear to have been having adverse effects. Two Spanish studies and a Taiwanese study found as follows:
a) In the Gascon et al. (2011) study, it was found that exposure to mothers' elevated PBDE levels via breastfeeding (but not during gestation) had effects including an 80% increase in relative risk of attention-deficit problems and a 160% increased relative risk of poor social competence.57 (Note that “poor social competence” is a way of describing a basic characteristic of people with autism.)
b) Researchers found “an association between PBDE concentrations in colostrum (early breast milk) and impaired infant cognitive development." The authors also pointed out that “in the group of children breastfed for a longer period the association between BDE-209 exposure and neuro-development impairment was somewhat stronger…. Further, associations in the longer breastfeeding group may be underestimated…" (going on to explain the reason for that).58
c) According to a study by four Taiwanese scientists, “Neonatal (relating to the first month after birth) exposure to BDE-209 has been found to have developmental neurotoxicity, including hyperactivity; learning and memory defects; a reduction in habituation;….”59 It should be noted that neonatal exposures to this chemical would have been predominantly a result of transfers via breastfeeding; remember the 2007 study’s finding that increases of body burdens of PBDEs since birth were over six times as high in 4-year-old children who had been breastfed as in 4-year-olds who had been formula fed.54
None of the above three studies showed results broken down according to narrowly-segmented levels of exposure.58a Narrower breakdowns might well reveal far more serious effects at the highest exposure levels, since a U.S. study (Schecter et al.) found that PBDE levels in the top 5% of the population were 10 to 100 times the median levels.60
PBDE levels in human milk in the USA versus those in Europe
(link to source of this at reference 66d)
When observing the above Europe-versus-US differences in PBDE exposures, remember that two of the studies described a few paragraphs up found increased attention deficit and impaired infant cognitive development, as well as 160% increased risk of poor social competence, associated with increased child PBDE levels in studies in Europe.
More about effects of PBDEs (specifically related to reduction of male hormones and therefore to male neurological development) can be found at www.male-development.info.
3.b.6: Information about sources of HBCD flame retardant in human milk, including a 17% increase in milk HBCD per item of stereo or video electronics in the home, and increasing levels of HBCD as of the most recent data, can be found in Appendix X.
Dioxins are part of the same chemical family (organohalogens) of which PCBs, PBDEs and HBCDs are members. Remember the expert statement earlier indicating extremely high transfer of this class of chemicals to infants via breast milk.17 Even though levels of dioxins in breast milk in a number of countries have been declining since about the 1960’s, dioxin has still been found to be present in breast milk in concentrations scores to hundreds of times higher than the relatively safe reference dose for dioxins established by the EPA (RfD -- 0.7 TEQ/kg bw/day); the still-high levels of dioxins have been found in studies from many countries carried out during the 2000’s,73a and a 2016 study found that levels of dioxins and dioxin-like PCBs in human milk in China have been rising dramatically.73c Dioxin has also been found to be present in breast milk at over 100 times the concentration in infant formula.75
Shown here is a chart from a publication of the government of Australia, showing dioxin concentrations in fat tissues over a life course, assuming breastfeeding for 6 months.75a A similar chart based on U.S. data (but more detailed, showing various durations of nursing) can be seen in a 2002 study, the lead author of which was a scientist with the EPA; that can be seen in Figure 3 of the article accessible via the link at reference no. 75b. That study, agreeing with the substance of the chart on the left, is the only one on the subject, "dioxin exposures of infants," brought up by the search tool on the EPA's website when a search was done using that phrase, in August of 2018.75c
The authors of the Australian study stressed the conclusion that, for most of a breastfed person's life, the lipid concentrations of dioxin are not higher than they would be without breastfeeding. When considering the significance of that, one should review Figure 2 and accompanying text, indicating the special neurodevelopmental vulnerability of the first year of life. Also, a major toxicology textbook published in 2011, with 21 contributing authors, states as follows regarding dioxins (as well as PCBs): “These studies have indicated that … the most susceptible period of exposure is during development and nursing.” Also, “early developmental exposures to these chemicals are particularly devastating.” (p. 551, bottom)131
Most of the rest of our discussion of dioxins is similar to the above discussions of PCBs and PBDEs, to which dioxins are chemically related: Many studies have found correlations of dioxin levels, especially in breastfed and male children, with traits of autism, ADHD, and learning disabilities. (see Appendix L.) The effects associated with the developmental exposures to dioxins were especially found in the long term; and the elevated levels of dioxins among breastfed offspring were still very prominent even into young adulthood. Since most of the rest of our discussion of dioxins is lengthy and similar to what has been said about their chemical relatives, PCBs and PBDEs, it has been placed in Appendix L.
Only one of the studies about effects of dioxins will be discussed here, because it has especially wide significance. First, bear in mind (from earlier) that testosterone is known to be important to male neurological development. The study of interest was of results of accidental exposures to dioxins in Seveso, Italy. A 2011 study (by Mocarelli and 12 others)76 of the aftermath of that accident measured characteristics of sons of mothers who had been exposed to increased levels of dioxins before their sons’ births, resulting in what the authors called “modest elevations” of the mothers’ dioxin levels. When the sons’ sperm quality and hormone concentrations were examined at ages 18 to 26, those who had been breastfed (and only those who had been breastfed) showed seriously adverse effects in all of the four different reproduction-related areas that were measured; by contrast, those who had been formula-fed showed no effects.
The findings of this study strongly suggest that, with environmental exposures such as often occur in modern industrial societies, if infants are subjected only to the original exposures, the effects may be insignificant; but when infants are in addition subjected to concentrated secondary exposures via breast milk, the long-term consequences can be very harmful. The Lung et al. study (late in Section 4.a.1) also illustrates this pattern.
To aid in understanding why prenatal exposures should have had effects on breastfed but not on formula-fed children, bear in mind that
(a) the “placental barrier” has that name for a reason (even though it is less effective in relation to some toxins than to others); protection provided by the placenta means that prenatal exposures are likely to be minor; and
(b) dioxins (like many other toxins) accumulate in the body, are stored in maternal fat, and are later mobilized and excreted in greatly concentrated form in breast milk. (see Section 2.b.1)
Remember from Section 2.b.3 the example that the U.S. ATSDR used to illustrate that, of exposures to similar toxins received prenatally, very little is received by the fetus compared with what is received postnatally via lactation: following an exposure to PCBs (chemical relatives of dioxins) that were received by a rat mother before pregnancy, 1600 times as much was received by the sucklings after birth compared with their prenatal exposures.23b Human studies, also, have found lactational transfers of developmental toxins to be hundreds of times greater than gestational transfers -- see here.
Above: Estimated dioxin levels in young children's blood according to feeding history.
For considerable authoritative evidence about dioxins as contributing to impaired neurological development, see Appendix L.
Section 3.d: Mercury is often already high in infants at birth, then a mother’s long-term accumulations are rapidly transferred in breast milk; “clearly documented toxic effects on the immature brain” occur during the postnatal period.
Mercury is the fourth developmental toxin that is normally present in human milk at levels greatly exceeding established safe doses. Breast milk typically has mercury concentrations that are about eight times the WHO guideline for drinking-water quality81 and four times the concentration that is allowed in U.S. bottled water,82 and often higher. And mercury in infant formula in North America has been found to be less than 1% as high as in human milk; 83 in three different European countries, mercury was found to average 3%, 6%, and 10% as high in formula as the 8 parts per billion average for breast milk.83d, 83b, 83c
According to both a national survey87 and a 2011 study reporting research by health departments of three U.S. states, about 8% of U.S. infants at birth already have mercury levels exceeding the EPA’s relatively-safe Reference Dose;86 and that is before the substantial transfers that will typically then take place via breastfeeding. An exposure assessment in New Jersey indicated that more than 20% of women of reproductive age exceed the limit for safe exposure.84a (But that percentage is above the U.S. national average, suspected to be so because of greater seafood consumption in New Jersey.)
Ethylmercury, as has often been used in vaccines, has been widely vindicated as being a cause of autism; but ethylmercury is only one of a number of species of mercury, and several other forms are authoritatively recognized to be neurological toxins. Methylmercury is listed by the International Program of Chemical Safety as one of the six most dangerous chemicals in the world's environment.83a Data from a major U.S. government survey indicates that methylmercury comprises about 82% to 94% of the mercury in women whose mercury levels are in the top quarter.92 Methylmercury is efficiently absorbed by milk-fed infants,92a and it accumulates in the brain.83a According to the EPA, “There is general agreement that the nervous system continues development in post-natal life and that methylmercury can adversely affect the developmental processes.” 90
A 2014 review article stated, "Mercury poisoning and autism show very similar symptoms, both mental disturbances and physical disorders," citing two studies in support of that, and also providing many examples in support of its statement about the "many parallels between the effects of Hg (mercury) intoxication on the brain and the brain pathology found in individuals with ASD."
A research team reviewed 91 studies that examined the potential relationship between mercury and ASD from 1999 to February, 2016. The authors stated that, "of these studies, the vast majority (74%) suggest that mercury is a risk factor for ASD, revealing both direct and indirect effects."90c
As was the case with PCBs (Section 3.a.4), observed effects of mercury make an especially good fit with ASD, ADHD and the reduced school performance of boys in recent decades (Section 7), because of its apparently much greater adverse effects on males than on females, which is a common characteristic of all three of the above impairments. See below.
Preview of Figure 8
Apparent effects of mercury on school performance of boys vs. girls in a general population
In addition to childhood exposures that may lead to postnatal effects along the above lines, transfers of mercury to infants also take place prenatally, but those are minor compared with the postnatal transfers via lactation. According to what may be the only study that compared mothers’ total mercury levels in early versus late gestation, average levels of total mercury of over a hundred women were the same at gestational week 37 as at gestational week 12.85a In a Brazilian study of 100 mothers and newborns, mercury levels in newborns were found to be less than one third as high as in the mothers.85e On the other hand, mercury excretion via breast milk has been found in many studies to rapidly draw down the mother's accumulated burden.20 Remember from earlier the 1998 German study that found that concentrations of mercury in breast milk of 85 lactating women at just two months after birth had declined by an average of over 70% from their levels at time of birth.19 See below a chart indicating mercury buildups in the children as part of the drawing down of mercury from the mothers.
Mercury (Hg) measured in mothers and their breastfed infants, and changes following lactational transfers, were shown in this chart from the study cited at reference 101.
Note that (a) the authors attributed about 40% of the increases in infant mercury levels to thimerosal, which was used in vaccines at the time of the study; and
(b) the changes in mercury levels during the first 180 days, as shown, were percentage changes compared with the levels at birth.
Bear in mind the evidence (presented earlier in this section and in Appendix A) linking mercury with neurodevelopmental harm, especially resulting from exposures occurring during early infancy; and remember that mercury in infant formula has been found to normally be less than one-tenth as high as in human milk.
Above: mercury levels found in children with ASD, compared with levels in controls,
as found in an Egyptian study
In addition to the differences between ASD cases and controls illustrated above, the authors also noted that mercury levels were higher in the group of mentally-retarded patients. The authors also acknowledged that some other studies had arrived at contrasting findings. There are clearly several other neurodevelopmental toxins to which children are also exposed at levels greatly exceeding established safe levels (see elsewhere in Section 3, especially Section 3.g), and there would normally be variations as to which toxins would be at higher or lower levels -- and more or less harmful -- in different study areas.
Most of the rest of our discussion of mercury and its effects on development is largely "more of the same" of the sorts of things presented earlier about PCBs, PBDEs, and dioxins: Highly-authoritative sources telling about the known toxicity of mercury to neurological development, findings of effects of exposures during the postnatal period while no significant effects of prenatal exposures were found, studies providing evidence of mercury's adverse cognitive and behavioral effects linked with exposures at general background levels (especially effects related to ASD), about very predominant exposure to mercury via breastfeeding, etc. Rather than presenting all of that here, we have placed it in Appendix A, where the reader is invited to go for the continuation on this subject.
Section 3.e: Pesticides and lead:
The four toxins discussed above are the four that are contained in breast milk in concentrations known to far exceed established relatively safe levels. But there are two additional important toxins contained in breast milk about which there is also substantial evidence indicating harmful developmental effects, even though there may be no established thresholds that permit comparisons with safe levels in these cases.
For information about pesticides, their known harmful neurodevelopmental effects, their substantial presence in human milk, and their absence in infant formula in the U.S., see Section 4.a.3
Lead: The following is a brief summation of information that is explained much more fully, with citations of authoritative sources, at www.breastfeeding-and-lead.info.
According to the EPA, harmful effects of lead "may occur at blood lead levels so low as to be essentially without a threshold;" the CDC and the American Academy of Pediatrics concur with this position. So, essentially, zero is the only known safe level of lead. And significant lead is transferred via breastfeeding: a 1994 study of over 500 women at ages 65 to 87 found that having breastfed was still associated with an average 11% reduction in women's lifetime accumulations of lead.19a So, of the total stores of lead that a fully-grown body had accumulated over many decades of exposure to environments, loss of an average of 11% of that was associated with transfers to a woman's small, developing infants within a matter of months.
A U.S. Food and Drug Administration study for the period 2006-2011 tested 34 samples of milk-based infant formula and found no detectable lead in any of the samples.19b
Associations have been found in many studies between heavy metal (lead and mercury) exposure, at levels often occurring in developed countries, and autism and ADHD. According to a document of the World Health Organization, "Early postnatal exposure (to lead) appears to be more effective than exposure prenatally" in reducing IQ. For much more complete information on this topic, including references to authoritative sources, see www.breastfeeding-and-lead.info.
Section 3.f: Long-term effects of these toxins becoming apparent in later childhood and adulthood, while often not being noticed earlier:
Despite the major, widespread presence of recognized developmental toxins in human milk (as presented earlier in this section, and summarized in upcoming Section 3.g), doctors' associations and the U.S. CDC unreservedly promote breastfeeding except in rare cases. That may be related to their consideration only of effects in early childhood, while ignoring long-term effects. When the topic of developmental toxins comes up in a web page on breastfeeding of the U.S. CDC, the document states,"despite the presence of chemical toxins (in human milk).... effects on the nursing infant have been seen only where the mother herself was clinically ill from a toxic exposure."37a One might wonder why the CDC focuses on "effects on the nursing infant," and says nothing about delayed, long-term effects that are recognized to take place.
There is, in fact, a very large amount of evidence indicating harmful effects of infant exposures to toxins that become apparent later in childhood or life. There have been findings of delayed and long-term effects of exposures to specific toxins that are high in human milk; and these long-term effects often follow exposures that produced no detectable effects during infancy. Authoritative general statements on this topic will be presented later, but first a few specific examples will be mentioned, regarding apparent long-term (often delayed) effects of toxins that are high in human milk:
In a Michigan cohort (Jacobson et al. 1990) that was studied for effects of PCBs, developmental exposure was not related to IQ at 4 years of age but was at age 11; similar results were obtained in another study in Oswego, New York. (Stewart et al. 2003b)38a Remember other evidence in Section 3.a.1 indicating that adverse effects of developmental exposures to PCBs were very significant at age 9 in one case and increased with age in another study. In Section 3.c, note the Lee et al. study's finding of negative cognitive effects of background exposures to dioxins in children at ages 12 to 15. In Figure 9, note that lower mental capacities in children at age 15 correlated well with higher average time-of-birth DDT levels according to geographic location. The Adams study (Appendix A) and the Davidson study (Figure 8 and accompanying text) both found apparent adverse effects of background mercury exposures in child populations that were in middle and late childhood.
A 2006 article by the neurology experts, Grandjean and Landrigan, states that many industrial chemicals (which include PCBs, dioxins, mercury and flame retardants) "can cause developmental neurotoxicity at low doses... (resulting in) permanent changes in brain function that might become detectable only when the animal reaches maturity." 38t
One effect of both PCB and lead exposures that could have especially great (but unknowable) impact in the long term: Both have been found to disrupt the blood-brain barrier (see Figure 5.0 and accompanying text). This would cause increased vulnerability of the brain to environmental toxins indefinitely into the future, with such increased vulnerability having resulted from exposures to toxins during infancy.
PCBs and dioxins are also classified as carcinogens,38q and cancer is well known to typically become observable mainly after long latencies following exposures to carcinogens. Dioxins have been found to have other effects that would ordinarily be observable only in the long run, including heart disease, skeletal malformations, and reproductive abnormalities.38r
Reproduction-related effects of toxins on infants would be especially likely to have long-term effects that would not be observable in the near-term. In a 2001 experiment with rats (Kaya et al.38u), developmental exposure to PCBs was found to result in reduction of testosterone, but with noteworthy differences according to time period. In the lowest-exposure group (which was the group with greatest similarity to typical human exposures), reduction in testosterone stemming from effects of exposure to PCBs during infancy rose from 12% in infancy to 34% in adulthood. Normal early development of the glands that produce testosterone in adulthood was probably disrupted by the exposure to PCBs during infancy, but in a way that was unlikely to have been observed during infancy. (Remember that according to an authoritative consensus statement, "The concordance between human and animal neurotoxicity assessment is remarkable as demonstrated for lead, mercury and PCBs."38v
See Section 13.b about major increases in mortality that have taken place among those born after 1970. Mortality is normally a long-term outcome in contemporary developed countries, and it is of interest that it has increased among those born during the period when breastfeeding was increasing. (See Figure 7)
There are many statements by authoritative sources such as the EPA, WHO, a commission of the National Academy of Sciences, a neurology expert, as well as additional studies, telling about “long-latency delayed neurotoxicity” and much more. Rather than continuing at length on this topic here, that other evidence has been placed in Appendix I.
Section 3.g: Each of the toxins dealt with above has been found to be present in typical human milk in concentrations exceeding the established safe level for that individual toxin; and their combined effects are likely to be more than merely additive. Infant formula has little or none of these toxins.
-- PCBs, in human milk in concentrations about 20 times the maximum allowed by law in U.S. public water supplies;37g and present in human milk in doses 63 to 270 times the minimal risk level established by the U.S. Agency for Toxic Substances and Disease Registry;23 a 2013 publication of the UN Environment Programme includes a statement that is compatible with the figures just above, and also provides additional support for the statement about dioxin levels just below.23f
-- dioxins, in human milk in concentrations exceeding the EPA’s RfD (reference dose, or estimated reasonably safe dose) by scores to hundreds of times;37j
-- Mercury concentrations in human milk typically eight times the WHO guideline for drinking-water quality,37k
-- PBDEs: Several studies have found typical breast milk levels of PBDEs to exceed the EPA's RfD or equivalent European standards, in some cases many times over, as summarized in the first paragraph of Section 3.b; and
-- lead: See Section 3.e above.
Pesticides: There do not appear to be applicable established safe levels for pesticides, but there can be little doubt that they are at far higher levels in human milk than in infant formula; see Section 8.
Not surprisingly, a report of the UN Development Program refers to "the overwhelming evidence of breast milk contamination above recommended levels by harmful chemical substances."40a
Absence of these toxins in infant formula:
According to a major document of the American Academy of Pediatrics published in 2012, referring specifically to PCBs, PBDEs, and major types of pesticides, "Infant formula is free of these residues...." (then going on to explain how that result comes about).24b Specifics regarding low or undetectable levels of these and other toxins in infant formula can be found in the relevant subsections of Section 3, above. There is also other authoritative evidence indicating that infant formula in the U.S. is essentially free of pesticides.139d, 139g
Effects of combined exposures to these toxins likely to be far greater than the sum of the individual hazards:
According to the former chairman of the Canadian Institute of Child Health, the combined (synergistic) effects of toxins such as those mentioned above are likely to be "more than multiplicative." He provided an example of the effect of PCBs on hepatic porphyrins in the presence of dioxin; a dose of PCB that alone will cause a 1.5 times increase in porphyrins will, in the presence of dioxin at a dose that alone produces no measurable effect, result in a 650-fold porphyrin increase.37e See also the statement from the Stein et al. study (in Section 3.a.3) about greatly increased effects of mixtures of toxins. Then note
a) how far both PCBs and dioxins exceed the EPA's RfDs for those toxins considered individually (see early in this section), and
b) seven studies have identified atypical urinary porphyrin profiles in children with autism, and increasing levels of porphyrin in children have been found to correlate with increasingly severe ASD, in five studies. 37e1
Also see below how high both dioxins and PCBs have been found to be in human milk worldwide, compared with established safe levels.
As indicated above, average levels of dioxins and PCBs in human milk in most countries are already far above established safe levels when the toxins are considered individually.
Effects resulting from synergy of combined exposures and from above-average exposures would be in addition to the effects of these already-high average levels of the individual toxins.
See text just above this chart for an example of the "more than multiplicative" health effect of dioxin combined with PCBs.
When inquiries have been sent to appropriate scientists and to the American academies of pediatrics and of family physicians regarding the above, asking if they are aware of any pathway other than breastfeeding by which infants are widely exposed to developmental toxins in doses exceeding established safe levels, none of the several respondents has suggested even one other pathway, for even one toxin. And nobody has expressed any doubt about the validity of the above evidence about toxins at high levels in human milk. (See Section 6).
Section 3.h: Despite some reductions in pollution, emissions of developmental toxins that enter human milk continue at hazardous levels, from sources to which humans are closely exposed:
Production and/or use of some of the pollutants of concern (PCBs, lead, and some pesticides and PBDEs) have been greatly reduced in developed countries in recent decades, but this has occurred much less in most less-developed countries. However, continued presence of these and other pollutants continues to be at harmful levels even in developed countries, because of (a) their persistence in the environment, (b) in the cases of PCBs and flame-retardants (including PBDEs), their extensive past incorporation into products and buildings that continue in use (see Sections 3.a and 3.b), and (c) their ongoing unintentional release, especially as products of widespread, everyday combustion, including in vehicular emissions, to which humans are especially closely exposed.37e2 Dioxins and mercury are also products of fossil fuel combustion such as that involved in residential and commercial heating, again sources to which humans are closely exposed. (see Appendix P) Leaded gasoline and paint are still widely used in less-developed countries; in developed countries, leaded fuel is still used in piston-engine aircraft, to emissions of which many people are exposed; and lead is also present in most lipstick, which is likely to be ingested by most women.37e3
Also counteracting the effects of declines in some of the above toxins, there have been great increases in the predominant pathway for all of these toxins to developing infants. (see Section 5 below)
Section 3.i: The toxins discussed here constitute the great majority of the toxins that have, in recent research, been identified as likely causes of autism and learning disabilities.
Going through this article, the reader may get the impression that an almost endless number of toxins have been identified as likely neurodevelopmental toxins. But that is not true; the number about which there is greatest reason for concern has been authoritatively narrowed down to ten, and that group included all of the ones discussed here. Landrigan and others, in a 2016 editorial, listed the environmental compounds considered to be the most likely causes of autism and learning disabilities, and that list includes all of the toxins that are discussed in this article: 37e4 Lead, mercury, PCBs, endocrine disruptors (which include dioxins), and PBDEs have all been discussed earlier in this section. Pesticides will be discussed in Section 4.a.3, and automotive exhaust will be discussed in Section 4.a.4.
Notice, in the descriptions of all of these toxins, that each of them (in line with the pattern described in Section 2.b) accumulates in and becomes concentrated in adipose tissue and therefore in human milk.
Section 4: Studies finding positive correlations between breastfeeding and autism prevalence or other mental impairment:
Section 4.a.1: Studies finding autism to be higher in association with greater duration of breastfeeding:
A 2011 study that investigated data from all 50 U.S. states and 51 U.S. counties found that "exclusive breast-feeding shows a direct epidemiological relationship to autism," and also, "the longer the duration of exclusive breast-feeding, the greater the correlation with autism." 116
Note that, according to the EPA, “Epidemiologic studies of exposed human populations provide the most convincing evidence of human health effects.”145a Also, a dose-response relationship between an exposure and a health outcome is considered to be especially significant evidence to support a finding of cause and effect. One example of a dose-response relationship, as found in an epidemiological study by a well-published scientist (R.J. Shamberger), was quoted in the previous paragraph. This finding was even more significant in that it was based on investigation of a very large, diversely-populated geographic area (all 50 U.S. states), and it also applied in relation to numerous smaller-scale units (51 counties).
Additional support for a causal connection between breastfeeding and autism was provided by three additional, relatively recent studies, with a dose-response relationship being apparent in the overall view of those three studies, with increasing odds of autism in parallel with greater durations of breastfeeding.
-- In a 2011 Canadian study of a population of over 125,000 (Dodds et al.), using discharge from the hospital as the dividing line for breastfeeding exposure, there was a 20% higher autism rate among the breastfed children than among non-breastfed children.117
-- In a 2009 U.K. study, the duration of breastfeeding that was assessed was four weeks versus less than four weeks. 65% of the autism cases had received exclusive breastfeeding for at least four weeks; that should be compared with only about 28% of the general U.K. infant population receiving that much breastfeeding. That meant a 130% higher odds of having had that much breastfeeding, among those with autism. 118
-- In a 2010 American study in Kentucky by an MD/PhD team, the duration of breastfeeding used for comparison was six months. 37% of autism cases had received that much breastfeeding, compared with 13% of the controls, indicating an approximately 185% (37%/13%) greater likelihood that the autism cases would have had more breastfeeding. The p-value was .003, implying three chances out of a thousand that the finding was a result of chance occurrence. 119
In the Dodds study, significantly increased odds of autism were found in relation to breastfeeding as determined by three different ways of analyzing the data, and the odds of autism in relation to breastfeeding increased with each improvement in the analysis.117
47% greater risk of ASD among breastfed children, in another U.S. study:
Other relevant risk information was found in a 2015 study that drew its data from the Autism Genetic Resource Exchange (AGRE), which is funded by Autism Speaks and the National Institute of Mental Health. The author reported, based on data from 902 ASD cases and 296 controls, that "maternal self-report of lactation practice (lactation without bottle use) was associated with increased risk of ASD diagnosis.... OR (Odds Ratio) = 1.47, 95% [1.05 - 2.08]."16r That is, there was a statistically-significant 47% increase in risk of autism associated with breastfeeding. Both "any lactation duration" and "dedicated lactation practice" were associated significantly (p < 0.05) with ASD risk. (p. 230)
Increased risk of autism was associated with greater duration of breastfeeding, and even greater risk of autism was associated with exclusive compared with non-exclusive breastfeeding:
In a 2015 U.S. study (Husk et al.120a), the authors found an increased-odds figure of 1.03 for current autism diagnosis for each additional month of "any" breastfeeding. For exclusive breastfeeding, the adjusted odds ratio was greater than the ratio for any breastfeeding (implying a dose-response relationship), at 1.04 for each additional month of exclusive breastfeeding.
Broadening the range of the data that they analyzed to include children ever diagnosed with ASD, the authors arrived at an adjusted odds ratio of 1.3 (30% increased odds of autism) associated with six months of exclusive breastfeeding versus no breastfeeding.120b
First note that "developmental delay" is a broader category of which ASD can be a sub-category.16s2 A small study in the U.S. Great Lakes area (Lynch et al., 201225g) was the source of the data that follows. A close look at Table 1 of the study indicates that
-- Of the 33 children who were breastfed,* 9 had mental delay, and
-- 100% of the children with mental delay had been breastfed.**
The mothers were participants in the New York State Angler Cohort Study, a high-fish-eating group, so their body burdens of PCBs and mercury would have been expected to be above average; see Section 3.a and 3.d about neurodevelopmental toxicities of PCBs and mercury and their absence in infant formula.
