Theravive Home

A new study published in the Lancet looked at autism in England with regard to assessing under-diagnosis in a population-based cohort study of prospectively collected primary care data.

“People often think of autism as a childhood condition, and historically, autism has mostly been diagnosed in children,” study author Elizabeth O’Nions told us. “Many autistic adults were not diagnosed when they were children, either because autism didn’t exist as a diagnosis when they were young, or the criteria were narrower, meaning that only a very small number of people were considered to be autistic at the time.”

Widening of diagnostic criteria and growing public awareness means that the proportion of children and young people getting a diagnosis is much higher than it used to be. Data from NHS digital shows that autism is now diagnosed in more than 3% of children and young people in England.

Research that has actively identified autistic people in the population suggests that the proportion of people who are autistic is similar across different age-groups. But a lot of adults who suspect they are autistic tell us that they have struggled to get a diagnostic assessment, or that they had to wait a very long time for one.

As a result, health services are not aware that they are autistic, so won’t necessarily offer them the support and adjustments that they need.

“We wanted to estimate how many undiagnosed autistic adults there might be in England by looking at the proportions of people of different ages who have a diagnosis in their electronic health records,” O’Nions told us. “We looked at the numbers separately for males and females, and for people with and without intellectual disability (often called ‘learning disability’ in the UK), using anonymized data from a subset of English primary care practices.”

The research team then compared these numbers to the numbers of people they think might meet the diagnostic criteria for autism if they were assessed. They used two estimates of the real numbers – a lower estimate of around 1% of the population, and an upper estimate of around 3%.

The lower estimate of around 1% was informed by epidemiological research published in 2011 that actively looked for autistic people in the population. However, the diagnostic criteria for autism have been updated since the study took place, and the measures used may have missed autistic women.

“The upper estimate of around 3% came from the numbers of children and young people who had been diagnosed in our dataset,” O’Nions told us. “We know that children and young people are most likely to have been assessed because people are very aware of autism in younger people, and there are more diagnostic services available.” 

Extrapolating the data that we had from a subset of primary care practices to population statistics for the whole of England, the team estimated that, as of the end of 2018, there were between 150,000 and 500,000 autistic people aged 20 to 49 who were undiagnosed, and between 250,000 and 600,000 autistic people aged 50 and over who were undiagnosed.

“Even assuming the lower estimate of the real numbers of autistic people (around 1%), this would mean more than nine out of 10 autistic over-50s were undiagnosed,” O’Nions told us. “If we assume that the mid-point of our estimate is correct (i.e. around 2% of the population are autistic), then there may be 750,000 undiagnosed autistic people aged 20 and above in England. Our estimates also suggest that many autistic adults with an intellectual disability may not have an autism diagnosis. Recognizing that someone with learning disability is autistic is really important, because this knowledge can help people to understand and support them better.”

It’s important to note that the study’s figures give an indication of the possible numbers of undiagnosed autistic people. To calculate real figure, the researchers would need to do in-depth research to actively look for autistic people in the community.

It’s possible that some people might have been diagnosed privately and not told their primary care provider, which would mean that their diagnosis wouldn’t be listed in the data. The team also wasn’t able to study rates of diagnosed autism in gender diverse people in particular, as gender diversity is not well recorded in electronic health records.

“We weren’t surprised to find that there were far more children and young people diagnosed compared to adults and older adults,” O’Nions told us. “But we were surprised by just how few older adults had a diagnosis. Publicly available data from NHS digital shows a similar pattern. The figures we have estimated suggest that there are a large number of undiagnosed autistic adults in England. There is a need to both improve access to adult diagnostic services and provide more support for people after diagnosis.”

It is also important that there is more awareness of autism in older adults, explained O’Nions. Autistic people are often be discriminated against because they may act in ways that are confusing to non-autistic people, meaning that they are misunderstood.

“If a person doesn’t have a diagnosis, services won’t necessarily make the adjustments that are needed to help them access care and support in a way that is not too overwhelming,” O’Nions told us. “As a result, they may give up trying to access support because the fear of being misunderstood creates too much anxiety.”