Thursday 30 August 2012

What are the early behavioural signs of autism?

Baby stare @ Paul Whiteley
"Autism is a developmental disorder presenting in early infancy". I think most people would agree with this statement; even those who know very little about autism and its intricacies. I can remember years ago using various videos (yes before DVDs et al there were videos) to demonstrate to groups of students how autism might manifest itself in young children and how presentation can differ from child to child depending on lots of different factors.

I struggle to remember the name of the particular program, but one video in particular gave a very personal, very moving account of a young girl with autism, showing home videos of the child growing up accompanied by a mum's description of her and the presentation of her autism. In one scene, the child is shown in her cot aged somewhere between 3 and 6 months old. It was always striking to me how 'detached' the child seemed even at that early age and based on just a few short minutes of commentary and pictures. The lack of eye contact pinpointing the potential for something in her future development.

Fast forward back to modern times (yes with DVDs and Blue-Ray and portable media) and an interesting paper by Barbaro & Dissanayake* brings me to this post looking at some of the various research done on potential early behavioural markers of autism spectrum conditions. I hasten to add that I have sort of touched upon this topic in a very early post on this blog (here) but consider this a bit of a more detailed update.

The Barbaro paper indeed highlights the lack of eye contact in their analysis of potential early markers of autism, alongside another important skill again already discussed on this blog, pointing. 

But what of the other research done in this area?

  • As one might imagine, there is a considerable body of research examining potential early markers of autism. Some names pop up more consistently than others in this area of investigation, one of them being Dr Robyn Young. Dr Young's papers seem to confirm quite a few things that people have suspected for a while. So for example, parents of children with autism are pretty good at spotting when development is not quite as it should be**. Other papers have also said as much, bearing in mind that the experiences when communicating those concerns to certain healthcare professionals might not exactly be 'optimal' in some cases (see this post). 
  • Using home videos as potential screening tools is another area where some important data have been collected. This paper by Dr Young and colleagues*** describes how even with a relatively small participant group, several discriminating features were apparent in pre-verbal children who went on to receive a diagnosis of autism including gaze aversion and issues with 'proto-declarative showing'. Notice again the emphasis on social communicative functions (as per the ADTree classifier work), bearing in mind that pointing generally comes a little later than eye contact in child development terms. Similar joint attention findings were also reported in this study by Watson and colleagues****.
  • Another quite famous name in this area of research is that of Dr Sally Ozonoff (yes, she of the MIND Institute) and her observations down the years. In particular, her paper reviewing the potential usefulness of family home movies and the early development of autism***** deserves mention. 
  • It is however when it comes to examining onset patterns in autism that Dr Ozonoff really comes into her own as per this paper****** (full-text) describing the dimensionality of onset patterns over dichotomous categories.  Again, parents are seen as key partners in 'predicting' autism (see this paper******* full-text). Interestingly, Dr Ozonoff in quite a few of her papers makes mention of the fact that her data does not seem to suggest that symptoms are present at birth as was suggested by Kanner, but rather emerge over time and show a regressive or diminishing factor. A good example is in this study******** (full-text) which prospectively looked at the early signs of autism and suggested that the period between 6 and 12 months seems to be an important one; again re-emphasising the social-communicative features to be paramount. 
  • I should also make reference to this paper by Turner-Brown and colleagues********* which has received some publicity on the First Year Inventory for screening at 12 months of age, just to keep up to date.

I admit that I've only really scratched the surface with this post and indeed only paid lip-service to important concepts like regression; aspects of which have been covered in other posts (see here). I would perhaps refer to an additional reference on the topic of regression and onset patterns with a link to this paper by Cohen-Ophir and colleagues********** (full-text). They reported on three case studies charting a slightly different progression into autism outside of the categorisations of (a) slow / subtle regression and/or (b) rapid regression (or indeed (c) the delays + regression phenotype) suggestive of initial developmental delay, a period of rescuing skills falling into typical development, followed by regression into classic autism. 

There are also a few other areas of potential interest to the issue of early behavioural presentation neatly summed in this paper by Clifford and colleagues************ on temperament in high risk infants and this paper by Flanagan and colleagues************ on head lag (I know this is more presentation rather than behavioural).

So to answer the question to the title of this post, what are the early behavioural signs of autism? It seems that social-communicative factors such as issues with eye contact and later pointing might be possible red flags, but also realising that the heterogeneity noted in the presentation of autism to some degree also mirrors onset patterns and the presentation of early behaviours. 


* Barbaro J. & Dissanayake C. Early markers of autism spectrum disorders in infants and toddlers prospectively identified in the Social Attention and Communication Study. Autism. June 2012.

** Young RL. et al. Parental identification of early behavioural abnormalities in children with autistic disorder. Autism. 2003; 7: 125-143.

*** Clifford S. et al. Assessing the early characteristics of autistic disorder using video analysis. JADD. 2007; 37: 301-313.

**** Watson LR. et al. Communicative gesture use in infants with and without autism: a retrospective home video study. American Journal of Speech-Language Pathology. July 2012.

***** Palomo R. et al. Autism and family home movies: a comprehensive review. Journal of Developmental & Behavioral Pediatrics. 2006; 27: S50-S68.

****** Ozonoff S. et al. The onset of autism: patterns of symptom emergence in the first years of life. Autism Research. 2008; 1: 320-328.

******* Ozonoff S. et al. How early do parent concerns predict later autism diagnosis? Journal of Developmental & Behavioral Pediatrics. 2009; 30: 367-375.

******** Ozonoff S. et al. A prospective study of the emergence of early behavioral signs of autism. Journal of the American Academy of Child & Adolescent Psychiatry. 2010; 49: 256-266.

********* Turner-Brown LM. et al. The First Year Inventory: a longitudinal follow-up of 12-month-old to 3-year-old children. Autism. July 2012.

********** Cohen-Ophir M. et al. Autism in early childhood: an unusual developmental course - three case reports. Case Reports in Psychiatry. 2012.
DOI: 10.1155/2012/946109

*********** Clifford SM. et al. Temperament in the first 2 years of life in infants at high-risk for autism spectrum disorders. JADD. August 2012.

************ Flanagan JE. et al. Head lag in infants at risk for autism: a preliminary study. American Journal of Occupational Therapy. 2012; 66: 577-585.