Statistical significance: In the above two studies, as is very common especially in small studies, "statistical significance" is a topic that comes up with regard to more than one chance out of 20 that the results could have resulted from chance. Common practice is to not pay serious attention to findings that don't meet this (arbitrary) standard; but there are good reasons, with highly-authoritative support, for respecting much of the evidence that doesn't meet that standard, especially when it is compatible with evidence from other studies. See Appendix T for much more on this topic.
160% increase in risk of a basic trait of ASD was associated with toxins ingested via breastfeeding but not with prenatal exposure to the same toxins:
A 2011 study (Gascon et al., 2011) found that gestational exposure to maternal PBDEs as determined in a study had no significant adverse effect, but exposure to those same mothers' PBDE levels via breastfeeding did have substantial effects, including an 80% increase in relative risk of attention-deficit problems and a 160% increased relative risk of poor social competence.16m (Note that poor social competence is a basic trait of autism, and attention deficits are a frequent characteristic of those with ASD.)
Effects of PCBs, PBDEs and DDT on children's cognitive functions that are related to ASD symptoms:
See below the autism-related effects that PCBs, PBDEs and DDT have been found to have on children.
Then see below the main pathway by which developing children are exposed to these toxins.
Below: Fig. 5.1, repeated
Human milk as the predominant pathway for PCBs, PBDEs and pesticides to developing children: There is clearly a reluctance to discuss the central role of breastfeeding in this regard, but it is a matter that should be addressed: See Section 3.a.2 about PCBs in that regard, Section 3.b.3 about PBDEs, and Section 4.a.3 about pesticides, as well as Section 6 about breastfeeding's apparent distinctiveness as a pathway for toxins to developing infants in concentrations exceeding established safe levels.
For more studies concerning effects of PBDEs related to ASD, see Section 3.b.5.
For more studies concerning ASD-related effects of PCBs, see Section 3.a.3
For more information about learning-disability effects of PCBs, see Section 3.a.4.
Emissions from municipal incinerators were found to be associated with autism-related outcomes far more in children who had been breastfed for six months than in the general population:
As introduced earlier, authors of a 2013 study in Taiwan (Lung et al.119g) worked with data from over 21,000 children, 953 of whom lived within three kilometers of a municipal incinerator. The study reported about effects of local incinerators on autism-related developmental outcomes in children, as indicated by parental responses to questions. (The authors cited other studies as having found parental observations to be valid.) Among children in the general population within the designated distance from the incinerators, only one area of significant effect was found; but among children who had been breastfed for at least six months, adverse effects associated with the local incinerator on measured outcomes were reported in all four developmental areas that were asked about. Also, the magnitudes of the effects (β values) were found to be far greater and the statistical significance far higher in breastfed children in all four cases, compared with the data for the one effect observed in children in the general population.119j
When discussing the relationships between incinerator emissions, breastfeeding, and adverse effects on neurological development, the authors commented that their findings suggested that "toxins can be transmitted via breastfeeding by mothers who are exposed." They referred specifically to dioxins, PCBs and mercury as having been found to be significantly present in incinerator emissions, and cited two studies as having reported about dioxins in milk of women living near incinerators.
More will be presented later about this study and about another study that carries a similar, important lesson: Direct exposures that frequently occur in modern environments may have little or no effect on children, whereas indirect (but concentrated) exposures to those same toxins, received via breastfeeding, may have substantial effects. More on this will come in Section 4.a.2.c.
Section 4.a.2: Other studies, including very large ones, linking breastfeeding with risk of autism and cognitive deficits:
Section 4.a.2.a: The CHARGE study in California, the especially large SEED study (carried out at six different sites across the U.S.117c), and another important U.S. study have found that children later diagnosed with ASD or ADHD had been breastfed for average durations of 6.7, 7.3 and 6.5 months.
Median durations of breastfeeding in the general U.S. population in the relevant years were found in all of the four large surveys to be about 3 months.
The CHildhood Autism Risks from Genetics and the Environment (CHARGE) study in California -- "the first comprehensive study of environmental causes and risk factors for autism and developmental delay"117c3 -- is one major source of this data. The average length of breastfeeding for children with autism/ASD was found to have been 6.7 months.
Very similar data is available from the Study to Explore Early Development (SEED), which the CDC refers to as "the largest study in the United States to help identify factors that may put children at risk for autism spectrum disorder (ASD) and other developmental disabilities."117d Mean duration of breastfeeding of children in this study who were later diagnosed with ASD was found to be 7.3 months.117g
A U.S. study of associations between breastfeeding history and later diagnoses of ADHD, published in 2016 (Stadler et al.), found that children diagnosed with ADHD had been breastfed for a median of about 6.5 months.117h3 The authors described their study as the first to investigate a well-characterized cohort with carefully validated ADHD diagnoses, which they considered to set the quality of their study above that of all prior studies.
Substantial online search for other studies that gathered data about breastfeeding duration for children later diagnosed with ASD did not bring up any other studies of any substantial size.
Looking at the above two ASD-related studies, their findings of 6.7 and 7.3 months of breastfeeding for children later diagnosed with ASD indicate fairly close agreement. The SEED study sample is especially broadly representative, because of its size and its wide geographic distribution. Also, considering the average age of children in the SEED study (4.7), that study's sample was a respectable representation of the population that would eventually be diagnosed with ASD.117z, 117c2
(For discussion of what data on this matter might look like from countries other than the U.S., see Appendix V.)
The above breastfeeding durations (among children later diagnosed with ASD and ADHD) should be compared with the much shorter average breastfeeding durations (2.3 to 3.2 months) in the U.S. general population at the time of those studies' data-gathering:
Since breastfeeding rates have varied significantly over the years, making valid comparisons with data from the three studies above requires data about U.S. breastfeeding durations from the same general period as the birth years that would have applied to the above studies: approximately 2003-2006.117h5 Such information is available from four large surveys, results of which will follow. (Data from those studies' control groups is of questionable validity, as will be explained below.)
As can be seen in the bottom row, last two columns, of the chart below provided by the NIH, a little less than half of U.S. mothers breastfed for three months or longer during those approximate years; put another way, half or slightly more than half of all mothers breastfed for less than 3 months.
Above table condensed from Table 10 (p.70) in NIH's Health, United States, 2013, at www.ncbi.nlm.nih.gov/books/NBK209224/pdf/Bookshelf_NBK209224.pdf
A repetition of that same ongoing national survey (for babies born 2008-2010) found that 3 months or less was the breastfeeding duration for 50% of U.S. mothers.117t Also, a chart from a study in the journal of the American Academy of Pediatrics shows a mid-point of breastfeeding duration in 2002 at about 3.2 months.117j.
The data in the study just mentioned came from the National Immunization Survey (which asks questions about breastfeeding as well as about vaccination rates), with a sample of about 20,000 and with participation rates of 65% to 76%.117q The data for the surveys discussed just before that one came from the National Survey of Family Growth, which also interviews about 20,000 people nationwide in each survey, with a response rate of about 73%.117u
The Pregnancy Risk Assessment and Monitoring System (PRAMS) study provided data from 11 states and New York City for the period of interest, with response rates of 70% or higher. This study reported that 50.2% of mothers in 2004-2006 breastfed for 10 weeks or more,117v which is to say for 2.3 months or more. 1000-3400 women are sampled per year in each state,117x providing a total sample of over 15,000.
Note that the data mentioned above came from four large U.S. surveys (counting the 2002-2004 and 2005-2007 surveys in Figure 5.a.2 as two), which were apparently the only large surveys that gathered this information; each had a good participation rate and was carried out during or close to the relevant comparison period (2003-2006). So this is an unusually strong group of sources, coming to a general agreement that an average of approximately 3 months of breastfeeding would have been normal in the overall U.S. during those years; a shorter average duration could have been reasonable to expect in a survey of just certain parts of the country (the PRAMS study). Remember from earlier that the ASD- or ADHD-diagnosed or developmentally-delayed children were determined in substantial studies to have been breastfed for over twice as long as the general U.S. population durations indicated just above.
Control groups in the above studies: The CHARGE, SEED, and Stadler studies, which were the sources of the above 6.7-, 7.3- and 6.5-month breastfeeding duration figures for impaired children, also included data for control groups that were intended to provide standards with which the data for the impaired children could be compared; but it is very questionable whether those control groups fulfilled the purpose of enabling valid comparisons, at least in the matter of breastfeeding duration. In the SEED study as reported by Soke et al. in 2017, the CHARGE study as reported by Lyall et al. (2015), and the Stadler study, first note that the control groups as reported by those authors contained 1223, 391, and 183 participants,117g in contrast with the total of over 75,000 in the combined four surveys discussed above (see text below Figure 5.a.5); those much larger data samples were in general agreement on a much shorter median breastfeeding duration for the general population, as compared with the durations reported for the control groups in the SEED and CHARGE studies. Sample size of a study or survey is a recognized major determinant of the validity (or lack of validity) of findings; it is normal procedure to weight the various findings of different studies according to the individual studies' sample sizes, in trying to determine the central tendency of a combined total. A weighted average of the U.S. population's breastfeeding duration based on all of the above surveys and studies would still be about 3 months.
There are various other factors that help explain why the breastfeeding durations for the above-discussed control groups would differ from those found in four large surveys of the U.S. population. Among them is the time-consuming nature of participation in the SEED, CHARGE and Stadler studies; this would greatly affect the representativeness of that small percentage of parents who are willing to go through all of the requirements of being in control groups in such studies, as compared with those who answer just a few questions in a high-participation survey. Parents of impaired children, on the other hand, would have sufficient personal interest to motivate them to agree to participate in the study, therefore providing high participation rates that permit valid data to be gathered concerning average breastfeeding histories of impaired children. This and related matters, including the combined use of medians and means in this discussion, are dealt with in more detail, with citations of sources, in Appendix S.
Section 4.a.2.b: In two separate studies, dose-response relationships were found between breastfeeding and a brain function that has been found to be strongly related to ASD.
According to a Brazilian research team, "Neuropsychological studies have indicated the existence of a strong relationship between alterations in the executive functions (EF) and ASD," with six studies cited in support of that statement.120c A widely-published U.K. expert on autism concurs with this, adding that the prefrontal cortex, which mediates executive function, has "extended postnatal development" and should therefore be "particularly sensitive to exogenous influences" (such as toxins) after birth.16s1
A 2004 study (Vreugdenhil et al.,16k discussed earlier in Section 2.d) tested children in the Netherlands for executive function in relation to their histories of different types of infant feeding. It was a prospective study, a type that allows unusually accurate measures of exposures. The authors investigated mental abilities of 9-year-old children in relation to their histories of breastfeeding versus formula feeding. Quoting from the study, "Children that were breast-fed for a long period scored significantly lower on the TOL (a test of executive function) than their FF (formula-fed) counterparts." Also, "the TOL scores were negatively associated with the duration of breast-feeding."16k (Remember that dose-response patterns, such as the one just described, are considered to be good evidence of causality.)
Review of Fig. 4
These authors discussed their findings in relation to postnatal vulnerability (to PCBs and dioxins) of the autism-related prefrontal cortex. They also pointed out in relation to their findings about child mental abilities that "recent animal behavioral studies do give evidence of negative effects of lactational exposure to low levels of PCBs," citing three studies with monkeys in support of that statement.16k (Bear in mind that monkeys are on the primate family tree, along with homo sapiens.) And remember how high lactational exposures to PCBs are in breastfed infants, increasing with duration of breastfeeding, in contrast with the absence of such exposures in formula-fed infants. (see Figure 4)
A U.S. study published in 2016 (Belfort et al.) arrived at very similar findings about dose-response effects of breastfeeding on executive function. Duration of breastfeeding was assessed during infancy (again helping insure accuracy compared with most other studies); and the children were later tested at ages 7 through 10 years, making possible findings regarding long-term effects. The authors observed that "for each additional month of any breastfeeding," the child's scores on two separate tests of executive function declined, with "both point estimates suggesting more problems with longer breastfeeding...."16l In line with the normal very uneven male-female ratios among children diagnosed with ASD and ADHD, the study's results "indicated worse performance on parent-reported behavioral regulation and metacognition in relation to longer breastfeeding duration in boys but not girls." Remember that dose-response relationships are considered to be especially strong evidence of causality; the presence of such a relationship, combined with a finding of adverse effects on boys but not girls, is of special interest in a discussion of origins of ASD and ADHD.
Section 4.a.2.c: Children who are exposed to toxins in the environment are likely to receive little or no effect from direct exposures but serious effects from indirect exposures; the latter can result from those same toxins' being accumulated and concentrated in mothers' bodies before being excreted to infants in breast milk.
As introduction to this topic, note that a 2002 study group measured concentrations of dioxins in tissues of 27 infants that had died unexpectedly; it was found that the closer the infant had been to first in birth order, the higher the dioxin concentrations in the deceased infants’ tissues, “thus showing” (as stated by the study’s authors) “that the mothers can decontaminate themselves by breast feeding.”130 This, especially when combined with the considerable evidence in Section 2.b (including about transfers of most of a grown person's long-term accumulations to a baby in a matter of months), illustrates that lactational transfers can result in substantial increases in infants' total exposures to neurodevelopmental toxins. And those increases come at a time of high vulnerability of the developing brain (Section 2.a).
The Lung et al. study of effects of incinerator emissions in Taiwan was discussed in Section 4.a.1. As related there, very minimal effects of the incinerator pollution were found in children in the general population; but among children who had been breastfed for at least six months, adverse effects associated with the local incinerator were reported in all four developmental areas that were asked about. Also, the magnitudes of the effects and the statistical significances were far higher in breastfed children in all four cases.119j All four of the areas in which breastfed children were seen to have worse development were areas that are traits of autism, as follows:
-- deficits in social development and communication are core characteristics of ASD,
-- language impairment is a very common trait of those with ASD,119h and
-- clumsiness is a common trait of those with ASD;119h consider this in relation to the study's findings of associations of breastfeeding with adverse effects on both gross motor development and fine motor development, as found at various times of measurement in this study.
To sum up what apparently happened here, illustrating a pattern that is likely to be occurring very widely:
Common exposures to toxins directly from the environment are likely to be relatively mild, producing only minor or negligible effects on developing children;
but, as a result of accumulation and concentration that takes place in maternal adipose tissue (see Section 2.b), total eventual exposures of developing infants to those toxins via lactation may reach more hazardous levels, very possibly causing adverse effects.
This same pattern was illustrated in another study that followed up on accidental exposures to dioxins in Seveso, Italy. (As preface to what follows, be aware that male steroidal hormones, especially testosterone, are known to be important to development of the brain -- see Section B.1 of www.male-development.info) A 2011 study (by Mocarelli and 12 others)119k of the aftermath of that accident measured characteristics of sons of mothers who had been exposed to increased levels of dioxins before their sons’ births, resulting in what the authors called “modest elevations” of the mothers’ dioxin levels. When the sons’ sperm quality and hormone concentrations were examined at ages 18 to 26, those who had been breastfed (and only those who had been breastfed) showed seriously adverse effects in all of the four different reproduction-related areas that were measured; by contrast, those who had been formula-fed showed no effects of the environmental exposures.
To summarize: Both of the above studies strongly suggest that environmental exposures, which can and do occur in modern industrial societies, can either have or not have harmful effects on developing children depending on the feeding type received during infancy. As indicated by very substantial evidence, there is a concentrating effect that takes place, as part of the lactational process, of transferring a mother's long-term accumulations of lipophilic (fat-soluble) substances to infants. (see Section 2.b). Depending on their feeding type, infants can receive either
(a) just the original, direct exposure from the environment, which is likely not to be harmful by itself, or
(b) that original exposure plus a far greater (very possibly harmful) additional exposure via lactational transfers from their mothers.
As indicated in both of the above studies, the additional exposures via lactation can have substantial long-term consequences, even when the original exposures by themselves have little or no observable consequences. That is as should be expected, considering the example that the ATSDR provides to illustrate the pattern of far greater exposures received by infants via lactation than are received prenatally, following prenatal exposures of the mother: following an exposure to PCBs received by a rat mother before gestation, 1600 times as much was received by the sucklings after birth compared with their prenatal exposures.23b Human studies have found lactational transfers of developmental toxins to be hundreds of times greater than gestational transfers -- see here.
To review some important points:
Emissions of pollutants in the environments of developed countries became increasingly widespread in the latter half of the 20th century (see Section 1.b). In the U.S. in the late 1960's and especially in the 1970's, breastfeeding began to increase greatly (see Figure 7). In line with the concentrating effect of lactation (see Section 2.b), the increase in breastfeeding led to elevated exposures of infants to those pollutants; these secondary exposures were much greater than the direct, primary exposures. The additional exposures may well have made the difference between major adverse effects and only minor or negligible adverse effects of the environmental pollution that had recently been increasing. (See the Lung and Mocarelli studies discussed earlier in this subsection, in which only breastfed children were found to have received harmful effects of environmental exposures.) It might not be entirely coincidental that many non-communicable disorders in children began to increase greatly in the 1970's, when breastfeeding rates were rapidly increasing.
Section 4.a.2.d: PCBs from breast milk have been found to greatly reduce activity levels of children. Harmless? Not at all:
Activity is necessary for normal development of the brain, according to highest authority.
According to the authors of the study that was the source of this chart (Jacobson et al. 16t), "reduced activity was associated with both 4-year PCB body burden and its principal determinant, exposure to PCB-contaminated breast milk." (There is ample evidence indicating that PCB contamination is a normal characteristic of contemporary human milk; see Section 3.a.2)
The most-exposed children were found in this study to be ”about 2-3 times more likely to be rated 'usually quiet and inactive,' and less than half as likely to be 'in action during much of the period of observation.' " The children who were observed were mostly from Great Lakes fish-eating populations, with slightly-above-average PCB exposures.16t1 Again, these were fairly typical environmental exposures, not poisonings.
At least four laboratory experiments have also found reduced -- often greatly reduced -- activity to be an effect of PCB exposure.16t2 One of the experiments was by a team of researchers who said that the concentrations of the toxins found in their test animals' brains were "about the same order of magnitude as observed in infants less than 1 year old."
There is substantial evidence indicating that inactivity leads to reduction in cognitive development: A 2011 meta-analysis of 59 studies found "a significant and positive effect of physical activity on children's achievement and cognitive outcomes, with aerobic exercise having the greatest effect."16z10 At least part of the reason for this is that increased blood flow optimizes transfer to the brain of nutrients needed to fuel neurological growth and function. Additional evidence supporting this generalization, including from a 2003 meta-analysis of 44 studies and from a committee of the U.S. Institute of Medicine, is so substantial that it would take up too much space here, so it has been placed in Appendix U.
Seeing that activity is important to neurological development, it bears repeating that child activity has been found to be reduced by exposure to PCBs, at environmentally-relevant levels of exposure (Figure 5.d just above, and accompanying text). And such exposure in turn is a normal result of breastfeeding. (see Section 3.a.2).
Section 4.a.2.e: The profile of autism matches the observed effects of inactivity: major deficits in perceptual abilities -- but not necessarily in intelligence.
In data from a 2003 meta-analysis of 44 studies (Sibley et al.),16z1 it is worth noting that perceptual skills were most affected by inactivity.16z2 On the other hand, the meta-analysis found that effects of inactivity on math tests, verbal tests and memory were at levels that could be described as small or negligible (p. 250 of 16z1).
It should be noted how well the basic nature of autism fits with this pattern of effects of inactivity, as indicated just above: In the cases of both autism and effects of inactivity, there are major deficits in perceptual skills but often little or no deficits in intelligence-related skills. Essentially all of the basic traits of ASD could be related to problems of perception (specifics to follow), but about half of all people with ASD are of normal or above-average intelligence.16z3
Problems of perception that are important traits of ASD:
-- Current estimates indicate that more than 80% of children with ASD exhibit co-occurring sensory processing problems;16z12
-- hyper- or hypo-reactivity to sensory input is a diagnostic criterion for ASD;16z12
-- Impairment in communication, a core trait of autism, could clearly stem from problems in quality of auditory perceptions. (Note that understanding meaning from sounds is different from detecting beeps in a hearing test.)
-- Impairment in social interaction is a core trait of autism that would be related to visual-perceptual difficulties, especially difficulties in perception of facial expressions and recognition of faces, problems that have been extensively documented in people with ASD.16z11
-- Self-injurious behaviors are common among children with ASD.16z13. Some children with ASD apparently do not feel (perceive) pain when they are being physically harmed, judging by the fact that they continue to self-injure. This could reasonably be considered to be a failure of perception.
Summarizing some main points from above:
a) inactivity is closely linked with PCB exposures (see Figure 5.d and accompanying text),
b) deficits in perceptual skills, which are inherent in autism (see above), appear to result from inactivity during development,16z2 and
c) PCB exposures in infants, which apparently cause inactivity and lack of development of perceptual abilities, mainly result from breastfeeding; (see Section 3.a about PCB exposures in relation to breastfeeding)
(a) Fairly common pesticide exposures in children have been closely linked with autism risk in many studies;
(b) breast milk is the predominant pathway for pesticides to most infants, at a stage of especially high developmental vulnerability.
The above two statements, both of which are supported by substantial, authoritative evidence (see below), should be considered at the same time. Taken together, those two statements (backed by the evidence below) say a great deal about how risk of autism could be greatly, easily reduced.
There have apparently been no studies that have investigated associations directly between autism and infants' ingestions of pesticides via breast milk; but logical combinations of studies enable us to see a real link between that exposure and that endpoint.
Substantial evidence linking common pesticide exposures with origins of autism:
a) When summarizing data from 37 unique studies, the authors of a 2014 review article (Rossignol et al.) found that 34 studies (92%) reported an association between estimated exposures to environmental toxicants and ASD. Most of the reviewed studies were said to have had good study designs, and “the toxicants that appeared to have the strongest association with ASD were pesticides and air pollutants.”139 (emphasis added)
b) Another 2014 review article generalized concerning effects of pesticides that "most give rise to neurotoxicity."137d Based on seven epidemiological studies determined to be of high quality, elevated risk of autism associated with pesticide exposure was found "with large enough impact and statistical precision to rule out sampling error."
c) Since there have been so many different studies of human populations that have found associations between pesticide exposures and autism prevalence, that strongly implies that the autism-linked pesticide exposures are relatively common, rather than isolated poisonings; bear in mind that pesticide levels in the breast milk of urban mothers have been found to be similar to levels in milk from mothers in agricultural areas.141
d) In a 2007 study, autism prevalence was found to be six times as high as normal near California agricultural fields where organochlorine pesticides were applied; autism prevalence varied in correlation with distance from the fields and with poundage of pesticides applied.136
f) A 2012 study found that the time of peak correlation of pesticide exposure with autism incidence (in the study described just above) was in the year after birth;137 (remember from Section 2.a that the year after birth is a time of very high vulnerability of the developing brain to toxins); correlation with pesticide exposure during the year after birth strongly suggests exposure via the pathway of breastfeeding, especially in communities with high breastfeeding rates such as in the above study; much more on this will follow.
Evidence indicating that infant exposures to pesticides are predominantly via breastfeeding, including in an agricultural community:
The evidence on this topic is extensive. A 2011 study examined many different “exposure prediction factors” that could contribute to children’s levels of metabolites of organophosphorous pesticides in an agricultural community.139a The authors found that, at six months of age, current breastfeeding was a stronger predictor of exposures to this major group of pesticide types than eleven other factors considered. The only factors that were stronger predictors (child care less than 60 meters from an agricultural field and home use of pesticides during the previous six months) applied to only small minorities of the children studied (6% and 2%). Therefore, for the vast majority of young infants, even those living not far from agricultural fields, current breastfeeding was found to be the strongest predictor of their levels of exposure to a major group of pesticides, among the many possibilities that were considered.
For a much more complete presentation of the evidence indicating that breastfeeding is the predominant pathway for pesticides to infants, see Section B.3 in www.pesticides-and-breastfeeding.info. Included there is
-- the expert statement, “specific pesticides ... are passed on to the infant via breast milk, resulting in infant exposure that exceeds the mother’s own exposure by 100-fold...."139e
To summarize briefly:
(a) Pesticides in infants, originating from common background exposures, are important risk factors for autism, by substantial evidence;
(b) lactational exposures to pesticides come during infants' vulnerable developmental periods;
(c) breastfeeding is apparently the predominant pathway for transfers of pesticides to most infants.
Combining (a) through (c) above leads to a summary statement to the effect that breastfeeding leads to increased autism risk, based on substantial evidence. Connecting a few dots here provides a useful overview on this topic and is compatible with other evidence in this section that points in the same direction.
For information about effects of pesticides in increasing the risk of ADHD, see Section B.2 of www.pesticides-and-breastfeeding.info.
Section 4.a.4: Traffic pollution during the first year after birth, spreading toxins that will subsequently be excreted in breast milk, is closely associated with autism:
A 2013 study (Volk et al.) in California found traffic pollution during the year after birth to be strongly associated with later diagnoses of autism; by contrast, associations of prenatal traffic pollution with autism have been found to be relatively weak. Vehicular pollution contains PCBs, PBDEs and mercury, all of which accumulate in women's bodies and are excreted in concentrated doses in human milk. Another 2013 study found similar results regarding postnatal traffic pollution in relation to hyperactivity. For considerable detail on this topic, see www.traffic-pollution-autism.info.
Section 4.b: Autism risk and toxins in breast milk both decline greatly as a child's birth order goes from firstborn to later-born. Those two declines’ taking place in parallel might not be a coincidence.
In a 2008 American study by eleven scientists, studying a birth cohort of over 250,000, a typical fourth child’s risk of autism was found to be half as high as that of a firstborn, and the odds of being diagnosed with autism decreased from first to later children.120 A California study121 and a major study in Australia122 provided general confirmation of the above relationship.
Probably related to the above: The average duration of breastfeeding is greater for earlier-born children than for later-born, as found in various studies in different countries,123, 125, 126 and milk received by later-born infants has toxin levels that have been found in many studies to be far lower than milk received by first-borns,127, 127a, 127b, 128, 128a, 129, 129a 85c,128b apparently as a result of excretion of long-term accumulations to earlier-born infants during previous nursing.124
At a quick first glance, the chart below could be a graphical representation of the decline that has been found in autism rates according to birth order.120
Neurological toxins typically ingested in infancy decline with birth order, as indicated above and in the text above this chart.
As mentioned briefly earlier, a 2002 study investigated concentrations of dioxins in 27 infants that had died unexpectedly; it was found that the closer the infant had been to first in birth order, the higher the dioxin concentrations in the deceased infants’ tissues, “thus showing” (as stated by the study’s authors) “that the mothers can decontaminate themselves by breast feeding.”130 (See in the chart above additional evidence of reduction of a mother's accumulations of toxins with each additional course of breastfeeding.) Considering the obvious benefits of the mother’s clearing out much of her lifetime accumulations of dioxins by transferring them to her infant, the reader should remember that, when referring to the “particularly devastating” effects of dioxins and PCBs, a major toxicology textbook also points out that “the most susceptible period of exposure is during development and nursing.”131 Clearly, feeding a grown person’s long-term accumulations of dioxins to a small, vulnerable, developing infant is to be recommended since, as stated by the authors of the study, various physicians’ associations recommend breastfeeding. No mention is made of the fact that, when repeatedly asked how they have determined that the toxins in human milk are not causing harm that outweighs the benefits of breastfeeding, those associations never respond.
