1. Prevalence Rates and Home Schooling.

I received a thoughtful email about the impact of home schooling on the CDC prevalence rate and autism research in general, given that many children with ASDs may be home schooled. Here is my response:

Regarding the CDC:
The prevalence was obtained from health records and, in some States, also educational records. States that used educational records had higher prevalence rates, and those records only included public school records. So theoretically, the prevalence would be even higher once home/private school cases are added. While education records may have included some children in private/home schools (many children in home school still receive special education services in some States and would therefore be identified by the CDC teams), many cases are likely being missed.

Interestingly however, the new CDC numbers are in line with the national autism prevalence study published in Pediatrics. This study was not based on educational or health records reviews, but instead it was based on detailed phone screenings of a representative sample of US families. Both of these studies however, would miss some children with ASD that are undiagnosed (and maybe home schooled) due to limited contact with health workers (pediatricians, etc). These children would not have any records showing that they have ASD symptoms and these parents would also respond ‘no’ to the basic phone screening question “have your child ever been diagnosed with an autism spectrum disorder?”

Regarding Research in General:
Fortunately, most research on autism is not conducted via the school systems. Most research is conducted at medical and university centers with families recruited from the community. In my neuropsychology assessment experience, I would say that at least 30% of the ASD kids we see are home schooled, and many of these children are active participants in our research programs. So the news is a bit better for general research, in that it is unlikely that home schooled kids are underrepresented in those studies.

2. Vaccines.

I really dislike writing anything about vaccines, mostly because regardless of how factual I aimed to be, any mention of vaccines is usually followed by a dozen of  ’friendly’ emails. But I’ve received several emails asking how the CDC numbers affect the vaccine theory. The CDC study does not address this issue at all, and the data say little about this theory. However, some reasonable conclusions can be made.

- If the increases in diagnoses among 8 year olds from 2002 to 2006 are due to real increases in true prevalence

and

- If vaccines play a role in the incidence of autism

- Then a 50% increase in the prevalence during the 4 year period should be accompanied by a noticeable change in vaccination practices during key years.

Specifically, the 2002 CDC  study was based on children born in 1994 and the new CDC study was done with children born in 1998.  Thus, given the striking increases in prevalence rates among the 1998 children, you would expect that compared to those born in 1994, children born in 1998 received higher vaccination dosages, received more harmful dosages, or simply were vaccinated at a higher rate. I have some data on vaccination rates:

I took a look at the CDC vaccination rates for MMR for those born in 1994 and 1998 by the time they were 2 years of age. You can take a look at the data here. The National vaccination rate for MMR for those born in 1994 was 90%. For those born in 1998, the vaccination rate was also 90%. For the states included in the CDC autism study, the vaccination rate for those born in 1994 was 90% and for those born in 1998 was also 90%. At the State and National level, there were no changes in vaccination rates for kids born in 1994 and 1998 that could help explain the 50% jump in autism prevalence.

12/28/09 UPDATE: Please note that in the paragraphs above I presented a simple logical argument for the vaccines debate. If vaccines played a role in the 1994 to 1998 autism rate change, then there must be a change in vaccination practices between 1994 and 1998-2000. Potential changes may have involved higher vaccination rates, changes in vaccine cocktails or contents, changes in schedules, etc etc. I then provided data for vaccination rates for one single vaccine as an example: MMR. Clearly such data are very limited and does not cover all possible changes that may have taken place during that time.

3. What’s up with Missouri?

Missouri had the highest autism rates of all states assessed (albeit it was a tie with Arizona in many measures), with rates that were often more than twice that of other States. One of my readers asked whether this was due to demographic differences in the target counties in Missouri. For example, is it possible that the data from Missouri came mostly from urban St Louis with a higher proportion of ethnic minorities or lower SES families? The data does not seem to support this theory. The Missouri sample was close to 70% white, and other States with significantly higher % of ethnic minorities in urban settings had significantly lower autism rates (e.g., Colorado – all from metro Denver with only 55% white; Florida – all from Miami with only 23% white; Georgia – all from metro Atlanta with only 38% white). We do not know why the high prevalence of autism in Missouri and Arizona, but it is very unlikely that it is due to demographic differences between these States and the other States included in the study.