--------- Barbaro J, & Dissanayake C (2012). Early markers of autism spectrum disorders in infants and toddlers prospectively identified in the Social Attention and Communication Study. Autism : the international journal of research and practice PMID: 22735682

Tuesday 28 August 2012

ADTree reloaded: classifying autism based on 7 ADI-R items?

A lengthy quote to begin this post:

"Deploying a variety of machine learning algorithms, we found one, the Alternating Decision Tree (ADTree), to have high sensitivity and specificity in the classification of individuals with autism from controls. The ADTree classifier consisted of only 7 questions, 93% fewer than the full ADI-R, and performed with greater than 99% accuracy when applied to independent populations of individuals with autism, misclassifying only one out of the 1962 cases used for validation".

Interested? The quote comes from this recent paper by Dennis Wall and colleagues* (full-text) following up some related research from this group based on another gold standard autism assessment schedule, the ADOS (see here).

I'm not going to dwell too much on the study details aside from saying:

  • The ADI-R (Autism Diagnostic Interview - Revised) is probably one of the more time and resource-intensive interview questionnaires used for the assessment of autism, coming in at 93 questions long and asking about current behaviour and 'most abnormal' (aged 4-5 years) if appropriate.
  • An abbreviated ADI-R would therefore be quite a useful measure in these austere times; particularly one which could maintain the same level of accuracy as completion of the full schedule.
  • Similar to their previous study, Wall and colleagues applied various machine learning algorithms (n=15) to ascertain whether any might be able to determine which ADI-R items are most relevant to diagnosis based on the AGRE dataset.
  • Once again, the Alternating Decision Tree (ADTree) model performed best: perfect sensitivity (1.0), a low false-positive rate (0.013) and "overall accuracy of 99.9%".
  • Seven items of the ADI-R comprised the ADTree model: (i) comprehension of simple language at most abnormal (4-5 years), (ii) reciprocal conversation (regarding the ability to facilitate the flow of conversation), (iii) use of imaginative / pretend play at most abnormal (I have post about pretend play coming up fairly soon), (iv) social imaginative play with peers at most abnormal, (v) direct gaze at most abnormal, (vi) group play with peers at most abnormal (spontaneous games or activities) and (vii) age when abnormality was first evident.
  • Testing accuracy was again carried out on participant data from the Boston Autism Consortium (AC) and the Simons Simplex Collection (SSC) from where the quite compelling data for the success of the ADTree model was derived.

So the ADI-R has been boiled down to 7 pertinent items. The ADOS boiled down to 8 distinguishing module 1 items. I think most people would stand up and take note of these findings even if further replication is still required (based on different geographical groups for example). Don't get me wrong, there is still a large degree of skill required to deliver the ADI-R and ADOS and maintain your reproducibility and diagnostic prowess so I don't think this combined data will be putting people out of work just yet; certainly not with the number of people estimated to be coming through the diagnostic process. That and the fact that these are assessment instruments and so are subservient to a final clinical opinion for an autism diagnosis or not.

Aside from the grand findings I am interested in the types of behaviours which are noted to be important for diagnosis. All very 'social-communicative' (sounds familiar) and not at all heavy on the 'restricted and repetitive behaviour' side of things. Indeed, looking back at the ADOS ADTree paper, I might be wrong but only one element, 'functional play with objects' seems to have any strong relation to issues with repetitive behaviours. This could be that we aren't asking the right questions about this area of behaviour, but I would hedge my bets that more likely is the stress on the social-communicative side of presentation as being key to diagnosis. I think I might have to look at this further in future posts.

Finally, I have previously talked about the lack of instruments to appropriately and accurately assess 'change' in autism (as a function of maturation or intervention or anything else). Y'know with these combined data, I think we might have the outline of something really quite useful...


* Wall DP. et al. Use of artificial intelligence to shorten the behavioral diagnosis of autism. PLoS ONE. 2012; 7: e43855.

---------- Dennis P. Wall, Rebecca Dally, Rhiannon Luyster, Jae-Yoon Jung, & Todd F. DeLuca (2012). Use of artificial intelligence to shorten the behavioral diagnosis of autism PLoS ONE : 10.1371/journal.pone.0043855

Monday 27 August 2012

The kynurenic acid hypothesis of schizophrenia

I'm back to tryptophan again in this post. No matter how hard I try, I just can't seem to get away from this interesting aromatic amino acid and its seemingly far-reaching effects on many aspects of human health.

Kynurenic acid @ Wikipedia
Today I'm exploring an interesting hypothesis looking at how a metabolite of tryptophan, kynuernic acid, might hold some connection to cases of schizophrenia in the so-called kynurenic acid hypothesis of schizophrenia*.

OK, probably best to start with a few caveats. Me = not an expert on schizophrenia, is probably the biggest caveat and so apologies in advance for any errors or important omissions you might find in this post. Schizophrenia, from a conceptual point of view, probably shares some similarity with quite a few other behaviourally defined conditions like autism, insofar as being defined as a spectrum condition. This, together with the risk/effect of certain comorbidity, almost certainly implies that finding 'specifics' in terms of universal theories of causation or effect, are probably going to be difficult at best. Many roads might lead to Rome.

Schizophrenia unlike autism however, does not seem to be a developmental condition present from early infancy (despite the history linking the two conditions). This is not to say however that there may not be a strong genetic component to schizophrenia if you like 'waiting in the wings' to express itself as maturation and environment play their hands. Instead schizophrenia has been associated with different stages of symptom presentation: a prodromal period, an acute phase and a relapse phase (see here) with some marked inter-individual variation on the timing of these phases. One therefore has to be quite careful when ascribing markers or generic theories to schizophrenia based on all these factors.

Back to the kynurenic acid theory, and the finding of elevated kynurenic acid (kynurenate) in post-mortem brain samples from people with schizophrenia** represents one of the first discoveries of some possible connection. The connection between elevated levels of kynurenate in certain brain areas is matched by a suggestion of reduced glutamate receptor function. Reports of elevations of kynurenic acid in cerebrospinal fluid in cases of schizophrenia followed*** and not just once (here and here). 

Then things start to get a little more speculative as questions start being asked as to why elevated kynurenic acid is there in the first place. The enzymes (and their cofactors) along the pathway to the formation of kynurenic acid have come under scrutiny as for example, per this preliminary report from Holtze and colleagues**** (full-text) on SNPs in the kynurenine 3-monooxygenase (KMO) enzyme. That and reduced levels of mRNA and lower enzyme activity have been found*****. Another enzyme suggested to show some involvement in this tangled hypothesis is that of indoleamine 2,3-dioxygenase (IDO). The suggestion is that the immune system may be able to affect the functioning of IDO in cases of schizophrenia (here) and hence increase production of kynurenic acid with some interesting knock-on effects based on the antagonistic effects on things like N-methyl-D-aspartate (NMDA) and its receptor.

