"Rates of death are higher for autistic individuals compared to the general population"

It's world autism awareness day today (2 April). The theme this year (2019), according to the United Nations, is "Assistive Technologies, Active Participation", something rather timely given the recent realisation that quite a few people diagnosed with autism are not properly represented [1] in various fields. Indeed, it continues a theme on the under-represented and under-studied populations within the autism spectrum being highlighted today of all days (see here)...

To mark this years autism awareness day, I'm turning to a topic that is probably about as important as it gets with regards to autism or any label: premature death. Specifically, highlighting how premature premature death can actually be when autism is diagnosed. I appreciate that this is not great PR for autism. But it's real-life for too many; and needs to be talked about as much as possible so people sit up and (hopefully) do something about it...

So...

The quote heading up this post - "Rates of death are higher for autistic individuals compared to the general population" - taken from the findings reported by Ye In (Jane) Hwang and colleagues [2] is perhaps not altogether accurate. A more precise meaning is required: as a group, those diagnosed with autism or autism spectrum disorder (ASD) are at some elevated risk for a premature (untimely) death than the non-autistic population (see here). Not exactly great dinner party conversation but vitally important nonetheless.

This is a topic that has cropped up time and time again in the peer-reviewed research literature and beyond (see here and see here). The grim trend crosses geographical boundaries and seems to cover the entirety of the autism spectrum. This time around New South Wales in Australia provided the source data with the aim "to report the rates and risk factors for mortality and cause of death in individuals on the autism spectrum (n = 35,929 age range 5-64) with and without concurrent intellectual disability (ID)."

The results: "Mortality rates for those on the autism spectrum were 2.06 times that of the general population." In line with other data (see here), those who also presented with intellectual (learning) disability alongside autism were at a higher risk of death. Epilepsy is also reported as being an important issue too (see here). A few other details are highlighted in the Hwang paper including the observations that "injury and poisoning" were some of the top-ranked causes of death as per other data (see here and see here).

What's more to say on this topic? Another call to action (see here)? Some more big news headlines (see here)? Another 'long-term plan' (see here)? All well-and-good, but what's actually being done to tackle such frightening statistics here and now and actually improve and extend lives in the here and now? What's currently helping autistic people to 'age well' [3]? And just in case you're of the inclination that epilepsy and intellectual (learning) disability are some how not a 'core' part of some autism, you're probably wrong in many cases (see here and see here)...

If the Hwang and other data aren't enough to satisfy you on this topic, another research paper with another (similar) set of grim findings has also recently been published [4]. Faced with such data, discussions about a puppet depicting one face of autism (see here) for example, don't seem all that important do they? Indeed, if half as much time was dedicated to highlighting the mortality stats associated with autism and making sure this makes headlines - over a 20 year period "6.4% of individuals died at an average age of 39 years" - may be more would be done to tackle them, including providing "adequate access to health care for individuals with autism spectrum disorder."

Food for thought, and hopefully important fuel for action...

----------

[1] Russell G. et al. Selection bias on intellectual ability in autism research: a cross-sectional review and meta-analysis. Molecular Autism. 2019; 10: 9.

[2] Hwang YIJ. et al. Mortality and cause of death of Australians on the autism spectrum. Autism Res. 2019 Feb 25.

[3] Hwang YI. et al. Aging Well on the Autism Spectrum: An Examination of the Dominant Model of Successful Aging. J Autism Dev Disord. 2018 May 2.

[4] Smith DaWalt L. et al. Mortality in individuals with autism spectrum disorder: Predictors over a 20-year period. Autism. 2019. Feb 28.

----------

"To examine associations between early developmental exposure to ambient pesticides and autism spectrum disorder"

The quote heading this post - "To examine associations between early developmental exposure to ambient pesticides and autism spectrum disorder" - reflects the aim of the study published by Ondine von Ehrenstein and colleagues [1]. Said aim was accomplished by examining data from "California’s main agricultural region, Central Valley, using 1998-2010 birth data from the Office of Vital Statistics" with regards to autism diagnoses (see here) and pesticide use and exposure. The results: "Findings suggest that an offspring’s risk of autism spectrum disorder increases following prenatal exposure to ambient pesticides within 2000 m of their mother’s residence during pregnancy, compared with offspring of women from the same agricultural region without such exposure." Also: "Infant exposure could further increase risks for autism spectrum disorder with comorbid intellectual disability."

The von Ehrenstein findings also come with an accompanying editorial [2] both published in the British Medical Journal (BMJ). That editorial provides a good overview of the findings and, importantly, some of their limitations. The editorial notes for example, that: "the study included only children with a diagnosis of “autistic disorder” by the California Department of Developmental Services, based on criteria described in the Diagnostic and Statistical Manual of Mental Disorders version IV-R." It cautions that the von Ehrenstein findings might not generalise to "milder forms of autism spectrum disorder." This 'lack of generalisation to milder forms' is a real reversal of what is typically seen in the peer-reviewed autism research literature (see here).

