Is autism being overdiagnosed?

Our Autism Champion Dr Conor Davidson examines the rise in autism diagnoses in the UK.

This is one of the questions I am most often asked in my role as Autism Champion, by both psychiatry colleagues and laypeople. Most people have a sense that ‘there is more autism about’ these days. It is now rare to find someone who doesn’t have a friend or relative affected by autism. A recent article in the Guardian newspaper highlighted the issue of a large rise in autism diagnoses in the UK. But does this mean that autism is being overdiagnosed?

I first became interested in the field of autism nearly 20 years ago. There is no doubt in my mind that, back then, autism was underdiagnosed. I encountered lots of patients in the mental health system whose difficulties, in retrospect, would have been better explained by autism. They tended to have diagnoses like schizoid personality disorder or simple schizophrenia, neither of which are in widespread clinical use anymore. I’ve also seen a number of female patients with a diagnosis of borderline personality disorder who I now suspect had undiagnosed autism. In those days, clinicians rarely considered the possibility of autism in women and girls as it was seen as mainly affecting boys.

Things have changed since then. We have a much better understanding now of how autism typically presents in women and girls, and the M:F sex ratio has narrowed from 9:1 to closer to 2:1. My adult autism clinic in Leeds now receives more female than male referrals. Time trends in autism incidence show that the bulk of the increase in recent years is due to more females, particularly adult females, being diagnosed.

By and large, I think this is something to be celebrated. People who were overlooked or misdiagnosed in previous decades are now being correctly identified as autistic. Most people diagnosed with autism report positive benefits in terms of self-acceptance, self-understanding, connecting to the ‘neurodivergent community’, and accommodations in education and workplace. It’s hugely encouraging that psychiatrists are taking such an interest in autism now (as evidenced by the enormous popularity of the National Autism Training Program for Psychiatrists), as an accurate diagnosis of autism in mental health settings can help with care and treatment planning, to access more support in the community, and allow appropriate reasonable adjustments to be made.

That said, there is emerging evidence that conversion rates (i.e. proportion of autism assessments that result in an autism diagnosis) are variable in different parts of the country. This is more likely due to differences in service and clinician factors (such as referral, triage and assessment processes and individual clinicians’ diagnostic threshold) than population differences between geographical regions.

It's important that the clinical community work towards better harmonisation of assessment processes and diagnostic threshold, as neither underdiagnosis nor overdiagnosis are desirable outcomes. As a psychiatrist, my main concern about the latter is that people with treatable mental health conditions - such as anxiety disorder or OCD - could be mislabelled as autistic and miss out on potentially helpful treatments. Last year a new national autism practitioners’ network was set up, with the aim of sharing best practice and reducing unwanted variation in practice. I encourage those of you who work in the field of autism to sign up to it.