Fair enough. But are the findings actually related to schizophrenia or purely epiphenomenal? There have been a few clues suggesting specific effects from elevated kynurenic acid as per this paper****** on various cognitive functions related to schizophrenia. Indeed extrapolating from rodent studies seems to have been quite a popular thing to do with kynurenic acid in mind as per other papers (here, here and here). There are some interesting themes to this work focused on things like the timing of kynurenic acid exposure; so, adolescence seems to be quite a sensitive period. I assume this makes such work all the more 'attractive' given the timing of symptom onset (see here) in many cases of schizophrenia.

On balance, the collected evidence does seem to be at least pointing the way to kynurenic acid elevations as being related to cases of schizophrenia. But it doesn't just stop there. I've already mentioned a possible role for immune function in accounting for kynurenic acid levels in schizophrenia. Various infectious agents have been suggested to show some 'connection' to activation of the kynurenine pathway including influenza A (here) and an old friend, Toxoplasma gondii (here). Of course one has to be quite careful not to put all your inflammatory eggs in one basket when it comes to immune function and inflammation.

Accepting all this collected data, the question then turns to what can be done about kynurenic acid and any excessive production and what implications that might have for the presentation of schizophrenia. I think I've probably said it before but I will repeat myself: when we talk about medication to treat/manage this condition or that condition or any condition, interventions don't just generally affect one system and one system alone, they most likely with affect lots of different systems and exert quite a few effects. Antipsychotics for example, have been suggested to be quite good antiparasitics also (see here) bearing in mind the T.gondii link suggested with cases of schizophrenia. So it is with kynurenic acid in mind, as Myint and colleagues******* demonstrated in their study looking at the effects of antipsychotics on the kynurenine pathway among other things.

Perhaps even more surprising is the suggestion that medications which target some of the processes involved with inflammation such as the COX-2 inhibitors might also show some potential with kynurenic acid and schizophrenia in mind (here) together with a growing evidence base on the use of non-steroidal anti-inflammtory drugs (NSAIDs) for cases of schizophrenia (here). At this point I will stress that I am not providing or intending to provide medical advice nor endorsement about these strategies.

I've focused on schizophrenia and kynurenic acid in this post but recognise that this might not be an exclusive relationship bearing in mind all the issues previously cited on diagnosis and symptom presentation. Indeed I was particularly drawn to this paper******** by McFarlane and colleagues looking at the Dangermouse that is the BTBR mouse (see this post), and how hidden away in all the findings of this mouse model, issues with KMO might just have some function in cases of autism. Perhaps an area ripe for further investigation and in particular, when overlap exists in the dual presentation of autism and schizophrenia?

This post has been quite a brief overview of the potential role of kynurenic acid in cases of schizophrenia and as such I've only scratched the surface of the potential meaning of this work and tie-ups with other areas and other theories. What I hope I've demonstrated is that once again, looking at amino acid chemistry might hold some valuable clues about behaviourally-defined conditions, and in particular how the aromatic amino acids seem to be potentially big players in such conditions.


* Erhardt S. et al. The kynurenic acid hypothesis of schizophrenia. Physiology & Behaviour. 2007; 92: 203-209.

** Schwarz R. et al. Increased cortical kynurenate content in schizophrenia. Biological Psychiatry. 2001; 50: 521-530.

*** Erhardt S. et al. Kynurenic acid levels are elevated in the cerebrospinal fluid of patients with schizophrenia. Neuroscience Letters. 2001; 313: 96-98.

**** Holtze M. et al. Kynurenine 3-monooxygenase polymorphisms: relevance for kynurenic acid synthesis in patients with schizophrenia and healthy controls. Journal of Psychiatry & Neuroscience. 2012: 37: 53-57.

***** Wonodi I. et al. Downregulated kynurenine 3-monooxygenase gene expression and enzyme activity in schizophrenia and genetic association with schizophrenia endophenotypes. Archives of General Psychiatry. 2011; 68: 665-674.

****** Akagbosu CO. et al. Exposure to kynurenic acid during adolescence produces memory deficits in adulthood. Schizophrena Bulletin. December 2010.

******* Myint AM. et al. Reversal of imbalance between kynurenic acid and 3-hydroxykynurenine by antipsychotics in medication-naïve and medication-free schizophrenic patients. Brain, Behavior & Immunity. 2011; 25: 1576-1581.

******** McFarlane HG. et al. Autism-like behavioral phenotypes in BTBR T+tf/J mice. Genes, Brain & Behavior. 2008; 7: 152-163.

Sunday 26 August 2012

Autism and immune dysfunction: don't forget your roots?

You're going to have to humour me a little with this post based on this opinion piece by Moises Velasquez-Manoff which appeared in the New York Times recently. I normally try and stick to discussing peer-reviewed published research on this blog but given the content of this article titled: "An immune disorder at the root of autism" (no question mark by the way), I couldn't resist having a more detailed look at some of the discussions which do include reference to some peer-reviewed research.

So what does it say?

I don't think I've missed anything out. The main message seems to be that increasing autism rates seem to be correlated with population increases in hygiene and less exposure to bacterial and viral infections and the knock-on effects to immune function. Actually this is not the first time that Mr Velasquez-Manoff has talked about some of these concepts are per this article from 2008 on worm therapy for inflammatory bowel disease (IBD) which is also the topic of his book (no plug intended).

There is no doubt that this is an interesting article which has some degree of support (published scientific support) behind elements of it. Outside of the journalist flair that accompanies the article - "perhaps one-third, and very likely more — looks like a type of inflammatory disease" (evidence please!) autism and immune function is an area of great research interest and speculation. Still however questions remain about how widespread immune dysfunction is in autism, the chicken or egg question of which came first: autism or immune dysfunction, and indeed whether immunological issues are a core part of autism, some cases of autism, or perhaps just another face of comorbidity.

[Update 31 August 2012: In response to some discussion about the lack of scientific references accompanying the article by Moises Velasquez-Manoff, the author has posted a reference source list on his website].


Patterson P. Maternal infection and immune involvement in autism. Trends in Molecular Medicine. 2011; 17: 389-394.

