Monday 31 March 2014

ALV003 glutenase and more

Today's post is a bit of mash-up but hopefully containing some common threads. If you're also reading the title of this post and wondering what it might have to do with autism research, stay with me...
For mash get smash... @ Wikpedia 

The chopping up duties of ALV003
To start, we have the paper by Lähdeaho and colleagues [1] who reported results based on the "ability of ALV003, a mixture of 2 recombinant gluten-specific proteases given orally, to protect patients with celiac disease from gluten-induced mucosal injury". Those with their ear to the coeliac (celiac) disease research ground will probably have heard something about ALV003 from Alvine Pharmaceuticals before. This glutenase mix - a protease compound designed to degrade gluten - comprising cysteine endoprotease B-isoform 2 and prolyl endopeptidase has already started to make some scientific waves following the results of a couple of studies reporting tolerability [2] and initial efficacy [3] in cases of coeliac disease (CD). The theory behind it being that chopping up those dastardly gluten proteins/peptides might break an important link in the chain of biological events observed in coeliac disease.

The data from Lähdeaho et al suggest that "the glutenase ALV003 appears to attenuate gluten-induced small intestinal mucosal injury in patients with celiac disease in the context of an everyday gluten-free diet containing daily up to 2 g gluten". In other words, taking ALV003 together with a gluten-supplemented meal seemed to have some protective effect on the state of the gut mucosa compared to the effects of a placebo.

With my autism research hat on, and particularly the green beret marked gluten and casein-free (GFCF) diets, I did wonder whether there may be something to be potentially learned from these results. I'm not for one minute implying that ALV003 should be indicated for autism nor am I suggesting that autism is coeliac disease despite some potential overlap. But I did wonder in light of something like the opioid-excess theory of autism [4] and some continued interest in all-things digestive enzymes and autism (see here), whether there may be some investigations to do in this area.

IgA deficiency and autoimmunity
Next up is the paper by Ludvigsson and colleagues [5] who observed that: "Individuals with IgA deficiency have a higher prevalence of several other autoimmune disorders". IgA deficiency is something that has cropped up on this blog before, again with an autism slant (see here). Harking back to some potentially important work by the late Reed Warren [6] suggestive that some on the autism spectrum may present with an IgA deficiency, extrapolating the results from Ludvigsson (yes, he of the 'not quite coeliac disease but something close in cases of autism' study) might have some important implications.

In particular, Ludvigsson et al  report quite a dazzling connection between IgA deficiency and the risk of various autoimmune conditions, such that there was: "a 35-fold higher PR [prevalence ratio] for celiac disease and 10-fold higher for type 1 diabetes". I'm only an amateur statistician but even I understand that a frequency of CD in IgA deficient vs. non-deficient population controls amounting to 6.7% and 0.2% respectively, is something potentially very important. Outside of the fact that IgA deficiency can affect the serological diagnosis of CD [7] I find these results to be interesting. That some people on the autism spectrum presenting with IgA deficiency may similarly find themselves in the cycle of greater risk of autoimmune conditions potentially offers some important lessons for screening and management of such issues. Also noted in a fairly recent presentation from one Alessio Fasano (see here for the video), science could do a lot worse than look to something like gluten as also being one of the important [modifiable] drivers in this autoimmunity relationship too.

I've probably bored you enough with this post and the speculations contained within. As per my recent ramblings on disappearing anti-gliadin antibodies following use of a gluten-free diet (see here) the important point to get across is that alongside a diagnosis of autism (and other comorbidity) there may be some important biochemistry that needs to be examined at least for some on the autism spectrum. Looking past the presented behavioural symptoms to see the person first and the autism label second is a pretty important part of this strategy so as to ensure that such health issues if present for some on the autism spectrum, don't disadvantage individuals further.

Now, how about some light reading paralleling pathogens and gluten peptides in CD [8] with suitable musical accompaniment by Sheryl Crow. Have some fun!


[1] Lähdeaho ML. et al. The Glutenase ALV003 Attenuates Gluten-Induced Mucosal Injury in Patients with Celiac Disease. Gastroenterology. 2014 Feb 25. pii: S0016-5085(14)00242-X.

[2] Siegel M. et al. Safety, tolerability, and activity of ALV003: results from two phase 1 single, escalating-dose clinical trials. Dig Dis Sci. 2012 Feb;57(2):440-50.

[3] Tye-Din JA. et al. The effects of ALV003 pre-digestion of gluten on immune response and symptoms in celiac disease in vivo. Clin Immunol. 2010 Mar;134(3):289-95.

[4] Shattock P. & Whiteley P. Biochemical aspects in autism spectrum disorders: updating the opioid-excess theory and presenting new opportunities for biomedical intervention. Expert Opin Ther Targets. 2002 Apr;6(2):175-83.

[5] Ludvigsson JF. et al. Association Between IgA Deficiency & Other Autoimmune Conditions: A Population-Based Matched Cohort Study. J Clin Immunol. 2014 Mar 2.

[6] Warren RP. et al. Brief report: immunoglobulin A deficiency in a subset of autistic subjects. J Autism Dev Disord. 1997 Apr;27(2):187-92.

[7] Shahnaz A. et al. Tissue transglutaminase antibody levels predict IgA deficiency. Arch Dis Child. 2013 Nov;98(11):873-6.

[8] Bethune MT. & Khosla C. Parallels between Pathogens and Gluten Peptides in Celiac Sprue. PLoS Pathog 2008; 4(2): e34.

---------- Lähdeaho ML, Kaukinen K, Laurila K, Vuotikka P, Koivurova OP, Kärjä-Lahdensuu T, Marcantonio A, Adelman DC, & Mäki M (2014). The Glutenase ALV003 Attenuates Gluten-Induced Mucosal Injury in Patients with Celiac Disease. Gastroenterology PMID: 24583059 Ludvigsson JF, Neovius M, & Hammarström L (2014). Association Between IgA Deficiency & Other Autoimmune Conditions: A Population-Based Matched Cohort Study. Journal of clinical immunology PMID: 24584841

Saturday 29 March 2014

Inflaming inflammation and psychiatry

The systematic review published by Mitchell & Goldstein [1] kinda says it all when it comes to our current view of the topic of inflammation and psychiatry, and in particular inflammation and neuropsychiatric conditions such as autism, schizophrenia and attention deficit hyperactivity disorder (ADHD): "There is preliminary evidence for elevated markers of inflammation in this population".
All at sea (JMW Turner) @ Wikipedia 

I've talked quite a bit on this blog about how, as strange as it might seem, heightened inflammation and inflammatory processes/markers do seem to be over-represented in a fair few diagnoses with a neuropsychiatric element to them (see here and see here for example). Granted, one has to be careful not to over-generalise to an entire population (remember: autisms not autism) and the chicken-and-egg question of which came first inflammation or condition still hangs over this 'association.' But the realisation that inflammatory processes may link immune function to behavioural presentation [2] is not to be sniffed at and may yet offer some intriguing new avenues for things like intervention [3].