Thank you for subscribing to the RSS feed of Child-Psych.org. Please visit our website to join the conversation. ©2010 Child Psychology Research Blog. All Rights Reserved.

While reading the same issue of the journal Pediatrics, I came across a related study that examined the factors that predict whether a mother intended to vaccinate her daughter against HPV. I was very interested in this study because it could shed light into the decision processes behind parental refusal to vaccinate. That is, is the decision to refuse vaccination only a function of weighting risks vs. benefits? Or is such a decision also associated with other factors that may not be related to the issue of vaccine safety at all?

As background, HVP is a sexually transmitted infection that is associated with a number of serious conditions including cervical cancer. A vaccine for HPV was approved by the FDA in 2006, which has the potential to significantly decrease the prevalence of HPV-related conditions. In this new study, the authors examined 7,207 nurses who had at least 1 daughter participating in a large longitudinal study of adolescence. The authors were interested in examining what demographic and attitudinal variables were associated with intention to vaccinate their daughters against HVP.

The results:

1. Intention to vaccinate was associated with the age of the daughter. 48% of mothers of girls between 9 to 12 intended to vaccinate, compared to 86% of mothers of teens between 16 to 18 years of age.
2. Income was highly associated with intention to vaccinate with those making over $40,000 being 2 times more likely to report an intention to vaccinate their daughters than those making less than 40K
3. Having a history of HPV or HPV related disease doubled the changes of intention to vaccinate.

A number of beliefs about vaccines also predicted whether the mothers intended to vaccinate the daughters. A few interesting findings:

4. Mothers who believed that vaccinated girls would practice riskier sex were 38 times less likely to report an intention to vaccinate their daughters than mothers who did not have this belief.
5. Mothers who believed their daughters were at risk of HPV were 77 times more likely to intend to vaccinate than other mothers
6. Mothers who believed that the vaccine was the best protection against cervical cancer were 234 times more likely to intend to vaccinate their daughters than mothers who did not have this belief.

Initially these results were not that surprising. It seems clear that intention to vaccinate was highly associated with specific beliefs mothers have about the effects, nature, and potential consequences of the vaccine. What I found most interesting is that these results were based on a very unique and not highly representative sample. That is, these mothers were all nurses, and theoretically, they would be better educated than the general population about vaccines. Thus, it was very surprising that even among these highly educated mothers, the decision to vaccinate their daughters was affected by factors not necessarily directly associated with “vaccine safety” but with other ‘values’ or “beliefs” factors, such as the belief that if they vaccinated their daughters, the teens would be more likely to have risky sexual relations.

Kahn, J., Ding, L., Huang, B., Zimet, G., Rosenthal, S., & Frazier, A. (2009). Mothers’ Intention for Their Daughters and Themselves to Receive the Human Papillomavirus Vaccine: A National Study of Nurses PEDIATRICS, 123 (6), 1439-1445 DOI: 10.1542/peds.2008-1536

Thank you for subscribing to the RSS feed of Child-Psych.org. Please visit our website to join the conversation. ©2010 Child Psychology Research Blog. All Rights Reserved.

The journal of the American Academy of Pediatrics recently published a study that examined the association between vaccination refusal and pertussis infections in Colorado. In this study, the authors conducted a case-control examination of children between 2 months and 18 years who were members of a large health plan (Kaiser Permanente Colorado). The study included 156 cases of pertussis and 595 control non-pertussis children.

The results:

1. The authors found time trend in the cases of pertussis, with a large increase in cases of pertussis after the year 2000.
2. Of the 156 children who acquired pertussis, 12% were children of parents who refused vaccination.
3. In contrast, only 0.5% of the children who did not get pertussis were children of parents who refused vaccination.
4. Among children of all ages, refusing vaccination increased the risk of pertussis infection by 2,280%!!!