I don't want to linger too much on the von Ehrenstein study but a few points are worth noting. The question of what specific pesticides *might* influence risk of offspring autism is a complicated one. The researchers reported that "small to moderately increased risks for the disorder in offspring with prenatal exposure to the organophosphates chlorpyrifos, diazinon, and malathion, the pyrethroids permethrin and bifenthrin, as well as to glyphosate, avermectin, and methyl bromide" were observed. They (and the editorial) talk about how further research is required to "examine the joint effects of multi-exposure mixtures to more effectively protect human health." This is important in the context of other research (see here and see here) and also to avoid any big media headlines (see here) about specific preparations 'causing' autism (which has not yet been authoritatively established). Given that various different classes of pesticides have somewhat slightly different biological actions, there is a lot of work to be done on the possible biological mechanisms/targets and any synergistic effects.

Also: "children with autism spectrum disorder and co-occurring intellectual disability were examined as a separate outcome." The fact that von Ehrenstein et al observed something like a 'stronger association' between pesticide exposure and those at the "more severe end of the autism spectrum" (the description according to one media source) is another important point requiring further study. I say this in the context that 'comorbidity' might not always be the best description for symptoms and diagnoses accompanying autism (see here).

The authors conclude that their findings - with appropriate caveats - have implications. Namely: "Exposure of pregnant women and infants to ambient pesticides with a potential neurodevelopmental toxicity mode of action should be avoided as a preventive measure against autism spectrum disorder." I don't think too many people would disagree with the sentiments of 'avoiding pesticide exposure' particularly for pregnant women and those who might also be more vulnerable to their potential effects. The questions now turn to genetics and biology and the question of why...

Music to close: Hostiles on the Hill (apparently)...

----------

[1] von Ehrenstein OS. et al. Prenatal and infant exposure to ambient pesticides and autism spectrum disorder in children: population based case-control study. BMJ. 2019; 364: I962.

[2] Bakian AV. & VanDerslice JA. Pesticides and autism. BMJ. 2019; 364: I1149.

----------

The NHS Long Term Plan and autism

The NHS (National Health Service) here in Blighty has recently released its Long Term Plan [1]. 'Long term' covers a period of 10 years and how, faced with ever-mounting financial and resource pressures, the NHS is continually trying to serve the needs of the population it serves in an ever-changing world.

Needless to say that lots of media outlets have covered the Long Term Plan (see here for example) and the various priorities that it includes. 'Prevention' and 'Early Detection' are words used quite a bit both in the reporting and the document itself, as the focus moves slightly away from hospitals to other front-line services like general practitioners (GPs) and community care agencies.

Mental health also figures in the Long Term Plan, as this area continues on its 'parity of esteem' journey (see here). Being careful not to label autism as a mental health condition, I note that autism and learning (intellectual) disability also gets a mention in the Plan for quite a few reasons...

So, starting on page 52 of the report: "Action will be taken to tackle the causes of morbidity and preventable deaths in people with a learning disability and for autistic people." It's about time that this was 'tackled'. I say that because the statistics on early mortality with autism in mind are truly, truly shocking (see here and see here). It's also of interest that the LeDeR (Learning Disabilities Mortality Review) initiative is mentioned in this context too (see here). How exactly such inequalities will be tackled is however, not precisely detailed in the Long Term Plan.

Then: "The whole NHS will improve its understanding of the needs of people with learning disabilities and autism, and work together to improve their health and wellbeing." Allied to improving uptake of annual health checks that should be available, the document talks about working with partners to "bring hearing, sight and dental checks to children and young people with a learning disability, autism or both in special residential schools." It's a start given what said overlooked issues might be involved with (see here) but what about this who aren't in special residential schools? The Plan also mentions how: "By 2023/24, a ‘digital flag’ in the patient record will ensure staff know a patient has a learning disability or autism." I believe this would also solve a few 'issues' with regards to the (estimated) prevalence of autism and/or learning disability here in Blighty (see here and see here).

Also: "Children and young people with suspected autism wait too long before being provided with a diagnostic assessment." Yes, yes they do (see here). The Plan therefore sets out to "test and implement the most effective ways to reduce waiting times for specialist services." You've got to be kinda careful with the wording here because, as far as I can see, there is no commitment to a timescale of diagnosis unlike the commitment to reducing those with autism being accommodated at inpatient units for example: "By March 2023/24, inpatient provision will have reduced to less than half of 2015 levels (on a like for like basis and taking into account population growth) and, for every one million adults, there will be no more than 30 people with a learning disability and/or autism cared for in an inpatient unit".

I'll leave readers to decide whether this Long Term Plan represents something 'good for autism' or just skirts around some of the bigger issues. Personally I see some positives and some 'missed opportunities'. Positives? Well as I said, anything that can impact on those shameful early mortality figures in the context of autism is a good thing. I do have questions about how issues like suicidality, that contribute quite a bit to the early mortality stats, are for example, going to be addressed, but if lives are going to be saved and hopefully enhanced, I'm all for that. Reducing waiting times for assessments is also a good thing, as is the idea of the 'digital flag' to (hopefully) help enhance the doctor-patient interaction where autism is a feature.