Promoters of breastfeeding point to a very small set of studies to support their position: a study carried out in the Sultanate of Oman (Al-Farsi et al.), another in which the participants in a parent-created survey responded to Google ads (Schulz et al.), and a 1989 study in Japan that found early weaning among infants who would later be diagnosed with autism (Tanoue et al.). This is not exactly an impressive array of modern-day, professionally-conducted studies that could be generalized to conditions in most countries; but a closer look shows it to be even worse.
In the Sultanate (Oman) in which the Al-Farsi study was carried out, there were only 1.4 cases of diagnosed ASD per 10,000 children up to 14 years old; only 19% of the control mothers had finished high school, and every child studied was reported to have been breast-fed. So there would be little or no basis for seeing applicability of this study to modern societies, considering the nature of the population, the prevalence of autism in that study area, and the environment in which the study was carried out. The authors acknowledged the obvious, that "these findings should therefore be generalized only with respect to Omani children."131a.
It is very much of interest that, when authors mention studies that found breastfeeding to be associated with reduced autism, the two studies just described seem to be the two (out of very few) mentioned most: one that was said by its authors not to be generalizable outside the Sultanate in which it was conducted, and one that was "produced by concerned parents," recruiting participants by means of Google ads, and apparently paying no attention to selection of suitable controls. The citing of these studies as evidence for an inverse association between breastfeeding and autism is an indication of the extreme scarcity of evidence supporting that position.
Early weaning in relation to autism: Regarding the 1989 study in Japan that is sometimes cited on this topic (Tanoue et al.): this study is an excellent example of likely reverse causation; there is ample evidence indicating a strong likelihood that emerging autism would cause early weaning, rather than causation being in the opposite direction as some people surmise. First bear in mind that a breastfed infant's feeding has major impacts on sensitive parts of a mother's body.
Keep the above in mind when reading that
-- A 2013 study found that children with ASD were "five times more likely to have a feeding problem, including extreme tantrums during meals." 119c Consider how unpleasant it would be for a mother to continue breastfeeding an infant who has tantrums during feeding.
-- In a U.S. study, of 16 mothers who breastfed infants who were later diagnosed with ASD, "several" reported physical trauma as a result of their infants' overly-persistent sucking, which the study's authors related to the persistent, repetitive behavior that is a diagnostic of ASD.119d
-- Women often give "sore, cracked or bleeding nipples" and "breastfeeding too painful" as reasons for discontinuing breastfeeding; 119b including within one week after birth.119s,119r Such early weanings could well result from infants' overly-persistent sucking (very possibly related to emerging autism -- see previous item), by babies with emerging traits of autism.
-- In a 2017 study of breastfeeding histories of children diagnosed with autism, a major pattern was found that "breastfeeding was interrupted early, generally because the infant refused to breastfeed. 'Right from the start, he rejected the breast...'" 119x This implies that early weaning occurred because of early appearance of neurological problems in the infant, rather than neurological problems' occurring because of early weaning.
-- Two studies have been cited as having reported that "children with development delays and ASD had more difficulty with the muscle and breathing coordination required for nutritive breastfeeding."119n Medical personnel and/or a caring mother would very likely observe the difficulties and do what is needed to nourish the baby.
To sum up the above: Negative traits of autism are likely to be emerging soon enough after birth to cause early weaning.
So, aside from the fact that this (1989) Tanoue study is almost ancient history, especially in relation to a disorder the diagnostics of which have greatly evolved in recent decades, the authors' finding of early weaning among children later diagnosed with ASD is subject to very conflicting interpretations.
Another study (Burd et al., 1988, also showing ASD associated with early weaning) is sometimes provided as evidence to indicate benefits of breastfeeding in relation to autism; a proper response to that, including (among others) the point that the comparison group consisted of other impaired children, is somewhat lengthy, so our response to that study can be found at the end of www.pollutionaction.org/comments.htm.
In a recent study that may provide the best evidence of beneficial effects of breastfeeding on autism risk, the likelihood of a favorable outcome was found to decrease with an increase in breastfeeding intensity.
In a study in Spain published in early 2017 (Boucher et al.),17d based on data including what is shown in the chart just below, the authors stated, "longer duration of breastfeeding was independently associated with a small improvement in global cognitive function and with slightly fewer autistic traits." The authors acknowledged that the effect that they found of breastfeeding duration on autistic traits was "small and minimally clinically relevant."17e
The meanings of the numbers shown in the above chart are difficult for an average reader to interpret, but the "P" values, marked with asterisks, are easier to understand and are noteworthy. Two asterisks (**) in this chart indicate a high level of statistical significance; * indicates a lower level of statistical significance, and no asterisk indicates lack of statistical significance. Note in the above chart that the evidence for a beneficial (but "small") effect of breastfeeding with regard to autistic traits goes from highly significant with "any" breastfeeding, to substantially less significant with predominant breastfeeding, down to statistically non-significant with exclusive breastfeeding. So it seems that the greater the intensiveness of breastfeeding, the weaker the evidence for autism-related beneficial effects of breastfeeding was found to be.
It would seem that when mothers are being urged to endure cracked nipples, lost sleep and other difficulties and pain in order to exclusively breastfeed their babies, they should be informed of the following:
a) According to the authors of this study, their evidence indicates only that "slightly fewer autistic traits" (not reduced autism) are related to longer duration of breastfeeding; and
b) the evidence for breastfeeding's benefits in this area, limited as those benefits were, was found to become weaker (down to non-significant) as the intensity of breastfeeding was increased.
It should also be noted that this study was carried out in a country (Spain) where the developmentally-toxic pollution that typically gets into human milk is very possibly the lowest that is to be found in any major country in the world. (See Appendix P)
At least as important, the very few other studies that have found favorable effects of breastfeeding in relation to autism have all had serious weaknesses either in their quality or in their relevance to contemporary autism in developed countries (see early in Section 4.c about those studies.)
If anyone can provide any other studies that imply benefits of breastfeeding with regard to autism risk, please send them to firstname.lastname@example.org and they will be posted here, along with our response.
Although the focus of this article is on neurological disorders, the following should be mentioned regarding general developmental problems:
a) according to the U.S. Surgeon General, essentially all of the studies finding benefits of breastfeeding have been of a type (observational) that leading authorities on medical evidence consider to be of low quality (see Section 10); and
b) over 50 studies (in addition to those cited earlier in this section) have found adverse health effects of breastfeeding, often including dose-response effects. (see www.breastfeeding-studies.info)
c) each of five developmental toxins dealt with in Section 3 has been found to be present in typical human milk in concentrations greatly exceeding the established safe level for that individual toxin; and their combined effects are likely to be more than merely additive. Infant formula has little or none of these toxins. (See Section 3.g)
Section 5: There have been major increases in breastfeeding in parallel with major increases in child disorders; the increases were very rapid for a decade or so and slower later, in the cases of both breastfeeding and the child disorders.
Following a moderate increase beginning in 1965, and a decade of rapid increase beginning in the early 1970's, overall growth of breastfeeding in the U.S. has been very large, especially in extended and exclusive breastfeeding. Breastfeeding for at least six months increased ten-fold from 1971 to 2007.
Bear in mind that this greatly increased feeding type is apparently by far the predominant pathway of infant exposure to several developmental toxins, in high concentrations; many knowledgeable scientists and health-related organizations have not disputed this point upon being queried about it (see Section 6 below).
For some indication of the significance of the major increases in transfers via this pathway (the major exceedances of officially-established safe levels of each of the toxins typically present in breast milk) see Section 3.g.
As can be seen in the above chart, breastfeeding rates showed
a) rapid increases from 1972 to 1983, followed by
b) a slowdown, and then much slower growth.
Disabilities in American children followed that same general pattern, with
a) especially large increases in disabilities among children born in the mid-80’s compared with children born in the early 1970’s (see Figure 11 and accompanying text); and
b) increases in years following those rapid increases were much slower.113a A CDC statement of 2008 reported that the more recent rate of increase in enrollments in special education for specific learning disabilities was less rapid than it had been earlier, following the “marked” increases of the 1970’s and 1980’s.108a
For most childhood diseases, historical health data for earlier decades is not sufficiently complete to show whether there was a minor decline that corresponded to the mild dip in breastfeeding during the 1980's; but data for childhood cancer does appear to be sufficiently detailed in that regard; see Figure 12.3 and accompanying text.
Note in the above chart that the overall breastfeeding rate increases in relation to the 1970 rates were especially dramatic in the case of longer-term breastfeeding.
(a) the above-discussed transfers of toxins to infants take place at a time of the developing brain’s known vulnerability to such toxins (see Section 2);
(b) typical breast milk contains five different developmental toxins each in concentrations that greatly exceed established safe levels, plus numerous pesticides, while the principal alternative feeding contains little or none of those toxins (see Sections 3 and 8); and
(c) all five of those principal toxins have been found to either reduce scholastic achievement specifically of males or to reduce hormones that are important to neurological development and motivation specifically of males; males are the sex that is much more affected by ADHD, autism, and some other increasingly-diagnosed disorders. (see Section 7)
There is much more indicating that the parallel tracking of breastfeeding increases with increases of child disorders might not have been merely coincidental; see Section 9.
As can be seen below, breastfeeding also increased considerably in most European countries in the late 20th century, with little data about when the increases began.
In addition to major increases in what is almost certainly the predominant pathway of developmental toxins to infants (breastfeeding -- see above and Section 6 just below), concentrations of some major toxins were also rapidly increasing in the environment and in human milk during those years: PBDE levels were doubling in humans approximately every three to five years during most of the period discussed here,114d and HBCD levels have been rapidly increasing well into the 2000’s. (see Figures 5.1 and 5a and accompanying text)
For more on the topic of increases in both breastfeeding and child health problems in the latter decades of the 20th century, see Section 13.b.
Section 6: A possibly unique pathway of widespread infant exposure to developmental toxins in doses exceeding established safe levels:
As discussed in Section 3.g with considerable authoritative supporting evidence, breastfeeding is a pathway for ingestion of multiple developmental toxins by infants, with PCBs, PBDEs, dioxins and mercury at especially high levels. But it is more than just a pathway; it may be the only pathway by which infants are widely exposed to developmental toxins in doses exceeding established safe levels. The author of this article has written relevant letters of inquiry to
a) the American Academy of Pediatrics,
b) the American Academy of Family Physicians,
c) the entire science team at the major autism-advocacy organization, Autism Speaks.
Those letters asked whether the addressees or their organizations were aware of
a) any toxins that are believed to widely reach infants in doses well in excess of a recognized safe level (e.g., EPA’s RfD), aside from the several such toxins that are ingested by means of breast milk, or
b) any other pathway of infant exposure to toxins in concentrations exceeding established safe levels.
As of six or more months later, none of the four replies that were received suggested any other such toxins or other pathway.
Related to this are the very large differences between concentrations of toxins in breast milk and those in the main alternative infant feeding. The toxins being discussed here are present in infant formula in concentrations less than 7% as high, and usually less than 1% as high, as their concentrations in human milk. (See near the beginnings of each of the subsections of Section 3 above.)
Section 7.a: Effects of all of the above developmental toxins specifically on male children, who are very disproportionately diagnosed with neurological disorders as well as falling behind in education:
According to an MD who is author or co-author of 248 studies, articles and editorials, a large number of them on the subject of ASD (Peter Szatmari), "the sex ratio in ASD is perhaps the most consistent finding in the field but perhaps the least understood."17h A research team who are collectively authors of about 300 studies says essentially the same thing about the "constant" finding of the preponderance of males in ASD.17j A very considerable amount of speculation and research on the subject17k has not led to any agreement as to the origins of the one-sided ratio.
The following is a brief summation of a publication at www.male-development.info.
The very high ratios of males to females among children affected by the increasing neurological disorders (especially ASD and ADHD) have been basically unexplained. Disproportionate problems among boys and young men have also reached well into the general population, in education and psychologically, far beyond those with diagnosed disorders. Again, there is nothing but speculation as to the cause of this disparity. Most of the toxins discussed here have been scientifically observed to have sex-specific effects on male learning ability, male capacity for higher cognitive processes, and development of the male brain. (See Section 7.b, next) Looking back at when the boys would have been infants who would have been the first to widely have these problems as they grew older, one can see the mid-to-late 1970’s as the likely birth years of the boys and young men who would later have been the first ones having the problems of concern here. Breastfeeding rates were very rapidly increasing during that period (see Figure 7); human milk is clearly a major source of infant exposure to developmental toxins that are known to have adverse effects specifically on male development. Apparently nobody knows about existence of any other major pathway for developmental toxins to infants in doses exceeding established safe levels. (see Section 6).
Adverse effects of mercury (as well as of the other toxins described earlier) are not limited to diagnosed disorders. This chart (from Davidson et al., 2010)98 shows effects observed in a general population of students in a study carried out in the Seychelles, showing apparent effects on scholastic achievement from moderately elevated mercury levels due to fish consumption.
It is significant to see evidence (in this set of charts) of widespread, adverse cognitive effects -- primarily on males -- associated with variations in postnatal levels of mercury.
For the complete text of which the above is a summary, go to www.male-development.info.
Since an understanding of what underlies the preponderance of diagnosed males may help in explaining the origins of ASD and ADHD, about which so little is known, something that helps explain this disproportion should be of considerable interest. Infants are very commonly exposed to several recognized developmental toxins each of which has adverse effects on males while having little or no effect on females; those exposures deserve careful attention (see below).
a) Mercury: see Section 3.d,
b) PCBs: See chart on right and Section 3.a.4;
c) PBDEs: Lactational exposures to four types of PBDEs were found to have adverse effects on males but not on females (Hoffman et al., 201617s) Some types of PBDEs have been found to inhibit production of testosterone (Costa et al., 200817m), and testosterone is important to male neurological development;17n, 17o
d) dioxins: Boys but not girls with elevated dioxin exposure were found to have decreased communication scores.17p Boys but not girls had significantly lower neurodevelopmental scores in relation to elevated total dioxin exposures via breastfeeding.(Nishijo et al., 2014)118h
e) lead: See below and www.breastfeeding-and-lead.info, Section F.
Also, human milk, which typically contains all five of the above chemicals in very significant concentrations (Section 3.g), has itself been found to have adverse effects on cognition specifically in males, in a dose-response relationship, in a long-term study. (see the Belfort et al. study in Section 4.a.2.d).
Even though toxins that accumulate in human milk seem generally to have distinctly male-adverse effects, such a pattern appears to be unusual for environmental chemicals in general; on the contrary, female sex has been shown to be a risk factor for clinically-relevant adverse drug reactions.17r So, considering that these five chemicals each have a property (adverse effects specifically on males) that fits very well with a fundamental characteristic of both ASD and ADHD, that is reason for even greater concern that each of those toxins is present in typical human milk in concentrations that exceed established safe levels.
The following is a brief summation of information that will be found in more complete form, with references to authoritative sources, in a web document at www.pesticides-and-breastfeeding.info.
Although pesticides have apparently not been officially designated as neurodevelopmental toxins, as was the case with the toxins discussed earlier, there are good reasons to believe that they should be so designated. According to a European/American team of scientists, many pesticides used in agriculture “target the nervous system of insect pests,” which is of special concern because of the “similarity in brain biochemistry” between insects and humans. By highly-authoritative sources, human milk has been found to normally contain many pesticides, and infant formula in the U.S. has been found to contain essentially none. Exposures of children to pesticides at current background levels correlate well with reduced mental capacities as well as with ASD, ADHD, and other neurodevelopmental problems, in many studies
There are many reasons to see breast milk as being by far the predominant source of pesticide exposures to infants during the early-postnatal period of rapid brain development. Two leading experts on toxins involved in child development have reported that “specific pesticides ... are passed on to the infant via breast milk, resulting in infant exposure that exceeds the mother’s own exposure by 100-fold on the basis of bodyweight." A 2011 study found that the largest source of exposures to pesticides for most infants was breastfeeding, even in an agricultural community.
To read the much more complete form of the above, with references to authoritative sources, go to www.pesticides-and-breastfeeding.info.
The time when toxic exposures take place is critical in determining their effects; vulnerability is greatest when development is most active,12,13 especially before six months after birth, when the blood brain barrier becomes complete.4b
(For more on the topic of especially great vulnerability of neurological development to toxins during the early postnatal months, see Appendix F)
Breastfeeding, transferring several different developmental toxins to infants, each in very significant amounts (see Section 3.g) is by far most prevalent during the early months after birth; also, the concentrations of some of those toxins are apparently much higher during earlier breastfeeding. (see Section 2.b.2) The combination of the above leads to far greater total transfers of toxins soon after birth than at any other stage of development.
Given all of the above: Could there be a connection between
(b) the unexplained major increases in impairments in specific functions that are especially vulnerable to toxins during the early postnatal period?
The charts above provide an introduction to increasing neurological impairments that appear to align well in time with increases in breastfeeding; but there are other kinds of disorders, also, that have increased in a way that similarly aligns well with increases in breastfeeding.
Increases in U.S. child obesity since the 1970's
Another example of this kind of parallel trend would be diabetes, which also increased to epidemic proportions among children and young adults during the decades of increasing breastfeeding. (see Figure 12.2 and accompanying text.) For other examples, see www.breastfeedingprosandcons.info.
The reader will probably notice that examples of what actually occurred over the years, such as the above, conflict with what would have been expected based on studies that are cited by promoters of breastfeeding. It should be noted that former U.S. Surgeon General Regina Benjamin pointed out that essentially all of the studies that have found health benefits of breastfeeding have been of the observational type;a1a and evidence from such studies is considered by the leading authorities on medical evidence to be of low quality. This topic will be discussed in more detail in Section 10. (The trend in obesity in relation to effects of breastfeeding will be discussed in Section 13.b.)
There are no doubt other toxins or sources of toxins that have also substantially increased in the environment during the period when the disabilities have increased greatly, but breastfeeding as a pathway for developmental toxins is distinctive in several important respects:
a) Some of the toxins known to be substantially present in breast milk have been found to affect males predominantly, which makes a very good fit with the otherwise unexplained disproportionately-high percentage of males affected by autism and ADHD. (see Section 7)
b) Many studies have found breastfeeding duration to be positively correlated with prevalence of autism, including in dose-response relationships. (see Section 4)
c) Several toxins in breast milk are widely received by infants in doses well in excess of recognized safe levels. (see Section 3.g). Based on the overall response to letters inquiring on this topic, addressed to many organizations and scientists with appropriate expertise, nobody seems to disagree with the above italicized statement or believe that children are widely exposed to any toxins exceeding safe levels by any pathway other than breastfeeding. (See Section 6) The principal alternative infant food contains very little or none of those toxins. (see Section 3.g)
d) Breast milk is a predominant or sole source of food for many or most infants during the especially sensitive early months after birth. This means that exposures to toxins in the milk during that very vulnerable period can be harmful in a way that could probably never be equaled by normal exposures to any other toxins. A study by Chien et al. found that over 95% of breastfed infants' exposures to mercury came from breastfeeding,96 and a study team led by Johnson-Restrepo calculated that 91% of a typical U.S. breastfed infant’s total exposure to PBDEs was from breast milk.56e Similar extremes of exposure to PCBs would also come from breastfeeding, as indicated in Section 3.a.
Section 9.b: In those early-postnatal months, other neurological development (in addition to that described in Section 2.a) is taking place that has not been going well during the period of major increases in early-postnatal transfers of toxins to infants:
The first three months after birth are the period in which growth of connections in the prefrontal cortex is especially rapid;114f bearing in mind that development is especially vulnerable to effects of toxins while the development is active (see Section 2.a), one should expect that those first months would therefore be the time when normal formation of that autism-related brain region (see Section 4.a.2.b) is unusually vulnerable to effects of environmental toxins. The following areas are controlled or strongly affected by the prefrontal cortex:
-- sustained attention,
-- formation and retrieval of memory,
-- aspects of intelligence, including ability to formulate and carry out behavioral plans,
-- gaze control; 114g
It should be apparent from the above list that defects in development of the prefrontal cortex could be highly relevant to (a) ADHD (sustained attention), (b) learning disability (memory, perception, and comprehension), and (c) ASD (all of the above functions, with gaze control/eye contact being especially distinctive as a problem in those with ASD).
That means that exposure of an infant to neurodevelopmental toxins during the first three months after birth, when connections are forming rapidly in the prefrontal cortex, could be of especially great significance regarding those increasingly-diagnosed disorders.
For additional information about special sensitivity to neurodevelopmental harm during the early months after birth, see Appendix F.
According to a leading authority on development of the brain,114e there are also specific areas of neurological development that are known to take place in the first six months after birth, which will be listed below; most of those are areas in which there have also been strong indications of faltering development of children in recent decades, especially in males. Those include the following:
-- behavioral indices of attention;
deficits in attention are clearly central to ADHD, which has greatly increased in recent decades; and attention problems are also a frequent trait of autism;
-- behavior in response to novel stressors;
agitated responses to novel situations or stimuli are well known to often be traits of those with the increasingly-diagnosed ASD;
-- circadian rhythm;
this ties in with the frequent problem of children with autism not wanting to sleep at night;
-- the motor system;
clumsiness is a frequent trait of those with ASD.
As common durations of breastfeeding increased to six months and more during the most recent decades (see Figure 7), the above areas of neurological development would have been receiving increased exposures to the developmental toxins that are known to be high in breast milk (see Section 3). The downward trends in development of children in those areas, paralleling the increases in exposures to those toxins, might not be coincidental.
For more about vulnerability to toxins during the early months after birth, see Appendix F.
Section 10: Poor evidence to support breastfeeding; studies and outcomes assessed only in early childhood.
Section 10.a: Authoritatively recognized low quality of the kind of studies (observational) that have found benefits of breastfeeding
Essentially all of the studies that have found benefits of breastfeeding have been of the observational type; according to a 2011 statement by former U.S. Surgeon General Regina Benjamin, "research on the health outcomes of different modes of infant feeding is limited to observational studies."a1a The leading authorities on medical evidence have determined that evidence from observational studies is predominantly of low quality, with evidence from only exceptional ones reaching a medium level of quality. One such determination has been provided by Dr. Gordon Guyatt and an international team of 14 associatesa2 (see Fig. 12 below); Dr. Guyatt is chief editor of the American Medical Association’s Manual for Evidence-based Clinical Practice, in which 26 pages are devoted to examples of studies (almost all of which were observational) that were later refuted by high-quality studies.a2a
(Source of and link to chart at a2a)
Notice that observational studies by their very nature would start out with quality of evidence considered to be low, and the quality would be downgraded even further in the following cases:
-- a) if there is serious risk of bias, which there almost always would be in the case of studies of outcomes of breastfeeding (see 3 paragraphs down), or
-- b) if there is publication bias, which there normally would be, given the widely and firmly held belief in the value of breastfeeding as well as the clear position on breastfeeding of the U.S. government; note that the U.S. government is the source of many millions of dollars in research grant money, which would be well known to researchers who hope to be funded in the future. (Go to www.cdc.gov/breastfeeding/ to see the U.S. government's unequivocal promotion of breastfeeding.)
A similar assessment of the low quality of evidence from observational studies has been provided by the other chief authority on medical evidence, Dr. David Sackett,a2c writing about “the disastrous inadequacy of lesser evidence,” in reference to findings from observational studies.a2b Dr. Sackett was said by the British Medical Journal to have been "widely regarded as 'the father of evidence-based medicine.'"a2d
When people choose or decide not to choose an intervention, such as breastfeeding, that introduces "confounding" due to different types of people choosing one or the other alternative. Specifically, mothers of higher socio-economic status (with better health-related circumstances) are much more likely to choose to breastfeed than mothers of lower SES;2 similarly, women who adhere to medical guidelines (and who therefore also have better health habits in general) are also unusually likely to breastfeed. (Section 10.b below) In both cases, this causes confounding as to what are the real causes of the better health outcomes. Better child health is found to be associated with breastfeeding, and inferential conclusions are typically drawn on the basis of those associations; but the associations could well be caused by the confounders. Defenders of observational studies say that they can control or adjust for confounders, but according to Rory Collins, an epidemiologist at Oxford University, “you can’t measure these things with precision so you will tend to under-correct for them. And you can’t correct for things that you can’t measure.” 152 According to a 2014 study by a team of eight researchers, "it is generally considered impossible to completely mitigate the potential for bias associated with observational studies through study design or analytic method because residual unidentified confounding factors can rarely be ruled out, and statistical adjustment or matching procedures are often inadequate."150a
Section 10.b: The adherer effect: This effect recognizes that people who faithfully follow medical instructions are different from those who don't; the presumed relevant difference is that people who follow medical instructions diligently are better at following guidelines for good health in general, leading to health benefits completely aside from effects of any particular treatment that is being tested. There can be no doubt that "adherers" among mothers would be very predominantly breastfeeders, considering the strong promotion of breastfeeding that comes from the medical community150b as well as from many seemingly-informed private citizens.
The adherer (or "compliance") effect stood out in the case of hormone replacement therapy (HRT). This therapy was found in many observational studies to be beneficial for women who faithfully complied with it. As of the mid-1990s, the American Heart Association, the American College of Physicians, and the American College of Obstetricians and Gynecologists had all concluded that the beneficial effects of HRT were sufficiently well established that it could be recommended to older women as a means of warding off heart disease and osteoporosis.152 A 2004 meta-analysis concluded, after examination of 31 studies on the topic, that the pooled estimate of effect from the best-quality observational studies implied a relative reduction of 44% - 50% in coronary heart disease with "ever" use of HRT.152i, 152i The authors stated that "the preponderance of the evidence strongly suggests that women taking postmenopausal estrogen therapy are at decreased risk for CHD. The consistency of the findings is more apparent in the better designed and analyzed studies."152j
But when results came in from randomized studies, the study type recognized to be of high quality, it was found that outcomes of the HRT therapy were actually worse rather than better. According to a New York Times article quoting Dr. Jerome Avorn, a Harvard epidemiologist, the observational studies may have inadvertently focused their attention specifically on the “Girl Scouts in the group, the compliant ongoing users, who are probably doing a lot of other preventive things as well.”152 There are good reasons why the leading authorities on medical evidence have such low opinions of observational studies (see earlier in this section); such studies are thrown off by confounders such as the adherer effect.
People who faithfully follow medical instructions (such as recommendations to breastfeed) are different from those who don't; their health-related habits are better.
Their health -- and probably the health of their children -- is also likely to be better, even if specific recommendations that they follow are useless.
Figure 12.1 (source of chart at ref. 151)
Good adherence to recommendations --- even taking sugar pills (and breastfeeding) — is associated with better health outcomes.
The above chart drew its data from "the placebo arms from eight studies."151 In addition to the 2006 meta-analysis of studies summarized in the above chart, other studies have arrived at similar findings of favorable health outcomes among those who diligently follow health instructions that merely direct patients to take medically ineffectual pills.151a, 153
The authors of the Simpson study (a chart from which is shown above), when summarizing their review of other studies, stated, "For participants with good adherence to placebo or beneficial drug therapy, the risk of mortality was about half that of participants with poor adherence." Their assumed explanation was that "the presence of good adherence is a marker for overall healthy behaviour."151 Specific aspects of healthy behavior would include diet, exercise, regular follow-up with healthcare professionals, immunizations, and screenings, all of which would also provide health benefits to an infant under an adhering mother's care. And the adhering mother would normally also be faithfully following the general and medical recommendations to breastfeed, so that better child health would be associated with breastfeeding even if the better health actually resulted entirely from the mother's general adherence to good health-related behavior.