And these numbers may actually be an underestimate of the real risk of vaccine refusal, since the authors found a possible bias in diagnostic practices. Specifically, parents who agreed to vaccinate their children were 2 times more likely than vaccine-refusing parents to visit the doctor for upper respiratory infections. One interpretation of this finding is that parents who accepted the vaccination are more likely to seek medical services when their kids are sick than parents who refused vaccination. If this is the case, it is likely that cases of pertussis in the unvaccinated kids may have been missed (as these kids were less likely to visit the doctor) so that the rate of pertussis among unvaccinated kids is actually higher than what was observed in this study.

In sum, this study indicated that children of parents who refused vaccination were over-represented among cases of pertussis in Colorado during the past decade and that vaccination refusal increased the risk of pertussis by more than 2000%

These findings made me think about the underlying reasons that drive some parents to refuse vaccination: the fear that by vaccinating their kids they are increasing their kids’ risk of having a severe side effect. But I’m wondering if these parents are properly weighting the risks. Specifically, I wonder how many of these parents have reliable information about both sides of the equation; so that the real and known risks of vaccination are properly weighted against the real and known risks of non-vaccination (such as a 2,000% increase in the risk of acquiring pertussis)?

Glanz, J., McClure, D., Magid, D., Daley, M., France, E., Salmon, D., & Hambidge, S. (2009). Parental Refusal of Pertussis Vaccination Is Associated With an Increased Risk of Pertussis Infection in Children PEDIATRICS, 123 (6), 1446-1451 DOI: 10.1542/peds.2008-2150

Thank you for subscribing to the RSS feed of Child-Psych.org. Please visit our website to join the conversation. ©2010 Child Psychology Research Blog. All Rights Reserved.

Infection with CMV is major cause of auditory and cognitive impairment in newborns. CMV is actually very common, but when acquired during pregnancy the risk to the developing fetus is very high. Approximately 27,000 CMV infections among pregnant women occur in the United States each year

This study included 464 females between the ages of 14 to 40. The authors randomly assigned 234 subjects to receive the CMV vaccine and 230 subjects to receive placebo. 14% of the females receiving the placebo got CMV infection within one year compared to only 8% of the females receiving the vaccine. 97 infants were born to the mothers in the placebo group and 3 of these infants were infected with CMV (3.1%). In contrast, 81 infants were born to the mothers receiving the vaccine and only 1 infant became infected (1.2%). Surprisingly, the group receiving the vaccines had significantly lower rate of pregnancy than the group receiving the placebo.

Regarding harmful effects of the vaccine to the infants, serious effects were observed in 7 of mothers in the vaccine group and in 8 newborns of mothers in the placebo group. Thus, the vaccine did not lead to more serious events than the placebo. No long-term health effects of the vaccine were reported but the results are provided for events occurring up to 48 months after birth, so the long-term effects are currently unknown.

The study was financed by The National Institute of Allergy and Infectious Diseases, with the support of Sanofi Pasteur and Chiron (now Novartis).

Pass, R., Zhang, C., Evans, A., Simpson, T., Andrews, W., Huang, M., Corey, L., Hill, J., Davis, E., Flanigan, C., & Cloud, G. (2009). Vaccine Prevention of Maternal Cytomegalovirus Infection New England Journal of Medicine, 360 (12), 1191-1199 DOI: 10.1056/NEJMoa0804749

Thank you for subscribing to the RSS feed of Child-Psych.org. Please visit our website to join the conversation. ©2010 Child Psychology Research Blog. All Rights Reserved.

One bio-social model of the etiology of autism suggests that a genetic predisposition, in the form of limited ability to remove toxins from the body (e.g., mercury), combined with exposure to such toxins, leads to an increased risk for developing autism. As I previously described in this study, researchers have examined urinary porphyrins as a measure of mercury exposure, and some studies have found increased levels of porphyrins in children with autism. In this study, the authors examined urinary porphyrin metabolites (a proposed measure of heavy metal toxicity) and plasma sulfates (a proposed measure of detoxification capacity) among children with autism.