Stick with me kid: adhesion molecules and autism

Spider web by Luc Viatour /
A paper by Charity Onore and colleagues* from the MIND Institute caught my eye quite recently. The topic was adhesion, and in particular cell adhesion related to immune function; and an analysis of some of the primary players in this sticky process such as the selectins and other interesting molecules including CD31 otherwise known as PECAM-1 (see here for an explanation) in relation to autism. From the outset, I should clarify that this is not a post on the neuronal cell adhesion research that has been undertaken on autism (see here). As if I would know where to start on that research area...

Regular readers of this blog might already understand that I have a continued interested in the output from the MIND Institute with autism in mind, much of which has been summarised in another paper from Dr Onore** on the role of the immune system in cases of autism. Indeed just reading the abstract to this summary paper makes me further realise how complex and 'real' the immune system findings are to autism - some cases of autism. These are not just funny abbreviations like IL-6 or INF but very real elements to what is a very complicated condition. An interesting opinion piece arrives at similar conclusions.

Anyway, back to the topic/paper at hand, cell adhesion, and probably best if I try and translate a few of the concepts outlined in the recent paper as well as summarising what was found, just so you know how the findings might fit in.

  • 'leukocyte transendothelial migration' basically refers to the process of getting leukocytes (white blood cells) to the site of infection or injury. If you can bear some pretty heavy biochemistry, this paper by Liu and colleagues*** (full-text) offers quite a detailed description of what's potentially involved in the process.
  • 'Levels of sPECAM-1 and sP-selectin were significantly reduced in the ASD group'. The selectins are kinda the first response in recruiting leukocytes to where they're needed. From what I can gather, they 'tether' white blood cells (leukocytes) to the endothelial cells to facilitate rolling to the site of injury and inflammation (see here). I suppose it's kinda like a velcro response although perhaps not so permanent (see here). Lower levels of [plasma soluble] sP-selectin might indicate some issue with the ability to perform these adhesive duties bearing in mind that the other selectins, L-selectin and E-selectin also play an important role. Given the link between elevations of sP-selectin as a marker for platelet / endothelial activation (among other things), one has to suggest that these results are perhaps reflective of hypo-activation compared with controls. Similar findings have been reported previously**** (full-text).
  • Lower levels of platelet endothelial adhesion molecule-1 (PECAM-1) (see here for a good overview) were also reported. PECAM-1 seems to perform a similar role with regards to adhesion. Interestingly, this is again not the first time that lower levels of PECAM-1 have been reported in cases of autism*****. 
  • "Soluble PECAM-1 levels were negatively associated with repetitive behavior and abnormal brain growth in children with ASD (p = .03)". In other words, low levels of PECAM-1 were linked to elevated levels of repetitive behaviours and unusual head growth (again as per the Tsuchuiya findings).

I can't claim to be a world's expert on this area of autism research so won't go to far further into the possible meaning. The authors conclude that these adhesion molecules "modulate the permeability and signaling at the blood-brain barrier". That is true as per reviews like this one from Kalinowska & Losy****** on PECAM-1 and neuroinflammation. Unless I'm reading this wrong, the only issue being that elevated levels of PECAM-1 seem to be related to such issues; the current results were kinda sailing against that tide at least in their cohort as a group.

Taken at face value, these results may suggest that the various ways and means that the body deals with inflammation may be perturbed in some cases of autism. Without actually knowing whether inflammation (acute or chronic) was present in the described cases of the current study, it is slightly difficult to make too much of the findings as they stand. Reiterating that the involvement of the immune system in cases of autism is far from simple - some say presentation of under-activation (as per this article on the immunoglobulins by the MIND Institute) others say over-activation (see here) - one has to be cautious about making sweeping generalisations. Realising also that the immune system is not just one system, as per the innate vs. adaptive classifications.

A final note. One added bonus to the current paper was the relationship with the Autism Phenome Project (APP) which very importantly realises that autism is an umbrella term to describe lots and lots of different presentations; not necessarily all the same thing however in terms of aetiology or trajectory (see the 'bloomers' post). Indeed based on the previous results reported for the APP looking at brain enlargement and regression in autism (see here), there is perhaps some overlap in participants taking part in the current study also.

To finish, the sad news of the death of a true icon of the age, Neil Armstrong, brings to mind an apt song, Man on the Moon by REM. Truly one small step, one giant leap.


* Onore CE. et al. Levels of soluble platelet endothelial cell adhesion molecule-1 and P-selectin are decreased in children with autism spectrum disorder. Biological Psychiatry. June 2012.

** Onore CE. et al. The role of immune dysfunction in the pathophysiology of autism. Brain, Behavior & Immunity. 2012; 36: 383-392.

*** Liu Y. et al. Regulation of leukocyte transmigration: cell surface interactions and signaling events. Journal of Immunology. 2004; 172: 7-13.

**** Iwata Y. et al. Serum levels of P-selectin in men with high-functioning autism. British Journal of Psychiatry. 2008; 193: 338-339.

**** Tsuchuiya KJ. et al. Decreased serum levels of platelet-endothelial adhesion molecule (PECAM-1) in subjects with high-functioning autism: a negative correlation with head circumference at birth. Biological Psychiatry. 2007; 62: 1056-1058.

***** Kalinowska A. & Losy J. PECAM-1, a key player in neuroinflammation. European Journal of Neurology. 2006; 13: 1284-1290.

---------- Onore CE, Nordahl CW, Young GS, Van de Water JA, Rogers SJ, & Ashwood P (2012). Levels of Soluble Platelet Endothelial Cell Adhesion Molecule-1 and P-Selectin Are Decreased in Children with Autism Spectrum Disorder. Biological Psychiatry PMID: 22717029

Thursday 23 August 2012

De novo mutations, older dads and autism (again)

Those with an interest in autism (and/or schizophrenia) will have probably already seen the headlines discussing the study by Augustine Kong and colleagues* on  the rate of de novo genetic mutations and a father's age as being potentially important for conditions like autism and schizophrenia. For those like me who are still struggling with all things mutation, quite a nice summary of the current research is offered in this Nature commentary accompanying the study.

I'm going to briefly focus on the potential implications specifically for autism in this post, accepting that (a) de novo mutations have been looked at with schizophrenia in mind (see here) and (b) increasing parental (and grand-parental) age has been linked to an increased risk of schizophrenia (see here).