A 2014 study cites three other studies from 2005 and later (in addition to the studies mentioned above) in support of its statement, "A growing body of evidence suggests that good adherence to treatment is an important predictor of positive treatment outcomes, regardless of whether the treatment is an active therapy or an inert placebo." And also, "The fact that good adherence among the placebo group is beneficial across a variety of medication trials, health outcomes, and patient samples suggests that this is a robust and reliable effect."154
In addition to general problems with observational studies (such as confounding by the adherer effect, as just described), there are special problems when such studies are carried out in the area of breastfeeding. Women with higher levels of certain toxins have been found to breastfeed for much shorter durations than the average; those in the top 5% for DDE were found to breastfeed for an average of only 9 weeks, typically reporting early weaning because of "not enough milk" or "difficulty feeding," while those in the bottom 6% of those toxins were found to breastfeed for an average of 35 weeks.155 So the category of children classified as "early weaned" would consist to a great extent of children who received the strongest doses of toxins, and those who were breastfed longer received the least polluted milk. So better health results would be found in children who were breastfed longer who were also those whose mothers had the least-polluted milk; the especially strong doses of toxins received from mothers who weaned early would be coming during the early weeks and months that appear to be the most sensitive period for disruption of neurological development (see Appendix F). The cause of the differing health outcomes in relation to breastfeeding duration would likely not be the duration of breastfeeding but instead would be the differing levels of toxins in mothers who breastfed for longer versus shorter durations.
Coming at the topic from a different angle, "Children who become ill or fail to gain weight while being breast-fed are usually supplemented or weaned."155 So poor health of the child leads to early weaning, the reverse of the interpretation that is typically (erroneously) made of breastfeeding studies.
Section 10.c: The confounder of socio-economic differences: The adherer effect overlaps with effects of socioeconomic differences. According to data provided by the U.S. Surgeon General, college graduates appear to be about twice as likely to breastfeed for six months or more as high school graduates.2 Mothers of lower socioeconomic status are much more likely to smoke (which is a health risk factor for nearby infants3) and "to eat what’s affordable rather than what the experts tell them is healthful, to have poor medical care and to live in environments with more pollutants, noise and stress;"152 such characteristics would obviously carry over to adverse health effects in babies of the non-breastfeeding mothers who are disproportionately of lower income and education. Conversely, health advantages are normal at the higher socio-economic levels, a common position of the families in which a high percentage of mothers breastfeed.
Meaningful long-term outcomes versus misleading short-term outcomes:
There is another lesson to be learned from the case of hormone replacement therapy, which Dr. Avorn (introduced above Figure 12.1) refers to as the "estrogen debacle." (Tens of thousands of deaths may have resulted from hormone replacement therapy.152). That lesson is the consequences of drawing conclusions from short-term outcomes. It was long-term studies that eventually found the adverse outcomes of HRT, after evidence from short-term studies had seemed thoroughly convincing to so many doctors and prestigious organizations. That pattern has considerable relevance to breastfeeding; it appears that the vast majority of the studies that have found benefits of breastfeeding have drawn their observations to a close after early childhood, before long-term effects could be observed. The next section will deal with long-term effects of breastfeeding.
Section 11: Evidence that supports breastfeeding draws predominantly on observations of apparent short-term benefits; the evidence about longer-term effects strongly indicates worse long-term health outcomes in relation to breastfeeding history.
Remember from Section 3.f that long-term effects of toxins found in human milk have been detected well past early childhood, often only past early childhood. Many studies were cited in that section, in which long-term effects of these toxins were found in areas related to IQ and other neurological functions, with the adverse effects sometimes found to increase with age. EPA scientists referred to “long-latency delayed neurotoxicity” when discussing effects of toxins such as are contained in human milk; (those toxins are minimal or absent in infant formula -- see Section 3.g); similar statements about long-term effects of these toxins have been made by the U.S. ATSDR, WHO, and a commission of the U.S. National Academy of Sciences. (Details and citations of sources for all of the above can be found in Section 3.f)
The U.S. Surgeon General's Call to Action to Support Breastfeeding 2011 makes only brief reference to long-term effects of breastfeeding (p. 1 at a1a), but what that document says on this topic is revealing if one looks at it closely. After referring to short-term effects of breastfeeding (as found in observational studies), former Surgeon General Regina Benjamin went on to discuss associations of formula feeding with longer-term effects. (p. 2) In that regard, her document asserts that "formula feeding is associated with higher risks for major chronic diseases and conditions... which have increased among U.S. children over time." Given that, it makes sense to consider the trends in recent decades in childhood diseases in relation to trends in infant feeding types during the same period; after all, there have been substantial changes in both in recent decades. The Surgeon General's document does not go into any detail regarding changing levels of formula feeding in relation to increases of the diseases she specifies, so we will do so here:
First, we need to remember the time trend of breastfeeding in the U.S., which began rapidly increasing in the early 1970's. (See chart from the Surgeon General's document on left.)
Then we should briefly review what has actually happened in prevalence of important childhood diseases during the decades when breastfeeding was increasing in the U.S.
Abbreviated review of Fig. 11a
Observational studies, as cited by promoters of breastfeeding, tell us that breastfeeding reduces risk of obesity. As discussed above and earlier in Section 10, there are several serious problems with drawing conclusions from such studies. This chart, showing what actually happened in U.S. childhood obesity when breastfeeding increased greatly, illustrates how separated such studies can be from reality.
One of the diseases mentioned by the Surgeon General in her reference to increases of diseases over time (type 2 diabetes) was authoritatively characterized as follows:
According to the president of the American Diabetes Association, as of 2002 type 2 diabetes had "changed from a disease of our grandparents and parents to a disease of our children." At that time it was on its way to being what she called a "new epidemic" among children and young adults.152a,150 According to the only readily-found study on the history of the increase of childhood diabetes, "the rising incidence of the condition was not widely recognized until the 1980s;"152b and the first reference found by this author (after substantial search) to inclusion of young adults within this epidemic was in 2002. All of this points toward a beginning of the major increase in childhood diabetes coming some time after the start of the rapid increase in breastfeeding. (For much more information on this topic, including remarkable parallels between highs and lows of childhood diabetes and highs and lows of breastfeeding rates geographically and according to time, see www.breastfeeding-and-diabetes.info.)
The next increasing disease to which Surgeon General Benjamin referred, when discussing presumed adverse effects of formula feeding, was asthma. Newacheck and Halfon (2000) reported that the prevalence of disabilities related to asthma among U.S. children (based on the National Health Interview Survey) increased 232%, or more than tripled, during the period from 1969–1970 to 1994–1995.152c (Data points were 10 years or more apart; the first point that showed an increase was in 1979.) Notice how similar that is to the increases in breastfeeding that took place during those years. (see Figure 12.2 just above) By contrast, serious cases of asthma among adults were stable during the portion of that period (1980 and later) for which data is readily available.152d (For more information on this topic, see www.breastfeeding-and-asthma.info.)
Surgeon General Benjamin also referred to the increase in leukemia as relevant to consideration of the "risks" of formula feeding. Since this discussion is about general health effects of breastfeeding versus formula, it would be much more meaningful here to discuss increases in childhood cancer in general. We should remember from Figure 12.2 (not far above) the following:
-- very rapid increases in breastfeeding from 1972 to 1982,
-- a subsequent dip lasting for a decade or so,
-- surpassing of the earlier peak 1-1/2 decades later, and
-- increases then continuing at slower rates.
See the similarities between the above pattern and that of childhood cancer (below):
In parallel with increases in breastfeeding, there were:
-- substantial increases in childhood cancer from 1975 to 1985,
-- then a minor dip for a decade or so,
-- rising above the earlier peak 1-1/2 decades later, and
-- a mild increase in the final decade.
The changes in childhood cancer incidence (described above) resemble a somewhat-smoothed-out version of the changes in breastfeeding rates, with the turns in child cancer diagnoses lagging several years behind the turns in breastfeeding rates.
It should not be surprising that average childhood cancer diagnoses would lag only very few years behind related exposures, since latencies of childhood cancers after exposures to carcinogens can be very short152g, 152f because of the rapid cellular proliferation taking place at that stage (see Figure 2); latencies even for some adult cancers have been determined to be only 4/10 of a year.152h
Note that there are also many close correlations of high childhood cancer incidence with higher breastfeeding rates, by geographic location as well as by time period; for more information on this topic, see www.breastfeeding-and-cancer.info .
Remember that it was former Surgeon General Regina Benjamin, a strong promoter of breastfeeding, who brought up the subject of increases that have been taking place in childhood diseases in relation to alleged risks of formula feeding. (She probably did so without having first looked at how the trends of breastfeeding and child diseases have actually compared in recent decades.) As indicated above, the increases that have taken place in important childhood diseases have apparently all tracked closely with increases that have taken place in breastfeeding. The close parallels might be seen as amazing coincidences except for the fact that the infant feeding that tracks so closely with the disease increases contains at least five different toxins in concentrations exceeding established safe doses. (See Section 3, especially Section 3.g)
Aside from what is revealed by the historical record, over 50 studies (in addition to those cited above) have found adverse health effects of breastfeeding, often including dose-response effects. (see www.breastfeeding-studies.info) A high percentage of those studies have made their assessments well past early childhood.
For more about long-term effects of toxins contained in human milk, see Appendix L.
Section 12: Randomized trials, although still subject to bias, are recognized as being superior to observational studies, and they provide far better evidence about effects of breastfeeding.
It is considered to be improper to dictate to mothers whether or not they should breastfeed, so there have been no truly randomized studies of breastfeeding versus formula feeding. (Studies in Belarus had randomized promotion, but the mothers still chose to breastfeed or not.) But there have been two randomized studies at the edges of the breastfeeding-versus-formula question, which have been revealing:
A study was conducted of effects of alternative feedings of preterm infants, in which the infants were randomly assigned to be fed either banked human milk or a combination of human milk and infant formula, as supplements to the mothers' milk. Developmental quotient scores (at nine months) of infants fed partially with formula were higher than those of infants fed with banked human milk. The higher scores for partially-formula-fed infants also applied in all of the six subgroups that were studied: boys, girls, babies ventilated for less than or more than 24 hours, and infants of sizes appropriate for or small for gestational age. The authors noted that "it was surprising that a brief period of dietary manipulation (median 30 days) could have such prolonged consequences."147a To summarize the above: When preterm infants were randomly assigned to be fed either entirely with human milk or partly with infant formula and partly with human milk, the infants fed entirely with human milk had lower developmental scores in all of the six subgroups that were studied.
The next chart is from another randomized trial, one that shows results that are in sharp conflict with the usual observational studies that have found benefits of breastfeeding.
Below: Prevalences of various allergies, with and without exclusive breastfeeding
One could search in vain for any reference to this study, or to other studies negative to breastfeeding (such as in Section 4), in the Policy Statement on breastfeeding of the American Academy of Pediatrics. 146a But one could find ample reference to "the protective effect of exclusive breastfeeding” in that AAP statement.
Bear in mind the doubling and tripling of allergy rates that have occurred in recent decades, especially among children and young people,145 during a period when breastfeeding rates have increased at least that much.
In addition to randomized studies, another type of study that is an improvement over the usual observational studies is indicated by the EPA as follows: “Epidemiologic studies of exposed human populations provide the most convincing evidence of human health effects.”145a For results of an epidemiological study that found positive correlations between breastfeeding and autism, and between more extended breastfeeding and greater prevalence of autism, based on data from all 50 U.S. states and 51 U.S. counties, see Section 4.
Still another way of avoiding the weaknesses of observational studies is sibling studies, as a way of trying to make sure that comparisons are valid; a 2014 sibling study found that asthma was “one outcome for which breastfeeding duration is consistently associated with poorer childhood health and wellbeing across all three models.”152e For much more about the relationship between asthma and allergies and breastfeeding, see www.breastfeeding-and-asthma.info.
Related to the effects of shielding developing infants from allergenic foods, discussed above, also note the following:
Breastfeeding also protects infants from bacterial challenges,
by means of the antibodies in human milk; this is not disputed. But again there is good evidence indicating that the eventual outcomes of the protection have turned out to be adverse rather than favorable.
While developing infants were receiving greater shielding from bacteria by means of increasing breastfeeding (late 1960's to present), auto-immune and allergic diseases among children increased greatly. 146 (as did many other diseases -- see Section 13 just below). Decline in immunity in relation to increased shielding is in line with the highly-regarded hygiene hypothesis. According to that hypothesis, reduced exposure to common infections leads to increases in those disorders, presumably because of lack of challenges that would otherwise be stimulating the normal development of the child's own immune system. Included among the increasing auto-immune and allergic diseases is asthma, which increased sufficiently during the decades when breastfeeding rates were greatly increasing that it came to be referred to as an epidemic.149
Section 13: Final remarks
(Some of the following is the same as the final part of the introductory summary, but it is worth repeating now that the information leading up to and supporting it has been presented.)
Section 13.a: A reasonable question to consider:
The above brings us to some important matters to think about:
(a) non-communicable disorders have been greatly increasing among children in recent decades -- for basically unknown reasons (Section 1),
(b) important developmental processes taking place after birth are authoritatively recognized to be vulnerable to toxins ingested postnatally (Section 2.a),
(c) toxins known to be typically at high levels in human milk have been found to lead to effects similar to symptoms of the increasing disorders (Section 3); and breastfeeding has been increasing while the disorders have been increasing (Figures 7 and 7.a), and
(d) positive dose-response relationships have been found between breastfeeding and autism in several published studies, and strong positive associations between breastfeeding and autism have been found in several other studies (Section 4),
it is therefore reasonable to ask the following question of the medical organizations that promote breastfeeding:
How has it been determined that developmental toxins in human milk have not been contributing to increases in disorders that could outweigh the benefits of breastfeeding?
The U.S. medical associations that promote breastfeeding (pediatricians, family physicians and obstetricians and gynecologists) do not respond after being repeatedly asked the above question (with variations in the wording).
Since there is substantial peer-reviewed scientific evidence to support (a) through (d) above, it would seem to be appropriate for those physicians’ organizations to consider that evidence before they recommend feeding infants a food that has been authoritatively determined to contain several developmental toxins in concentrations far exceeding established safe levels. (see Section 3.g) That would be especially called-for given that child neurological and other disorders have increased in parallel with the major increases in breastfeeding. (see Sections 1, 5 and 13.b ahead) And a response to the above question would be even more in order considering the apparent absence of widespread exposure of infants, by any pathway other than breastfeeding, to developmental toxins in doses exceeding established safe levels (see Section 6).
If careful study had been carried out on such an important matter of public health, a written record of the analysis of the important evidence ought to be available to the public. But there is apparently no such record, which implies that the breastfeeding recommendations are based on something other than careful consideration of the important evidence. Doctors, like the rest of us, are subject to groupthink and other conformity-inducing social influences, and there can be no doubt that there is a powerful popular movement in favor of breastfeeding.
The apparently unmet need for careful consideration of scientific evidence on both sides of this topic is especially great since there is an alternative feeding that worked well for an entire generation born in the mid-20th century U.S., before the generations were born that have had major increases in non-communicable disorders. (see below)
A 2008 study found that 78% of women would stop breastfeeding if they were aware of just one toxin in their milk, even at low levels.132 Most American mothers do breastfeed, indicating that most of them are not aware of the presence of any of the toxins in their milk, not to mention five toxins, each present in concentrations far exceeding established safe levels see Section 3.g), plus others in very significant concentrations, all of which are either low or absent in U.S. infant formula, according to authoritative assessments.
There is considerable case law that applies to a parent’s right to be informed about toxins contained in breast milk. According to a decision of the Supreme Court of Washington (state), citing two earlier cases, “Important decisions must frequently be made in many nontreatment situations in which medical care is given….The physician's duty is to tell the patient what he or she needs to know in order to make them” (the decisions).133 (A doctor’s advice about infant feeding clearly influences an important health-related decision.) The Supreme Court of Wisconsin stated in 2012, "a growing number of courts require physicians to disclose what a reasonable person in the patient's position would want to know."133a (emphasis added) As of 2016, courts in the U.K. and in half of U.S. states had adopted such a standard.133b As indicated in the above-cited study,132 verifying what should be intuitive, most women do consider knowledge of presence of toxins in breast milk to be a crucial determinant of their decisions about whether to breastfeed. Doctors therefore apparently have a legal obligation to tell mothers about those toxins, as part of consultations on breastfeeding.
The "baby boomers" (the post-World-War-II generation) were rarely breastfed, according to the American Academy of Family Physicians.134 That generation had increases in life expectancy that were as great as those of the previous, more-breastfed generations,134c despite a high rate of smoking among boomer women.134d But the favorable trend took a downturn among those born after the births of the rarely-breastfed boomer generation.
Observe in this chart (from a 2017 study) mortality data for 30-year-old people born in the second half of the 20th century. For the U.S. whites represented on the red line, subtract 30 from the mortality-data years shown, which will show that those whose mortality was distinctly increasing were born in 1970 and later. That increase was in sharp contrast with the mortality declines that had been the norm for those born in earlier years. See how closely that shift parallels the historical trend in breastfeeding, which increased rapidly in the 1970's. (see Figure 7)
Since these recent developments seem to be at odds with the conventional wisdom about benefits of breastfeeding, it justifies a close look as to what may be underlying factors. First to consider is that, according to highly authoritative sources, typical contemporary human milk contains five different developmental toxins in concentrations exceeding established safe doses, with two of those toxins exceeding the EPA's reference dose by scores to hundreds of times; and it is very likely that the effects of the combined exposures are far greater than the effects of exposures to those toxins individually. (The above is discussed in Section 3.g)
Note that an alternative type of infant feeding is readily available that, compared with human milk, contains little or none of the developmental toxins discussed above (see Section 3.g); that same alternative infant feeding was also the standard feeding type for the generation born in the U.S. “throughout the mid-20th century,” as stated by the American Academy of Family Physicians.134 And that generation had health outcomes that
a) were very respectable compared with the health of the generations that preceded them, and
b) would compare especially favorably with the compromised health of the much more highly-breastfed generations that followed. (see beginning of Section 13b, above, and Figure 13c above, and also below)
We should look at factors that might underlie the noticeable departure from the long-established pattern of increases in lifespans. Increasing obesity, starting in childhood, was the primary increasing health problem seen by one team that researched this topic; life-shortening consequences of obesity include diabetes, heart disease, cancer, and disabilities. This research group (as the CDC has done134h) referred to obesity as an "epidemic."134c
When looking at the next chart below, excerpted from a publication of the CDC, one can see evidence of existing low, stable obesity prevalence in children born during the 1960's, as shown in the still-low 1971-74 figure for 6-11-year-olds. This was followed by a long-term upward trend beginning among children born in the 1970's and later; see the 1976-1980 figure for evidence of substantial increase in obesity that began taking place among the (6-to-11-year-old) children as their birth years started to be in the early 1970's.
Also see the subsequent tripling and quadrupling of obesity that took place in later years. And remember from earlier the rapid, major increases in breastfeeding rates that began to be substantial during the early 1970's and in general continued upward in later years. (See Figure 7)
For the other two age groups for which the CDC showed data from the 1970's, there were also no significant increases in obesity until the period when the birth years of those age groups' members would have been in the 1970's and beyond.134e For discussion of specific toxins that are related to obesity and that are typical contents of human milk, see Section 13.c later.
To review important points from the above:
-- 1) Life expectancies of children born during the low-breastfed, mid-century "boomer" generation continued to increase in line with the previously established pattern of increases.
-- 2) It was only after this generation was born that the longstanding, continuing improvements in health outcomes stopped and changed directions. Both of the leading explanations for downturns in life expectancy (obesity and misuse of drugs taken to deal with pain) are compatible with being results of developmental toxins ingested via the breastfeeding that rapidly increased in the 1970's.
(As opposed to the increases in child obesity in children born in the 1970's and later, which would be expected to lead to increased risk of adult obesity, there was also increased obesity distinctly in adults in the boomer generation; but that apparently had its origins in occupational transition from farm and factory work to sedentary work. That is discussed in Appendix W.)
According to an article in the journal Pediatrics, review of the relevant scientific literature suggests that overweight among preschool children, as well as among older children, may be associated less with increased energy intake and more with low physical activity; in the Framingham Children’s Study, inactive preschool children gained substantially more subcutaneous fat by the time they reached first grade than did more active children.134t Given that, it is significant that infant exposure to PCBs via breastfeeding has been closely associated with reduced physical activity of children, in a dose-effect relationship. (see chart on right)
It is also relevant to remember that PCBs have been found in human milk in doses 63 to 270 times the minimal risk level established by the U.S. Agency for Toxic Substances and Disease Registry,23 which is compatible with data from the UN Environment Programme.23f See illustrative chart below.
Reinforcing the link between obesity and PCB exposures: A 2011 review article summarized, "these findings indicate that relatively low, and for most populations environmentally relevant (PCB) exposure levels, ... may be associated with obesity development;" and in their conclusion, the authors stated that findings of the reviewed studies indicate that "endocrine disruptors (the broader category that includes dioxins and PBDEs as well as PCBs) may play a significant role in the current obesity epidemic."134u
There were increases in other adverse health trends that also started in the 1970's, in addition to that of obesity, increases that make the health of the low-breastfed boomer generation look good by comparison, in multiple respects:
-- As reported in a 2008 publication of the U.S. Center for National Health Statistics: “Over the past three decades in the United States, behavioral and learning disorders have emerged as major chronic conditions affecting the development of school-aged children and adolescents.”5 So 1978 would be the approximate year when large numbers of children increasingly affected by those disorders would have been first entering school; those beginning-schoolchildren with declining health would have started being born in approximately 1972-73.
-- A 2012 study by a team of scientists from the U.S., France, U.K., Denmark and New Zealand (Barouki et al.) referred to the “many… major diseases – and dysfunctions – that have increased substantially in prevalence over the last 40 years.” 6 Diseases that have apparently been increasing since the 1970's, such as childhood cancer6h and (the especially common) asthma,6m can be present in children during the first year after birth;6k, 6n therefore the early 1970's would have been relevant birth years for children at the beginning of the increasing trend in diseases and dysfunctions that (according to this research group) was first observed after about 1972.
-- A 2017 study by 10 U.S. and Canadian researchers states that "The prevalence of neurodevelopmental disorders such as autism and Attention-Deficit/Hyperactivity Disorder (ADHD) has increased over the past four decades," citing five studies other than the ones cited above.6p
In all of the above cases, those children whose health trends were declining would have started being born after the low-breastfed mid-century generation had been born.
See (in this chart from a document of the U.S. Surgeon General) how infant feeding was changing during those years when child diseases were increasing. And remember from Section 3.g authoritative information about toxins typically present in human milk.
See (below) data concerning what was taking place in the environment after World War II, which would have had a major impact on the contents of human milk in recent decades (see Section 2.b):
.(Source of chart at 6d)
Clearly, chemicals were dramatically increasing in the environment during the middle of the 20th century, leading up to the increases in diseases and dysfunctions. However, notice above that chemical production quadrupled over a 20-year period (1947-1966) without increases in adverse health outcomes being observed. This suggests that chemicals that are merely out in the environment don't necessarily reach developing children in doses sufficient to cause harm.
But: Notice in the chart before this one the time of rapid increases in activity in the main pathway for developmental toxins to infants in concentrated form. (Also see Section 2.b) Could greatly increasing transfers via that pathway have had something to do with the many health declines that were reported to have begun in the 1970's?
As long as the main pathway for exposure of infants to these chemicals was decreasing (through about 1960 -- see Figure 7), overall effects of these increasing toxins on the general population of children were so minor that any effects apparently went unnoticed. Remember from above (and see more below) that general increases in adverse child health conditions were reported by many sources as having started in the 1970's. As a minor deviation, one report indicated the second half of the 1960's as birth years of children at the beginning of adverse health trends; but that is still fully compatible with adverse effects' occurring only when children are exposed to toxins via breastfeeding. (See left-hand chart in Figure 7, showing a 30% increase in breastfeeding 1965-1969.)
Health outcomes of children who were born after the seldom-breastfed generation, continued:
-- Studies in the New England Journal of Medicine and the Journal of Allergy and Clinical Immunology point to the 1970’s as the beginning of doubling and tripling of allergy rates, especially among children (allergies occur commonly in infants) and young people.145
-- Ample evidence points to the 1970’s as the time when major increases in childhood diabetes began.146
The earliest beginnings of the child health declines: According to a 2007 article in the Journal of the American Medical Association, “the number of children and youth in the United States with chronic health conditions… has increased dramatically in the past 4 decades.” 144 Asthma, the most common chronic health condition among children,144a can start at any age;6g so the above estimate (4 decades before 2007) would place the birth years of the children who were born at the bottom of asthma's increasing trend line in the late 1960's. Bear in mind that even the flat levels of breastfeeding during 1962-1965 (see below) would have led to some increasing transfers of toxins to infants, given the increases of chemicals that had been taking place in the environment.
Repeat of Figure 7, for convenient reference
A different study of historical U.S. infant feeding was very compatible with the above patterns, stating that
a) breastfeeding in the U.S. continued a long-term decline until about 1960, and
b) breastfeeding increased after the mid-1960s. 134a But, as shown in the above charts, the increase was relatively minor during the late 1960's and became rapid only in the early 1970's.
A central point: Toxic chemicals in the environments of mothers and infants can be increasing without having noticeably increasing effects on infants, as long as there is only minor activity in the major pathway transferring the toxins to infants in concentrated form (breastfeeding -- see Section 2.b).
The above brings up a pattern that has been observed elsewhere in scientific studies:
-- In the Lung et al. study (Section 4.a.1), only minor effects of pollution from nearby incinerators were found among non-breastfed local children; but substantial autism-related effects were found in breastfed children in those same local areas;
-- In the Mocarelli et al. study, boys whose mothers had been exposed to elevated levels of dioxins during gestation were found in adulthood to have either minimal effects, if not breastfed, or substantial adverse effects if they had been breastfed.
All of the above strongly suggests that environmental exposures, which can and do occur in modern industrial societies, can either have or not have harmful effects on developing children depending on the feeding type received during infancy. As indicated by very substantial evidence, there is a concentrating effect that takes place as part of the lactational process, in which a mother's long-term accumulations of lipophilic (fat-soluble) substances are transferred to infants. (see Section 2.b) Depending on their feeding types, infants can receive either
(a) just the original, direct exposure from the environment, which may well not be sufficient to be harmful by itself, or
(b) that original exposure plus a greater, concentrated additional exposure via lactational transfers from their mothers, very likely yielding total exposures that are sufficient to be harmful.
Links of breastfeeding with the major increases in obesity, with its life-shortening consequences, were brought up briefly near the beginning of Section 13.b. It is worth exploring the associations of specific known contents of breast milk with obesity.
Review of Figure 4
Many studies have found PCB exposures to be associated with obesity. 134f According to one major study, "Cross sectional studies of adult PCB exposures have consistently shown a positive association with measures of obesity."134g
Studies have also shown effects of PCBs on brain development, including in areas that could impact satiety and metabolism.134o Remember Figure 5.d reviewed (and accompanying text) convincing evidence of effects of PCBs in reducing activity, which in turn is expected to increase risk of obesity.