The sample included 28 children between the ages of 2 and 16 recruited from the Dallas/Fort Worth community. These children had a diagnosis of autism (Childhood Autism Rating Scale scores above 30). This group was categorized into a mild (CARS
38.5) autism. Blood and urine samples were collected for analysis. A laboratory standard control sample of typically developing children was also included in the plasma analysis only.

The mild ASD group had significantly lower ratios of key porphyrin levels when compared with kids with severe ASD. The following group means are statistically significantly different from each other:

Porphyrin
Pentacarboxyporphyrin
MILD ASD = 4.92±2.79 VS. SEVERE ASD = 6.0±1.58

Precoproporphyrin
MILD ASD = 17.3±8.9 VS. SEVERE ASD = 24.05±7.28

Ratios:
Pentacarboxyporphyrin/uroporphyrins I and III
MILD ASD = 0.19±0.06 VS. SEVERE ASD = 0.27±0.08

Precoprophyrin/uroporphyrins I and III
MILD ASD = 0.68±0.28 VS. SEVERE ASD = 1.09±0.36

Coproporphyrins I and III/ uroporphyrins I & and III
MILD ASD = 9.54±2.97 VS. SEVERE ASD = 12.64±3.47

(Precoproporphyrin+ pentacarboxyporphyrin) / (uroporphyrins I & and III+ heptacarboxyporphryin)
MILD ASD = 0.73±0.26 VS. SEVERE ASD = 1.1±0.29

The authors concluded that increased urinary porphyrin metabolites were associated with more severe autism based on CARS scores. Urinary porphyrin are believed to represent mercury toxicity.

Furthermore, the authors found significantly lower levels of plasma sulfates (Plasma cysteine, Plasma reduced glutathione, Plasma oxidized glutathione) among children with autism when compared to typically developing kids. Decreased sulfation (low plasma levels of sulfates) can indicate limited detoxification capacity, especially of heavy metals such as mercury.

In conclusion, the study presents some compelling evidence regarding differences in urinary porphyrin metabolites between kids with mild and severe autism symptoms, as well as differences in plasma sulfates between kids with autism when compared to neurotypical kids. This is consistent with the author’s theory that mercury exposure plays a role in autism. However, given the controversy regarding the mercury-autism association, I would like to clarify that these results do not indicate that mercury causes autism. Yes, it is possible that mercury toxicity and reduced detoxification capacity plays a role in autism, both, in the onset and severity. However, it is also possible that Autism itself results in reduced detoxification capacity and mercury toxicity via other mechanisms, but that such toxicity is not associated at all with the development of the disorder. For example, erythrocyte sedimentation rate and C-reactive protein are tests used in arthritis. Atypical results on these tests are found in people with arthritis. Yet, neither one is related to any of the proposed causes of arthritis, and instead reflect a consequence of the disorder itself (inflammation).

Thank you for subscribing to the RSS feed of Child-Psych.org. Please visit our website to join the conversation. ©2010 Child Psychology Research Blog. All Rights Reserved.

The journal of Toxicology and Environmental Health just published a study conducted in Australia examining the possible link between mercury and autism. The study examined urinary porphyrins as a measure of mercury exposure in children with autism. Porphyrinuria, or the excess urinary excretion of porphyrin, is purported to reflect heavy metal exposure and in particularly mercury. Two previous studies (Nataf et al., 2006, and Geier & Geier, 2007) used this urinary measure and reported increased levels of porphyrins in children with autism when compared to typically developing children. In the current study the authors examined urinary samples of 41 patients with ASD (detailed diagnostic procedure information was not provided). The age of the sample ranged from 1 to 16 (average 6). There were 30 boys and 11 girls. The authors did not include a control group. That is, no local comparison group of typically developing children was used. Instead, the authors compared the levels of porphyrins of the Australian sample with the control samples (typically developing kids) used by the two previous studies as well as to ‘normative’ laboratory ranges obtained from a French laboratory and the normative ranges published in a 1996 European study (Minder and Schneider-Yin, 1996).