The Mitchell / Goldstein review very sensibly notes that: "the data are inconsistent" and sound the oft-heard call for "Larger, prospective studies [which] are needed to realize the goal of inflammatory markers informing clinical practice". But, there is not escaping that immune function may be doing so much more than just protecting us for the myriad of beasties which wish to either call us home or make a meal of us or both.

As if to further prove the point, I also chanced on the study by Charity Onore and colleagues [4] who alongside some notable names on the autism research scene looked at the concept of maternal immune activation on the inflammatory status of offspring. The name Charity Onore has (again!) been talked about on this blog and some particularly interesting work looking at adhesion molecules and autism (see here); an area which has since been followed up (see here). The focus on the concept of the maternal immune activation (MIA) model is nothing new when it comes to autism research for example (see here and see here) denoting how elevated inflammatory responses at certain critical times during the nine months that made us may have some important consequences for offspring.

The latest Onore paper details how in a mouse model of MIA (based on the artificial stimulation of mother mice using poly I:C as the immunostimulating agent) "macrophage cytokine production in adult offspring" was the experimental target. Macrophages, also known as the 'big eaters' of the immune system, represent an important part of immune function undertaking some important tasks. Again, they have been mentioned in other blog entries, notably with the initial chatter about GcMAF in connection to autism (see here).

Anyhow, based on the MIA model, Onore et al  reported that analysis of adult offspring mice samples ("femurs were collected and bone marrow-derived macrophages were generated") for the presence of various cytokines with or without conditions of stimulation (LPS... see here) produced some interesting results. So for example, "a higher production of CCL3", a chemokine, was recorded. Indeed, elevations in CCL3 otherwise known as MIP-1α, was recorded both with and without stimulation leading authors to conclude: "a general increase in production of this chemokine". Their closing remark on their cumulative results being: "Collectively, these data suggest that MIA can produce lasting changes in macrophage function that are sustained into adulthood". I should add 'in mice' to the end of that sentence too.

Whilst there is still much to do in this potentially important area of research, what I perhaps would like readers to take from this post are a few things: (i) inflammation and by inference, immune system involvement, is an emerging area when it comes to behaviour and development, and (ii) whilst not ruling out the suggestion that there may be a variable genetic fragility to various neurodevelopmental conditions (perhaps even overlapping common ground) one would be wise to look at the interaction with our environment [5] as being an equally important variable [6] particularly during our earliest days of being. Oh, and that inflammation or inflammatory processes might be relevant to other tissues outside of just the grey-pinkish matter floating in the skull...

Music to close. Having seen the brilliant Franz Ferdinand play live this week, I'd refer you to perhaps their most famous song: Take me out. Or if you prefer, the live version from the venue on Thursday noting that this is not my own recording...


[1] Mitchell RH. & Goldstein BI. Inflammation in Children and Adolescents With Neuropsychiatric Disorders: A Systematic Review. J Am Acad Child Adolesc Psychiatry. 2014 Mar;53(3):274-296.

[2] Sommer IE. et al. Efficacy of anti-inflammatory agents to improve symptoms in patients with schizophrenia: an update. Schizophr Bull. 2014 Jan;40(1):181-91.

[3] Dada T. et al. Mouse model of intrauterine inflammation: Sex-specific differences in long-term neurologic and immune sequelae. Brain Behav Immun. 2014 Jan 31. pii: S0889-1591(14)00015-4.

[4] Onore CE. et al. Maternal Immune Activation Leads to Activated Inflammatory Macrophages in Offspring. Brain Behav Immun. 2014 Feb 21. pii: S0889-1591(14)00053-1.

[5] Meyer U. Prenatal poly(i:C) exposure and other developmental immune activation models in rodent systems. Biol Psychiatry. 2014 Feb 15;75(4):307-15.

[6] Goyal DK. & Miyan JA. Neuro-Immune Abnormalities in Autism and Their Relationship with the Environment: A Variable Insult Model for Autism. Front Endocrinol (Lausanne). 2014 Mar 7;5:29.

---------- Mitchell RH, & Goldstein BI (2014). Inflammation in Children and Adolescents With Neuropsychiatric Disorders: A Systematic Review. Journal of the American Academy of Child and Adolescent Psychiatry, 53 (3), 274-296 PMID: 24565356 Onore CE, Schwartzer JJ, Careaga M, Berman RF, & Ashwood P (2014). Maternal Immune Activation Leads to Activated Inflammatory Macrophages in Offspring. Brain, behavior, and immunity PMID: 24566386

Thursday 27 March 2014

The new CDC autism prevalence rate

Whilst quite a lot of media space is currently being devoted to talking about the study by Rich Stoner and colleagues [1] and sweeping generalisations like the BBC headline: Autism 'begins long before birth' with seemingly only little appreciation of the small-scale nature of the study and little details linked to samples being post-mortem tissues, other important autism-related news is also out there.
CDC US autism prevalence estimates @ Autism Speaks 

I'm talking about the latest autism prevalence estimates from the US CDC (Centers for Disease Control and Prevention) for 2010.

The latest report [2] reveals that the overall estimated prevalence for 8-year olds in the United States having an autism spectrum disorder (ASD) in 2010 is:

1 in 68.

Back in 2012 (has it been that long?) I talked about the publication of the estimates based on the 2008 surveillance data which highlighted an estimated 1 in 88 children presenting with an autism spectrum condition [3]. That the latest figures further revise upwards the prevalence estimates follows a trend over quite a few years now as per the graph included in this post taken from the Autism Speaks coverage of this news. The intervening years have seen other reports talking about increased prevalence estimates for autism as per reports here and here and here both in the US and other parts of the world, taking into account factors such as ethnicity too. In light of these estimates, some of which talked about 1 in 48 children with autism I don't think anyone should be too surprised by the latest CDC estimate.