This is not the first time this year that findings related to de novo mutation and paternal age in relation to autism have surfaced in Nature as discussed in this post. The latest study from Kong and colleagues suggests a few things:

  • The entire genome of 78 families (mum, dad, child) of Icelandic-origin was sequenced.
  • They studied the small genetics changes, SNPs, in and between mums and dads and their children, taking into account the age of the parents. Apparently of the offspring included for study, 44 had received a diagnosis of an autism spectrum disorder and 21 diagnosed with schizophrenia.
  • Their analysis suggested (i) fathers passed on about four times more mutations than mothers, (ii) approximately 2 new mutations were present in children for every year of increase in the father's age at conception - so a father conceiving at 20-years old passed on 25 random mutations compared with a father conceiving at 40-years old who passed on 65 mutations, (iii) mums were reported to pass on about 15 mutations irrespective of age; thought due to the fact that women carry their eggs through life whilst sperm is constantly being produced and therefore potentially subject to the rigours of ageing and environment.

There has, understandably, been some discussion about the implications of this research with regards to autism and more generally about the trend towards couples starting families when they're getting on a bit. Kari Stefansson, lead author on the paper discusses the trend (in Iceland) towards increasing paternal age when conceiving a child - 27.9 years in 1980 to 33 years in 2011 - as a case in point and potentially what implications this may have. That being said, I have to say that I raised a smile when reading about this recent research which suggested that children with older fathers might actually be genetically-programmed to live longer as a function of increasing teleomere length with age in sperm**. I am comparing apples and oranges here but want to show how such things are never straight-forward.

Another debate has also seemingly arisen from the Kong study with regards to the numbers of children with autism being currently diagnosed/estimated, and whether the figures reflect better case ascertainment, etc. or are reflective of a true increase in the numbers of people with autism. To quote from Fred Volkmar (here) "This study provides some of the first solid scientific evidence for a true increase in the condition of autism". Earlier this year I discussed the latest estimates of autism in the United States produced by the CDC and how an estimated 1 in 110 8-year olds with autism became 1 in 88 8-year olds with autism. If accurate, what Dr Volkmar seems to be suggesting is a kind of bridge between the genetics-environment camps which will almost certainly impact on the autism numbers game: older parents, older fathers passing on a greater number of mutations potentially increasing the risk of autism or schizophrenia. I pass no judgement on this proposal by the way.

There are a few final comments to make on this research and the area in general. Regular readers of this blog might already know that I am becoming a bit of a fan of the area of epigentics and the promise that it might hold for conditions like autism. Noting the recent research from Richard Anney and colleagues ("no single SNP shows significant association with ASD or selected phenotypes at a genome-wide level") it's all well and good suggesting that older fathers might increase the risk of autism, but without such details as to which SNPs are consistently and reliably at work, there is likely to be a whole lot more involved in the aetiology of autism than just the structure of the genome. That and my continuing questioning as to whether age is the only factor in causing mutation or whether other environmental 'exposures' might also exert an effect.

Additionally and finally, I've said it before and I'll say it again: autism is not a homogeneous condition. We might talk about autism as being diagnostically homogeneous insofar as a person presents with this triad (soon to be dyad) of symptoms and the autism diagnosis box is ticked, but symptom onset and presentation varies from person to person (endophenotype to endophenotype?) and is further complicated by elevated risk for various comorbidity, psychological, cognitive and also more somatic comorbidity. I see nothing in the current research that rules out the possibility that more mutations passed to offspring as a result of older dads might not also be tied into an increased risk for other things (autoimmune features and conditions, epilepsy, learning disability, depression, anxiety, etc., etc.). Until we can, with some degree of confidence, peel back the diagnostic and risk layers, this issue will be in the background of any autism research.

Don't get me wrong, I am very interesting in the latest results from Kong and the now replicated findings of advancing paternal age and risk in some cases of autism. I do however believe that within the context of all the other research published and on-going in autism, this might be but one small piece of a much larger puzzle on aetiology and underlying pathology.

To finish, Jim Morrison has been on my musical radar of late, so how about some Riders on the Storm?


* Kong A. et al. Rate of de novo mutations and the importance of father’s age to disease risk. Nature. 2012; 488: 471-475.

** Eisenberg DTA. et al. Delayed paternal age of reproduction in humans is associated with longer telomeres across two generations of descendants. PNAS. 2012; 109: 10251-10256.

---------- Augustine Kong, Michael L. Frigge, Gisli Masson, Soren Besenbacher, Patrick Sulem, Gisli Magnusson, Sigurjon A. Gudjonsson, Asgeir Sigurdsson, Aslaug Jonasdottir, Adalbjorg Jonasdottir, Wendy S. W. Wong, Gunnar Sigurdsson, G. Bragi Walters, Stacy Steinberg, Hannes Helgason, Gudmar Thorleifsson, Daniel F. Gudbjartsson, Agnar Helgason, Olafur Th. Magnusson,, Unnur Thorsteinsdottir, & Kari Stefansson (2012). Rate of de novo mutations and the importance of father’s age to disease risk Nature DOI: 10.1038/nature11396

Tuesday 21 August 2012

Resurrecting XMRV in Chronic Fatigue Syndrome?

The letters X-M-R-V in relation to Chronic Fatigue Syndrome (CFS) have had their fair share of controversy in recent years. For those who don't know about XMRV - xenotropic murine leukemia virus-related virus - there is ample discussion about this topic on the web or you can browse my past posts covering the story here and here. I'm sure that a feature film or TV mini-series will be made eventually about XMRV and CFS including the back-story of Dr Judy Mikovtis and the whole WPI saga. I'll leave you to make up your own film title.

After all the excitement of contamination, missing notebooks and scientific retractions, you'd think this would be a 'leave well alone' area for many researchers not really wanting to get involved, given the career-tarnishing effects it seems to have. It was with some interest therefore that I stumbled across this fairly recent article by Paolucci and colleagues* (full-text) and their findings on XMRV and CFS adding to the mix.