The authors reported that the ratio of uroporphyrin to coproporphyrin (CP to UP) was statistically significantly higher in the Australian ASD group when compared to all control samples of the previous studies. The comparison effect sizes were very large, namely: 1.60 when compared to the Geier & Geier sample, 1.54 when compared to the Nataf et al sample, and 1.46 when compared to the 1996 normative sample. Effect size is a measure of the differences between the means of two groups that is not as sensitive (not affected by) the very large differences in sample sizes between these studies. Effect sizes in this range (1.46 to 1.60) indicate that the differences between the groups were very substantial.

A final thought. I can not discuss the merits of this particular urinary measure as reflective of mercury exposure or mercury toxicity, since this is beyond my understanding of heavy metal metabolism, thus I would simple make one final comment regarding the methodology of this study. I am very surprised that the authors did not include a local comparison sample, especially since their intention was to replicate the previous findings within a local Australian group. This is a strong limitation of the study. The results indicate that the Australian ASD group is statistically significantly different than the typically developing group used in the two previous samples (one from the USA and another from France) in regards to CP to UP ratio, but the results provide no information as how this ASD group compares to typically developing children in the sample geographical regions.

However, the data is compelling in showing a strong difference in the CP to UP ratio between the Australian sample and typically developing children in the USA and France. Does this mean that mercury causes autism? Not at all. Readers should be careful not to make conclusions about causal mechanisms from these type of association studies. There are multiple possible interpretations of the data, including the possibility that heavy metal exposure may play a role in the development of ASD. However, it is also possible that CP/UP ratio differences between these groups are the result, rather than the cause, of physiological differences in Autism. The results of this study do not support one interpretation more than the other. The results simply indicate that the groups are different in CP to UP ratio. The data do not tell us why these groups are different, or whether mercury causes autism.

Geier, D. A., and Geier, M. R. 2007. A prospective study of mercury toxicity
biomarkers in autistic spectrum disorders. J. Toxicol. Environ. Health,
A 70:1723–1730.

Minder, E. I., and Schneider-Yin, X. 1996. Age-dependent reference values
of urinary porphyrins in children. Eur. J. Clin. Chem. Clin. Biochem.
34:439–443.

Nataf, R., Skorupka, C., Amet, L., Lam, A., Springbett, A., and Lathe, R. 2006.
Porphyrinuria in childhood autistic disorder: Implications for environmental
toxicity. Toxicol. Appl. Pharmacol. 214:99–108.

Thank you for subscribing to the RSS feed of Child-Psych.org. Please visit our website to join the conversation. ©2010 Child Psychology Research Blog. All Rights Reserved.

Although most children with autism present very early signs and symptoms and a linear developmental trajectory, a small subset of children present a trajectory characterized by normal development followed by a loss of acquired skills or a failure to use the acquired skills. This pattern has been termed autistic regression. Possible explainations for this phenomenom have varied from a genetic effect on brain restructuring and pruning during the early stages of life, to enterocolitis due to vaccinations, to epilepsy. In this study, the authors explored differences in developmental outcomes for children with and without regressive autism, and the association between regression and enterocolitis and epilepsy. This study examined a population cohort born in the UK in 1990 and 1991. Out of 56,946 children in this cohort, 218 had and ASD diagnosis by age 10. A subset of these children were evaluated via ADOS and ADI and divided into a broad autism (N=105), narrow autism (N=53), and no autism (N=97). The narrow autism group met full criteria for autism based on ICD-10. The broad autism group met clinical consensus for autism but not full ICD-10 criteria. These children were then evaluated for history of epilepsy, gastroinstestinal problems, and developmental regression. 39% of children with narrow autism had a history of regression during development. This compared to 11% of children with broad autism, and 3% of children with no autism. On average this regression occurred around the 25 month of age. There were no differences in IQ or adaptive functioning between those with or without regression. However, those with regression classified in the broad autism group had significantly more symptoms than those without regression also classified in the broad autism group. Regression was not associated with gastroinstestinal symptoms or with epilepsy.