As per my 2012 post on the 1 in 88 figure, I'm sure that the latest CDC estimates will generate further discussions about the hows and whys of the increasing numbers of children being diagnosed with an ASD. I'm willing to take on board arguments about better awareness and improved diagnostic vigilance as accounting for some of the change, but those factors only go so far. Yet again, I'm going to link to the Nature piece from Karen Weintraub [4] on the autism prevalence puzzle and how blanket commentary like 'we're just better at diagnosing autism' is starting to wear a little thin particularly in light of the continued reliance on DSM-IV criteria in this latest estimate.

Perhaps now is the time to start further widening the research agenda when it comes to the question of what is driving the increase in cases of autism to include a few additional points: (a) that autism is probably a plural condition, so 'autisms' over autism which is exquisitely exemplified by the recent BCKDK research, (b) genes and genetic influences are [variably] important to the autisms (see previous point), and (c) genes represent our blueprint but are not necessarily our destiny as per the rise and rise of the science of epigenetics combined with the notion that we don't walk around with all our genes permanently fixed to the 'on' position all the time. That final point in particular opens the door to environment - some facets of environment - also variably impacting on genes, the function of genes, and indeed autism and autism risk. And as we've seen in recent times, how perhaps we should be devoting a little more effort to looking at certain external factors [plural and cumulatively] as potentially playing some [variable] role in the increasing numbers of children being diagnosed with an ASD.

Just sayin'.


[1] Stoner R. et al. Patches of Disorganization in the Neocortex of Children with Autism. NEJM. 2014; 370: 1209-1219.

[2] Autism and Developmental Disabilities Monitoring Network Surveillance Year 2010 Principal Investigators; Centers for Disease Control and Prevention. Prevalence of autism spectrum disorders--Autism and Developmental Disabilities Monitoring Network, 14 sites, United States, 2010. MMWR Surveill Summ. 2014 Mar 30;63. SS-2.

[3] Autism and Developmental Disabilities Monitoring Network Surveillance Year 2008 Principal Investigators; Centers for Disease Control and Prevention. Prevalence of autism spectrum disorders--Autism and Developmental Disabilities Monitoring Network, 14 sites, United States, 2008. MMWR Surveill Summ. 2012 Mar 30;61(3):1-19.

[4] Weintraub K. The prevalence puzzle: Autism counts. Nature. 2011; 479: 22-24.

----------- Autism and Developmental Disabilities Monitoring Network Surveillance Year 2010 Principal Investigators (2014). Prevalence of Autism Spectrum Disorder Among Children Aged 8 Years — Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2010 Morbidity & Mortality Weekly Report, 63

Wednesday 26 March 2014

Vision impairment and ADHD?

Science often has the ability to surprise.

So it was when I first read the paper by Dawn Decarlo and colleagues* (open-access here) which suggested that: "children with vision impairment may be more likely to be diagnosed with ADHD [attention deficit hyperactivity disorder] than children in the general population". I should point out that this observation should not be totally unexpected given some media for other studies by the authors (see here).
Lady in a green jacket @ Wikipedia  

The Decarlo study was based on a question of "whether the prevalence of parent-reported ADHD diagnosis in a cohort of pediatric low vision subjects was similar to that reported for the general population". It seems researchers were a little bit worried that there may be some mixing up of the signs and symptoms of possible ADHD as a consequence of a child's visual impairment and sought to further look into the issue.

So, bearing in mind the paper is open-access, a few study details:

  • Alabama, USA was the study setting, and the medical records of participants (N=245) with low vision were drawn from two sites. Researchers garnered various details about participants covering basic demographic information and that relevant to vision including "primary ocular diagnosis, presence or absence of nystagmus, and best-corrected visual acuity in each eye". They also asked participants' parents whether or not their child had ever received a diagnosis of ADHD and had state and national ADHD prevalence estimates to hand.
  • Results: the Alabama state ADHD prevalence estimate (based on 2007 data) was 14.3% and the US national prevalence 9.5%. The parent-reported prevalence rate of study children ever having a diagnosis of ADHD was 22.9% (56.245). Examination of an additional cohort of 19 children with almost total vision loss for a diagnosis of ADHD was 10%.
  • Albinism - a condition concerned with the production of melanin - was the most frequently reported eye problem among participants (16%) and similar to the issue of optic nerve hypoplasia (ONH), saw a prevalence of ADHD diagnosis at 20% of these cases. Indeed, ONH has previously been linked with 'behavioural problems' in other research** and indeed, some of the features of autism***.
  • Researchers concluded: "the prevalence of parent-reported ADHD among children with vision impairment in the present study is considerably higher than that reported by the Centers for Disease Control and Prevention".

There is obviously more to do in this area insofar as ascertaining whether these results are more widely applicable to cases of paediatric vision impairment or just a statistical blip. Noting for example, that the State of Alabama has an ADHD prevalence estimate some way above the US national estimate, might imply that ADHD is more generally apparent as a diagnosis there, so children were more likely to receive a diagnosis as a result of better screening facilities for example. Indeed, in my review of some of the [worldwide] prevalence rates of pediatric ADHD rates (see here), the figures coming back were nearer the US national average than the Alabama state estimate although potentially rising.

The lack of an external control group looking both at asymptomatic controls or even those children with something like hearing impairment in the Decarlo study, means that some caution must be exercised before leaning too far into the suggestion of any relationship.

The question of what mechanism might be behind any association is also important. I could suggest that the quite recent report of a link between solar intensity and ADHD (see here) might potentially offer some clues if one is to assume that vision is a common denominator across these areas of research. That being said, it is quite a tentative link and as yet, I can offer no further support for this suggestion.

Perhaps more relevant are the discussions about shared genetic/environmental influences which affect vision and potentially, risk of ADHD, as being linked to any association. Taking you back to the paper by Antoine and colleagues**** which was discussed in another post (see here) on inner ear conditions being linked to hyperactivity, one might assume similar mechanisms to be plausible. That also the paper by Kim and colleagues***** reported issues with vision function and colour vision to be linked with cases of adult ADHD is something else to throw into the mix.

Whatever the reason(s) for such a correlation, I am intrigued by the Decarlo findings and wait attentively for any further movement in this area of investigation.

Music to close, and having recently heard the song Purple Rain on the radio, I have a rekindled respect for Prince...