It is a short paper but nevertheless contains some interesting facts which I'll try and summarise:

  • Twelve participants diagnosed with CFS and some 40 controls (including asymptomatic, those confirmed as HIV positive or hepatitis C infection and transplant recipients) were investigated for the presence of XMRV "and polytropic MLV-related provirus". This last bit taps into some now author retracted research published by Lo and colleagues** (full-text).
  • I can't claim to understand all the ins-and-outs of the process for trying to detect XMRV or its friends but PCR was the name of the game, looking for fragments of DNA from the respective viruses.
  • Results: nothing for the control participants when it came to the potential presence of XMRV or its friends. For the CFS participants, in the majority nothing also except in two participants who "were positive for gag PCR and integrase PCR respectively". Further study of positive participant number 1 suggested that "the retrovirus was more closely related to the polytropic MLV rather than to XMRV (identity 100% vs. 96%)". For participant 2, further analysis "confirmed the similarity with the XMRV sequence".
  • Given the methods used by the authors, the contamination issue that overshadowed the previous research were excluded.

I'm not making too much of this paper at the current time given that other contemporary studies have reported no presence of XMRV in cohorts of participants with CFS (see here and here) alongside the relatively small participant group included as part of the Paolucci study. As per my previous reporting of this area of research, detecting such viruses is not by any means an easy task and no doubt there are still gaps in the methodology and science that need patching up. Even with the virus detected, there is a substantial leap to make before one is able to suggest the virus might be 'causative' of CFS or anything else.

What we can perhaps infer from the Paolucci paper is that the XMRV chapter of CFS research history might not be ready for the scientific scrapheap just yet. The big 'final say' study from virus hunter Ian Lipkin - yes, he of the autism and Sutterella research - has, at the time of writing, still not found its way into the peer-reviewed literature as far as I can see despite some signs it would be here already.

So we wait and see...

While waiting perhaps you would like to hear which Disney song currently fills my house. I've got a dream y'know..


* Paolucci S. et al. Xenotropic and polytropic murine leukemia virus-related sequences are not detected in the majority of patients with chronic fatigue syndrome. New Microbiologica. 2012; 35: 341-344.

** Lo SC. et al. Detection of MLV-related virus gene sequences in blood of patients with chronic fatigue syndrome and healthy blood donors. PNAS. 2010. 107: 15874-9 [RETRACTED].

--------- Paolucci S, Piralla A, Zanello C, Minoli L, & Baldanti F (2012). Xenotropic and polytropic murine leukemia virus-related sequences are not detected in the majority of patients with chronic fatigue syndrome. The new microbiologica, 35 (3), 341-4 PMID: 22842604

Monday 20 August 2012

Autism and microglia

An interesting paper has appeared by Beumer and colleagues* with the grandiose title: The immune theory of psychiatric diseases: a key role for activated microglia and circulating monocytes. It's an intriguing paper linking specific actions of the immune system on brain areas in connection to various conditions including depression and schizophrenia via a biological entity that seems to be gaining some research interest: microglia.

On the basis of this publication I therefore resisted no longer and decided to set off on the voyage of discovery that is a possible role for microglia in cases of autism. Please understand that I undertake this post with some trepidation as my skill set does not readily encompass the hows and whys of microglia, so be at the ready with that rather large pinch of salt.

Quite a good overview of microglia is here by Kettenmann and colleagues** (full-text) alongside some discussion on their 'constant gardening' effect in this piece by Virginia Hughes*** (full-text). Apparently constituting as much as 10% of the cells of the central nervous system, microglia derive from cells of the immune system and activate in response to injury or inflammation. More than that however, there is some emerging evidence that microglia might also play a role in the pruning process which the developing brain undertakes during the early years and perhaps beyond.

With autism specifically in mind, I've already linked in previous posts to an interesting blog piece from Paul Patterson on microglia potentially eating synapses in autism as a consequence of some interesting research suggesting increased microglial activation in some cases of autism****. I draw your attention also to another interesting blog post on the SFARI website which strengthens a possible connection between microglia and a mouse model of Rett syndrome, noting also the controversial issue of bone marrow transplant as a potential route to "generating healthy microglia". I might add that I am not getting too obsessed with stem cells and the such like, but in light of Prof. Patterson's recent paper, one has to keep an open mind.

Other research on microglia and autism:

  • Starting from the early days of the PubMed catalogue on this topic a few familiar names crop up with microglia in mind. Andrew Zimmerman and colleagues***** (he of the 'can you grow out of autism' research) talked about neuroglial activation and neuroinflammation quite a few years ago. They reported "marked activation of microglia" alongside several other inflammatory related findings in the brains of people with autism. Similar follow-up studies from this research group also talked about paradoxical lower levels of quinolinic acid in cerebrospinal fluid which  brings us back to everyone's favourite aromatic amino acid, tryptophan. I'll leave that for now.
  • Propionic acid (PPA), the topic of quite a recent blog post, also gets a look-in with microglia in mind as per the paper by Derek MacFabe and colleagues******. They report that during their 'inject PPA into rodent brain' experiments, increases in activated microglia was one of the effects contributing to the neuroinflammatory process described. Whilst I would love to be able to say this might tie into the short chain fatty acid findings also talked about with PPA and autism in mind (here), it is probably not unexpected that PPA administration so directly and at such quantities would provoke a reaction from microglia given their activation at the first sight of any foreign invaders.
  • Other research speculates on the role of microglia and autism (here, here and here), in some cases speculating on the potential environmental factors which may exacerbate microglial priming. I was also drawn to this paper by Schwarz & Bilbo******* who discuss how gender differences in the colonisation and function of glia may offer some interesting explanations for the gender differences seen in conditions like autism, acknowledging that glia includes more than just microglia.

There is still some way to go with regards to the microglia-autism story but so far the research base could be described as very interesting. Inflammation, as I have said before, seems to be playing many hands in at least some cases of autism. Whilst the chicken-and-egg debate about immune function and autism continues, there is little doubt that for some, immune involvement - whether mediated/described by microglia activation, circulating cytokines or other markers of inflammation and immune over- or -under activation, seems to be a facet of symptom presentation. The puzzle is still far from complete however.

Finally, there is the question about what can be done about over or chronic activation of microglia. Reiterating my caveat about not giving medical advice or anything approximating such advice and aside from the bone marrow transplant data already touched upon, I was interested to read the various literature on the effect of administration of the tetracycline antibiotic, minocycline on microglial activation as per studies likes this one and this one (both full-text). Accepting that certain antibiotics might be a bit of a double-edged sword for some cases of autism (see this post) as well as such compounds having other, more recognised antimicrobial action - potentially important in some cases of autism - it does make me wonder whether targeting such interventions at activated microglia at a critical time might be a research area worth investing in.

To finish, I'm in the mood for some Ska. So without further ado, the Specials and Too Much Too Young.