Thank you for subscribing to the RSS feed of Child-Psych.org. Please visit our website to join the conversation. ©2010 Child Psychology Research Blog. All Rights Reserved.

A study in the latest issue of the journal Autism examined the possible role of acetaminophen use after MMR vaccine and autism. The authors provided an excellent review of the MMR vaccine-autism research, which indicates that although some clinical studies have found a link, most epidemiological studies have failed to find an association between MMR and autism. The authors noted that acetaminophen is commonly used to treat the adverse reaction of MMR vaccinations such a fever and rash. In addition, one study (Alberti et al, 1999) showed that some low functioning children with autism process acetaminophen differently. Thus, the authors proposed a innovative hypothesis: Does acetaminophen use after MMR vaccination increase the risk for Autism?

The authors recruited parents of typically developing children and children with autism via internet advertisement. The total sample included 83 parents completing the survey for children with autism and 80 parents completing the control survey. The surveys included a variety of questions about the child such as age, gender, what medications were used to prevent or treat reactions to the MMR vaccine, including aspirin, acetaminophen, or ibuprofen. The survey of parents of children with autism included additional questions about their children diagnosis such as whether a regression in development was observed.

The authors found that parents of children with autism reported:
- more adverse effects of MMR vaccine, including fever, diarrhea, irritability
- increased presence of concurrent illnesses with MMR vaccine
- more acetaminophen use after the MMR vaccine among children who had a reaction to the vaccine, children who had a regression in development, and those under 5.
- more acetaminophen use between 12 and 18 months of age
No association was found between ibuprofen use and autism.

A few caveats:
The finding that parents of children with autism reported more adverse effects of MMR vaccine or more concurrent illnesses with MMR vaccine is not overly informative. Note that these results were based on parental reports via internet with no possibility of verification of accuracy of such reports. So it is completely plausible to argue that given the extended media coverage of the vaccine-autism link, some parents of children with autism are more attuned to their children histories after vaccination and thus are more likely to remember or report complications. What is informative and actually interesting is that there was a significant difference in reports of acetaminophen use as compared to ibuprofen use. This difference can not easily be explained on the basis of some report bias. The problem is that acetaminophen use was reported much more frequently than ibuprofen use by all parents. So it could be argued that since parents of children with autism were more likely to report complications (even if just a by-product of recall bias), the use of acetaminophen will also appear to be different between the two groups.

Thank you for subscribing to the RSS feed of Child-Psych.org. Please visit our website to join the conversation. ©2010 Child Psychology Research Blog. All Rights Reserved.

A review of: PALMER, R., BLANCHARD, S., WOOD, R. (2008). Proximity to point sources of environmental mercury release as a predictor of autism prevalence. Health & Place DOI: 10.1016/j.healthplace.2008.02.001

This fascinating, yet bound to be controversial, study hit the news yesterday as it was made available (pre-publication) by the Journal Health & Place. The study is simple, straightforward, elegant, with some powerful findings. In fact, the findings are somewhat daunting given the simplicity of the design. The researchers reviewed the amount of mercury release reported by industrial facilities and power plants in the State of Texas in 1997 from data provided by US Environmental Protection Agency Toxics Release Inventory. They compared these data against autism rates in 1997 and 2002 as measured by schools’ autism classifications provided by the Texas Education Agency. Using a specialized geographical analysis system, the authors were able to locate each source of mercury and calculate the distance between each mercury source and each school. The results:

Industrial release of mercury and distance to industrial sources independently predicted increased rates of autism. The association with industrial release of mercury was not linear, instead the statistical model fit suggested an accelerated risk. This association remained statistically significant after controlling for specific variables such as SES, urbanicity, and race.

Power plant release of mercury and distance to power plant independently predicted increased rates of autism. In this case the association was linear (not accelerated). Again, this association remained statistically significant after controlling for other variables.

It is easy to dismiss these findings as inconsequential because they are ‘correlational’ in nature, or do not really prove anything. Researchers are too often guilty of selective acceptance of research: those studies that fit the consensus are accepted while those that don’t are dismissed for their methodological flaws – even though the studies we accept are equally flawed.