* Decarlo DK. et al. Prevalence of attention-deficit/hyperactivity disorder among children with vision impairment. J AAPOS. 2014 Feb;18(1):10-4.

** Teär Fahnehjelm K. et al. Optic nerve hypoplasia in children and adolescents; prevalence, ocular characteristics and behavioural problems. Acta Ophthalmol. 2013 Oct 7.

*** Parr JR. et al. Social communication difficulties and autism spectrum disorder in young children with optic nerve hypoplasia and/or septo-optic dysplasia. Dev Med Child Neurol. 2010 Oct;52(10):917-21.

**** Antione MW. et al. A Causative Link Between Inner Ear Defects and Long-Term Striatal Dysfunction. Science. 2013; 341: 1120-1123.

***** Kim S. et al. Visual function and color vision in adults with Attention-Deficit/Hyperactivity Disorder. J Optom. 2014 Jan-Mar;7(1):22-36.

---------- Decarlo DK, Bowman E, Monroe C, Kline R, McGwin G Jr, & Owsley C (2014). Prevalence of attention-deficit/hyperactivity disorder among children with vision impairment. Journal of AAPOS : the official publication of the American Association for Pediatric Ophthalmology and Strabismus / American Association for Pediatric Ophthalmology and Strabismus, 18 (1), 10-4 PMID: 24568975

Monday 24 March 2014

A specific female ASD phenotype is emerging...

The title of this brief post is a quote taken from the abstract of the paper by Frazier and colleagues [1] who, following an analysis of participants included in the Simons Simplex Collection, concluded that autism research and practice might well consider looking at differences in the presentation of autism spectrum disorders (ASDs) between the sexes.
The Lady of Shalott @ Wikipedia 

Including nearly 2500 people with autism including over 2100 males and 304 females, examining "autism symptoms, cognitive and motor functioning, adaptive behavior, and associated behavior problems", researchers were able to build up a quite generalised picture of symptom presentation across the sexes.

Bearing in mind their assertion that "diagnostic instruments captured autism similarly in males and females" they reported some interesting trends: "Females with ASD had greater social communication impairment, lower levels of restricted interests, lower cognitive ability, weaker adaptive skills, and greater externalizing problems relative to males".

Of course we have been here before as per a previous blogpost on this topic albeit not exactly with the same profiles being reported. Compared with the results from Mandy and colleagues [2] for example, the finding of less frequent issues with repetitive stereotyped behaviour in girls seems to be one of the few consistencies reported [3] again and again [4] provided we assume that restricted interests and repetitive behaviours fall into the same behavioural domain. Bear in mind also that Mandy et al were looking at a participant group comprising "high-functioning ASD" so I'm perhaps comparing apples and pears. The results from Head and colleagues [5] (open-access) however do suggest there is more to do in this area with their assertion that: "females with ASD have more developed social skills than males with ASD".

Frazier et al conclude by suggesting that the disparity in cases of autism according to gender may very well, to some extent, be due to some of those sex differences in behavioural presentation and the need to "evaluate whether observed sex ratios in high-functioning cases are reduced if female-specific indicators of restricted interests are included". Certainly with the arrival of DSM-5 and all that SCD (social communication disorder) chatter, it will be interesting to see the data on which sex is more likely to be diagnosed with autism, and which with SCD. Oh, and then there are the results from Sébastien Jacquemont and colleagues [6] and the female protective model (see here) ...

To close, no song this time around but rather a link to a forgotten classic animated film: The Lord of the Rings which is currently doing the rounds among the female members of my brood. Indeed, I am very proud to say that Barbie and dolls and related stuff has given way to discussions about the finer points of Tolkien's classic books. And the primary influence behind this shift in interests? The Lego DS game... who said technology was all doom and gloom for child development?


[1] Frazier TW. et al. Behavioral and cognitive characteristics of females and males with autism in the simons simplex collection. J Am Acad Child Adolesc Psychiatry. 2014 Mar;53(3):329-340.e3.

[2] Mandy W. et al. Sex differences in autism spectrum disorder: evidence from a large sample of children and adolescents. J Autism Dev Disord. 2012 Jul;42(7):1304-13.

[3] Werling DM. & Geschwind DH. Sex differences in autism spectrum disorders. Curr Opin Neurol. 2013 Apr;26(2):146-53.

[4] Kreiser NL. & White SW. ASD in Females: Are We Overstating the Gender Difference in Diagnosis? Clin Child Family Psych Rev. 2014; 17: 67-84.

[5] Head AM. et al. Gender differences in emotionality and sociability in children with autism spectrum disorders. Molecular Autism 2014, 5:19

[6] Jacquemont S. et al. A Higher Mutational Burden in Females Supports a “Female Protective Model” in Neurodevelopmental Disorders. Am J Human Genetics. 2014. February 27.

---------- Frazier TW, Georgiades S, Bishop SL, & Hardan AY (2014). Behavioral and cognitive characteristics of females and males with autism in the simons simplex collection. Journal of the American Academy of Child and Adolescent Psychiatry, 53 (3), 329-340000 PMID: 24565360

Friday 21 March 2014

Dioxin exposure and autistic traits?

As promised in a previous post, today I'm turning my attention to the paper by Muneko Nishijo and colleagues [1] and their conclusion of "a specific impact of perinatal TCDD [2,3,7,8-tetrachlorodibenzo-p-dioxin] on autistic traits in childhood, which is different from the neurotoxicity of total dioxins (PCDDs/Fs) [polychlorinated dibenzo-p-dioxins/furans]".

TCDD @ Wikipedia 
With all the recent chatter about [surrogate] environmental markers and the numbers of cases of autism spectrum disorders and environmental toxicants and autism risk it is indeed timely that the Nishijo paper comes to publication now. Environmental factors, however you wish to define this, are certainly no stranger to autism research, and are fast finding a place in the autism research psyche, perhaps in part due to the rise and rise of the science of epigenetics (see here) as a bridge between genetics and environment. Genes, or rather the blueprint that is your genome, might not necessarily be your destiny and all that jazz...

