Saturday, 21 May 2011

Hey sugar: diabetes and autism

I have a few mantras that I try and stick to on this blog. Probabilities, not absolutes, drive science is perhaps the primary one. Not too far behind is the fact that based on the current evidence, autism, in all its forms, does not seem to confer any protection against the development of other diseases/infections/conditions co-morbid to autistic symptoms. It is sometimes easy to forget that autism as a concept is really just a descriptive label covering a number of clinically-relevant characteristics. Underneath that label is a person with genes and an environment which confer the same variable protection/risk against all the different things out there which can potentially affect health, quality of life and also mortality the same as everyone else. The characteristics and severity of autism (and its co-morbidities) might be a barrier to either communicating needs or accessing appropriate healthcare; and if so, this is what we need to work on if we are ever going to truly have a society of equality. In this post I want to talk about quite an important condition - diabetes - and its links (or not) to autism spectrum conditions.

There are a number of good definitions of diabetes on the web. Diabetes UK has probably the most understandable description that I have found, discussing the primary forms of the condition, type-1 and type-2 diabetes. Importantly they also discuss some of the risks and myths around diabetes. The stats first: diabetes affects an estimated 4-5% of the general population, and the incidence of diabetes (type-2) seems to be increasing. Treatment depends on the type of diabetes. Type-1 diabetes is where the body does not produce insulin and so must be regularly supplemented. Type-2 diabetes is characterised by some endogenous insulin production but either not enough or the development of insulin resistance.

Insulin injections are the most common management option for type-1 diabetes, but there are other modes of delivery; the oral dosage form does have a few issues not least what happens to proteins and peptides in the gut. There is a link between diabetes and things like being overweight/obese and according to Diabetes UK, the risk of diabetes is increased where a 'severe mental health problem' exists. The risk of several other 'conditions' is elevated when diabetes is present; one of the more interesting for me was this paper suggesting quite a high rate of IgA antitissue transglutaminase (tTG) following on from my recent post on such findings in autism.

Whilst not wishing to imply that shared risk factors cover the entire autism spectrum, there is perhaps some scope for looking more closely at people with autism and their risk of developing diabetes. I have previously blogged about being overweight / obese in relation to autism and concluded that a similar risk of such issues seemed to be reported in autism and the more general not-autism population. Likewise co-morbidity of mental health problems in autism has also come under discussion and evidence suggests that there is perhaps a greater risk present in some cases of autism. Added to the fact that autism is generally not a life-limiting diagnosis in terms of early mortality (outside of mortality caused by accidents, etc), alongside other issues such as the use of various pharmacotherapies which carry some increased risk of potential risk factors for diabetes (i.e. weight gain), one could assume that autism carries at least some risk of developing diabetes. So what does the research literature tell us about autism and diabetes?

Well, quite a bit. There is some suggestion that there is an increased history of autoimmune conditions to include diabetes, in other family members of people with autism compared with control populations and evidence is not just from one study either. Interestingly, it appears that parental type-1 diabetes is the more common association over type-2. This study suggested that maternal diabetes during pregnancy might be one of several risk factors for autism. A paper delivered at the recent IMFAR 2011 meeting has put the association back on the agenda (association at least).

Getting an actual prevalence figure of diabetes in autism is slightly more elusive. There is some data from a few years back, reporting that rates of type-1 diabetes may be elevated in autism compared to population estimates (0.9% vs. 0.3-0.7%). This study also suggested that within a sample of nearly 6000 adults with 'developmental disability' over 11% presented with diabetes, although the terminology used does not necessarily translate as just autism. I would perhaps be inclined on the basis of the available data to suggest that diabetes, or at least type-1 diabetes, seems to show some association with autism perhaps a slightly greater levels than those present in the general population.

One of the main problems is how to ensure that appropriate monitoring for conditions like diabetes is put in place for people with autism. One would perhaps expect that those who are on the more severe end of the autism spectrum, who are perhaps not living independently and potentially in receipt of pharmacotherapy, would be subject to some kind of monitoring for diabetes, just as part of good practice and good medicines management. As for those higher-functioning people, who live independently and may not access healthcare quite so routinely, the question must be asked: who is monitoring their health?


  1. It is important to recognize, when examining numbers, that the term "developmental disability" encompasses Down's Syndrome. Diabetes is known to be more common in people with DS. To the best of my knowledge, no study has confirmed a link between autism and diabetes, and there was a study in Denmark that tried. This is born out by my personal experience as the parent of a child who has both autism and type 1. We are few and far between!

  2. Thanks for dropping by and the comment Susan. I agree that 'developmental disability' is a nebulous term covering a lot of ground and would by no means suggest that this is 'autism only'. There is a large gap in the evidence base with regards to diabetes (both types) and autism likely because of a few reasons, not least that diabetes is not a 'symptom' of autism and screening is therefore not necessarily indicated. My point in this post is that the data does not yet imply that autism confers any protection against co-morbid diabetes. Indeed, particularly in the more able population who are perhaps not routinely in contact with healthcare providers, there seems to be very little evidence of proactive screening programmes for diabetes despite such (at least) equal risk with the general population.

  3. You're right and I agree completely. The fact that my son was diagnosed by a psychiatrist -- after I'd dragged him to multiple doctors first -- bears that out. Doctors are eager to place all issues into a "behavioral" bucket when they have a patient with autism. It can be a big problem for a population already challenged by self-advocacy. Thanks for this article, I enjoyed it much and forwarded it on.

  4. If my understanding is correct from titled, "Night-Shift Nurses Have Higher Risk to Develop Type 2 Diabetes, Study Says" that lack of sleep and the irregularity of sleeping can cause 2 types of diabetes