Dioxins are chemically related to PCBs; one of the two major categories of PCBs is classified as "dioxin-like." PBDEs have been found in a laboratory experiment to cause weight gain;134m and remember from Section 3.a Dr. Merzenich's statement that PBDEs are known to be close chemical relatives of PCBs and should be expected to have similar effects. Also remember from Section 3.g that all three of these chemicals are present in human milk in concentrations greatly exceeding established safe levels, while either absent in infant formula or present in only very minor concentrations.
Remember from "Increasing obesity" in Section 13.b that obesity was stable and low among children born during the 1960's but it rapidly more than tripled among those born in the 1970's and later. However, among young children born in the 2000's, it declined following major efforts intended to address the obesity epidemic; the decline occurred after what the CDC described as "a combination of childhood obesity efforts under way at the national, state, community and family levels." Federal efforts include a childhood obesity program of the CDC and USDA’s revision of the WIC food package,134k including limiting children older than two years to low-fat milk.134n (The WIC program is aimed at lower-income families but reaches over half of all American children. 134n) There was also considerable national promotion of physical activity. But even after some apparent success of these considerable measures to address the basic causes of obesity (efforts that were not in place in earlier years), new cases of obesity were still developing in over twice as many children as had been the case before the obesity-reducing measures were put into effect. There is clearly something affecting children other than the causes normally associated with obesity.
In reporting about matters that could be related to the above, publications of the CDC point out that the increase in infant and child obesity was very large (34%) during a decade when breastfeeding also increased greatly (35%);134h on the other hand, the increase of obesity was much more moderate (10%) during a decade when the increase in breastfeeding was much more moderate (17%).134j (The CDC was merely reporting the above data while not bringing up the possibility of a causal relation between those two variables.)
So bottle-feeding/ non-breastfeeding/ formula-feeding can be seen as a favorable alternative to breastfeeding in modern industrial times. Human milk would probably be the ideal food for infants based on human evolution, given the environments that homo sapiens inhabited for over 99% of our history. The problem is the pollutants that have entered human milk mainly since the middle of the 20th century.
It appears that well-meaning people have built up a powerful bandwagon on the basis of what seemed to make sense, based on what was known at an earlier time. They found that their ideas could be confirmed by studies; but those studies were very much subject to confounding by underlying other factors. Remember from the discussion of "observational studies" (in Section 10) that essentially all studies that have found benefits of breastfeeding (as stated by the U.S. Surgeon General) are of that type, a type that leading authorities on medical evidence consider to be basically of low quality.
The American Academy of Pediatrics, despite the very ample evidence of developmental toxins in average human milk (see Section 3), does not mention environmental toxins even once in its policy statement, "Breastfeeding and the Use of Human Milk."37b One might wonder whether the phrase, "the whole truth," has any meaning to some people when they are comfortably sitting on a bandwagon. Such telling of half-truths would be excusable if parents were generally aware of the important information that is being withheld, but that is certainly not the case here.
Between the adherer effect and the fact that most breastfeeding studies have paid no attention to long-term adverse effects, it should not be surprising that many studies have found benefits of breastfeeding. What should be surprising is that, despite the above, over 50 peer-reviewed studies have found overall adverse health effects of breastfeeding. (see www.breastfeeding-studies.info)
For additional information about trends in health of American children since 1970, see www.breastfeeding-health-effects.info. For more information about the toxins that have been found in human milk versus those in infant formula, see www.breastfeeding-vs-formula.info. For additional information about doctors’ legal obligations to inform parents about toxins contained in breast milk, see www.medical-liability.us.
In the U.S. CDC's publication, GUIDELINES FOR THE IDENTIFICATION AND MANAGEMENT OF LEAD EXPOSURE IN PREGNANT AND LACTATING WOMEN, the "Research needs" section near the end expresses a need for "development of new therapeutic agents or mechanisms to remove lead from breast milk." 98g3
That is certainly a sensible objective; but, considering the ample evidence of presence of other developmental toxins that are also present in human milk, in most cases in concentrations far exceeding established safe levels (see Section 3.g), a proper list of research needs should also include the following:
Development of means to remove PCBs from breast milk (see Section 3.a)
Development of means to remove flame retardants from breast milk (see Section 3.b)
Development of means to remove dioxins from breast milk (see Section 3.c)
Development of means to remove mercury from breast milk (see Section 3.d)
Development of means to remove pesticides from breast milk (see Section 3.e)
Seeing a well-justified list such as this and considering the difficulties that would be involved in achieving more than minor reductions in levels of those toxins in the near or medium term, and very possibly even in the long term, one might well become discouraged. If so, it would be good to bear in mind that there is an alternative infant feeding that has very little or none of those toxins; and it is a feeding type that, as stated above, was overwhelmingly normal for the generation born in the mid-20th-century U.S.,134 which was a generation that did not have the major increases in non-communicable diseases that have become commonplace among children born since breastfeeding became widespread. It should not be surprising that a far better history of child health prevailed during the period when breastfeeding was rare, considering the levels of developmental toxins present in recognized excessive doses in human milk, as opposed to in the main alternative. (see Section 3.g)
As the author of the above, my role has not been to carry out original research, but instead it has been to read through very large amounts of scientific research that has already been completed on the subjects of environmental toxins and infant development, and then to summarize the relevant findings; my aim has been to put this information into a form that enables readers to make better-informed decisions related to these matters. The original research articles and government reports on this subject (my sources) are extremely numerous, often very lengthy, and are usually written in a form and stored in locations such that the general public is normally unable to learn from them.
My main qualification for writing these publications is ability to find and pull together large amounts of scientific evidence from authoritative sources and to condense the most significant parts into a form that is reasonably understandable to the general public and also sufficiently accurate as to be useful to interested professionals. My educational background included challenging courses in biology and chemistry in which I did very well, but at least as important has been an ability to correctly summarize in plain English large amounts of scientific material. I scored in the top one percent in standardized tests in high school, graduated cum laude from Oberlin College, and stood in the top third of my class at Harvard Business School.
There were important aspects of the business school case-study method that have been helpful in making my work more useful than much or most of what has been written on this subject, as follows: After carefully studying large amounts of printed matter on a subject, one is expected to come up with well-considered recommendations that can be defended against criticisms from all directions. The expected criticisms ingrain the habits of (a) maintaining accuracy in what one says, and (b) not making recommendations unless one can support them with good evidence and logical reasoning. Established policies receive little respect if they can’t be well supported as part of a free give-and-take of conflicting evidence and reasoning. That approach is especially relevant to the position statements on breastfeeding of the American Academy of Pediatrics and the American Academy of Family Physicians, which statements cite only evidence that has been
(a) selected, while in no way acknowledging the considerable contrary evidence,a1 and
(b) of a kind that has been authoritatively determined to be of low quality. (See the paragraphs dealing with observational studies near the end of Section 10 above.)
When a brief summary of material that conflicts with their breastfeeding positions is repeatedly presented to the physicians’ associations, along with a question or two about the basis for their breastfeeding recommendations, those associations never respond. That says a great deal about how well their positions on breastfeeding can stand up to scrutiny.
The credibility of the contents of the above article is based on the authoritative sources that are referred to in the references: The sources are mainly U.S. government health-related agencies and reputable academic researchers (typically highly-published authors) writing in peer-reviewed journals; those sources are essentially always referred to in footnotes that follow anything that is said in the text that is not common knowledge. In most cases a link is provided that allows easy referral to the original source(s) of the information. If there is not a working link, you can normally use your cursor to select a non-working link or the title of the document, then copy it (control - c usually does that), then “paste” it (control - v) into an open slot at the top of your browser, for taking you to the website where the original, authoritative source of the information can be found.
The reader is strongly encouraged to check the source(s) regarding anything he or she reads here that seems to be questionable, and to notify me of anything said in the text that does not seem to accurately represent what was said by the original source. Write to email@example.com. I will quickly correct anything found to be inaccurate.
Comments or questions on the above are invited, including criticisms if they are specific, and will usually receive a response. At the next link are past comments and questions from a number of readers, including eight doctors, followed by our responses. Some of the doctors have been critical but others have been substantially in agreement with us (including one with children with asthma, one who says she has delivered thousands of babies, and one with a son with autism); they put into briefer, everyday language and personal terms some important points that tend to be immersed in detail when presented in our own publications. Topics discussed in that section include about having breast milk tested for toxins and about means of trying to achieve milk that is relatively free of toxins, including the “pump and dump” option. To read the above, go to www.pollutionaction.org/comments.htm
In criticisms, please point out any specific passages that you feel are not accurately based on authoritative sources (as cited) or that do not logically follow from the evidence presented. Note that the author of this article feels no obligation to present the pro-breastfeeding case as long as the medical associations and other promoters of breastfeeding fail to inform parents about the developmental toxins that are, without dispute, present in high concentrations in human milk. Please e-mail criticisms or other comments to firstname.lastname@example.org. Most will receive a reply.
For a more complete statement about the author and Pollution Action, please go to www.pollutionaction.org.
Shenandoah, VA, USA
Appendix A: More on adverse developmental effects of mercury:
According to an EPA report to Congress, referring to a crucial part of organization of the brain, “Neuronal migration, a process specifically affected by methylmercury,… continues until five months after birth.”93 (italics added) Note that those five months of recognized special vulnerability of a critical part of brain formation occur during a period of major exposure of breastfed infants to methylmercury (see Section 3.d).
A 2007 study determined the levels of mercury, lead, and zinc in baby teeth of children with autism spectrum disorder and found that children with autism had significantly (2.1-fold) higher levels of mercury in their baby teeth, but similar levels of lead and zinc.95 As the authors pointed out, baby teeth are a good measure of cumulative exposures to metals that occurred during early infancy. That provides strong indication that the mercury exposures that may well have caused the autism took place after birth, at a time when breastfeeding would have been by far the dominant source of mercury to the infant. (See "Predominant exposure to mercury via breastfeeding" in Section 3.d)
In a Saudi study, the number of children a mother had breastfed was found to have a highly significant negative effect on the mercury level in her breast milk, indicating substantial transfers of long-term accumulations of mercury to earlier infants.85c
For other studies finding associations between child mercury levels and autism, bear in mind the dose-response relationship between mercury levels and variations in severity of autism as found in the Adams et al. study described later in this Appendix, and see other studies in Section H of http://www.breastfeeding-and-lead.info.
According to a publication of WHO, “Methylmercury is highly toxic, particularly to the nervous system; the developing brain is thought to be the most sensitive target organ for toxicity.” 90a See Figure 2 and accompanying text for indication of how extensively the brain is developing during the year after birth. Then remember from Section 3.d about methylmercury's major proportion among mercury forms in human blood, and therefore in human milk.
At least ten published studies have found high levels of mercury in those with ASD.94 A publication of Harvard University’s Center on the Developing Child, when discussing “prenatal and early childhood exposures to substances that have clearly documented toxic effects on the immature brain,” mentions only three leading examples of such neurodevelopmental toxins, one of which is mercury.88 The authors gave mercury in fish as a specific example of a source of toxins of concern, indicating that exposures expected to cause neurological harm can consist merely of common dietary exposures. Related to such "early childhood" exposures to mercury possibly having “toxic effects on the immature brain”: observe the fundamentally immature state of the brain during the first year after birth, as illustrated in Figure 2 earlier; also remember (from the beginning of Section 3.d) how far mercury in typical human milk exceeds established safe levels.81, 82
A 2013 study (Adams et al.), by a team of 12 researchers investigating children of ages 5 to 16, found that mercury stood out from among various metals that were studied for their associations with autism. Levels of several metals including mercury were found to be associated with autism and were also “strongly associated with variations in the severity of autism.” But mercury was the variable that was “most consistently significant” in relation to increased autism in both red blood cells and whole blood. 89 The p-value for that relationship was .0003, meaning there were estimated to be only three chances out of 10,000 that the finding regarding autism-related effects of mercury was a random occurrence.
As another indication that background exposures to mercury can have adverse neurodevelopmental effects: in a Spanish study (Freire et al., 2010) it was found that, in 220 preschool children with mercury exposure that was merely above average for the area, exposure was negatively associated with memory (-8.4 points), verbal area (-7.5 points), and General Cognitive Score (-6.6 points).89a (However, above-average mercury level in this population was higher than in most locations, since fish was a larger proportion of the area diet than average.)
Predominant exposure of infants to mercury is via breastfeeding: A 2006 study (Chien et al.) determined that, of the three sources of infant mercury exposure, ingestion (breast milk), inhalation, and dermal exposure, the largest contribution was from breast milk, providing 96 to 99.6% of the total exposure.96 The U.S. ATSDR shows compatible data indicating over-150-times-greater intake of mercury via food than via air,96a and provides other evidence indicating that absorption of mercury is predominantly via food.96b Combining the above with awareness of the typical concentrations of mercury in human milk,81,82 it is apparent that breastfed infants' intakes of mercury would therefore generally be predominantly via their intakes of food.
In a study by a highly-published scientist (P. Grandjean) and his team, it was found that total mercury concentrations in infants that had been breastfed for one year were three times as high as those in infants that had not been breastfed.100 Another study found more than doubling of infant mercury levels due to 6 months of breastfeeding.101
Reductions of nursing mothers' levels of mercury, PCBs and dioxins during lactation have been found in many studies (see Section 2.b.2)
Corresponding increases in infants during lactation also take place in the cases of other neurodevelopmental toxins present in human milk: PCBs, PBDEs and dioxins -- see Sections 3.a, 3.b and 3.c.
When observing here the major increases in mercury levels in infant blood during lactation, bear in mind that methylmercury becomes concentrated in the brain, to many times the levels in the blood.83a (Hair measurements, such as were used in this study, are considered by the ATSDR to reflect mercury levels in the blood.83e) So the relative increases in infants' brains' levels of mercury related to breastfeeding would be even higher than the blood-level increases.
Transfer of mercury from the mother also takes place prenatally, but that is minor compared with postnatal transfers via lactation. According to what may be the only study that compared mothers’ total mercury levels in early versus late gestation, average levels of total mercury of over a hundred women were the same at gestational week 37 as at gestational week 12.85a In a Brazilian study of 100 mothers and newborns, mercury levels in newborns were found to be less than one third as high as in the mothers.85e Gestational transfer of mercury to the fetus clearly occurs, but apparently at a rate that is similar to what the mother would normally be accumulating; by contrast, mercury excretion via breast milk has been found in many studies to rapidly draw down the mother's accumulated burden.20 Remember from earlier the 1998 German study that found that concentrations of mercury in breast milk of 85 lactating women at just two months after birth had declined by an average of over 70% from their levels at time of birth.19
In a Taiwanese study, negative correlation was found between methylmercury levels in 3-year-old children and their scores on a test of expressive language. Prenatal mercury exposure, however, did not show significant influence on neurological development.92b
Bear in mind the evidence (presented in Section 3.d) linking mercury with neurodevelopmental harm, especially resulting from exposures occurring during early infancy, and remember that mercury in infant formula has been found to normally be less than one-tenth as high as in human milk. (see first paragraph of Section 3.d)
For much more on developmental effects of mercury, see www.mercury-effects.info.
Appendix B: Please go to www.pollution-effects.info/appendixBandC.htm
Appendix C: Please go to http://www.pollution-effects.info/appendixC.htm
Appendix D: Another reason why learning disability can be seen to be linked with auditory harm associated with the early, most active months of breastfeeding:
Remember the effectiveness of breastfeeding in transferring PCBs to infants (see Figure 4), as well as the apparent effects of PCBs on hearing function.31, 30a Bear in mind that normal hearing tests (assessing ability to hear beeps of sound) would not detect inability to understand meanings of sounds; and remember from Section 3.a that PCB exposures via lactation (at levels similar to some human exposures) were found to lead to “dramatically altered organization of the representations of sound” in the brains of half of the rats tested. Most children with learning disabilities have their primary deficits in basic reading skills, and research indicates that “disability in basic reading skills is primarily caused by deficits in phonological awareness”114c (phonological means related to speech sounds). A child’s awareness of variations in speech sounds (and therefore learning ability) would clearly suffer from “altered organization of the representations of sound,” such as was found in the brains of rats that were lactationally-exposed to PCBs.
If the results of the animal experiment are relevant to effects on developing humans (which is a very strong possibility), a child’s learning ability would be expected to be related to the PCB exposure that occurs via human milk (see Figure 4); that kind of exposure has increased greatly (Figure 7) while learning disability increased by over 280,000 children, over the 14-year period during which breastfeeding and the accompanying transfers of PCBs were very rapidly increasing. (Figure 11) And that exposure is by far greatest during the early months after birth, the specific period when the auditory function in the human brain is developing and therefore especially vulnerable. This is in line with the increases that have been recorded.
Appendix E: Effects of PCBs, continued:
A publication of the EPA cites six studies, of both humans and animals, as indicating links between PCBs and ADHD; only one other toxin (lead) was mentioned as being implicated in this regard.25e Another EPA publication states, "Studies in humans have suggested effects similar to those observed in monkeys exposed to PCBs, including learning deficits and changes in activity associated with exposures to PCBs."25f
A Dutch study detected hyperactivity and slower mean reaction times in relation to the current PCB levels in the children at 42 months of age.25c (Bear in mind a 2014 study's finding that children at 45 months of age who had been breastfed for 6 to 12 months had over 9 times the PCB concentrations compared with non-breastfed children.30a) According to a 2012 review article, "a substantial body of epidemiologic literature has provided evidence that cognitive deficits are associated with elevated PCB exposures."25a In a year 2000 article, a scientist with Japan's National Institute for Environmental Studies points out that PCB exposure even at low levels during the perinatal period (shortly before or after birth) can influence development in a way that can "cause irreversible neurological damage."25b
According to the U.S. ATSDR, "Transfer of PCBs via breast milk can be considerable and, like prenatal exposure, has the potential to contribute to altered development."25d Also, "Monkeys exposed from birth to age 20 weeks to PCB mixtures of congeneric composition and concentration similar to that found in human breast milk showed learning deficits long after exposure had ceased." (then citing five studies in support of this statement) (p. 384 of above) In addition, the ATSDR document says that "There is evidence that PCBs play a role in neurobehavioral alterations observed in newborn and young children from women with PCB burdens near background levels...." (p. 382 of above)
According to a 2016 study, "recent epidemiological data show that PCB body burdens continue to be associated with impaired reproductive and neurological health in humans." 23a The authors cited 12 studies published between 2009 and 2015 in support of their statement about adverse neurological and reproductive effects of PCBs as found in recent studies of populations.
A 2008 American study (Stewart et al.), carried out by a team of six researchers who are authors or coauthors of over 500 published studies among them, also provided relevant results; this study was based on testing of children at the very meaningful age of 9. (Many studies have tested children at much younger ages, before long-term effects of the toxins became apparent, and then declared that the PCBs had no detectable effects on the children.) This study's results, showing specifics about effects of relatively typical developmental PCB exposures, are indicated in the charts below:
(Note that on scales of this type, below 70 would normally be considered retarded)
Continuing exposures to PCBs at harmful levels: The exposures to PCBs that were linked with the cognitive declines shown in the charts above were not poisonings in the usual sense, they resulted merely from background exposures in the U.S. Great Lakes region; the exposures were to PCBs that are widely present in the environment, including from eating fish, from emissions from PCB-containing products, building materials, and other residuals from the widespread uses of earlier decades, and from various new sources, including diesel and (probably) gasoline-powered vehicle emissions (see Section 4.a.4), paint, cosmetics, textiles, paper, leather, printing ink, and other sources27 especially including color-printed paper and as by-products of pigment manufacturing.27b
The “p-values” for the first three charts just above all met the standard for statistical significance by wide margins. The authors indicated that they found IQ deficits to be correlated with “prenatal and/or perinatal” (up to four weeks after birth) exposures of the children, which was in line with the time-of-birth time of their measurements; they made no later measurements that would have permitted correlations with exposures that occurred longer after birth. But there are good reasons to believe that early-postnatal measurements even well past the perinatal period, if carried out, would have shown even greater effect than those at time of birth, as follows:
a) postnatal exposures to such chemicals have been found to be many times greater than prenatal exposures (see Section 2.b); also see the charts (in Figure 4) showing greatly increased postnatal PCB levels in breastfed children in relation to time-of-birth levels;
b) after birth is a stage of development at which vulnerability to toxins continues to be high, and for some important areas of development it is a period of much greater sensitivity than during gestation (see Sections 2.a and 9);
c) as seen in the Mocarelli et al. study, it was found in one important case that prenatal toxic exposures by themselves had no apparent effect on the child in adulthood; but prenatal exposures did affect children who were later breastfed. The placenta serves as a protective barrier against many types of toxins; toxins received by the mother prenatally are stored in her body fat and later mobilized and transferred in concentrated form via breastfeeding. Notice (below), in an animal study cited by the U.S. ATSDR, that lactational transfer of maternal PCBs to pups was found to be many hundreds of times greater than prenatal transfer."23b
Notice "freedom from distractibility" as one of the areas measured in the above charts; that is clearly something that would be relevant to a core characteristic of ADHD. A 2015 study also found close associations between exposures to PCBs (in early infancy), at common background exposure levels, and ADHD-related behaviors.27a While discussing effects of PCBs other than on IQ and memory, note that a research team cited nine studies to support its statement about "chronic effects (of PCBs) on a range of social and anxiety-related behaviors." 23a
According to the Washington State Department of Ecology, “PCBs have been shown to impact normal brain development in addition to producing other toxic effects. PCBs have been well studied in laboratory animal and human epidemiological studies.” Continuing, “These studies indicate that exposures to PCBs are associated with impairments in brain function resulting in deficits in IQ, memory, language and school performance.” 24 (italics added) In that same Department’s ranking of 15 different environmental toxins for developmental effects on infants and children (Table 13, column 1), PCBs were the only toxin that joined lead in having the highest ranking for toxicity of its "Developmental Effects."
Adverse exposures to PCBs are not limited to an insignificant minority. According to a research team who are authors or co-authors of over 340 scientific studies among them, writing in 1993, "Based on current breast milk concentrations nationwide, it is estimated that at least 5% and possibly more of the babies born in the United States are exposed to quantities of PCBs sufficient to cause neurological effects."6c That was published in a year when breastfeeding for six months was half of its current rate, and before exclusive breastfeeding had become as common as it became in the 2000's. (see Figure 7) PCBs in the environment may have declined since 1993, but they continue at harmful levels, and comparably toxic chemical relatives of PCBs have greatly increased in the environment since that time. (see Merzenich quote and also Section 3.b)
Another study found adverse neurological effects of PCBs as measured in breast milk of women who had no known special exposures to PCBs, nothing except general background exposure.23d
A 1980 study in the New England Journal of Medicine, quoting data gathered during the period when PCBs were still widely intentionally produced, showed PCB levels in human milk to be only about 14 times the Allowable Daily Intake at that earlier time;23e seeing this in relation to the figures quoted above in a 2010 publication, there is an implication that PCB hazards have not declined greatly in recent decades, as has sometimes been reported.
Appendix F: Additional information about harmfulness of postnatal exposures to developmental toxins, especially during early weeks and months after birth: (Section 2.a discussed high sensitivity of development to toxic exposures during the postnatal period overall.)
-- A report of the U.S. Public Health Service refers to the “early months after birth” as a “critical period” for development of the nervous system, during which infants are particularly vulnerable to harmful effects of toxins.93a
-- The type of dioxin that is the predominant type in human milk (OCDD) declines in its concentration in breast milk by 50% during the first month after birth.93d
-- One major form of PBDE (BDE 209), which the EPA says has high developmental toxicity,93k was found in a study to be seven times as high in colostrum (breast milk during the first few days) as in mature breast milk.56f
-- A web page of the NIH states that neonatal hypothyroidism, which it says can cause intellectual disability, can result from thyroid levels that are “only slightly low.”93e (“Neonatal” refers to the first four weeks after birth.) Note that thyroid levels are known to be reduced by dioxins,93g PCBs,93h PBDEs,93j and mercury, all of which are typically present in breast milk in concentrations greatly exceeding established safe levels (see Section 3.g). Dioxins, PCBs and certain PBDEs are present in breast milk in especially high concentrations in the early weeks after birth. (see above and 73a, 93d, 56f) Several other authoritative sources concur with the statement of serious consequences of neonatal hypothyroidism, including statements in an Oxford Journal about “devastating functional consequences” and “mental retardation resulting from neonatal thyroid hormone deficiency.”93f,
-- Remembering that neonatal refers to the first few weeks after birth, note that according to a Taiwanese study, “Neonatal BDE-209 exposure has been demonstrated to have neurotoxic effects in most in vivo studies.” Continuing, “Neonatal exposure to BDE-209 has been found to have developmental neurotoxicity, including hyperactivity; learning and memory defects; a reduction in habituation;….” 59 BDE 209 is present in especially high concentrations in traffic pollution, postnatal exposure to which has been found to be closely correlated with autism (see www.traffic-pollution-autism.info, including Figure 3).
-- The maternal body burden of one major type of PCB (153) was found in animal tests to be reduced by about 60% by lactation during the first five days of nursing.85b Animal experiments in this area are considered by the U.S. ATSDR to be relevant to humans, as indicated in that agency's statement, "Results from animal studies support the importance of breast-feeding transfer to infants."23b Bear in mind that the ATSDR is the Agency for Toxic Substances and Disease Registry; this indicates the kind of breastfeeding transfer that the ATSDR considers to be important.
-- Overall growth of the brain is unusually rapid in the early weeks after birth (1% per day at first, declining to 0.4% per day after 3 months).114h Remember from Section 2.a about heightened vulnerability of neurological development during rapid growth of the brain.
-- As can be seen in Figure 2, neural connections for hearing functions and for visual functions are forming rapidly in the first few months after birth; therefore, for any functions that are dependent on processing of information received via hearing and/or vision (such as learning ability, communication, and social interaction), those early months are an especially sensitive period for effects of toxins. (see Section 2.b about especially great sensitivity to toxins while development of a structure is active.)
-- Concentrations of dioxins in breast milk in the first days after birth in the U.S. are in the upper part of the range (roughly 300 times the RfD) that has been found in the U.S.; but they taper off to 20 or so times the RfD over the course of a year of breastfeeding.74
-- A 1998 German study (Drexler et al.) found that concentrations of mercury in breast milk of 85 lactating women at two months after birth had declined by an average of over 70% from their levels at time of birth;84 and there have been compatible findings in many other studies.20
-- Formation of connections of the prefrontal cortex is known to take place in the first three months after birth, meaning that those early months would be a period of special sensitivity of that brain region to toxins; development of the prefrontal cortex is relevant to ASD and ADHD, including effects on attention, memory, aspects of intelligence, speech, gaze control and comprehension. See Section 9.b for more.
Associations of PCB levels in early infancy with ADHD-related behaviors at age 8
The 2015 study that was the source of this chart was carried out by a team of researchers who are authors or coauthors of over 1700 studies among them. Note the higher associations of childhood ADHD-related behaviors with PCB exposures when those exposures took place during early infancy, months 0 to 3; those associations of the more-exposed group, in the lower chart, are all statistically significant. Compare those with the very minimal associations when exposures occurred after those early months. When other researchers have failed to find effects of postnatal exposures to toxins, they were essentially always measuring exposures at times much later than the highly sensitive early months; or else they were assessing outcomes at an age that was too early to detect effects (such as symptoms ADHD and ASD) that typically cannot be properly determined until farther into childhood. See the text above this chart and in Section 2.d for considerable other evidence of special vulnerability during those early months.