In the spirit of fairness I have to say that these findings are strong. Their methodology and analytical process are not any different from what is commonly seen in social science or epidemiology research. Is it perfect? Far from it. Is it useful or informative? Definitively! The data speak very clearly: In Texas, mercury release from industrial sources and power plants in 1997, and school proximity to these sources, are associated with rates of autism in 2002 as measured by school special education classifications.

Does this mean that mercury causes autism? Not at all. In the last sentence of the previous paragraph you can not replace the words are associated with with the word cause. There is a major difference. The data, albeit strong, have limitations. For example, the most obvious (to me) alternative explanation is that mercury release and proximity to these sources is also associated with another mystery factor that is causing this apparent association and that in fact, mercury release has nothing to do with autism rates in 2002. Let’s hypothesize that these power plants and industrial sources also release another toxin – let’s call this toxin autisimic (this is a made up toxin). These sources release mercury and autisimic at the same rate, so for each pound of mercury released there is a pound of autisimic released. It is possible then that this autisimic toxin directly increases the risk for autism, and this could explain completely the strong (but now obviously inaccurate) association between mercury release and autism.

Does this study show that vaccines cause autism? Absolutely not. I know this question may sound ludicrous to some, but I pose it rhetorically because I am certain that some will make the wide leap and link these findings to the vaccine issue.

There are other problems and limitations with this study, such as how autism rates were calculated (using all children instead of only those born inor after 1997), whether the autism rates are truly climbing and not explained by other factors, whether there are other variables that could be explaining this relation, etc, etc — and yes, this study does not prove or directly indicate that autism is caused by mercury exposure (click here for a much more critical review of this study). But this study is compelling in showing an association between mercury exposure and autism rates, and scientists can not just ignore it under the basis of its imperfect design and inability to make causal links – if that is the case, then only carefully controlled laboratory studies, with poor external validity, should be published and accepted as contributors to our greater scientific knowledge. This is study is far, far, from perfect, and many changes should have been requested prior to publication, but I can say the same of 90% of what is published today.

Thank you for subscribing to the RSS feed of Child-Psych.org. Please visit our website to join the conversation. ©2010 Child Psychology Research Blog. All Rights Reserved.

DeSoto, M.C., Hitlan, R.T. (2008). Concerning Blood Mercury Levels and Autism: A Need to Clarify. Journal of Child Neurology, 23(4), 463-465. DOI: 10.1177/0883073808314718

The latest issue of the Journal of Child Neurology includes two letters regarding the mercury-autism controversy. For most people familiar with this issue, this summary will not present anything new, but for those interested in knowing the basics of this latest controversy, here is a micro-crash course on the mercury-autism link.

In 2004 authors Ip, Wong, HO, Lee, and Wong, published an article in the Journal of Child Neurology comparing the blood and hair mercury levels of children with and without autism in order to examine if a mercury-autism link existed. The results were as follow:

There was no difference in the mean mercury levels [between the group]. The mean blood mercury levels of the autistic and control groups were 19.53 and 17.68 nmol/L, respectively (P = .15), and the mean hair mercury levels of the autistic and control groups were 2.26 and 2.07 ppm, respectively (P = .79).

The conclusion, which was supposedly inconsistent with the theory that mercury causes autism, shocked some members of the autism community. I said ‘supposedly’ because whether or not the mercury theory is correct, the manner in which the results are presented goes beyond what the data can actually say. Specifically, the authors concluded that:

Thus, the results from our cohort study with similar environmental mercury exposure indicate that there is no causal relationship between mercury as an environmental neurotoxin and autism.

The problem here is that this data only speak to concurrent levels of mercury and autism diagnosis. So it is entirely possible that in some children pre- or post-natal exposure to this toxin results in neurodevelopmental changes leading to autism symptoms. This idea does not imply that mercury levels would remain high in affected children. It only implies that that these children were exposed to this toxin. A similar example would be exposure to alcohol during pregnancy leading to fetal alcohol syndrome (FAS). Children with FAS do not have higher levels of alcohol in their blood than typically developing kids. I am not endorsing the mercury theory, but I am stating that the conclusions as stated by Ip go beyond what their data say.