* Beumer W. et al. The immune theory of psychiatric diseases: a key role for activated microglia and circulating monocytes. Journal of Leukocyte Biology. August 2012.

** Kettenmann H. et al. Physiology of microglia. Physiological Reviews. 2011; 91: 461-553.

*** Hughes V. Microglia: the constant gardeners. Nature. 2012; 485: 570-572.

**** Morgan JT. et al. Microglial activation and increased microglial density observed in the dorsolateral prefrontal cortex in autism. Biological Psychiatry. 2010; 68: 368-376.

***** Vargas DL. et al. Neuroglial activation and neuroinflammation in the brain of patients with autism. Annals of Neurology. 2005; 57: 67-81.

****** MacFabe DF. et al. Neurobiological effects of intraventricular propionic acid in rats: possible role of short chain fatty acids on the pathogenesis and characteristics of autism spectrum disorders. Behavioral Brain Research. 2007; 176: 149-169.

******* Schwarz JM. & Bilbo SD. Sex, glia, and development: Interactions in health and disease. Hormones & Behavior. February 2012.

--------- Beumer W, Gibney SM, & Drexhage RC (2012). The immune theory of psychiatric diseases: a key role for activated microglia and circulating monocytes J Leukoc Biol. DOI: 10.1189/jlb.0212100

Friday 17 August 2012

Vitamin D and autoimmunity and autism

Heavens above @ Paul Whiteley
They're at it again. The very marvellous Saudi-Egyptian autism research tag-team that is Gehan Mostafa and Laila Al-Ayadhi and another very, very interesting article titled: Reduced serum concentrations of 25-hydroxy vitamin D in children with autism: relation to autoimmunity* (full-text here).

I'd kinda been waiting for an article like this to come along given the shifting Eye of Sauron onto a possible role for the sunshine vitamin in cases of autism. You can find a little more background on the whole vitamin D-autism research base in a previous post I wrote some while back (here). I might also at this point plug a few other previous posts focused on vitamin D: leaky gut? (here), mental health and vitamin D deficiency? (here) and synthesis of vitamin D and the "evil" cholesterol (here). OK, stop with the self-promotion already.

The latest paper is full-text but to briefly summarise:

  • This wasn't just a study to see whether children with autism were deficient in vitamin D, or rather the active form of the vitamin, 25-hydroxy vitamin D, but rather also some inspection of any link between vitamin D levels and a feature of autoimmunity given the proposed immunological effects of vitamin D and the previously discussed suggestion of a link between autoimmunity and cases of autism.
  • Fifty children with autism (aged 5-12 years) were compared against 30 asymptomatic control children. Sunshine exposure data were collected, supplementation controlled for, and serum levels of 25-hydroxy vitamin D were ascertained during the spring/summer months (April-September).
  • Levels of anti-myelin-associated glycoprotein (anti-MAG) auto-antibodies were also measured in both groups. The presence of anti-MAG antibodies is basically a sign that the body is attacking myelin associated glycoproteins; myelin being an essential part of proper neuronal functioning. Elevations in anti-MAG antibodies is a sign that brain tissue is being targeted by the bodies own immune system.
  • The results: as a group, the children with autism presented with lower levels of 25-hydroxy vitamin D than controls (P<0.001), with about 40% of the autism group being classified as vitamin D deficient (<10ng/mL). None of the controls fell into the deficient range (although about 20% were classified as being 'insufficent' <30ng/mL). These results were based on pretty much equivalent sunshine exposure between the groups. There was also a suggestion of a negative correlation between 25-hydroxy vitamin D levels and scores on the CARS.
  • Children with autism as a group also presented with significantly higher levels of anti-MAG auto-antibodies compared to controls (P<0.001). Indeed increased levels were found in about 70% of participants with autism.
  • Serum 25-hydroxy vitamin D levels also significantly negatively correlated with anti-MAG auto-antibodies in children with autism.

Accepting that the participant groups were relatively small and the findings based on children I assume of Saudi origin, it is quite telling that children with autism living in a country where sunshine is probably not in short supply were still in the majority deficient in the active form of vitamin D. Appreciating the health advice that is provided nowadays when it comes to being 'sun aware' I still find some difficulty in accepting the fact that people in this region might not be getting enough sunshine to synthesise vitamin D as a way of accounting for the results. Could it be that autism, some cases of autism, confer problems with the biological synthesis and metabolism of vitamin D or even with the functioning of vitamin D receptors? If so, what are the implications for supplementation for example?

Anti-MAG antibodies is a new one for me in the autism research arena. Linked to a variety of autoimmune peripheral neuropathies, anti-MAG antibodies have actually been reported previously in cases of autism** (by the same authors). In that study, the authors introduced the possibility that "autism could be, in part, one of the pediatric autoimmune neuropsychiatric disorders". I'm not going to get too bogged down in the potential impact of anti-MAG antibodies and autism at this point though as more reading on my part is required.

So in short, a few more potential pointers for further investigations from the Saudi-Egyptian research team and another intriguing connection between vitamins and autism.  

To finish, I'm in the mood for a little Stranglers at the moment and a lovely tune called Gordon Brown... sorry, Golden Brown to at least remember what the sun looks like.


* Mostafa G. & Al-Ayadhi LY. Reduced serum concentrations of 25-hydroxy vitamin D in children with autism: relation to autoimmunity. Journal of Neuroinflammation. August 2012.

** Mostafa GA. et al. Serum anti-myelin—associated glycoprotein antibodies in Egyptian autistic children. Journal of Child Neurology. 2008; 23: 1413-1418.

Mostafa G, & Al-Ayadhi L (2012). Reduced serum concentrations of 25-hydroxy vitamin D in children with autism: relation to autoimmunity Journal of Neuroinflammation
DOI: 10.1186/1742-2094-9-201

Thursday 16 August 2012

Are gut problems in autism linked to anxiety and sensory issues?

I'd been waiting to get hold of the recent article by Dr Micah Mazurek and colleagues titled: Anxiety, sensory over-responsivity, and gastrointestinal problems in children with autism spectrum disorders* (abstract here) for quite a few days before posting about it.

This paper was always going to catch my eye given the focus on gastrointestinal (GI) problems comorbid to cases of autism. That and the authors connecting issues with anxiety and sensory issues to the presence of such functional GI problems in cases of autism, which really neatly falls within the remit of my interest in our second brain and how it might do some much more than just digest food. Don't get me wrong, I know all about the difference between correlation and causation - as in one does not necessarily mean the other - and tread carefully when discussing Dr Mazurek's findings.