Negatives? Well, there doesn't seem to be a great amount of details mentioned and even less discussed about adult autism and what the Long Term Plan is going to do for the thousands of autistic adults (many of whom are not also described as 'learning disabled'). There are many pressing issues for this group (see here and see here), some of which cross-over with the primary tenets of the Long Term Plan (see here) with autism in mind. It strikes me that there is much more to do in this area.

So I guess we'll just have to see what happens...

Music to close: And given some recent news about Weezer, a sublime blast from the past...

----------

[1] The NHS Long Term Plan. January 2019.

----------

Autism and learning disability: "double disadvantage" on health status

The paper published by Kirsty Dunn and colleagues [1] continues a theme on how Census information - where data is systematically recorded about a specific population, normally a country - is influencing autism research (see here and see here). The Census in question is "Scotland's Census, 2011" and the rather forward facing step to include some questions about autism and learning (intellectual) disability in the Scottish Census for 2011.

This time around the same research group, who were very much involved in the publication of other Census based information on autism, set out with the aim to examine "the population prevalence of co-occurring intellectual disabilities and autism, and its impact on general health status." From a total population of over 5.2 million people who were included in the Scottish Census in 2011, they observed that about around 0.1% "had co-occurring intellectual disabilities and autism." Although perhaps not sounding like a lot of people, that percentage represented over 5000 people, including children and adults. Most of those with autism and learning disability were male.

Then to the headline conclusion related to the "double disadvantage" mentioned in the title of this post from the Dunn paper: "Their general health status was substantially poorer than for the rest of the population, more so for children/young people, and they had more limitations in their day-to-day activities." Researchers also mention how poor general health status was "apparent across the entire life course." They further recommended that: "Staff in services for people with either of these conditions need to be trained, equipped, resourced and prepared to address the challenge of working for people with this duality. This is essential, to address these substantial health inequalities."

I know the use of the words "double disadvantage" are probably not going to be well received by everyone, particularly those who insist that autism is more about difference than disability. I'm in partial agreement that, whilst a diagnosis of autism does have a profound influence on many aspects of life, not everything is necessarily to be seen as a 'disadvantage' for everyone. When it comes however to health inequality in the context of autism (and learning disability), I don't think many people would argue with the idea that both in the context of autism alone and autism + learning disability, various aspects of health - physical and mental - can and do suffer as a result (see here and see here for examples) and are, in that light, a disadvantage. With all this in mind, the call to action made by Dunn et al is not without some substantial merit...

----------

[1] Dunn K. et al. The prevalence and general health status of people with intellectual disabilities and autism co-occurring together: a total population study. J Intellect Disabil Res. 2018 Nov 28.

----------

In-utero exposure to antiepileptic drugs and offspring learning disability

There was something quite familiar about the findings reported by Laura Fuglslang Bech and colleagues [1] investigating any "association between in utero exposure to antiepileptic drugs and learning disabilities in the first year of compulsory education among offspring." Familiar because this class of medicines - antiepileptics - used to primarily treat/manage epilepsy and seizure disorders have been put under the spotlight in recent years as a result of various *associations* with offspring childhood developmental disorders/conditions when used during pregnancy (see here and see here).

Obviously one has to be quite careful when it comes to talking about this class of medicines. Antiepileptics, as I have mentioned before, are not typically dispensed willy-nilly, and on many occasions have proved to be not just a life-enhancer but also a lifesaver to those living with epilepsy. The balancing act comes when they are prescribed to women of child-bearing age and ability, and the possibility of them producing 'adverse' effects' on the developing child.

Bech et al focused on learning disability as a possible offspring outcome following pregnancy exposure to a range of antiepileptic medicines. One of those marvellous Scandinavian population registries provided the research starting point (based in Denmark) and lead to the participation of some 630 cases ("offspring exposed to antiepileptic drugs in utero") and 430+ controls ("unexposed offspring of mothers previously redeeming antiepileptic drug prescriptions").

Results: "Learning disabilities were identified among 7.1% cases compared with 3.7% for controls." Although not at first glance showing a huge difference between the groups, a little statistical analysis revealed that: "During any trimester, the adjusted OR [odds ratio] of the association between in utero exposure to antiepileptic drugs and learning disabilities was 2.20 (95% CI 1.16 to 4.17)." When taking into account different types of antiepileptic medicines and their possible risk profile, yet again one particular medicine stuck out: valproate. As the authors mention: "valproate carried a higher risk (OR 4.67, 95% CI 1.73 to 12.59)" compared with other antiepileptic medicines. This accords with other findings.

As with any observational study of this type, one has to be a little careful with the old 'correlation is not necessarily causation' thing. But set against a tide of research observing similar trends - particularly with respect to the use of valproate - and indeed, a rodent model of autism called the valproate rodent model of autism - it's getting ever harder to say that pregnancy valproate use in particular, is not associated with a variety of adverse offspring outcomes...

If in doubt, please speak to your prescribing physician. Please don't tinker yourself with things that you shouldn't tinker with...

----------

[1] Bech L. et al. In utero exposure to antiepileptic drugs is associated with learning disabilities among offspring. J Neurol Neurosurg Psychiatry. 2018 Aug 3. pii: jnnp-2018-318386.

----------