-- In another animal study, exposures of newborn mice to PBDEs at levels relevant to human exposures on postnatal days 3 and 10 were found to impair neurological development of the pups, but exposures on postnatal day 19 had no detectable effect.85m Mouse postnatal day 10 is the neurodevelopmental equivalent of at or just before the average human day of birth, whereas mouse postnatal day 19 (at which point no effects of the toxins was found) is the equivalent of about 180 days after human birth. So this implies a "window of susceptibility" of brain development to important toxins such as PBDEs that may close within five months after human birth, or possibly much sooner than that.
(Remember from Section 3.a.3 the substantial evidence indicating relevance of animal studies to effects of toxins on human development.)
(In chart below, note that T4 is one of the two major types of thyroid hormone.12b)
In this 2000 study (Crofton et al. 12e), a common PCB mixture (Aroclor 1254) was administered to groups of pregnant/lactating rats using a method (cross-fostering, or switching exposed and unexposed mothers after birth, before nursing) that made it possible to assess effects of gestational exposures separately from effects of lactational exposures. The top line in the chart shows postnatal T4 (thyroid) levels in the control (comparison) group, which had no exposure to the PCBs. The second line shows T4 levels in young rats that had received only prenatal exposures to the PCBs; prenatal exposures clearly had only a minor effect. The third line shows T4 levels of rats that received only postnatal, lactational exposure to the PCBs -- producing a major effect. The effect shown in the fourth line applies to rats that received exposures both prenatally and postnatally -- an effect that was very similar to the effect on mice that received postnatal-only exposure. It is apparent that postnatal exposure to PCBs (via lactation) had a major effect on the developmentally-critical thyroid hormone supply while prenatal exposures had very little effect.
A significant point regarding the above: For brain development, the average human day of birth would be equivalent to a rat age of 11 days;12f as seen in the above chart, brain development at rat postnatal day 11 (equivalent to human average day of birth) is a time when the effect of lactational exposure to PCBs was found to be a large and increasing reduction of thyroid hormones.
In another experiment, PBDE exposures were found to have a similar effect in reducing T4 during the crucial early-postnatal developmental period.12d Remember that thyroid hormones in humans, including T4, are essential for proper postnatal nervous system development.
Aside from the above evidence concerning sensitivity to effects of postnatal toxic exposures, over 30 other studies have found postnatal exposures to developmental toxins to have greater effects than prenatal exposures, in a wide range of health areas.16b
Appendix G: Other illustrations of the concentrating effect of lactation:
Lead is one of the chemicals that (a) is normally present in human milk and (b) (in recent years) has been essentially absent in infant formula.(Section 3.e) It doesn't accumulate in breast milk to multiples as high as is the case with the lipophilic toxins, but the increases in infant exposures via breastfeeding can nevertheless be substantial. In a Chinese study, the mean concentration of lead in breast milk of twelve occupationally-lead-exposed women was found to be almost 12 times higher than that for the occupationally non-exposed population. (Li et al., Transfer of lead via placenta and breast milk in human, Biomed Environ Sci. 2000 Jun, http://www.ncbi.nlm.nih.gov/pubmed/11055009 In the only readily-found international comparison of lead levels in breast milk, the level in China was found to be two to 40 times as high as that in most other countries, with only Saudi Arabia being in near second place. (Winiarska-Mieczan, Cadmium, Lead, Copper and Zinc in Breast Milk in Poland, Biol Trace Elem Res. 2014; 157(1): 36–44. at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3895183/ Table 6)
Breastfeeding's property of magnifying concentrations of toxins can be especially disturbing if it is combined with major elevations in environmental exposures to the toxins, which obviously do occur. For many years, Israeli dairy products contained 100 times the concentration of HCH (a pesticide) compared with similar products in the U.S.; the concentrations found in breast milk were estimated to be 800 times greater than those in U.S. dairy products. (Pohl and Tylenda, U.S. ATSDR, Breastfeeding exposure of infants to selected pesticides: a public health viewpoint, Toxicology and Industrial Health (2000) 16, 65-77) One way to look at the above is as follows: International variation accounts for a 100-fold difference in general-population exposure, and the concentrating effect of lactation accounts for 700 of the final 800-fold difference in infant exposures.
A U.S. study found that treatment of military homes with chlordane for pest control resulted in a fivefold increase of chlordane levels in the breast milk of nursing mothers. (Pohl and Tylenda, U.S. ATSDR, Breastfeeding exposure of infants to selected pesticides: a public health viewpoint, Toxicology and Industrial Health (2000) 16, 65-77) It should be kept in mind that the dose of pesticide to the infant from the breast milk would come in addition to the direct impact of the pesticide on the infant, who was likely also to be breathing it in, possibly crawling on its residue, and very likely engaging in hand-to-mouth activity.
Typical U.S. breastfed infants have been found to take in 242 times as much dioxin per pound of body weight per day as adults take in per day, on average. (Lorber et al., Infant Exposure to Dioxin-like Compounds in Breast Milk, Vol. 110 No. 6, June 2002, Environmental Health Perspectives at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=54708#Download,)
Appendix H: This has been removed.
Appendix I: Studies providing additional evidence of long-term or delayed effects of developmental toxins:
The EPA, when discussing effects of methylmercury and also effects of a component of various pesticides,38e tells about “long-latency delayed neurotoxicity.” The neurology expert, Bernard Weiss, discussing effects of developmental exposure to environmental toxins, says that often “manifestations of damage emerge only after compensatory processes have been exhausted…. Latency periods as long as 15 years have been reported...." (following methylmercury exposure)38f According to a publication of WHO, “Many diseases that are caused by toxicants in the environment require decades to develop." (Specific toxicants causing the diseases discussed were said to be lead, mercury, and endocrine disruptors;38g the latter category includes PCBs and dioxins.) A commission of the U.S. National Academy of Sciences says essentially the same thing about latencies with regard to effects of pesticides.38n
Also, the many studies in Section 4 that found positive correlations between breastfeeding rates and autism prevalence are relevant to a discussion of long-term effects of developmental exposures to toxins, since autism prevalence statistics are normally determined in relation to children age 8 and older. That is apparently due to autism's often not being diagnosed until age 8 and later.
To return briefly to the CDC statement that prompted the above discussion of long-term effects of exposures to toxins during infancy, that statement will be repeated here: The CDC document says,"despite the presence of chemical toxins (in human milk).... effects on the nursing infant have been seen only where the mother herself was clinically ill from a toxic exposure."37a Given the substantial evidence in the previous 6-7 paragraphs about long-term effects of exposures to toxins that are undisputedly high in human milk, including effects that are often observed only after years or decades, one may wonder why a thinking person would present a case based only on what is apparent during infancy. If one were to make allowance for possible intended meaning (in the CDC statement above) of "total eventual effects on people who were breastfed," that meaning would be easily refuted by the substantial evidence presented in this article, especially as found in Section 4.
When reading the following section, remember that, according to a consensus statement signed by an impressive list of 57 scientists, researchers and health professionals, "The concordance between human and animal neurotoxicity assessment is remarkable as demonstrated for lead, mercury and PCBs."36a
According to a team of researchers who are authors or coauthors of over 750 studies, when discussing effects of exposure to PBDEs as investigated in animal studies, "a deficit in learning and memory processes has been found at adulthood in neonatally exposed animals." 38p
. Effects of neonatal PBDE exposure on behavior only after long delays
Aging but not young mice showed odd behavior, after early-postnatal exposures.
The test used in the study that was the source of the above charts is included in the EPA's "Guidelines for Neurotoxicity Risk Assessment." 39s
In a 2006 Swedish study, neurological defects that resulted from developmental exposure to PCBs and PBDEs worsened with age;38m the brain PCB concentrations produced in the experiment were in the same order of magnitude as has been found in human infants and were administered at the equivalent of human time of birth,38c and the effects included defects in learning ability and memory in adulthood. In a 2007 study, peri-natal exposures to PCBs (as well as methylmercury) were found to greatly reduce certain brain components that are important to higher cognitive function, but only in measurements taken as of the rat age that was equivalent to human age 17; such damage was not observed in the assessments of rats at younger ages, including at puberty.38b, 38c An animal experiment (Kaya et al.) is described in www.male-development.info in which the effect of developmental exposure to PCBs in reducing testosterone was nearly three times as great in adulthood as it had been in infancy; bear in mind from that same website that testosterone is important to male mental functioning.. Also see the Colciago et al. study in that same website about the strong effect on learning ability in adult rats that has been found to result from developmental PCB exposure, in an animal experiment.
See Section 11 for additional evidence concerning long-term effects of toxic exposures that take place during the early postnatal period.
Appendix K: Emissions from incinerators that could be causes of autism and other neurological disorders:
PCBs, PBDEs, mercury, and lead are included among chemicals that are part of normal municipal waste matter that would go into an incinerator, especially in discarded electrical fixtures and electronics (PCBs, PBDEs and lead), old painted products (lead), discarded medical items, appliances and light bulbs (mercury), etc. And dioxins are known typical contents of incinerator emissions, created as part of combustion processes that include burning of plastics. All five of those chemicals have been found in potentially hazardous concentrations in typical human milk (see Section 3); it would be reasonable to assume that those chemicals would be present in higher concentrations in milk of mothers living closer to airborne sources of those chemicals, such as incinerators. For explanation of the concentrating effect of lactation, see Section 2.b.
Developmental exposures to the chemicals that would be emitted from incinerators (discussed above), and which then become concentrated by lactation on the way to the infant, have been linked in many studies with neurological deficits, including deficits that are related to autism. (see Section 3)
Appendix L: (Continuation of Section 3.c) Dioxins, with long-term harmful effects found at advanced ages; breastfeeding found to have been the main determinant of long-term dioxin levels:
One effect of dioxins, which has been verified in animal experiments, is reduction of testosterone production in males. This would be expected to have an effect on human reproduction, which is a significant concern at a time when most developed countries have declining populations, leading to concerns about viability of social security systems; declines of birth rates are occurring despite major expenditures on artificial reproductive technologies, with accompanying health risks. But testosterone, reduced by dioxins, is also important to neurological development. More about neurological effects of dioxins and other environmental toxins especially on males (who are disproportionately affected by autism, ADHD and other neurological disorders), can be found below and in Section 7.
In the 2011 study (by Mocarelli and 12 others)76 of the aftermath of accidental exposure to dioxins, the following was found: even at ages 18 to 26, average dioxin concentrations were still twice as high in the breastfed young men as in those who had been formula fed, in both the exposed group and in the comparison group.77
The high levels of dioxins in breastfed young men, even decades after infancy, is explainable partly by the known persistence of these chemicals in the body and partly by the extremes of exposure via breastfeeding. In the above Mocarelli study, it was determined that the median levels of dioxins in breastfed infants were doubled within 4-5 months after birth, while levels in formula-fed infants were reduced by half in that same period of time.76 A German study found that, at 11 months of age, dioxin toxicity-equivalent concentrations in breastfed infants had become about 10 times higher than in formula-fed infants.79
To aid in understanding why prenatal exposures should have had effects on breastfed but not on formula-fed children, bear in mind that
(a) the “placental barrier” has that name for a reason (more so in relation to some toxins than others); the placenta probably provides some protection to the fetus from dioxins during gestation; and
(b) dioxins (like many other toxins) accumulate in the body, are stored in maternal fat, and are later mobilized and excreted in greatly concentrated form in breast milk. (see Jensen quote and accompanying text below Figure 2)
The findings of the above (Mocarelli) study strongly suggest that, with environmental exposures such as often occur in modern industrial societies, if infants are subjected only to the original exposures, the effects may be insignificant; but when infants are in addition subjected to the concentrations of those exposures that come out in breast milk, the long-term consequences can be very harmful.
Another illustration of effects of dioxins was found in a 2007 study by an international research team (Lee et al.) that tested children at ages 12 and 15, thereby permitting good indications of long-term effects of developmental toxins. This study found that learning disability and attention deficit disorder were two and three times as high among children with elevated levels of dioxins, compared with children with undetectable dioxin levels.80 The dioxins associated with such dramatic increases in risk of neurological disorders were at common levels -- found in 27% to 31% of children. Remember from the Mocarelli study above that breastfeeding is by far the main determinant of dioxin levels even far past infancy.
A study published in 2014, investigating developmental effects of dioxins in the Netherlands, where background levels were at levels that the authors described as "similar" to what was found in other industrialized countries, found the following in boys at ages 8 to 12: "an increase in social problems (p<0.001), thought problems (p=0.005), and aggressive behaviour (p=0.001)," as reported by teachers, in relation to increasing postnatal background dioxin exposure.80b Bear in mind that problems in social interaction are a core characteristic of ASD, and the other two problem areas are often found in those with ASD. (The p values shown all indicate very high levels of statistical significance.) The relevant infant exposures took place in about 1990, decades past the peak level of dioxins in most industrialized environments.
The authors commented that "It is alarming to find a significant increase in aggressive behaviour and social problems ... in relation to background dioxin concentrations... both the teachers and parents reported abnormalities.... Furthermore, there is (scientific) literature evidence to support these findings of neurodevelopmental toxicity."
Boys with elevated dioxin exposure, but not girls with such exposure, were found in a study to have decreased expressive communication scores.118g (Remember the 4.5-to-1 male-to-female ratio of autism, a disorder in which impaired communication is a core trait.) In another study, both boys and girls with higher exposures to a specific type of dioxin via breastfeeding had "significantly higher Autism Spectrum Rating Scale (ASRS) scores," and boys had significantly lower neurodevelopmental scores in relation to elevated total dioxin exposures via breastfeeding.118h
Appendix N: Reduction in blood flow and transport of nutrients to developing brain due to lactational PCB exposure:
Inactivity (due to PCB exposure -- see Section 4.a.2.d) would result in reduction of the blood flow that is required for development of the brain. The amount of circulation-stimulating physical activity that would normally take place in developing infants can be approximated, starting with data provided by the highly-published experts, Adolph and Berger: they report that typically-active walking infants “...take more than 9,000 steps and travel the distance of more than 29 football fields" daily.16z4 Then remember from the text next to Figure 5.b earlier that children with the highest PCB exposures were observed to be less than half as active as other children. Such PCB-related inactivity in an infant would mean a reduction in a body's blood flow that normally would enable travelling the lengths of at least 15 football fields per day. (figuring an over-50%-reduction in relation to the data from Adolph and Berger). Thousands of heartbeats per day would not be taking place in inactive children, children who are apparently made less active by PCB exposures during infancy.
Fig. 5, repeated
If those heartbeats were to occur (as they would in normally-active children), they would also bring about major transport of nutrients to the actively-developing infant brain. But considerable brain-development-enabling blood circulation does not occur in children who are inactive due to PCB exposures, exposures that would result mostly from breastfeeding.
In support of a statement about "emerging literature that physical activity and high levels of aerobic fitness during childhood may enhance neurocognition," a team of nine researchers cites eight other studies and adds the findings of their own 2010 study.16z6
Appendix O: Accumulations in breastfed infants:
-- In a 2004 German study, it was found that, already at six weeks after birth, PCB levels in breastfed infants were between four and eight times the PCB levels in bottle-fed infants.18f
-- PBDEs are recognized by the EPA to be bioaccumulative.18g Other evidence indicates long-term major accumulation of PBDEs received via breastfeeding.56g A study team calculated that 91% of a typical U.S. breastfed infant’s total exposure to PBDEs was from breast milk.56e
-- A 2006 study (Chien et al.) determined that, of the three sources of infant mercury exposure, ingestion (breast milk), inhalation, and dermal exposure, the largest contribution was from breast milk, providing 96 to 99.6% of the total exposure.96 ( Mercury is well known to accumulate in organisms.)
-- A 1998 study of 330 mother-infant pairs found that "breast-fed infants of smoking mothers have urine cotinine levels 10-fold higher than bottle-fed infants whose mothers smoke."18c (Cotinine is a marker for smoke exposure)
-- In a laboratory experiment described in a study of milk excretion of chemicals, "levels of BDE-99 following maternal exposure in rats were found to be up to 17-fold higher in brain of pups compared to dams." (as described in reference 107)
-- In a study cited by the ATSDR, it was found that at 11 months of age dioxin toxicity-equivalent concentrations in infants who had been breastfed 6-7 months were about 10 times those in formula-fed infants. 79
-- As mentioned earlier, the U.S. ATSDR, when stating that transfers of developmental toxins are expected to be higher during breastfeeding than during gestation, illustrates that by describing a laboratory test in which the sucklings received 1600 times as much PCB as was received via transfer to fetuses, from the same original prenatal dose.23b In human studies summarized in a publication of the National Academies Press, the comparisons that were most relevant to the above pattern yielded ratios of about 280 to 1 and 775 to 1;18 the specific comparisons in those cases were of PCB concentrations in breast milk in relation to PCB contents in umbilical cord serum and cord blood; so the actual total ingestions of PCBs by infants would obviously be determined by how long breastfeeding continued; the postnatal/prenatal ratio for total exposure could very possibly be over 1000 to 1 with extended breastfeeding.
Appendix P: Pollution levels, with toxins that enter human milk, vary greatly: Much higher in some places, from typical (non-industrial) combustion sources to which humans are closely exposed; but in some areas human milk could still be relatively uncontaminated, similar to what it was normally like before the mid-20th century in most industrial countries.
According to the latest inventory of U.S. sources of dioxins provided by the EPA, combustion of fossil fuels and wood is a leading source of dioxin emissions, exceeded only by medical waste incineration and backyard barrel burning.37o Mercury in combustion emissions is best known in relation to power plant and other industrial emissions; but combustion of fossil fuels in general, including for home heating37p and vehicle operation,37r also emits substantial mercury (although only minimally in the case of natural gas). Although mercury emissions from residential heating and vehicle operation may be smaller than those from industrial emissions, they have the distinct health disadvantage of being emitted near ground level and close to inhalation zones of many more people.
As illustration of the importance of emissions from combustion for heating purposes: Background levels of dioxins in air were measured in a Minnesota location about 25 miles northwest of Minneapolis-St. Paul, with no major industrial or commercial activity in the area. Ambient air samples were collected in the winter and summer, checking for levels of dioxins. The maximum average level of one type of dioxin was 17 times higher in the winter than during the summer, and the maximum average level of another dioxin type was 342 times higher during winter than during the summer.(p. 427 of ATSDR document37n). The authors of the study attributed the differences to increased emissions from combustion sources during the winter, and that certainly seems to be a logical assumption regarding a northern non-coastal state with cold winters, such as Minnesota. The adverse effects of high emissions of dioxins (as well as mercury) during winters would be accentuated by urban concentration: Minnesota has 17 of the 21 coldest U.S. cities with populations over 50,000.37n1 High dioxin levels in areas of concentrated human presence (as indicated above), combined with possible effects of some of the very highest exclusive breastfeeding rates in the U.S.,37t would probably contribute to the apparent fact that (according to the U.S. Department of Education) Minnesota has by far the highest rate of autism in the U.S.37s
Having considered factors that greatly contribute to high levels of environmental toxins that enter breast milk, it will be instructive to look also at an example near the opposite extreme of developmental toxins in the environment (and therefore with very low levels of toxins in breast milk).
The above map of most of Europe shows levels of fine particulate matter (PM2.5) that are normal in various areas. PM2.5 levels are excellent indications of emissions from combustion of fossil fuels; and those combustion emissions would also include the developmental toxins discussed above (see first paragraph of this appendix). Spain clearly stands out as having far lower exposures to this pollution than most of Europe (and probably compared with most developed countries), due to a combination of less industry, below-average population density, and minimal combustion for heating purposes (given its latitude and the temperature-moderating effect of the surrounding seas). The end result would be unusually-low levels of neurological toxins in Spanish breast milk, compared with those in other developed countries.
But Spain also has much less of some important kinds of pollution compared with much or most of the non-industrial world, as well. DDT is still widely used in many developing countries, at least partly because it is an inexpensive way of fighting malaria.37v (For information about harmful neurodevelopmental effects of DDT and its substantial presence in human milk in many countries, see Section 8.) Also, many women in developing countries have high lead and mercury levels because of use of contaminated cosmetics,37u consumption of contaminated fish, and small-scale mining exposures.
So Spain is an example of a location where one might expect mothers -- and human milk -- to have some of the lowest levels of pollutants in the world. Results of this apparent advantage of human milk in Spain over that in other countries are discussed in Section 4.c
Appendix R: Disruption of the blood brain barrier is closely linked with autism:
(Below: comparisons between certain proteins in blood brain barriers in healthy controls (HC) versus in people with ASD)
In the study from which the above charts were drawn,98h in which alterations in blood-brain barriers (BBBs) in relation to autism were investigated, postmortem brain tissues from healthy control subjects (HC) and subjects with ASD were analyzed. The authors concluded that there were alterations "associated with BBB integrity" in brains of subjects with ASD; they also said that they saw evidence to support their hypothesis of "impaired BBB" in the subjects with ASD.
The authors went into detail with explanations for why the various proteins measured were elevated in brains of ASD subjects, including "compensatory mechanism" and proteins being created but not incorporated into filtering tissues; but for our purposes it is sufficient to know that the blood-brain barrier, a normally finely-balanced means of protecting the brain (admitting needed nutrients but screening out considerable unhealthy matter) has been found to be altered in people with ASD. And it is apparent that PCBs, highly present in typical human milk (See Section 3.a), disrupt the blood brain barrier. (see end of Section 3.a.6)
Appendix S: Other reasons why the breastfeeding duration figures for the control group as indicated in the Soke et al., Lyall et al., and Stadler et al. studies depart so far from what seems to be normal:
If participation in a study is burdensome, it should be expected that participation will be high among people who are personally touched by the disorder of concern, but for the controls (intended to represent the population base from which the impaired children came), participation will probably be limited to an unrepresentative minority. Comparisons of data for cases (impaired children) with that of controls will therefore probably not be meaningful.
In the study (SEED) reported about by the Soke group, of the families that were eligible and that were invited to join the study's general population group, more families refused to participate than agreed to join the study, at the recruitment stage. (p. 8 of 117x2) Additional loss of representativeness in that group's responses would result from attrition among those who enrolled, considering the numerous time-consuming procedures of that study: many forms, questionnaires, and interviews, plus medical history provision, biological sample provision, and typically a trip to the designated clinic, all of which were expected to take a total of 5-1/2 hours plus travel time. (p. 26 of 117x1) Information appears not to be available about the percentage of general-population participants in SEED who actually completed the study's procedures, but some idea of the attrition that can take place in a general population control group after enrollment in such a study can be gained from records of the CHARGE study; that study utilized data collection procedures similar to those of SEED. (p. 18 of 117x1) The CHARGE study enrolled 120 for the population group;117y but of those 120, data was reported for only 21, in a later study based on the CHARGE study.117c1 So major expected attrition after enrollment, combined with the original rejection of participation by most of the eligible people who were invited, seems to indicate that only a small fraction of invited families would end up providing the data for the population group; this casts considerable doubt on the representativeness of the data gathered from that sample. The SEED study provided a $175 incentive to encourage people to join and stay with the study (p. 26 of 117x1); but introducing financial incentives to participate may have done much to distort the characteristics of the group who joined and continued in the population group -- and detracted from the ability of that group to represent the population.
It should be borne in mind that the average age of the children studied in the SEED study was 4.7, and children in the CHARGE group were 2 to 5, but most children who will have ASD are not diagnosed with it until after age 4,117z very often years after age 4.117c2 So there will obviously be many children in the general population who have not yet been diagnosed with ASD but whose parents are observing traits that concern them about their child. The reader should consider whether the control (population) groups would attract a disproportionately high percentage of those parents with reason for serious concern. So their responses to many questions would be expected to be heavily biased toward what would normally come from parents of impaired children. This is an example of "volunteer bias" referred to by Dr. David Sackett as being observed repeatedly.117h7 (Sackett is often referred to as the father of evidence-based medicine.)
The representativeness of these studies' population groups (as regards breastfeeding data) should receive a low rating for another reason. High-quality surveys, such as the ones described in Section 4.a.2 that found a median breastfeeding duration in the general U.S. population of 3 months or less, have good participation rates; this helps insure that the responses are representative of the population. Data from a survey with participation as low as 60% is likely not to be considered to be usable.117x One might wonder what a participation rate that starts out below 50%, and declines from there (due to attrition during the study), would imply about the validity of a control group's responses.
The unrepresentativeness of these control groups is illustrated in data for the control group in the Stadler et al. study.117h3 69% of the control mothers were 4-year college graduates, which was over twice the U.S. national average, which in turn was higher than the average in Oregon, where the study was centered.117h6 And college graduates are known to breastfeed at about twice the rate of high-school graduates.117h4 So the long median breastfeeding history in this control group would be related to the unrepresentative nature of the people who were willing to participate as controls in a study such as this, rather than implying a beneficial effect of breastfeeding.
That would almost certainly apply also to other non-randomized studies, including the CHARGE and SEED studies: People who agree or volunteer to participate as controls in time-consuming studies such as these, when the topic doesn't personally affect their families (or a diagnosis has not yet been received), would be unrepresentative as controls.
The above is not to say that such studies' control groups are useless. If biological specimens, medical records, and answers to hundreds of questions are needed, data from such a control group would very often be the best that is available. But if certain questions have been addressed by large, high-quality, easily-answered surveys with high participation rates (such as those that found average breastfeeding durations to be about 3 months in the general U.S. population-- see Section 4), those surveys would be the logical sources for data concerning those questions.
Median versus mean: A minor concern is that, in the comparisons discussed in Section 4.a.2, most of the average breastfeeding durations presented were medians, but in some cases the type of average (mean vs. median) was not specified; and two figures for the SEED study were means. In this discussion, the key differences are of 100% or greater -- breastfeeding durations for ASD and ADHD cases versus those for the general population. By contrast, what appear to be the only studies that have provided both a median and a mean for breastfeeding duration (according to a substantial internet search) showed that the median number of weeks of breastfeeding was 21% less than the mean in one study117h1 and 25% more than the mean in another;117h2 and the data from the poll on this topic in a Babycenter chat group showed the mean and the median to be about the same.118b So it seems that those different types of averages can be used essentially interchangeably without affecting the overall comparison that is being made here.
Appendix T: Concerns, often unjustified, about statistical significance:
Some studies have reached findings that are very meaningful even though the results did not attain "statistical significance" (typically indicated by a certain "p-value" or "confidence intervals"); this is a standard the use (and misuse) of which has been extensively debated.117r In 2016 the American Statistical Association (the world’s largest professional association of statisticians) issued a statement saying that “While the p-value can be a useful statistical measure, it is commonly misused and misinterpreted.... This has led to some scientific journals discouraging the use of p-values, and some scientists and statisticians recommending their abandonment.”117m, 117n, 117n2 Some of the experts who participated in the American Statistical Association's consideration of this matter wrote separate commentaries stressing that the ASA's statement did not go far enough in condemning P-values’ pernicious influences on science.117m According to an international team of seven experts on statistics in science (in a 2016 article), "in most scientific settings, the arbitrary classification of results into 'significant' and 'non-significant' is unnecessary for and often damaging to valid interpretation of data."117n3
Dr. Herbert Needleman, whose studies of children exposed to low levels of lead prompted regulations that limited or banned lead in many common products,117o had some relevant comments on this subject. In a 1991 article by him and another prominent scientist (David Bellinger), 10 studies are summarized that failed, entirely or in large part, to find adverse cognitive effects of low-level exposures to lead; these negative results were usually because of the findings' not meeting the typical standard for statistical significance, a mistake that Needleman and Bellinger referred to as "overvaluing the status of the p value as a criterion." 117p
One or two of the studies described in Section 4 indicate associations between breastfeeding and autism that, according to common sense, would seem to strongly suggest a cause-and-effect relationship, even though an arbitrary standard of statistical significance was not reached. These studies should be seen especially in combination with
(a) the substantial evidence of developmental toxicity of chemicals known to be heavily and distinctly present in breast milk (see Section 3) and
(b) the many studies that have found associations between breastfeeding and autism that were highly statistically significant. (elsewhere in Section 4).
