Previously, it has been reported that autistic people live 16 years less on average than non-autistic people. 

However, this claim was based on misinterpreted results from a study that did not estimate life expectancy. 

The National Autistic Society supported researchers from University College London in undertaking new research. It explicitly aimed to estimate the average life expectancy of autistic people living in the UK using the method adopted by the Office for National Statistics. 

In this article, Aimee Fletcher from the National Autistic Society’s Evidence and Research team summarises the findings.

Methodology and analysis

Autistic people were involved in the design and conduct of the research as part of a steering group of experts by experience. Four autistic adults, alongside a representative from the National Autistic Society, gave feedback about the appropriateness and usefulness of the research questions. Input from this steering group also informed the final article on the research findings.

The research team used anonymised data from GP practices in the IQVIA Medical Research Database. This data came from the medical records of people who have received autism diagnoses throughout the UK between 1989 and 2019. The research focus was on diagnosed autistic people, who, as the researchers highlight, are not representative of the whole autistic population, given that many autistic people are undiagnosed. 

The researchers studied data from 17,130 autistic people without a learning disability and 6,450 autistic people who also have a learning disability. The team then identified comparison groups of people with similar demographic backgrounds (for example, same age and sex) but no autism diagnosis. 

The research team analysed the records of nearly ten million people to find records for comparison across the identified groups. They compared the records of 171,300 presumed non-autistic (not diagnosed as autistic and without a learning disability) to the records of autistic people without a learning disability. They also compared 64,500 presumed non-autistic people without a learning disability to the 6,450 records of people identified as autistic with a learning disability.

Findings

The researchers found significant differences in life expectancy between the different groups. 

On average, autistic men without a learning disability had an estimated life expectancy of 74.6 years, and autistic women without a learning disability had an average of 76.8 years.

For autistic people who also had a diagnosed learning disability, average estimated life expectancies were lower, with men having a life expectancy of 71.7 years and women having a life expectancy of 69.6 years. These estimates all have a degree of uncertainty, particularly for women due to the smaller numbers of women who have been diagnosed autistic.

Researchers compared results for both the diagnosed autistic groups (with and without a learning disability) to the life expectancies for their matched comparison groups who didn’t have these diagnoses. The average life expectancy in these groups was 79 to 81 years for non-autistic males and 83 to 84 years for non-autistic females. 

Comparing the life expectancies between the matched groups suggested that autistic people without a learning disability had a six-year shorter life expectancy on average. 

For autistic men with a learning disability, the life expectancy was seven years shorter than the non-autistic comparison group. For autistic women with a learning disability, it was 15 years shorter than comparison women. This last finding could demonstrate that many autistic women are undiagnosed unless they have high support needs (which might negatively impact life expectancy). 

Implications

This study is one of the first to provide evidence that diagnosed autistic people are more likely to die prematurely than non-autistic people in the UK. 

An important element of this study is that the researchers considered data collated about autistic people with a learning disability. These individuals are often overlooked in research, and including them provides additional findings and draws valuable conclusions. 

Whilst the findings suggest a higher life expectancy than had previously been thought for autistic people, the difference between autistic and non-autistic populations is significant and a cause for concern. 

The findings from this research suggest that there could be wider issues around autistic quality of life. For example, is the lower life expectancy due to many autistic people, with and without a learning disability, experiencing poorer physical and mental health outcomes compared to non-autistic people? 

Whilst exact causes of death were not available in the records analysed, there is evidence from other sources (see ‘further reading’ list) that issues such as social exclusion and discrimination negatively impact autistic people’s physical and mental health. Additionally, failure to provide adequate support can also result in avoidable death in autistic people. 
The researchers propose that their findings demonstrate an urgent need for inequalities (such as healthcare) for autistic people to be prioritised.

The lead investigator of this study, Professor Josh Stott (UCL Psychology and Language Sciences), said:

“Our findings show that some autistic people were dying prematurely, which impacted the overall life expectancy. However, we know that when they have the right support, many autistic people live long, healthy and happy lives. Although our findings show important inequalities, we were concerned about frightening statistics that are often quoted, and it is important to provide more realistic information.”

