18% patients with Multiple Sclerosis are Misdiagnosed (2019) Kaisey et al

I agree with @Woolie that there are lots of problems with drawing conclusions from this study.

MS is defined as multiple episodes of neurological deficit differing in time and site of involvement. That means that in the ideal situation everyone sent with a diagnosis of MS to a specialist centre does not officially have it. Because ideally they will be sent with their first attack. And the 18% figure is likely to be an artefact of the particular referral dynamics locally. Elsewhere it might be more or less.

I doubt it tells us anything very useful.

 

For UK patients. If an MS patient isn't obvious and ME is suspected the guidelines recommend to do as little testing as possible and so the MS diagnosis will be missed. But there won't necessarily be an ME diagnosis since it's likely the MS will progress faster, leading to a correct diagnosis, before the wait-and-see period the ME guidelines recommend

Basically, it's a complete mess because what's written down isn't reliable and most of what would be useful to prevent mistakes isn't written down. Great work, NICE.

 

A BMJ article about diagnostic error - according to the authors 10 to 15% seems to be the norm. This article also discusses various methods to capture this data:

https://qualitysafety.bmj.com/content/22/Suppl_2/ii21


ETA: one suggestion in the BMJ article is "secret shoppers" - in medicine - I've thought of this for a study of how practitioners react to the ME, or "CFS" diagnosis - probably been done though. With hidden cameras as some TV programs do might be revealing re treatment and attitudes.

ETA#2: The hidden camera thing has been done in medical settings.