Appendix U: Additional evidence indicating importance of physical activity to cognifive development:
-- In a 2003 meta-analysis of 44 studies (Sibley et al.16z1), it was found that, of eight different types of cognitive assessment carried out, seven showed beneficial effects from physical activity. (Table 2) At least part of the explanation for this is that movement would increase heart rate and blood flow, transporting appropriate nutrients to the growing brain. (For more on this, see Appendix N.)
-- In a publication of the U.S. National Academy press, by a committee from the U.S. National Research Council and the Institute of Medicine, a list is provided of conditions and substances that are "Needed for Normal Brain Development." In addition to the expected needs for specific substances (protein, fatty acids, micronutrients, etc.), also listed as needed are "Activity; Social interaction."16z5
Shown in this chart are indications of numbers of new neurons formed in brains of young adult mice that were either active or inactive. (The serial numbers below the bars represent the two different strains of mice that were tested.) It appears that physical activity either doubled or quadrupled generation of new neurons, depending on strain of mice tested, compared with being sedentary.
The results of an experiment with rats, assessing the effect of exercise on formation of neurons, are shown in the chart below.16t4 The white bars show formation of neurons overall, according to varying levels of activity; the black bars show formation of neural stem/precursor cells, the type of cells that will generate new cells in the future. So the black bars are clearly the ones to focus on as being far more important to long-term function of the brain. Note that four times as many of these key neurons were being formed among light runners as among the less-active comparison group (controls).
Observing that more running had less benefit than light running, the authors proposed the explanation that there were negative effects of stress resulting from the heavier-running rats' being subjected to more compulsion to get them to run more. But even those rats that had been most stressed still had over twice as much creation of neural stem/precursor cells as the control (less-active) rats. The authors also found that learning ability was significantly increased by exercise.
The authors of the above study pointed out that "running, as a kind of physical exercise, is known to be an important positive stimulus for neurogenesis," citing four studies in support of that.
In another animal experiment, mice that had been running for 6 weeks were observed to have two to three times as many newly-born neurons and 60% more proliferating cells than control mice. Among running mice, the number of newly-born neurons correlated with the total running distance. 16z9
Considering the apparent widespread acceptance of findings from an online survey on this topic (see next paragraph), it may be worth looking also at some data that came out as part of an online poll of mothers of children with ASD, in which the mothers were asked how long those children had been breastfed. (This 2016 poll was taken in a chat group affiliated with Babycenter.com.). Mothers responded with durations of breastfeeding for 18 ASD-diagnosed children, and the average duration (both mean and median) was 19 months.118b
It seems unscientific to devote significant attention to data from an online convenience sample, and that kind of poll would not be considered for citing here except for the fact that just such a voluntary online sample was the only source of data for a study (Schultz et al., 2006118c) that is widely cited, including in prestigious scientific journals, as evidence for a favorable effect of breastfeeding in relation to autism risk.118d
Since the 19-month average figure for breastfeeding duration for ASD-diagnosed children (in the chat-group poll) ought therefore to also receive respect (at least from some people), we should think about why such a long duration of breastfeeding could be equivalent to the 6.5 to 7-month durations mentioned in Section 4.a.2:
a) An English-language poll in an online chat group could be drawing participants from many parts of the world, and extended breastfeeding is much more common in many other countries than in the U.S.; the studies discussed in Section 4.a.2 were carried out in the U.S.; and
b) The longer duration of breastfeeding elsewhere could be the equivalent of a shorter duration in the U.S. in terms of developmental toxins transferred to the infant; bear in mind that contents of at least one important developmental toxin (PBDEs) have been found to be present in breast milk in North America in concentrations 10 to 100 times higher than in most other countries.66b Also remember the EPA statement that PBDEs have “adverse neurobehavioral effects following exposure during the postnatal period," 43 as well as other evidence of adverse developmental effects of PBDEs as described in Section 3.b.
Appendix W: Increases in certain health problems in boomer adults:
These have been mainly (a) cancers and other results of unusually heavy smoking on the part of boomer women,134d and (b) conditions that are known to stem from lack of exercise and from the obesity that results from lack of exercise (heart disease, diabetes, joint pain, and reduced mobility).
The lack of exercise is a logical result of occupations' having gone substantially from farm and factory work to sedentary work, by the time that generation was working; agricultural employment in the U.S. dropped by more than half between 1950 and 1970,135 and manufacturing employment also declined greatly (see chart).
But consider that, even with the unhealthy results stemming from major reduction in occupational exercise and from the increase in smoking, the boomer generation's expected average lifespan still continued to increase in line with the pattern of earlier generations. (This was as reported by the large Olshansky research team.134c) This strongly suggests that the lack of breastfeeding on the part of the mid-century generation did not lead to any decline in their health outcomes.
Appendix X: HBCDs and their adverse effects:
The EPA has assigned to HBCDs a “High hazard” designation (its highest designation) for developmental neurotoxicity, as well as other levels of hazard for its other toxic effects.40
HBCD ingestion by breastfed infants has been found to be seventeen times as high as HBCD ingestion by formula-fed infants. (All of the detectable HBCD ingested by the formula-fed infants came from drinking water.)73
In a 2010 U.K. study, HBCDs were found in every sample of human milk examined in the study; and average daily exposure of an infant to HBCDs via breast milk alone was found to be four times higher than the average adult’s combined exposures (per pound of body weight) from diet, inhalation and dust ingestion. In addition, infants at the 95th percentile had exposure over three times higher than that average exposure.41
In an article that drew data on brominated flame retardants in Europe and Asia from over 100 studies published in 2005-2007, the researchers concluded that “Generally, trends show ... increases in concentrations of HBCD wherever determined.”48 (Note that HBCD is the same as the HBCDD referred to in this chart.) A 2008 French study found that “concentration levels (of HBCDs) measured in quantifiable breast milk samples were around 10 to 100 times higher than the levels usually observed for other organic pollutants.”65 A study of pollutants in human milk in Belgium compared levels in 2009-2010 with levels in a previous study in 2006 and found that levels of HBCD in human milk increased by 153% during that relatively brief period.66
Sources of HBCD (HBCDD) in human milk:
HBCDs, also, are used as flame retardants, in construction materials and in home goods. They have been found in human milk and serum throughout the world, although they have received comparatively little attention in the United States.63 A 2012 U.S. study (Carignan et al.) found HBCD in all samples of breast milk tested, with a 39% increase in milk HBCD levels linked with a mother’s residence in a carpeted home, a 17% increase in milk HBCD per item of stereo or video electronics in the home, and a substantial decrease in milk of mothers who regularly chose organic foods.64
One noteworthy characteristic of HBCD exposure is the extremes of exposures within the general population. An American study, testing HBCD levels in dust of 19 Boston-area homes, found that the exposures varied by a ratio of 29,000 to 1;67 and a U.K. study found high-end exposure to HBCDs in house dust that was also about that high.68 Those high levels of HBCD in dust almost certainly heavily affect HBCD levels in mothers’ milk; remember from just above the close associations that were found between HBCDs in a mother’s milk and presence in the woman’s home of electronics and carpeting. In a 2008 U.S. study, house dust accounted for over 80% of estimated human intake of PBDEs (the other, more studied, brominated flame retardant);69 other studies have arrived at compatible findings,70 including with HBCDs.71
All of the above strongly imply that, with household HBCD dust levels found to vary by a ratio of 29,000 to 1 even within a very small study group, and with such dust clearly entering breast milk, some of the breast milk concentrations of this chemical that is designated by the EPA to be a “high hazard” for developmental neurotoxicity40 could probably harm the development of a significant number of infants.
Remember from three charts in Section 3.b (beginning with Figure 5.1) the recently increasing levels of PBDEs and HBCDs in humans and in breast milk, with “increases in concentrations of HBCD wherever determined,” as indicated in the most-recently published studies (published 2005-2007).
Also bear in mind the expert statement (quoted earlier) that "Significantly more (10 to 20 times) of a mother's body burden of persistent organohalogens (which includes HBCDs and PBDEs) is transferred to the infant via the milk than by the transplacental route."17
Appendix Y: Interview with Dr. Michael Merzenich, screenshot of document:
a1a) The Surgeon General’s Call to Action to Support Breastfeeding 2011, p. 33, at www.surgeongeneral.gov/library/calls/breastfeeding/calltoactiontosupportbreastfeeding.pdf
a2) Figure 2 in Guyatt et al., GRADE guidelines: 1. Introduction -- GRADE evidence profiles and summary of findings tables, Journal of Clinical Epidemiology, at http://www.jclinepi.com/article/S0895-4356(10)00330-6/pdf
a2a) Dr. Gordon Guyatt is chief editor of User’s Guides to the Medical Literature: A Manual for Evidence-based Clinical Practice, 2nd Edition (3rd is upcoming), copyright American Medical Association, published by McGraw Hill.
a2b) Writing in The Canadian Medical Association Journal, as quoted in “Do We Really Know What Makes Us Healthy?” New York Times, published: September 16, 2007 at http://www.nytimes.com/2007/09/16/magazine/16epidemiology-t.html?pagewanted=2&_r=0
a2c) In a review in the Journal of the Medical Library Association, only two guides are recommended for use by physicians in evaluating evidence in medical literature, one of which is the one edited by Guyatt et al., already referred to, and the other of which is by Dr. Sackett. (Journal of the Medical Library Association, Oct. 2002, User’s Guide to the Medical Literature: A Manual for Evidence-Based Clinical Practice, Review by Rebecca Graves, at httpi://www.ncbi.nlm.nih.gov/pmc/articles/PMC128970)
a2d) BMJ 2015;350:h2639, at https://www.bmj.com/content/350/bmj.h2639
1) The Surgeon General’s Call to Action to Support Breastfeeding 2011, p. 33, at www.surgeongeneral.gov/library/calls/breastfeeding/calltoactiontosupportbreastfeeding.pdf
2) The above Surgeon General's document, Table 2
4) Houtrow et al., Changing Trends of Childhood Disability, 2001–2011, Pediatrics Vol. 134 No. 3 September 1, 2014 at http://pediatrics.aappublications.org/content/134/3/530.abstract
4b) Center on the Developing Child at Harvard University, National Scientific Council on the Developing Child: Early Exposure to Toxic Substances Damages Brain Architecture, 2006, Working Paper No. 4; especially introduction, pp. 2, 7, 9; p. 2: "the time of greatest brain growth and most intensive construction of brain architecture is also the period that is most vulnerable to the relatively free passage of toxins into its cells." Link for this publication at http://developingchild.harvard.edu/resources/early-exposure-to-toxic-substances-damages-brain-architecture/ This Council is comprised of twelve leading scholars from all over the U.S.
5) Pastor et al., Diagnosed attention deficit hyperactivity disorder and learning disability: United States 2004-2006, National Center for Health Statistics, 2008, at http://www.cdc.gov/nchs/data/series/sr_10/Sr10_237.pdf
6) Barouki et al., Developmental origins of non-communicable disease: Implications for research and public health, Environ Health. 2012; 11: 42. PMCID: PMC3384466 at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3384466/
6a) GW Chance, Environmental contaminants and children’s health: Cause for concern, time for action, Paediatr Child Health 2001 Dec; 6(10): 731–743. at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2805986/
6b) Grandjean and Jensen, Breastfeeding and the Weanling’s Dilemma Am J Public Health. 2004 July; 94(7): 1075. PMCID: PMC1448391 at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1448391
A thorough 2014 study by a Swedish/Norwegian team also estimated mid-20th century as the approximate beginning time of the more serious pollution: Manzetti et al., Chemical Properties, Environmental Fate, and Degradation of Seven Classes of Pollutants, Chemical Research in Toxicology, 2014 at http://pubs.acs.org/doi/pdfplus/10.1021/tx500014w
6c) Colburn et al., Developmental Effects of Endocrine-Disrupting Chemicals in Wildlife and Humans, Environ Health Perspect., 1993, at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1519860/pdf/envhper00375-0020.pdf
6d) from Chemical & Engineering News, January 12, 1998, Copyright © 1998 by the American Chemical Society, at http://pubs.acs.org/cen/hotarticles/cenear/980112/prodshipgraphs.html
6f) Women, Infants and Children (WIC) Legislative History of Breastfeeding Promotion Requirements in WIC, at https://www.fns.usda.gov/wic/legislative-history-breastfeeding-promotion-requirements-wic
Also see summary of U.S. government breastfeeding promotion at https://maloney.house.gov/sites/maloney.house.gov/files/documents/olddocs/breastfeeding/20050505_CRS_Federal%20Legislation.pdf
6g) Asthma web page for Ireland at https://www.asthma.ie/content/what-age-does-asthma-start
6h) Cancer.gov web page at https://seer.cancer.gov/csr/1975_2014/results_merged/sect_28_childhood_cancer.pdf, showing Table 28.2, Age-Adjusted SEER Cancer Incidence Rates, 1975-201 and Figure 28.1
6k) Table 28.13 at cancer.gov website in footnote 6h.
6m) Grant et al., Observations on emerging patterns of asthma in our society, The Journal of Allergy and Immunology, Aug 1999, at www.jacionline.org/article/S0091-6749(99)70268-X/fulltext
6n) web page at http://asthmaandallergies.org/asthma-allergies/asthma-in-infants-and-young-children/
6p) Lam et al: Developmental PBDE Exposure and IQ/ADHD in Childhood: A Systematic Review and Meta-analysis, Environ Health Perspect/; DOI:10.1289/EHP1632 at https://ehp.niehs.nih.gov/ehp1632/
6r) Commission on Life Sciences, National Research Council: Pesticides in the Diets of Infants and Children, p. 43, National Academy Press, Washington, D.C. 1993, at http://www.nap.edu/openbook.php?record_id=2126&page=43
For a discussion and evidence on this topic, see Section 2 of www.pollutionaction.org/case-deaton-full.htm
7) Di Renzo et al., International Federation of Gynecology and Obstetrics opinion on reproductive health impacts of exposure to toxic environmental chemicals, International Journal of Gynecology and Obstetrics, 2015, at http://www.figo.org/sites/default/files/uploads/News/Final%20PDF_8462.pdf
9) Commission on Life Sciences, National Research Council: Pesticides in the Diets of Infants and Children, p. 43, National Academy Press, Washington, D.C. 1993, at http://www.nap.edu/openbook.php?record_id=2126&page=43
10) Pages 98 and 125 in Improving the Risk Assessment of Persistent, Bioaccumulative, and Toxic Chemicals in Breast Milk, Workshop Summary Report, 2013, prepared for U.S. EPA by ICF International, at http://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=262210
-- Also, according to the twelve U.S. scholars comprising the National Scientific Council on the Developing Child, “the time of greatest brain growth and most intensive construction of brain architecture is also the period that is most vulnerable to the relatively free passage of toxins into its cells.” (see the period of greatest brain growth, after birth, in Figure 2.) (Center on the Developing Child at Harvard University, National Scientific Council on the Developing Child: Early Exposure to Toxic Substances Damages Brain Architecture, 2006, Working Paper No. 4; especially pp. 1 and 2; link for this publication at http://developingchild.harvard.edu/resources/early-exposure-to-toxic-substances-damages-brain-architecture/
11) Rice et al., Critical Periods of Vulnerability for the Developing Nervous System: Evidence from Humans and Animal Models, EPA National Center for Environmental Assessment, at www.ncbi.nlm.nih.gov/pmc/articles/PMC1637807, p. 515
12) Pesticides in the Diets of Infants and Children, p. 60, Commission on Life Sciences, National Research Council, National Academy Press, Washington, D.C. 1993, at http://www.nap.edu/openbook.php?record_id=2126 “Toxic exposures at a particular time would differentially affect the structures undergoing peak development.”
Proceedings of the National Academies of Sciences of the United States of America, Vol. 112,
12b) EPA and U.S. Public Health Service: PUBLIC HEALTH IMPLICATIONS OF EXPOSURE TO POLYCHLORINATEDBIPHENYLS (PCBs), at https://www.epa.gov/sites/production/files/2015-01/documents/pcb99.pdf, p. 7
Schroeder et al., Thyroid hormones, T3 and T4, in the brain, Front. Endocrinol., 31 March 2014 | https://doi.org/10.3389/fendo.2014.00040
Also Porterfield, Thyroidal dysfunction and environmental chemicals--potential impact on
brain development, Environ Health Perspect <#>. 2000 Jun; at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1637839/
Also Zhou et al., Developmental Exposure to Brominated Diphenyl Ethers Results in Thyroid Hormone Disruption, Toxicological Sciences, Vol. 66, Issue 1, March 2002 at https://academic.oup.com/toxsci/article/66/1/105/1634285/Developmental-Exposure-to-Brominated-Diphenyl
12c) Defined at http://www.medicinenet.com/script/main/art.asp?articlekey=7898
12d) Zhou et al., Developmental Exposure to Brominated Diphenyl Ethers Results in Thyroid Hormone Disruption, Toxicological Sciences, Vol. 66, Issue 1, March 2002 at https://academic.oup.com/toxsci/article/66/1/105/1634285/Developmental-Exposure-to-Brominated-Diphenyl
12e) Crofton et al., PCBs, Thyroid Hormones, and Ototoxicity in Rats: Cross-Fostering Experiments Demonstrate the Impact of Postnatal Lactation Exposure, Toxicological Sciences, Volume 57, Issue 1, September 2000 at https://academic.oup.com/toxsci/article/57/1/131/1654970/PCBs-Thyroid-Hormones-and-Ototoxicity-in-Rats
12f) Calculated using the translation tool for humans to rats at www.translatingtime.net.
13) WHO: Principles for Evaluating Health Risks in Children Associated with Exposure to Chemicals, Environmental Health Criteria 237, at http://www.inchem.org/documents/ehc/ehc/ehc237.pdf
13a) National Institutes of Health, News Release: Risk of Autism Tied to Genes that Influence Brain Cell Connections, April 28, 2009, at https://www.nih.gov/news-events/news-releases/risk-autism-tied-genes-influence-brain-cell-connections
13b) Limperopoulos et al., Does cerebellar injury in premature infants contribute to the high prevalence of long-term cognitive, learning, and behavioral disability in survivors?, Pediatrics, 2007, at https://www.ncbi.nlm.nih.gov/pubmed/17766532/
13c) McCarthy and Wright, Convergence of sex differences and the neuroimmune system in Autism Spectrum Disorders, Biol Psychiatry, 2017, at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5285451/
15) WHO: Principles for Evaluating Health Risks in Children Associated with Exposure to Chemicals, Environmental Health Criteria 237, See Figure 2, at http://www.inchem.org/documents/ehc/ehc/ehc237.pdf
16) According to the U.S. National Research Council, of the National Academies, 83% of the human brain’s growth spurt is postnatal. National Research Council (U.S.). Committee on Toxicology, Recommendations for the Prevention of Lead Poisoning in Children, p. 19, at https://books.google.com/books? The following direct link might work: id=15grAAAAYAAJ&printsec=frontcover&dq=Recommendations+for+the+Prevention+of+Lead+Poisoning+in+Children&hl=en&sa=X&ved=0CB4Q6AEwAGoVChMI06yW25q2xwIVjQqSCh2dsQ2G#v=onepage&q=Recommendations%20for%20the%20Prevention%20of%20Lead%20Poisoning%20in%20Children&f=false
16k) Vreugdenhil et al., Effects of Perinatal Exposure to PCBs on Neuropsychological Functions in the Rotterdam Cohort at 9 Years of Age, Neuropsychology, 2004, Vol. 18, No. 1, 185–193 at http://psycnet.apa.org/journals/neu/18/1/185.pdf
16l) Belfort et al., Infant breastfeeding duration and mid-childhood executive function, behavior, and social-emotional development, J Dev Behav Pediatr. 2016 Jan; at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4829914/
16m) Gascon M. et al., Effects of pre and postnatal exposure to low levels of polybromodiphenyl ethers on neurodevelopment and thyroid hormone levels at 4 years of age. Environ Int. 2011 Apr;37(3):605-11. doi: 10.1016/j.envint.2010.12.005. Epub 2011 Jan 14 abstract at www.ncbi.nlm.nih.gov/pubmed/21237513 Full text at www.academia.edu/23477421/Effects_of_pre_and_postnatal_exposure_to_low_levels_of_polybromodiphenyl_ethers_on_neurodevelopment_and_thyroid_hormone_levels_at_4years_of_age..
Note in Results section that "breastfeeding determined age 4 PBDE 47 exposure."
16n) Field, Interaction of genes and nutritional factors in the etiology of autism and attention deficit/hyperactivity disorders: A case control study, Medical Hypotheses 82 (2014) at http://www.acpeds.org/wordpress/wp-content/uploads/S.-Field-ADHD-Autism-article-2014-in-Medical-Hypothesis.pdf
16p) U.S. ATSDR: PCBs, at http://www.atsdr.cdc.gov/ToxProfiles/tp17.pdf , p. 391
16r) P. 253 of Sullivan, Temporality of Risk Factors and the Gender Differential Related to Autism Spectrum Disorder Diagnosis (PhD thesis), Walden University Scholar Works, 2015, at http://scholarworks.waldenu.edu/cgi/viewcontent.cgi?article=1274&context=dissertations
16s) Lynch et al., The effect of prenatal and postnatal exposure to polychlorinated biphenyls and child neurodevelopment at age twenty four months, Reproductive Toxicology 34 (2012) 451– 456 , available free at Researchgate.
16s1) Pellicano, The Development of Executive Function in Autism, Autism Res, 2012, v.2012 at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3420556/
16s2) See CDC document at https://www.cdc.gov/ncbddd/autism/documents/ga-seed-newsletter-13-508.pdf
16t) Jacobson et al., Effects of Exposure to PCBs and Related Compounds on Growth and Activity in Children,
Neurotoxicology and Teratology, 1990, Vol. 12, pp. 319-326, at http://ac.els-cdn.com/089203629090050M/1-s2.0-089203629090050M-main.pdf?_tid=5f238b74-1d4f-11e7-8a29-00000aacb35d&acdnat=1491761419_84450b69c691bfdc5ffd9534957f7940
16t0) Eriksson et al., Polybrominated Diphenyl Ethers, A Group of Brominated Flame Retardants, Can Interact with Polychlorinated Biphenyls in Enhancing Developmental Neurobehavioral Defects, Toxicological Sciences, Volume 94, Issue 2, December 2006, at https://academic.oup.com/toxsci/article/94/2/302/1647585/Polybrominated-Diphenyl-Ethers-A-Group-of
16t1) Kostyniak et al., Relation of Lake Ontario Fish Consumption, Lifetime Lactation, and Parity to Breast Milk Polychlorobiphenyl and Pesticide Concentrations, Environmental Research, 1999, at http://www.sciencedirect.com/science/article/pii/S0013935198939391
Great Lakes fish-eaters have been found to have PCB levels about 35% higher than non-fish-eaters for some of the various forms of PCBs; note that this difference is very small compared with the typical wide variations in people's PCB levels (see the horizontal axis in Figure 5.b).
16t2) Two such studies referred to in Chen et al., A 6-Year Follow-Up of Behavior and Activity Disorders in the Taiwan Yu-cheng Children, American Journal of Public Health, Mar. 1994, Vol. 84 No. 3, p. 418 bottom, at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1614813/pdf/amjph00454-0081.pdf.
Also Johansen et al., Postnatal exposure to PCB 153 and PCB 180, but not to PCB 52, produces changes in activity level and stimulus control in outbred male Wistar Kyoto rats, Behavioral and Brain Functions, BioMed Central Ltd. 2011, at https://behavioralandbrainfunctions.biomedcentral.com/articles/10.1186/1744-9081-7-18
16t3) Clancy et al., Extrapolating brain development from experimental species to humans, 2007, Neuro Toxicology, at http://people.psych.cornell.edu/~blf2/pdfs/BCBLFNeurotox07.pdf
16t4) Xu et al., Forced running enhances neurogenesis in the hippocampal dentate gyrus of adult rats and improves learning ability, Acta Physiologica Sinica, October 25, 2006, 58 (5): 415-420, at http://www.actaps.com.cn/qikan/manage/wenzhang/2006-5-03.pdf
16u) Ginsburg, The Importance of Play in Promoting Healthy Child Development and Maintaining Strong Parent-Child Bonds, Pediatrics, January 2007, VOLUME 119 / ISSUE 1, at http://pediatrics.aappublications.org/content/119/1/182
16v) Autism Speaks web page at https://www.autismspeaks.org/what-autism
16w) Dickinson et al., A Randomised Control Trial of the Impact of a Computer-Based Activity Programme upon the Fitness of Children with Autism, Volume 2014 (2014), Article ID 419653,
16y) CDC web page on autism at https://www.cdc.gov/ncbddd/autism/facts.html
16z) Booth et al., Lack of exercise is a major cause of chronic diseases, Compr Physiol., 2012 Apr, PMCID: PMC4241367
16z1) Sibley et al., The Relationship Between Physical Activity and Cognition in Children: A Meta-Analysis, Pediatric Exercise Science, 2003, 15, 243-256 at https://peandhealth.wikispaces.com/file/view/Sibley+and+Etnier+2003.pdf Page 244 especially.
16z2) Sullivan et al., Using Effect Size—or Why the /P/ Value Is Not Enough, J Grad Med Edu, v.4(3); 2012 Sep PMC3444174 at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3444174/
16z3) Autism Speaks web page at https://www.autismspeaks.org/what-autism/faq
16z4) Adolph and Berger. 2006. “Motor Development,” in Handbook of Child Psychology: Volume 2: Cognition, Perception, and Language (Sixth edition), John Wiley and Sons, p. 181
16z5) National Research Council and Institute of Medicine: From Neurons to Neighborhoods, the Science of Early Childhood Development, National Academy Press, 2000, Table 8-1, p. 199, at https://www.nap.edu/download/9824
16z6) Chaddock et al., Basal Ganglia Volume Is Associated with Aerobic Fitness in Preadolescent Children, Dev Neurosci., 2010 Aug; 32(3): 249–256,PMCID: PMC3696376 at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3696376/
16z7) Chaddock et al., A neuroimaging investigation of the association between aerobic fitness, hippocampal volume, and memory performance in preadolescent children, Brain Research, Vol. 1358, October 2010, at http://www.sciencedirect.com/science/article/pii/S0006899310018317
16z8) Hippocampus, at http://www.caam.rice.edu/~cox/wrap/hippocampus.pdf
16z9) Bednarczyk et al., Prolonged voluntary wheel-running stimulates neural precursors in the hippocampus and forebrain of adult CD1 mice, Hippocampus. 2009 Oct;19(10):913-27. doi: 10.1002/hipo.20621.at https://www.ncbi.nlm.nih.gov/pubmed/19405143.