The Nishijo paper in a little more detail:

  • Set in Vietnam, which it has to be said, has seen more than its fair share of chemical exposures in recent history, the authors looked at the possibility of perinatal dioxin exposure being linked to the presence of autistic traits based on a sample of 153 infants. This follows other work by this group looking at dioxin exposure and more generalised infant neurodevelopment [2] as part of a wider research agenda. Dioxins for those who might not know, are categorised as environmental pollutants, and because of their biological persistence, are deemed pretty hazardous to human and other animal health (see the WHO fact sheet here). It's accurate that I mentioned Agent Orange in reference to the chemical load witnessed in Vietnam because Agent Orange was contaminated with TCDD - the chemical name for dioxin -  and there are some very scary quotes about TCDD being for example "perhaps the most toxic molecule ever synthesized by man". The US IOM report 'Veterans and Agent Orange' provides some sober reading on the topic.
  • Scare tactics aside, the authors assessed the levels of TCDD in breast milk as part of a larger analysis of various other dioxins (PCDDs, PCDFs) based on analysis of samples via GC-MS. Actually quite a good overview of their methods can be found in another paper by this group [3] (open-access). Data from these analyses were transformed to form something called the TEQ (Toxic Equivalent) which basically gives you some idea of the toxicity of these classes of compounds relative to TCDD. TCDD has a value of 1 so setting the gold standard of toxicity. 
  • Offspring were followed-up based on the use of the Bayley Scales of Infant and Toddler Development, Third Edition (Bayley-III) and specifically with autism in mind, the Autism Spectrum Rating Scales (ASRS).  
  • Results: exposure groups (mild and high) were determined according to a cut-off level of 3.5 pg/g fat of TCDD being detected and results were reported according to gender. So: "The high-2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) exposed groups... showed significantly higher Autism Spectrum Rating Scale (ASRS) scores for both boys and girls than the mild-TCDD exposed groups, without differences in neurodevelopmental scores". This indicates some kind of dose-dependent relationship between TCDD exposure and autistic traits in study participants. When it came to looking at any connection between other PCDDs/Fs and autistic traits, nothing significant was picked up. Ergo, TCDD exposure seemed to have specifically impacted on infant autistic traits.

These are interesting results, of that there is no doubt. I could start to go on about correlation not being the same as causation, or how the ASRS might not necessarily have been the best instrument to use in this particular instance given it being standardised on American children and not officially translated into Vietnamese. That also it is not a professionally-administered instrument is another potential gap. But I'm not going to let all that get too far in the way of the Nishijo findings.

I see from some of the additional data from this paper that there were some other differences noted across the high and mild TCDD exposure groups which may be relevant to the results. When comparing boys and girls in the high and mild exposure groups, I note that mean birth weight was lower in the high exposed group compared to the mild exposed group (average 2920g vs. 3298g respectively) for boys. Realising that low birth weight is not an exclusively autism-correlated phenomena (see here) one might however consider this to be something which could potentially have affected the results obtained.

As per the discussions about the geographical location of this study [4], one of the question which then needs to be asked is whether the possibility of a TCDD exposure link is something applicable to other areas and other cases of autism/autistic traits. I'm no expert on TCDD so cannot readily answer this question aside from directing you to some data from the US Environmental Protection Agency on sources of TCDD. It does appear that there are quite a few potential sources of TCDD; although food seems to be the most widely cited source of exposure in modern times. By saying that I'm not trying to panic anyone, given that many countries do monitor foods for dioxin levels (see here) and act accordingly when high levels are detected.

Still, if we assume that there may be many roads towards a diagnosis of autism or the presentation of autistic traits, and that those roads may not be the same in every part of the world, the Nishijo results make for an interesting addition to the research landscape. As to mechanisms of effect, well take your pick; TCDD is categorised an endocrine disruptor so potentially falling into the same type of effect as that discussed by Rzhetsky and colleagues [5]. I also note some research chatter about TCDD and GABA [6] with specific mention of GAD (see here for some description of what this is). Jumping on the epigenetics bandwagon however, one might also suggest some further inquiry in that area if results like those by Manikkam and colleagues [7] are anything to go by, bearing in mind the continued focus on sperm and eggs and autism risk. That all being said, I don't doubt that it's going to be a complicated effect from TCDD with various factors involved.


[1] Nishijo M. et al. 2,3,7,8-Tetrachlorodibenzo-p-dioxin in breast milk increases autistic traits of 3-year-old children in Vietnam. Mol Psychiatry. 2014 Mar 18.

[2] Tai PT. et al. Dioxin exposure in breast milk and infant neurodevelopment in Vietnam. Occup Environ Med. 2013 Sep;70(9):656-62.

[3] Nishijo M. et al. Impact of perinatal dioxin exposure on infant growth: a cross-sectional and longitudinal studies in dioxin-contaminated areas in Vietnam. PLoS One. 2012;7(7):e40273.

[4] Banout J. et al. Agent orange footprint still visible in rural areas of central Vietnam. J Environ Public Health. 2014;2014:528965.

[5] Rzhetsky A. et al. Environmental and state-level regulatory factors affect the incidence of autism and intellectual disability. PLoS Comput Biol. 2014 Mar 13;10(3):e1003518.

[6] Hays LE. et al. Evidence that GABAergic neurons in the preoptic area of the rat brain are targets of 2,3,7,8-tetrachlorodibenzo-p-dioxin during development. Environ Health Perspect. 2002 Jun;110 Suppl 3:369-76.

[7] Manikkam M. et al. Dioxin (TCDD) Induces Epigenetic Transgenerational Inheritance of Adult Onset Disease and Sperm Epimutations. PLoS ONE. 2012; 7(9): e46249.

---------- Nishijo M, Pham TT, Nguyen AT, Tran NN, Nakagawa H, Hoang LV, Tran AH, Morikawa Y, Ho MD, Kido T, Nguyen MN, Nguyen HM, & Nishijo H (2014). 2,3,7,8-Tetrachlorodibenzo-p-dioxin in breast milk increases autistic traits of 3-year-old children in Vietnam. Molecular psychiatry PMID: 24637425

Thursday 20 March 2014

Environmental exposure and autism continued

In a post not-so-long-ago I talked about the paper by Andrey Rzhetsky and colleagues [1] and their assertion that environment (various facets of environment) might correlate with the increasing numbers of cases of autism being diagnosed. As per what was said on that post, there were lots of media headlines generated about the findings; some balanced and some a little sensational.
Luc Viatour / @ Wikipedia 

One of the main caveats I had with the Rzhetsky study was the reliance on statistical models onwards to their conclusions; so using surrogate markers of exposure and modelling trends but not actually looking at people in any biochemical or genetic sense or importantly, the metabolites of various candidate exposure compounds appearing in biofluids. Today therefore, I'm talking about the paper by Braun and colleagues [2] (open-access here) and some of the accompanying research looking at various measured chemical exposures in relation to autism.