Dr Judith Brown, Head of Evidence and Research at the National Autistic Society, added:

“Without investment, improved understanding, inclusion and the correct level of support and care, autistic people will continue to see reduced life expectancy, with the most at-risk group in this study being autistic women with learning disabilities. This research should be a wake-up call for Government, the NHS, healthcare professionals and society as a whole that we must tackle the health inequality autistic people face.”

The results of this research demonstrate that some autistic people die prematurely compared to the general population, particularly autistic women with learning disabilities. These results suggest an inequality in quality of life for autistic people, which could be reduced by providing better support, increasing understanding of autistic needs and making inclusive changes that facilitate access to services.

Limitations

One of the limitations of this study was the absence of information about the cause of death of those included. The gap in this information from the records means that the researchers were unable to estimate how many years of life lost for autistic people may be attributed to different causes, such as co-occurring health conditions. 

The researchers recommend that population-based studies identify both diagnosed and undiagnosed autistic people in the community to provide more conclusive evidence of life expectancy regardless of diagnostic status. This information would help to identify specific risk factors that impact autistic life expectancy, such as lack of support. 

Another factor is that the database did not include data on gender diversity. Therefore, researchers could not explore these differences in their data analysis and interpretation.

The researchers note that nine out of ten autistic people may have been undiagnosed over the 30-year period studied, which links to their previous research into the high levels of underdiagnosis in England. For this reason, the researchers argue that their findings cannot be generalised for all autistic people. 

In the past, diagnosis typically occurred for autistic people who are likely to have greater-than-average support needs, particularly adults. Therefore, there may be undiagnosed autistic people unknowingly factored into the general population statistics, and these findings could overestimate the reduction in life expectancy experienced by autistic people. This study highlights the need for improved access to diagnosis and subsequent post-diagnosis support to enable previously missed autistic people to have better life outcomes.

Funding

This research was funded by the Dunhill Medical Trust, Medical Research Council, National Institute of Health and Care Research, and the Royal College of Psychiatrists. The funders had no role in study design, data collection, analysis, interpretation or writing dissemination papers.

You can read the full paper here: https://www.sciencedirect.com/science/article/pii/S2666776223001953.

Further reading

• Hirvikoski T, Mittendorfer-Rutz E, Boman M, Larsson H, Lichtenstein P, Bölte S. Premature mortality in autism spectrum disorder. Br J Psychiatry. 2016 Mar;208(3):232-8. doi: 10.1192/bjp.bp.114.160192. Epub 2015 Nov 5. PMID: 26541693
• Griffiths S, Allison C, Kenny R, Holt R, Smith P, Baron-Cohen S. The vulnerability experiences quotient (VEQ): a study of vulnerability, mental health and life satisfaction in autistic adults. Autism Res. 2019;12:1516–1528. 
• Botha M, Frost DM. Extending the minority stress model to understand mental health problems experienced by the autistic population. Soc Ment Health. 2020;10:20–34. 
• Branford D, Gerrard D, Saleem N, Shaw C, Webster A. Stopping over-medication of people with intellectual disability, Autism or both (STOMP) in England part 1 – history and background of STOMP. Adv Ment Health Intellect Disabil. 2019;13:31–40. 
• Care Quality Commission. Out of sight – who cares?: restraint, segregation and seclusion review; 2020. Accessed December 21, 2022. 
• Office for National Statistics UK. Outcomes for disabled people in the UK: 2021; 2021. London, United Kingdom. 
• Stewart GR, Corbett A, Ballard C, et al. Traumatic life experiences and post-traumatic stress symptoms in middle-aged and older adults with and without autistic traits. Int J Geriatr Psychiatry. 2022;37. 
• Doherty M, Neilson S, O’Sullivan J, et al. Barriers to healthcare and self-reported adverse outcomes for autistic adults: a cross-sectional study. BMJ Open. 2022;12:e056904. 
• Weir E, Allison C, Baron-Cohen S. Understanding the substance use of autistic adolescents and adults: a mixed-methods approach. Lancet Psychiatry. 2021;8:673–685
• White, A; Sheehan, R; Ding, J; Roberts, C; Magill, N; Keagan-Bull, R; Carter, B; Chauhan, U; Tuffrey-Wijne, I; Strydom, A; (2023). LeDeR Annual Report Learning from Lives and Deaths: People with a Learning Disability and Autistic People

See also: Under diagnosis of autism in England: a population-based study.