16z10) Fedewa et al., The Effects of Physical Activity and Physical Fitness on Children's Achievement and Cognitive Outcomes, Research Quarterly for Exercise and Sport, Volume 82, 2011 - Issue 3, at http://www.tandfonline.com/doi/abs/10.1080/02701367.2011.10599785?src=recsys
16z11) Schultz, Developmental deficits in social perception in autism: the role of the amygdala and fusiform face area, International Journal of Developmental Neuroscience, 2005, Volume 23, Issues 2–3 at http://www.sciencedirect.com/science/article/pii/S073657480400156X
16z12) Case-Smith et al., A systematic review of sensory processing interventions for children with autism spectrum disorders, Autism 2015, Vol. 19(2) 133–148, at http://journals.sagepub.com/doi/pdf/10.1177/1362361313517762
16z13) Autism spectrum disorder symptoms among children enrolled in the Study to Explore Early Development (SEED), Autism Dev Disord, Oct 2015, PMID:26048040 PMCID:PMC4573234, at http://europepmc.org/articles/PMC4573234
16z14) Liu et al., [Environmental risk factors for autism spectrum disorders in children], Zhongguo Dang Dai Er Ke Za Zhi. <#> 2015 Nov;17(11):1147-53. abstract at https://www.ncbi.nlm.nih.gov/pubmed/26575869
17) Jensen, A.A., Slorach, S.A.: Chemical Contaminants in Human Milk, CRC Press, Inc., Boca Raton, Ann Arbor, Boston, 1991, p 15.
For a more recent study finding disproportionate ratios between organohalogens in breast milk versus those in cord tissue and cord serum, see Needham et al., Partition of Environmental Chemicals between Maternal and Fetal Blood and Tissues, Environ Sci Technol. Feb 1, 2011; 45(3): 1121-1126, at http://pubs.acs.org/doi/pdf/10.1021/es1019614, Table 2, finding weight-based concentrations of organohalogens to be over 30 times higher in human milk than in umbilical cord tissue.
17a) See http://www.pesticides-and-breastfeeding.info/index.htm#secB, Section B.1.
17b) Table 1 of Rogan et al., Polychlorinated Biphenyls (PCBs) and Dichlorodiphenyl Dichloroethene (DDE) in Human Milk: Effects of Maternal Factors and Previous Lactation, AJPH (Amer. Journal of Public Health) February 1986, Vol. 76, No. 2, at http://ajph.aphapublications.org/doi/pdf/10.2105/AJPH.76.2.172
17c) p. 175 of above Rogan study.
17d) Boucher et al., Association between breastfeeding duration and cognitive development, autistic traits and ADHD symptoms: a
multicenter study in Spain, Pediatric Research Volume 81 | Number 3 | March 2017, at http://www.nature.com/pr/journal/v81/n3/full/pr2016238a.html
17e) They acknowledged that "the association between duration of any breastfeeding and autistic traits fell short of statistical significance in supplemental analyses" which were intended to deal with missing values; also they found "no significant association between breastfeeding duration and the occurrence of scores within the clinical range."
17f) Roberts et al., Perinatal Air Pollutant Exposures and Autism Spectrum Disorder in the Children of Nurses’ Health Study II Participants, Children's Health Volume 121 | Issue 8 | August 2013, Environ Health Perspect; DOI:10.1289/ehp.1206187 at http://ehp.niehs.nih.gov/1206187/#r35
17g) Medical Daily, June 18, 2013, at http://www.medicaldaily.com/prenatal-exposure-air-pollution-may-raise-childs-autism-risk-women-heavily-diesel-polluted-areas
17h) Peter Szatmari, How girls and boys differ when it comes to autism, Special to CNN at http://www.cnn.com/2012/04/04/health/mental-health/autism-sex-differences/index.html
17j) Whitely et al., Gender Ratios in Autism, Asperger Syndrome and Autism Spectrum Disorder, January 2010 DOAJ at https://www.researchgate.net/publication/51018620_Gender_Ratios_in_Autism_Asperger_Syndrome_and_Autism_Spectrum_Disorder
17k) Werling, The role of sex-differential biology in risk for autism spectrum disorder, Biol Sex Differ., 2016; 7: 58 at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5112643/
18) National Academies Press, Hormonally Active Agents in the Environment (1999), Chapter: 6: Neurologic Effects, at http://www.nap.edu/read/6029/chapter/8#178, p. 178. Describing studies measuring maternal concentrations of developmental toxins in 313 women in Michigan, this publication states, “The mean concentrations of PCBs were 6 ng/mL in maternal serum, 3 ng/mL in cord serum, and 841 ng/g in breast milk.”(p. 178) From a German study (Winneke et al., 1998), “Mean concentrations of PCBs were 0.55 ng/mL in cord blood and 427 ng/g in the fat of breast milk.” (p. 183) (1 mL is about the same as 1 g when discussing a substance whose weight is about the same as that of water.)
18a) Kommission “Human-Biomonitoring” des Umweltbundesamtes: Stoffmonographie PCB - Referenzwerte für Blut, Section 8.3., found within https://www.umweltbundesamt.de/sites/default/files/medien/377/dokumente/pcbblut.pdf, website of Umwelt Bundes Amt (German Federal Environmental Office). The text drawn on says, " "Die derzeit durchschnittlich vom Erwachsenen täglich aufgenommene Menge an PCB (ca. 0,02 μg PCB/kg KG ) liegt deutlich unter der ATD von 1 μg PCB/kg KG. Der gestillte Säugling erhält dagegen eine deutlich höhere PCB-Zufuhr (3 μg PCB/kg KG.", which Bing Translator very respectably translates as " "The amount taken daily average currently by the adults of PCB (approx. 0.02 μg PCB/kg bw ) is well below the ATD of 1 μg PCB/kg. The breastfed infant, however, receives a significantly higher PCB intake (3 μg PCB/kg bw.)"
(18b) Birnbaum and Slezak, Dietary Exposure to PCBs and Dioxins in Children, Environmental Health Perspectives, Volume 107, Number 1, January 1999, at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1566291/pdf/envhper00506-0029.pdf "As observed in other studies, nursing infants consume a daily TEQ intake that is 50 times higher than adults."
(18c) Exposure of young infants to environmental tobacco smoke: breast-feeding among smoking mothers. Mascola,et al., Am J Public Health. 1998 June; 88(6): 893–896. PMCID: PMC1508233 found at www.ncbi.nlm.nih.gov/pmc/articles/PMC1508233
(18d) Schecter et al., Polybrominated Diphenyl Ether (PBDE) Levels in an Expanded Market Basket Survey of U.S. Food and Estimated PBDE Dietary Intake by Age and Sex, Environ Health Perspect. 2006 October; 114(10): 1515–1520. Published online 2006 July 13. doi: 10.1289/ehp.9121PMCID: PMC1626425 at https://www.ncbi.nlm.nih.gov/pubmed/17035135
18e) Iughetti et al., Childhood obesity and environmental pollutants: a dual relationship, Acta Biomed 2015; Vol. 86, N.1:5-16, at http://www.mattioli1885journals.com/index.php/actabiomedica/article/view/3593/3316
18f) Lackmann et al., Organochlorine compounds in breast-fed vs. bottle-fed infants: preliminary results at six weeks of age, Science of the Total Environment, Vol. 329, Aug. 2004 at http://www.sciencedirect.com/science/article/pii/S0048969704001925
18g) EPA web page on PBDEs and TSCA at https://www.epa.gov/assessing-and-managing-chemicals-under-tsca/polybrominated-diphenyl-ethers-pbdes
19) Drexler et al., The mercury concentration in breast milk resulting from amalgam fillings and dietary habits, Environ Res. 1998 May;77(2):124-9. at http://www.ncbi.nlm.nih.gov/pubmed/9600805
19a) Muldoon et al., Lifestyle and Sociodemographic Factors as Determinants of Blood Lead Levels in Elderly Women, Am. J. Epidemiol. I, 1994 139 (6), p. 599, Table 2, abstract at http://aje.oxfordjournals.org/content/139/6/599.abstract. The 11% reduction comes from dividing the average difference between the blood levels of women who had breastfed and the blood levels of women who had not breastfed (0.59 mcg -- see Table 2) by the 5.4 mcg average blood levels.
20) Exploration of Perinatal Pharmacokinetic Issues Contract No. 68-C-99-238, Task Order No. 13 Prepared for EPA by: Versar, Inc. EPA/630/R-01/004, Section 220.127.116.11, at www.epa.gov/raf/publications/pdfs/PPKFINAL.PDF According to these researchers contracted by the EPA, "a wealth of information" indicates that lactational transfer of maternal mercury during the first 15 days of lactation is equal to about a third of the total transfer of mercury that takes place during gestation.
-- Wigle, D.T., MD, PhD, MPH: Child Health and the Environment, Oxford University Press, 2003, Ch. 5. p. 106 (typically available through Ebsco Host at local libraries) Stated that the half-life of methylmercury in the blood of lactating women is about half that in nonlactating women, apparently due to excretion in breast milk.
-- Mercury Study Report to Congress c7o032-1-1, EPA Office of Air Quality Planning & Standards and Office of Research and Development Volume VII, Section 18.104.22.168, at http://www.epa.gov/ttn/oarpg/t3/reports/volume7.pdf According to this EPA article, “Lactating women have shorter biological half-lives for methylmercury (average value 42 days), compared with nonlactating women (average value 79 days) (Greenwood et al., 1978). This is presumably a reflection of excretion of mercury into milk.”
-- Vahter et al., Longitudinal Study of Methylmercury and Inorganic Mercury in Blood and Urine of Pregnant and Lactating Women, as Well as in Umbilical Cord Blood, Environmental Research, Section A 84, 186}194 (2000) at http://www.detoxmetals.com/content/FISH/FISH/Hg%20in%20pregnant%20urine%20and%20cord.pdf According to this 1999 Swedish study, “there was a marked decrease in I-Hg (inorganic mercury) in (the mothers’) blood and urine during lactation, most likely related to the excretion of I-Hg in milk…. About 10% of the Hg (mercury) present in circulating blood (5 L]0.3 lg/L) would be transferred to the milk every day.” (Obviously, the mother also keeps taking in mercury.)
-- U.S. Hazardous Substances Data Bank of the National Library of Medicine's TOXNET system, at http://toxnet.nlm.nih.gov. Evidence from the Iraqi poisoning incident showed that lactation decreased blood mercury clearance half-times in women by 44%, indicating rapid excretion of mercury in breast milk.
-- P. Grandjean et al., Human Milk as a Source of Methylmercury Exposure in Infants, Environmental Health Perspectives, accepted Oct. 1993 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1567218/pdf. In this study by a prominent scientist (P. Grandjean) and his team, it was found that total mercury concentrations in infants that had been breastfed for one year were three times as high as those in infants that had not been breastfed.
-- Marques RC, et al., Hair mercury in breast-fed infants exposed to thimerosal-preserved vaccines. Eur J Pediatr. 2007 Sep;166(9):935-41. Epub 2007 Jan 20 This study found that mercury measured in infants’ hair increased 446% during the first six months of breastfeeding, while mercury measured in the mothers’ hair decreased 57%. These measurements included mercury from vaccines (still containing mercury at that time in Brazil, where the study was carried out), which the authors estimated accounted for about 40% of the infants’ exposure during those six months. Given that, combined with the finding in a Taiwanese study that over 95% of an infant’s exposure to mercury was from breastfeeding (Chien et al., 2006), the increase in the infants’ mercury levels attributable to breastfeeding was probably well over 200% during the first 6 months of breastfeeding.
-- Rice et al., Environmental Mercury and Its Toxic Effects, J Prev Med Public Health. 2014 Mar; 47(2): 74–83, doi: PMCID: PMC3988285 at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3988285 According to these experts, the average whole body biological half-life of inhaled mercury is approximately 60 days, but it is estimated that the half-life of mercury in the brain can be as long as 20 years. The above findings of doubling and tripling of mercury levels in breastfed infants would have been based on levels in the whole body, where half-lives are relatively short and accumulation relatively minor, as opposed to levels in the especially vulnerable developing brain, where accumulation would be far greater.
20a) Hirokadzu et al., Transfer of Polychlorinated Biphenyls to Infants from their Mothers, Archives of Environmental Health: An International Journal, Volume 35, 1980 - Issue 2 at http://www.tandfonline.com/doi/pdf/10.1080/00039896.1980.10667472?needAccess=true
20b) La Leche League International, Betty Crase: Pesticides and Breastfeeding, From: LEAVEN, Vol. 30 No. 3, May-June 1994, pp. 37-40, at http://www.llli.org/llleaderweb/lv/lvmayjun94p37.html
20c) Iszatt et al., Prenatal and Postnatal Exposure to Persistent Organic Pollutants and Infant Growth: A Pooled Analysis of Seven European Birth Cohorts, Environ Health Perspect v.123(7); 2015 Jul at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4492262/
21) Virgintino et al., Fetal Blood-Brain Barrier P-Glycoprotein Contributes to Brain Protection During Human Development, Journal of Neuropathology and Experimental Neurology, DOI: http://dx.doi.org/10.1097/nen.0b013e31815f65d9 50-61 First published online: 1 January 2008 at http://jnen.oxfordjournals.org/content/67/1/50.long
21a) Wang et al., Serum Concentrations of Selected Persistent Organic Pollutants in a Sample of Pregnant Females and Changes in Their Concentrations during Gestation, Environ Health Perspect. 2009 Aug; 117(8): 1244–1249, at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2721868/
22) P Grandjean and AA Jensen, Breastfeeding and the Weanling’s Dilemma Am J Public Health. 2004 July; 94(7): 1075. PMCID: PMC1448391 at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1448391/
-- Also, according to Encyclopedia Britannica, use of insecticides was considered to be responsible for a dramatic increase in agricultural productivity between 1945 and 1965. (https://www.britannica.com/technology/insecticide)
23) Oregon Department of Environmental Quality Environmental Cleanup Program, Oct. 2010, 10-LQ-023, p. D2-4 (attachment 2 of Appendix D, near very end) at
See relevant section below.
23a) Bell et al., Two-hit exposure to polychlorinated biphenyls at gestational and juvenile life stages: Sex-specific neuromolecular effects in the brain, Molecular and Cellular Endocrinology, 20 (2016) 125, at http://www.sciencedirect.com/science/article/pii/S0303720715301489
23b) U.S. ATSDR, Persistent chemicals found in breast milk, Appendix A, p. 180, at https://www.atsdr.cdc.gov/interactionprofiles/ip-breastmilk/ip03-a.pdf
23c) Hopf et al., Background levels of polychlorinated biphenyls in the U.S. population, Sci Total Environ. <#> 2009 Dec 1;407(24):6109-19. doi: 10.1016/j.scitotenv.2009.08.035. Epub 2009 Sep 20. at https://www.ncbi.nlm.nih.gov/pubmed/19773016
23d) DO Carpenter, Polychlorinated biphenyls and human health, in International Journal of Occupational Medicine and Environmental Health 11(4):291-303, February 1998, (referring to Rogan study) at https://www.researchgate.net/publication/13261771_Polychlorinated_biphenyls_and_human_health
23e) Rogan et al., Pollutants in Breast Milk, New England Journal of Medicine, Vol. 302, No. 26, 1980, Table 3.
23f) UNEP: Results of the global survey on concentrations in human milk of persistent organic pollutants by the United Nations Environment Programme and the World Health Organization, Geneva, 2013, at www.pops.int/TheConvention/ConferenceoftheParties/Meetings/COP6/tabid/3074/cti/Download/mid/9701/Default.aspx?id-85&ObjID=16190 At that page, scroll down about 30 items to the title of this document. See p. 19 in that document.
24) Washington State Department of Ecology, Multiyear PBT Chemical Action Plan Schedule, 2007, at https://fortress.wa.gov/ecy/publications/documents/0707016.pdf, p. 64.
24a) Birnbaum and Slezak, Dietary Exposure to PCBs and Dioxins in Children, Environmental Health Perspectives * Volume 107, Number 1, January 1999, at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1566291/pdf/envhper00506-0029.pdf
24b) American Academy of Pediatrics: Pediatric Environmental Health, 3rd Edition, 2012, p. 200.
24c) Aguilar, Bioaccumulation of polychlorinated biphenyls (pcbs) and dichlorodiphenylethane (dde) methyl sulfones in tissues of seal and dolphin morbillivirus epizootic victims, Journal of Toxicology and Environmental Health, at https://www.academia.edu/10778961/BIOACCUMULATION_OF_POLYCHLORINATED_
24d) Project TENDR: Targeting Environmental Neuro-Developmental Risks The TENDR Consensus Statement, Environ Health Perspect. 2016 Jul; 124(7): A118–A122. at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4937840/
25) Jens Walkowiak et al., Environmental exposure to polychlorinated biphenyls and quality of the home environment: effects on psychodevelopment in early childhood. Lancet 2001: 358: 1602-07 Abstract at www.thelancet.com/journals/lancet/article/PIIS0140-6736(01)06654-5/abstract
25a) Environ Health Perspect. 2012 Jul; 120(7): 944–951, Published online 2012 Apr 25. doi: 10.1289/ehp.1104553, Review: Tipping the Balance of Autism Risk: Potential Mechanisms Linking Pesticides and Autism, at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3404662/
25b) Industrial Health 2000, 38, 259–268 Review Article: The Effects of Dioxin on Reproduction and Development Junzo Yonemoto National Institute for Environmental Studies, Japan p. 262; at
25c) Koopman-Esseboom C. Effects of perinatal exposure to PCBs and dioxins on early human development. Erasmus University, Rotterdam, 1995
25d) p. 381 of U.S. ATSDR: Toxicological Profile for Polychlorinated Biphenyls (PCBs), 2000 at http://www.atsdr.cdc.gov/toxprofiles/tp17.pdf
25e) Pp. 233 and 238 of EPA: America's Children and the Environment, at https://www.epa.gov/sites/production/files/2015-06/documents/ace3_2013.pdf
25f) EPA: Learn about Polychlorinated Biphenyls (PCBs), at https://www.epa.gov/pcbs/learn-about-polychlorinated-biphenyls-pcbs#healtheffects
26) Stewart et al., The Relationship between Prenatal PCB Exposure and Intelligence (IQ) in 9-Year-Old Children, Environ Health Perspect. 2008 Oct; 116(10): 1416–1422, PMCID: PMC2569105 at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2569105
27) EPA: Biomonitoring, Polychlorinated biphenyls (PCBs), at https://www.epa.gov/sites/production/files/2015-05/documents/biomonitoring-pcbs.pdf For other details and sources, see Section 1.b.f of www.breastfeeding-vs-formula.info
27a) Verner et al., Measured Prenatal and Estimated Postnatal Levels of Polychlorinated, Biphenyls (PCBs) and ADHD-Related Behaviors in 8-Year-Old Children, Environ Health Perspec, Sept. 2015, at http://ehp.niehs.nih.gov/1408084/
27b) Frazier B; Environ Sci Technol 44: 2753-4 (2010)], as reported on an NIH ToxNet web page for Polychlorinated, Biphenyls at https://toxnet.nlm.nih.gov/cgi-bin/sis/search2/f?./temp/~ArMiU4:1, under the headings "Other Environmental Concentrations" and "Artificial Pollution Sources."
See also State of Washington Dept. of Ecology: Polychlorinated Biphenyls (PCBs) in General Consumer Products, at https://fortress.wa.gov/ecy/publications/documents/1404035.pdf
28) Verner et al., Measured Prenatal and Estimated Postnatal Levels of Polychlorinated Biphenyls (PCBs) and ADHD-Related Behaviors in 8-Year-Old Children, Figure 2, Environ Health Perspect; DOI:10.1289/ehp.1408084 , Vol. 123, Issue 9, Sept. 2015, at http://ehp.niehs.nih.gov/1408084
29) Quinn et al., Investigating Intergenerational Differences in Human PCB Exposure due to Variable Emissions and Reproductive Behaviors, Environ Health Perspect. May 2011; 119(5): 641–646. at www.ncbi.nlm.nih.gov/pmc/articles/PMC3094414
--Jacobson et al., Determinants of polychlorinated biphenyls (PCBs), polybrominated biphenyls (PBBs), and dichlorodiphenyl trichloroethane (DDT) levels in the sera of young children, Am J Public Health. 1989 October; 79(10): 1401–1404
-- Table 1 in Jusko et al., Prenatal and Postnatal Serum PCB Concentrations and Cochlear Function in Children at 45 Months of Age, Environmental Health Perspectives, 22 July 2014 (Advance Pub.) at http://ehp.niehs.nih.gov/wp-content/uploads/advpub/2014/7/ehp.1307473.pdf
-- Danish Health and Medicines Authority, Health risks of PCB in the indoor climate in Denmark, 2013, at http://sundhedsstyrelsen.dk/~/media/D290AF38C2114775804F1B6BDD6841C6.ashx
-- Ayotte et al., Assessment of Pre- and Postnatal Exposure to Polychlorinated Biphenyls:
Lessons from the Inuit Cohort Study, Environmental Health Perspectives • Volume 111 | Number 9 | July 2003, at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1241583/pdf/ehp0111-001253.pdf (finding 6.6-fold increase in infant PCB levels with over three months of breastfeeding, compared with no breastfeeding -- see Table 4)
-- Trnovec et al., Assessment of exposure to PCB 153 from breast feeding and normal food intake in individual children using a system approach model, Chemosphere, Dec. 2011, at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3228605/
-- Figure 6 of Crofton et al., PCBs, Thyroid Hormones, and Ototoxicity in Rats: Cross-Fostering Experiments Demonstrate the Impact of Postnatal Lactation Exposure, Toxicological Sciences, Volume 57, Issue 1, September 2000 at https://academic.oup.com/toxsci/article/57/1/131/1654970/PCBs-Thyroid-Hormones-and-Ototoxicity-in-Rats
30) Rogan et al., Polychlorinated Biphenyls (PCBs) and Dichlorodiphenyl Dichloroethene (DDE) in Human Milk: Effects of Maternal Factors and Previous Lactation, American Journal of Public Health, A1JPH February 1986, Vol. 76, No. 2, at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1646471/pdf/amjph00265-0062.pdf
30a) Jusko et al., Prenatal and Postnatal Serum PCB Concentrations and Cochlear Function in Children at 45 Months of Age, Environmental Health Perspectives, 22 July 2014 (Advance Pub.) at http://ehp.niehs.nih.gov/wp-content/uploads/advpub/2014/7/ehp.1307473.pdf
30c) Jacobson et al., Determinants of Polychlorinated Biphenyls (PCBs), Polybrominated Biphenyls (PBBs), and Dichlorodiphenyl Trichloroethane (DDT) Levels in the Sera of Young Children, AJPH Oct. 1989, Vol 79 No. 10, at
31) Kenet et al., Perinatal exposure to a noncoplanar polychlorinated biphenyl alters tonotopy, receptive fields, and plasticity in rat primary auditory cortex, 2007, The National Academy of Sciences of the USA, 7646–7651, PNAS, May 1, 2007, vol. 104, no. 18, at http://www.pnas.org/content/104/18/7646.full.pdf at http://www.pnas.org/content/104/18/7646.full.pdf
Re Dr. Merzenich: see http://www.brainhq.com/world-class-science/science-team/dr-michael-merzenich
32) Merzenich, What underlies the documented increase in autism incidence? Results of a new study, from On the Brain, by Dr. Michael Merzenich, 26 April 2007, at http://www.onthebrain.com/2007/04/underlies-documented-increase-autism-incidence-results-new-study
Also: Breastfeeding, Brain Development and Chemical Poisons: Neuroscientist Michael Merzenich, by Jeff Miller on May 18, 2007, at https://www.ucsf.edu/news/2007/05/3817/merzenich: "...the levels of PCBs to which the laboratory rats were exposed proportionately equal the levels now being found in the breast milk of women all over the country."
Also see Appendix Y for excerpts of transcript of interview with Dr. Merzenich.
32a) Science News/Autism Speaks, 20 April 2016, Meeting highlights from the Interagency Autism Coordinating Committee; summary of presentation by Dr. Paul Lipkin and Kiely Law of Interactive Autism Network, of preliminary findings of a study; www.autismspeaks.com
34) National Scientific Council on the Developing Child, Science Briefs: Prenatal and Infant Exposure to an Environmental Pollutant Damages Brain Architecture and Plasticity (2007). at http://www.policyarchive.org/handle/10207/20626
35) University of California at San Francisco: Breastfeeding, Brain Development and Chemical Poisons: Neuroscientist Michael Merzenich, By Jeff Miller, May 18, 2007 at https://www.ucsf.edu/news/2007/05/3817/merzenich See Appendix Y for excerpts of transcript of interview with Dr. Merzenich.
36) WHO, Persistent Organic Pollutants: Impact on Child Health, p. 6, at http://whqlibdoc.who.int/publications/2010/9789241501101_eng.pdf
36a) Collaborative on Health and the Environment’s Learning and Developmental Disabilities Initiative: Scientific Consensus Statement on Environmental Agents Associated with Neurodevelopmental Disorders, 2008, at https://www.healthandenvironment.org/docs/xaruploads/LDDIStatement.pdf
37) Class of PCBs causes developmental abnormalities in rat pups, UCSF News Center, Univ. of California San Francisco, by Jennifer O’Brien, April 23, 2007, at http://www.ucsf.edu/news/2007/04/5564/class-pcbs-causes-developmental-abnormalities-rat-pupsat
37a) CDC web page, Breastfeeding: Exposure to Environmental Toxins, at http://www.cdc.gov/breastfeeding/disease/environmental_toxins.htm
37b) AAP Policy Statement, Pediatrics, Mar. 2012, Vol. 129, Issue 3, Breastfeeding and the Use of Human Milk, at http://pediatrics.aappublications.org/content/129/3/e827.full#content-block
37c) Bennetto et al., Children with autism spectrum disorder have reduced otoacoustic emissions at the 1 kHz mid-frequency region, Autism Research, at http://onlinelibrary.wiley.com/doi/10.1002/aur.1663/abstract. Also an article in Medline Plus, Aug. 1, 2016, R. Preidt,"Hearing Test May Predict Autism Risk Sooner: Study" at https://medlineplus.gov/news/fullstory_160181.html
37d) Jusko et al., Prenatal and Postnatal Serum PCB Concentrations and Cochlear Function in Children at 45 Months of Age, Environ Health Perspect, 2014, at http://ehp.niehs.nih.gov/wp-content/uploads/advpub/2014/7/ehp.1307473.pdf
37e) GW Chance, Environmental contaminants and children’s health: Cause for concern, time for action, Paediatr Child Health. 2001 Dec; 6(10): 731–743. at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2805986/
Also see: State of Washington Department of Ecology, Multiyear PBT Chemical Action Plan Schedule, at https://fortress.wa.gov/ecy/publications/documents/0707016.pdf, p. 62
Also see: Eriksson et al., Polybrominated Diphenyl Ethers, A Group of Brominated Flame Retardants, Can Interact with Polychlorinated Biphenyls in Enhancing Developmental Neurobehavioral Defects, Toxicological Sciences, Volume 94, Issue 2, December 2006, at https://academic.oup.com/toxsci/article/94/2/302/1647585/Polybrominated-Diphenyl-Ethers-A-Group-of
37e1) Rossignol et al., Environmental toxicants and autism spectrum disorders: a systematic review, Transl Psychiatry, 2014 Feb; at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3944636/
37e4) Landrigan et al., A research strategy to discover the environmental causes of autism and neurodevelopmental disabilities, Environ. Health Perspect., 120 (2012), pp. 258-260