The Braun paper is open-access but here are a few details:

  • The aim: "To identify gestational EDC [endocrine disrupting chemicalexposures associated with autistic behaviors" was accomplished by screening 175 pregnant women who were part of the HOME study (looking at "the impact of low-level fetal and early childhood exposures to environmental chemicals on health developmental and behavioral outcomes") for various suspected EDCs in urine and serum mid-pregnancy.
  • Said EDCs included "phthalate metabolites", "polychlorinated biphenyls (PCBs)", "brominated flame retardants" (including the PBDEs) and "perfluoroalkyl substances" (see here).
  • Keeping the analytical data in mind, mothers then completed the Social Responsiveness Scale (SRS), a standardised instrument which quite accurately seems to perform when it comes to autism screening [3] when offspring were 4-5 years of age.
  • Results: "Most of the EDCs were associated with negligible absolute differences in SRS scores", so no smoking gun but... "maternal serum concentrations of trans-nonachlor and PBDE-28 were associated with higher SRS scores". Indeed it's worth pointing out that some of the EDCs actually seemed to negatively correlate with SRS scores as for example: "PBDE-85, PCB-178, β-HCH, and PFOA concentrations were associated with less autistic behaviors".
  • That also the authors "adjusted for numerous potential confounders including gestational tobacco smoke exposure, socioeconomic factors, perinatal factors, caregiving environment, maternal IQ, and maternal depressive symptoms" when analysing their results is an important strength of their study.

I was intrigued by the Braun results for several reasons. Bearing in mind this was a study looking at maternal biofluids and in-utero exposures correlating with autistic behaviours (not specifically diagnoses) there are some interesting details requiring follow-up. BDE-28 (one of the congeners of the PBDEs) crop up quite a bit in research terms as per the the findings from Daniels and colleagues [3] looking at human milk samples. As Daniels et al suggest however: "The consequences of exposure to PBDEs are unknown" so I don't think we can draw too many conclusions from that one; indeed as per one of my previous posts on PBDEs and autism, there are a few more studies which could be done. That also another PBDE congener (PBDE-85) was actually associated with lower SRS scores - "consistent with less autistic behaviors" - is another example of just how complicated this area actually is.

Although maternal phthalate metabolites were not seemingly connected to scored offspring autistic behaviours, I'm not quite ready to trash any connection in this area. The paper by Testa and colleagues [4] (open-access here) looking at "primary and secondary metabolites of DEHP [di-(2-ethylhexyl) phthalate] in children with ASD [autism spectrum disorder]" offers some preliminary evidence for a potential role for phthalates exposure and autism. Compounded by the data presented by Stein and colleagues [5] on "a decreased capacity for detoxification via glucuronidation of compounds in the DEHP metabolic pathway" in their cohort with autism, and there are still questions to be answered about this group of compounds. And just in case you are still furrowing your brow at this association, there is the data from Larsson and colleagues [6] (open-access here) to look at, bearing in mind the suggestion that "the use of soft PVC as flooring material may increase the human uptake of phthalates in infants" [7].

Pinpointing one or more single pollutants as being tied into the increasing numbers being diagnosed with autism is always going to be an uphill struggle. Not only do we have very little data on the multitude of 'chemicals' (remember the mis-use of that word) we're all exposed to as part of modern living (or indeed that our parents were exposed to) but examining the synergistic effects of various multiple exposures is going to be problematic at best. As per the Stein research, we've also got to consider how an individual processes or metabolises their pollutant load which must have a bearing on the physiological effect from any exposure. Thinking back also to the recent air pollution - autism work starting to ask about gene x environment interactions, there are multiple dimensions to this work which aren't necessarily going to be answered by just looking at what a person is exposed to.

Oh, and I'll be coming to the paper by Nishijo and colleagues [8] titled "2,3,7,8-Tetrachlorodibenzo-p-dioxin in breast milk increases autistic traits of 3-year-old children in Vietnam" at some point quite soon.

To close, Rock Lobster...


[1] Rzhetsky A. et al. Environmental and state-level regulatory factors affect the incidence of autism and intellectual disability. PLoS Comput Biol. 2014 Mar 13;10(3):e1003518.

[2] Braun JM. et al. Gestational Exposure to Endocrine-Disrupting Chemicals and Reciprocal Social, Repetitive, and Stereotypic Behaviors in 4- and 5-Year-Old Children: The HOME Study. Environ Health Perspect. 2014 Mar 12.

[3] Norris M. & Lecavalier L. Screening accuracy of Level 2 autism spectrum disorder rating scales. A review of selected instruments. Autism. 2010 Jul;14(4):263-84.

[4] Testa C. et al. Di-(2-ethylhexyl) phthalate and autism spectrum disorders. ASN Neuro. 2012 May 30;4(4):223-9.

[5] Stein TP. et al. Autism and phthalate metabolite glucuronidation. J Autism Dev Disord. 2013 Nov;43(11):2677-85.

[6] Larsson M. et al. Associations between indoor environmental factors and parental-reported autistic spectrum disorders in children 6-8 years of age. Neurotoxicology. 2009 Sep;30(5):822-31.

[7] Carlstedt F. et al. PVC flooring is related to human uptake of phthalates in infants. Indoor Air. 2013 Feb;23(1):32-9.

[8] Nishijo M. et al. 2,3,7,8-Tetrachlorodibenzo-p-dioxin in breast milk increases autistic traits of 3-year-old children in Vietnam. Mol Psychiatry. 2014 Mar 18.

---------- Braun JM, Kalkbrenner AE, Just AC, Yolton K, Calafat AM, Sjödin A, Hauser R, Webster GM, Chen A, & Lanphear BP (2014). Gestational Exposure to Endocrine-Disrupting Chemicals and Reciprocal Social, Repetitive, and Stereotypic Behaviors in 4- and 5-Year-Old Children: The HOME Study. Environmental health perspectives PMID: 24622245

Tuesday 18 March 2014

Soy infant formula and seizures in autism

Spring @ Wikipedia 
'Association' is a word I'm sure many people with a connection to autism will have heard a lot about. Y'know gene X or compound Y is the plat du jour when it comes to autism aetiology; more often than not carrying the caveat 'requires further investigation'. As to whether such investigations are ever truly carried out would perhaps be an interesting piece of research on autism research.