But back to the Ip et al. (2004) study. In 2007, DeSoto and Hitlan published an article in the Journal of Child Neurology after they found a major mathematical mistake in the original findings reported by Ip. Specifically, they found that the non-significant differences in blood levels reported by Ip was an error and significant differences were actually observed. I did the calculations myself, and yes in fact, the kids with autism had significantly higher mercury levels than the control group (t = 2.6163; df = 135; p>.01).

Then in the latest issue of the Journal of Child Neurology, Dr. Michael Aschner published the following short correspondence:

In a recent article DeSoto and Hitlan1 reanalyzed an original data set, concluding that a relationship exists between blood mercury levels and the diagnosis of autism spectrum disorder (ASD). The conclusion is based on the reanalysis of hair to blood mercury ratios. Hair to mercury concentration ratios while informative need to be considered within the context of a temporal relationship. As elegantly demonstrated by Grandjean and colleagues, mercury levels in the hair reflect a delayed average compared to the blood mercury level averages. That is, mercury hair concentrations at hypothetical time point T reflect blood mercury levels at T minus 1 to 2 months. Chelation therapy and changes in diet and fish consumption (both more likely to occur in the ASD group) in the 2 months preceding the mercury analysis are likely to affect blood, but not hair mercury sample concentrations. The analysis by DeSoto and Hitlan, which presumes that the 2 biomarkers are equally affected, is clearly erroneous. Thus, absent appropriate corrections for the temporal fluctuations in mercury levels, the conclusions should be interpreted with utmost caution and revalidated taking the above issue into account.

This was followed by a response from DeSoto and Hitlan explaining further that their 2007 study addressed directly the error in blood sample Mean differences as reported by Ip et al in 2004. DeSoto and Hitlan re-analyzed Ip’s data showing that in fact blood mercury levels in the autistic group were higher than in the control group. This finding, and the original Ip error, had nothing to do with hair-to-blood ratios as described by Dr. Aschner. Dr. Aschner criticism is much more applicable to a separate hair analysis performed by DeSoto and Hitlan showing that the autism group had lower hair mercury levels than expected based on their blood samples, which is inconsistent with the idea that chelation therapy may have affected the blood levels. I explain: blood levels reflect immediate levels of mercury while hair levels reflect levels 1 to 2 month prior to testing. Thus, if blood levels at Time 1 were 20 (I’m using random numbers as example), then hair levels at Time 2 (2 month later) should be 20. If after Time 1 the child undergoes chelation therapy, then theoretically, blood levels at time 2 (2 month later) should be lower than 20 but hair levels should remain 20. Thus at Time 2, blood levels should be lower than hair levels. This was not supported by DeSotos’ analyses. What does this mean? Simply that the differences in blood levels found between the autism and control group are unlikely to have been affected by chelation therapy in the autistic group since the hair-blood analysis was not consistent with what is expected if chelation therapy had occur. WARNING: this is not a statement supporting chelation therapy, it is just a clarification of a possible interpretation of this data.

Does this mean that vaccines cause autism? Not at all. Again, although we would like to think that data can give us the entire story, data speak in morphemes rather than words, or sentences. This data only indicate that this particular group of children with autism had higher mercury levels than typically developing children. This could mean many things, such as that the children with autism have difficulty processing mercury or that the children with autism were exposed to higher levels of mercury, and these explanations in turn, may or may not have anything to do with the symptoms observed in children with autism, both in terms of symptom presentation, severity, or causes. These data simply do not give us that information.

More recent articles about the Autism and Mercury controversy and the Autism and Vaccines controversy can be found here.

Thank you for subscribing to the RSS feed of Child-Psych.org. Please visit our website to join the conversation. ©2010 Child Psychology Research Blog. All Rights Reserved.