Skip to main content

People with autism are 'dying younger,' warns study

Monday 21 March 2016

"People with autism are dying earlier than the general population," BBC News reports.

A recent study in Sweden showed the average age of death for a person with autism spectrum disorder (ASD) is 54 years, compared with 70 for matched controls.

The study used records of 27,122 people diagnosed with ASD to look at how long they lived, what the main causes of death were, and how their chances of death were affected by whether they were male or female and the type of autism they had.

For the purposes of the study, ASD was split into two categories: low-functioning ASD, where a person with ASD also had learning difficulties, and high-functioning ASD, where a person with ASD had average or above average intelligence.

Researchers then compared them with an age- and gender-matched sample from the general Swedish population.

The researchers found all groups of people with ASD were 2.5 times more likely to have died during the study than people without.

The highest risk seemed to be in people with low-functioning ASD – particularly women, who had almost nine times the mortality risk of women the same age without ASD.

Leading causes of death included neurological disorders such as epilepsy, which has previously been linked with ASD, and suicide. People with high-functioning ASD had a ninefold increased suicide risk.

The researchers said their study demonstrated that much more needs to be done to support both the mental and physical health of people with ASD. 

Where did the story come from?

The study was carried out by researchers from Sweden's Karolinska Institutet and was funded by Stockholm County Council, Karolinksa Institutet, and the Swedish Research Council.

It was published in the peer-reviewed British Journal of Psychiatry.

The Guardian, Mail Online, The Daily Telegraph and BBC News covered the study primarily as part of stories covering the launch of a campaign by the charity Autistica.

The campaign has called for more research into the causes of death among people with autism and the need for action to address the situation.

The media coverage was accurate and contained useful quotes from independent experts.

What kind of research was this?

This was a case-control study, where records of people diagnosed with ASD were "matched" with those of similar people without a diagnosis of ASD. Case-control studies can show up differences between groups of people, but can't tell us what's behind those differences. 

What did the research involve?

Researchers took records of all 27,122 people diagnosed with ASD in Sweden between 1987 and 2009. They matched each person to up to 100 people from the Swedish general population who were the same age, sex and country of origin, but without ASD.

The researchers used the records to compare people's chances of having died during the study period and having died of specific causes.

The records came from Sweden's national databases. Researchers included Asperger's syndrome, autism and pervasive developmental disorder as ASD. 

The three categories are essentially based on intelligence, Asperger's being at the top of the scale and pervasive developmental disorder at the bottom.

People were categorised as having low-functioning ASD if they also had a learning disability. Otherwise, they were seen as having high-functioning ASD.

Researchers first calculated the overall chances of death (mortality) for all people with ASD, compared with all people without. They then looked at the results separately for people with low-functioning and high-functioning ASD, and for men and women.

They also looked separately at the chances of people with ASD having died from different categories of causes of death:

  • infections
  • cancers
  • hormonal disorders
  • mental and behavioural disorders
  • diseases of the nervous system, circulatory system, respiratory system, or genitourinary system
  • birth defects
  • external causes, with self-harm or suicide recorded separately

What were the basic results?

Overall, people with ASD were 2.56 times more likely to have died during the study period than people without (odds ratio [OR] 2.56, 95% confidence interval [CI] 2.38 to 2.76). The average age of death for people with ASD was 53.87 years, compared with 70.2 years for people without.

These stark figures break down to give some even more worrying numbers. People with low-functioning ASD on average died before they reached 40, at 39.5 years.

Overall, people with low-functioning ASD had a higher risk of having died – a more than fivefold risk, compared with a twofold risk for people with high-functioning ASD.

Women with low-functioning ASD had the highest risk of any group – an eightfold higher risk of death than a woman the same age without ASD.

Apart from infections, people with ASD were more likely than those without to have died from any of the causes of death considered. However, the two causes that stand out are suicide and epilepsy.

People with ASD were 7.55 times more likely to die by suicide. People with high-functioning ASD were at greater risk of suicide than low-functioning groups, and – unusually – women were more at risk than men. In the general population, rates of suicide are 3.5 times higher in men compared with women.

Deaths as a result of nervous system disorders – primarily epilepsy – were 7.49 times higher among those with ASD, and people with low-functioning ASD were most at risk. 

How did the researchers interpret the results?

The researchers said: "Our observation of excess cause-specific mortality in individuals with ASD may signify a generally increased biological vulnerability in ASD, as well as insufficient awareness, diagnoses and treatment of comorbid diseases within the healthcare system."

In other words, people with autism may be more vulnerable to getting certain diseases that can lead to death, and doctors may not be as good at diagnosing and treating diseases in people with ASD.

Looking at suicide as one example, the researchers suggested that people with ASD may be at more risk of getting depression, but also may be less likely to be diagnosed with depression and have support networks in place to help them with mental illness. This means they may be more likely to take their own lives rather than be successfully treated.

They concluded that, "Adequate and co-ordinated medical care for individuals with ASD and research into the phenomenon should be a target for a considerably broader audience of medical specialties than psychiatry and neurology." 


These are distressing figures for anyone with ASD, and their friends and families. But we need to remember what the figures actually represent: people with ASD in this Swedish population sample had an increased risk of dying during follow-up relative to the people without ASD.

These results do not mean that people with ASD have the certainty of a shortened life. Average figures don't tell you about what will happen to one individual.

Although some previous studies showed that people with ASD have a higher risk of dying sooner than those who do not have the condition, they were too small to look at the detail of the causes of death and the differences between men and women, as well as between people with high-functioning and low-functioning autism.

This study is big and based on reliable databases. However, because of the way that ASD was recorded in Sweden before 2001, it may represent more people with severe autism than in the average population.

People were only added to the record if they had been in contact with clinical psychiatric services. People with less severe autism may not have had a diagnosis recorded.

At present, we don't know enough to say what causes the increased chances of death for people with ASD. We don't know exactly what causes ASD, which makes it hard to see how possible causes, such as genes, might affect both the chances of getting ASD and of getting other conditions that can affect the length of your life.

Given that the chances of death were raised from almost all causes, it seems likely there could be a mixture of reasons for the increased risk. These could lie within ASD and other conditions that may be associated with it, and possibly also within society and the healthcare system.

For example, social and communication difficulties may mean that people with ASD have more difficulty accessing healthcare or discussing health problems, signs and symptoms with doctors.

It is important that people working in healthcare are aware of this risk association. This may help doctors and nurses or other carers identify potential problems, and put effective support systems and treatment in place.

Ultimately, we need much more research about why these shocking differences in lifespan occur and what can be done to address them.

Analysis by Bazian
Edited by NHS Website

Links to the headlines

Premature autism deaths are 'hidden crisis' says charity

The Guardian, 18 March 2016

People with autism die 16 years earlier than expected, study finds

The Daily Telegraph, 18 March 2016

People with autism die 16 years early: Suicide and epilepsy are leading to premature deaths

Mail Online, 18 March 2016

People with autism 'die younger', warns charity

BBC News, 18 March 2016

Links to the science

Hirvikoski T, Mittendorfer-Rutz E, Boman M, et al.

Premature mortality in autism spectrum disorder

British Journal of Psychiatry. Published online March 1 2016

Further reading

Woolfenden S, Sarkozy V, Ridley G, et al.

A systematic review of two outcomes in autism spectrum disorder – epilepsy and mortality

Developmental Medicine and Child Neurology. February 2012


Personal tragedies, public crisis – The urgent need for a national response to early death in autism (PDF, 4.2Mb)

March 2016