Perhaps a few study details first:

  • Quite a large sample of children aged between 2-17 years old (N=2973) diagnosed with an autism spectrum disorder were included for study derived from the Autism Treatment Network (ATN). This means that we can reliably assume that (a) participants were diagnosed with autism, and (b) diagnosis was confirmed by gold-standard instruments such as ADOS.
  • Parent report data was collected on various features pertinent to the current study including a subset of items derived from the Short Sensory Profile (SSP), the Anxiety Problem T-score from the Child Behaviour Checklist (CBCL) and a GI Symptom Inventory Questionnaire to measure GI health related to five chronic functional GI symptoms. Cognitive functioning was also assessed.
  • The results: about a quarter of the participants studied presented with at least one chronic GI problem - that is a GI problem "within the past 3 months, duration of 3 or more months"; most frequently constipation. Children with chronic GI issues (n=733) scored higher on the CBCL Anxiety Problem scale than those (n=2239) without. Children with a chronic GI problem also presented with greater sensory over-responsivity (SOS) than those without such issues. SOS included features such as tactile sensitivity, taste-smell sensitivity (see here), movement sensitivity and visual-auditory sensitivity (see here). 
  • Finally, logistic regression analyses seemed to indicate that both anxiety and SOS "significantly contribute to the prediction of chronic constipation, chronic abdominal pain, chronic bloating and chronic nausea".

To my mind, these findings make for some very persuasive reading. One could argue about whether parent report is the most accurate way of collecting such data. I would however highlight two studies in the autism research field which suggest that parent reporting might not be totally off the mark as per developmental history recall (here) and that very important study by Phillip Gorrindo and colleagues** on parent reporting of GI issues in autism.

What more to say that has not already been said. The possibility that functional GI issues might be over-represented in cases of autism is to my mind a point which requires little more research in light of research like this and this and this. Obviously not everyone with autism presents with such bowel problems and not everyone with such bowel problems presents with autism. Science still needs to determine whether there might be a specific autism phenotype where bowel issues are a primary somatic symptom as per the findings on lactose intolerance from Tim Buie and colleagues. But for those who, for whatever reason, would once argue that comorbid bowel features are merely a fluke finding to autism, some autism... show me your evidence.

Such bowel issues if present are more than likely going to affect people with autism at least to the same extent as they do to those with not-autism particular when in the paediatric population. I think back to the Autism Now series aired April 2011 and vaguely recall some description of a child with autism also with bowel problems and the way their behaviour seemed to manifest such issues. It certainly is not outside the realms of possibility that a person with autism with a comorbid bowel issue like constipation for example, might well have some accompanying stress and anxiety over going to the toilet for a number two with the expectation of the discomfort that this might well bring. In terms of more chronic generalised anxiety, well the research literature suggests that functional bowel problems and anxiety might already share the same bed as per studies like this one from Waters and colleagues***.

An important implication from the Mazurek findings is that if there is a connection between functional GI issues, anxiety and sensory issues, one would expect that attempts to resolve the GI issues may very well impact on the other symptom areas too and vice-versa. The authors talk about the use of cognitive behavioural therapy (CBT) as one potential avenue of intervention for these overlapping issues. Personally I'm not a great fan of CBT as being particularly useful to "treat" bowel problems in autism either directly or periperhally; to coin a phrase from a good friend of mine, bowel problems are probably not present as a result of some unconscious desire to obsessively collect bus tickets or anything like that. Rather better looking at a more physical - somatic causation is likely to yield more productive, long-term positive effects.

I've talked previously about how those trillion or so bacteria which make up our gut microbiota might very well have the ability to influence our behaviour - or at least in animal models. A friend of mine (thanks Dr Mark Wetherell) also put another interesting recent article my way by Dinan & Cryan**** which talks about the stress response potentially being affected by aspects of the gut microbiota. If we assume that the functional bowel issues noted in cases of autism might very well be partially present as a result of some bacterial issues - such as those here and here, it's not too difficult to suggest some tentative connection.

I hope I'm not getting too speculative when I also bring to your attention studies like this one from Messaoudi and colleagues***** and the use of probiotic formulations to reduce stress and anxiety responses in both animals and human volunteers. OK, no mention is specifically made about functional bowel problems in the Messaoudi paper, although similar probiotic formulations have been suggested to also affect bowel issues like constipation****** (full-text).

I do also wonder about the more traditional pharmacotherapies used to treat constipation and diarrhoea or indeed more serious underlying issues such as inflammatory bowel diseases and whether there could be potential positive effects from those also in cases of autism. This paper makes mention of the "symptomatic improvement in their behavior and general well-being after bowel clearance before colonoscopy" noted in cases of autism. I've often wondered whether interventions like the gluten- and casein-free diet which seem also to impact on bowel issues in cases of autism might be another case in point.

More research needed on the gut-brain axis in autism perhaps?


* Mazurek M. et al. Anxiety, sensory over-responsivity, and gastrointestinal problems in children with autism spectrum disorders. Journal of Abnormal Child Psychology. August 2012.

** Gorrindo P. et al. Gastrointestinal dysfunction in autism: parental report, clinical evaluation, and associated factors. Autism Research. 2012; 5: 101-108.

*** Waters AM. et al. Functional gastrointestinal symptoms in children with anxiety disorders. Journal of Abnormal Child Psychology. July 2012.

**** Dinan TG. & Cryan JF. Regulation of the stress response by the gut microbiota: Implications for psychoneuroendocrinology. Psychoneuroendocrinology. 2012; 37: 1369-1378.

***** Messaoudi M. et al. Assessment of psychotropic-like properties of a probiotic formulation (Lactobacillus helveticus R0052 and Bifidobacterium longum R0175) in rats and human subjects. British Journal of Nutrition. 2011; 105: 755-764.

****** Guerra PV. et al. Pediatric functional constipation treatment with Bifidobacterium-containing yogurt: a crossover, double-blind, controlled trial. World Journal of Gastroenterology. 2011; 17: 3916-3921.

---------- Mazurek MO, Vasa RA, Kalb LG, Kanne SM, Rosenberg D, Keefer A, Murray DS, Freedman B, & Lowery LA (2012). Anxiety, Sensory Over-Responsivity, and Gastrointestinal Problems in Children with Autism Spectrum Disorders. Journal of abnormal child psychology PMID: 22850932