Today I'm talking about another association, another variable to throw into the statistical risk mix derived from the paper by Cara Westmark [1] (open-access here) suggesting that where seizures or epilepsy present alongside (or as part of) autism, there may be a curious correlation to be had with the use of soy infant formula.

I'll readily admit that when first reading the title to this study I was more than a little reluctant to blog about it. I can't really explain why - whether it was yet another study of 'association' or something about what appears to be quite a well-used feeding strategy for infants where traditional methods are not indicated - there wasn't the initial appetite to talk about it. Following some light reading around the topic and what looks to be some quite strongly held views on the use of infant soy formula (see this BBC report) including mention of a possible relationship with ADHD (see here) I eventually decided that this might be something to cover. The press release covering the study also helped (see here and see here) in making my decision.

A few details:

  • The hypothesis: "the use of soy-based infant formulas could be contributing to seizure incidence in autism and other neurodevelopmental disorders" came about apparently on the basis of some earlier work by the authors [2] (open-access here) which suggested that a component of soy-feed (daidzein) given to rats might have the ability to induce seizures after a few days consumption. More discussion about this earlier trial can be found here.
  • So, based on an analysis of data derived from the Simons Foundation Autism Research Initiative (SFARI), Westmark and colleagues looked at the occurrence of seizures and if available, the types of seizures reported, alongside the frequency of soy infant formula use, using sex as a differentiating variable in cases of autism.
  • Results: "There was a 2.6-fold higher rate of febrile seizures in the soy-fed cohort (4.2% seizures with soy and 1.6% seizures without soy)". That being said, the results only passed significance for females who were fed soy formula; males were more likely to present with seizures after soy feeding "but [results] were not statistically significant".
  • Also: "A comorbid diagnosis of autism and epilepsy was more prevalent in males fed soy-based formula (odds ratio = 2.4, 95% confidence interval 1.1–5.2; P = 0.02) than females (odds ratio = 1.4, 95% confidence interval 0.056–14; P = 0.8)". This does sound a little counter-intuitive given what the last sentence said about febrile seizures and gender but I would draw your attention to the use of the diagnostic term 'epilepsy' as opposed to 'febrile seizure'.

I'm pretty sure that you can see from the collected data included in this paper there are some potentially interesting details which require further analysis. I'm for example interested in seeing more about the proposed mechanism to account for the correlation reported between soy formula use and an elevated frequency of seizures/epilepsy in relation to autism. The authors suggest "the effects of an underlying genetic mutation that lowers seizure threshold may be exacerbated, for example, by dietary exposure to high concentrations of phytoestrogens". Yes, possibly; although with various other autism-related research areas in mind, I'd be interested to see whether there may other effects from for example, the gut microbiota [3] and how they might also play their part.

Perhaps another question would be whether the underlying reason why soy infant formula was used in the first place might also have played some role in the results. The authors do approach this topic as per the sentences: "A possible criticism is that subjects, who were fed soy-based infant formula because they were allergic to cow's milk, had allergies that made them vulnerable to illnesses associated with fever-induced convulsions. Though the retrospective nature of the data does not allow us to make definitive conclusions regarding this point, we found that 2.7% of females and 1.7% of males in the study population reported allergies, but no subjects reported both allergies and febrile seizures suggesting that this criticism may not be valid". I'd perhaps chime in here and suggest that 'allergy' might be a bit of a red herring here as per the research looking at something like lactose intolerance in relation to autism (see here) and the still quite speculative role of milk peptides (see here). There is also the potential issue of something like those folate receptor autoantibodies to consider (see here) which also seem to be affected by old fashioned milk although I am speculating here. There is the odd study (case report) talking about something like lactose intolerance and epilepsy for example [4] but still a lot more scope for further investigation in this area.

As per my introduction to this post on the role of 'association' when it comes to autism research, one always needs to be a little careful when interpreting such data and drawing too many conclusions. I'm not saying that seizures or epilepsy in autism may not be linked to something like early infant feeding choice, but with the same logic, I'd be hard pressed to say there is conclusive evidence of any association yet. And remember, the plural condition of autisms are a very complicated set of conditions indeed.

Finally, quite by coincidence, just yesterday I received an email invite to submit to a Frontiers journal topic headed by Dr Westmark on a favourite topic of mine (diet and the brain) which reminded me of an equally interesting research area on the use of ketogenic diets with autism and epilepsy in mind... more food for thought? (without any medical or clinical advice given or intended)

Music to close. Bowie and Jagger and a great video with some contrasting dancing styles. Condolences also to Sir Mick and the family of L'Wren Scott.


[1] Westmark CJ. Soy infant formula and seizures in children with autism: a retrospective study. PLoS One. 2014 Mar 12;9(3):e80488.

[2] Westmark CJ. et al. Soy-based diet exacerbates seizures in mouse models of neurological disease. J Alzheimers Dis. 2013;33(3):797-805.

[3] Atkinson C. et al. Gut bacterial metabolism of the soy isoflavone daidzein: exploring the relevance to human health. Exp Biol Med (Maywood). 2005 Mar;230(3):155-70.

[4] Yaman H. et al. Epileptic seizures associated with lactose intolerance in a child: A causal relationship? J Pediatr Neurology. 2012; 10: 151-154.

----------- Westmark CJ (2014). Soy infant formula and seizures in children with autism: a retrospective study. PloS one, 9 (3) PMID: 24622158

Sunday 16 March 2014

Obesity and autism (again)

The paper by Phillips and colleagues [1] looking at the prevalence of 'unhealthy weight' in relation to developmentally-defined conditions including autism brought me to writing this update post on the topic of obesity and autism. As you might glean from that previous sentence, obesity and over-weightness(?) has been talked about before on this blog (see here and see here for example). My recent discussion in relation to findings of elevated plasma leptin in cases of autism (see here) adds to the interest in this topic, as does a post on the issue of sedentary behaviours and autism (see here).

I wanna dance... @ Wikipedia 
The Phillips data, based on the body mass index (BMI) calculation suggested a few important trends might need further action based on the primary issue that: "Obesity was elevated among adolescents with all DD [developmental disabilities] types, and was highest among the autism subgroup". Further: "Adolescents with either a DD or obesity had higher prevalences of common respiratory, gastrointestinal, dermatological and neurological conditions/symptoms than nonobese adolescents without DDs". This last point was, in particular, of interest to me because it kinda hinted that issues such as gastrointestinal (GI) problems noted on more than one occasion in relation to autism, may very well tie into overall physical health. By saying that however, I'm not insinuating that all GI problems are 'caused' by obesity when it comes to autism, because as I've talked about before, autism is a very heterogeneous condition in terms of presentation and comorbidity. Oh and perhaps I should introduce the possibility that BMI and autoimmune conditions might be linked at this point too given the 'overlap' between these concepts with autism.

Outside of discussions about the usefulness of the BMI measure, whether 'fat and fit' is a possibility or the growing interest in the notion of the TOFI (which may yet be important to some of those leptin-autism findings), the Phillips paper led to me to have a closer look at some of the research literature in this area over the past couple of years. So, in no particular order, here goes:

The study by Zuckerman and colleagues [2] looked at the frequency of overweight and obesity in a sample of 376 children diagnosed with an autism spectrum condition. They reported that approximately 18% of their cohort were categorised as being overweight and 17% were obese. When it came to looking at other health-related measures and variables for those children falling into one or both of those weight-related categories, one of the primary correlates was that of sleeping difficulties and onwards the use of something like melatonin (the molecular handy-person).

Then we have the study by Egan and colleagues [3] which followed a similar formula in relation to looking at the rates of overweight and obesity in their cohort of 273 children with autism. For them: 17% of their sample were categorised as overweight and nearly 22% had "a BMI percentile in the obese range". Just in case you want a visual representation of what ratio of height to weight corresponds to the obese range, have a look at the graphic at the end of the report shown here. I should also mention that the Egan data also suggested differences in the proportions of overweight and obesity according to differing diagnoses along the autism spectrum which potentially introduces some influence from factors such as cognitive or language ability into the equation. Indeed, even whether certain parts of the autisms might be prone to weight issues? Gut bacteria perhaps... nah.

On the topic of cognitive functions, the paper by de Winter and colleagues [4] more generally looked at the frequency of overweight and obesity in a sample of adults diagnosed with a range of intellectual disabilities including autism. Interestingly their measures were not just confined to BMI but also included "waist-to-hip ratio (WHR) and skin fold thickness". Waist-to-hip ratio has been suggested to be a better marker of health risk than BMI for some conditions (see here). The de Winter results: "Women, people with Down syndrome, higher age, less severe ID, autism, people who are able to eat independently, preparing meals and doing groceries independently, people with physical inactivity and use of atypical antipsychotics were significantly more at risk of being overweight or obese". Quite a few variables there, and certainly quite a few potential avenues of intervention too.

And finally we have the paper by Tyler and colleagues [5]. Again the focus was on adults, and specifically adults with autism (N=108). Based on looking at electronic patient records, various factors were included for analysis and their findings.. "rates of chronic disease included 34.9% for obesity, 31.5% for hyperlipidemia, and 19.4% for hypertension". The figures for obesity are certainly a lot higher than for the other studies but bear in mind this was an investigation based on adults.

What I think we can take from this collected data are a few details. First, and probably most obviously, autism as a diagnosis is not protective against being overweight or even obese. Yes, one has to be careful to bear in mind the important side-effects which various medicines can have on issues like weight gain (depending also on how long they are used for). But I'm pretty sure that not every person with autism falling into those overweight and obese categories for these studies were taking those kinds of medicines, and indeed, even something like melatonin (with some history with regards to autism) might even mitigate some weight effects.

Second, there is some indication that overweight and obesity may be over-represented when a diagnosis of autism is present. I have to say that I'm not totally sold on the idea that there is a vast difference between the frequencies reported in autism and those present in the general population as a whole but I'm willing to concede that autism might place someone towards the top end of any risk of weight issues. Whether this trend changes in line with what seems to be emerging from the examination of US population pediatric obesity rates [6] is fodder for speculation.

Finally, knowing what is out there in the peer-reviewed research literature and how overweight and obesity might place someone at elevated risk of other health complaints and even early mortality, the question must be: what can we do about it? This is perhaps a trickier question that you might first think. Parents, caregivers and professionals can implement strategies to minimise the risk of such issues in childhood (accepting that the relationship between weight and exercise is not necessarily as straight-forward as just do more and lose weight) but for adults... well, adults tend to have a lot more say on how their lives are run than children (see this discussion on Gillick competency). What this means is that even when presented with someone who may have profound difficulties in relation to autism and/or learning disability, one cannot just start making dietary changes or introducing exercise regimes without attempting to infer what they want to do with their lives. Indeed, as per what goes on in the population as a whole, not everyone enjoys doing exercise and not everyone really wants to think about how food choices can influence weight and subsequent health issues. So one needs to tread or even walk carefully.

And just as I post this entry, so I can also direct you to the peer-reviewed open-access mega review on all things autism and obesity from Curtin and colleagues [7]...


[1] Phillips KL. et al. Prevalence and Impact of Unhealthy Weight in a National Sample of US Adolescents with Autism and Other Learning and Behavioral Disabilities. Matern Child Health J. 2014 Feb 20.

[2] Zuckerman KE. et al. Overweight and Obesity: Prevalence and Correlates in a Large Clinical Sample of Children with Autism Spectrum Disorder. J Autism Dev Disord. 2014 Feb 2.

[3] Egan AM. et al. Obesity in young children with autism spectrum disorders: prevalence and associated factors. Child Obes. 2013 Apr;9(2):125-31.

[4] de Winter CF. et al. Overweight and obesity in older people with intellectual disability. Res Dev Disabil. 2012 Mar-Apr;33(2):398-405.

[5] Tyler CV. et al. Chronic disease risks in young adults with autism spectrum disorder: forewarned is forearmed. Am J Intellect Dev Disabil. 2011 Sep;116(5):371-80.

[6] Ogden CL. et al. Prevalence of Childhood and Adult Obesity in the United States, 2011-2012. JAMA 2014; 311: 806-814.

[7] Curtin C. et al. Obesity in children with autism spectrum disorder. Harv Rev Psychiatry. 2014 Mar-Apr;22(2):93-103.


ResearchBlogging.orgPhillips KL, Schieve LA, Visser S, Boulet S, Sharma AJ, Kogan MD, Boyle CA, & Yeargin-Allsopp M. (2014). Prevalence and Impact of Unhealthy Weight in a National Sample of US Adolescents with Autism and Other Learning and Behavioral Disabilities Matern Child Health J DOI: 10.1007/s10995-014-1442-y