Early diagnosis of Alzheimer's: 85% accurate?

So I set off looking for what this might have been based on - the content and source of the Press Release for the story – to see who made it and what was actually claimed. And if anyone said anything about the 15% of inaccurate diagnoses.

 

The 85% accuracy seemed to be good news so I assumed the 15% inaccuracy was not. If the scan implied ‘OK’ when someone did have Alzheimer’s – the false negatives - that would be reviewed when symptoms began to appear, but what if the inaccuracy went the other way when you are in fact OK but early Alzheimer’s is detected – the false postives. How much weight does the scan have in clinical decision making? Could you get a wrong diagnosis and all the effects that go with it – mostly tied to unnecessary treatment and unnecessary psychological harm. Can a diagnosis of dementia be easily rescinded? Depression in the elderly is under diagnosed and treated and can be an alternative diagnosis.

 

I got a bit further but didn’t uncover anything about the 15% inaccuracy.

 

The Alzheimer’s Society commented as below, adding some corporate key messages about dementia and a positive spin on new technology and its advantages. Their interest was piqued because “getting an early diagnosis enables people with dementia to understand their condition, access treatments that could help relieve their symptoms and give them time to plan for the future”.  It wasn’t them that started the story and they made no mention of 85% anything. They merely commented after the fact.

Published 8 March 2011

A new brain scan which can detect early signs of Alzheimer’s is being trialed at memory services in Croydon, Lambeth and Southwark.

 

An 'Automated MRI' software package compares a patient's brain scan against 1,200 others showing varying stages of Alzheimer's disease. This is the first time such software has been used by the NHS.”

 

The scan has been developed by scientists at the National Institute for Health Research Biomedical Research Centre for Mental Health at the Maudsley Hospital and Kings College London together with colleagues from the Karolinksa hospital in Stockholm.”

 

So the new scan technology is about to be trialled, thus it is not a proven technology yet, just being proved. It is also being trialled on people already with symptoms that have lead them to a memory clinic. If someone is already attending a memory clinic they have symptoms that raise concerns (possibly wrongly, but there is still a need for an expert clinic assessment) already. It is a story about testing the new scan technology on patients, mimicking how it would be used in clinical practice going further than seeing if, in principle, it works.

 

The press release could have been by one of three places - National Institute for Health Research Biomedical Research Centre for Mental Health at the Maudsley Hospital, or Kings College London or possibly Karolinksa hospital in Stockholm.

 

I can’t find any reference at all on the National Institute for Health Research Biomedical Research Centre for Mental Health at the Maudsley. So it’s not them. Nothing identifiable on the Karolinksa Hospital site so probably not them either.

 

In among, a random news web service www.physorg.com managed a bit more detail as follows:-

 

“The 'Automated MRI' software automatically compares or benchmarks someone’s brain scan image against 1200 others, each showing varying stages of Alzheimer’s disease. This collection of images is thought to be the largest of its kind in the world.

 

Normally in routine clinical practice, brain scans are used to simply exclude diseases that can mimic Alzheimer’s disease, but here automated MRI software is being used for the first time in a NHS setting (Memory Clinics) to make an early and accurate diagnosis of the illness. 

Early diagnosis of Alzheimer's is clinically difficult and patients with the early signs are frequently not treated until their symptoms become stronger. The new scan however can return 85 per cent accurate diagnostic results in under 24 hours.

An early diagnosis allows someone to plan their care before the condition worsens - helping to prevent institutionalisation, dramatically improving their quality of life. It is also a cost effective and efficient way to manage and organise treatment of the disease.

 

The scan has been developed by scientists at the IoP, together with colleagues from the Karolinska Hospital in Stockholm.

The system is being 'field tested' over the next 12 months with patients attending SLaM memory services in Croydon, Lambeth and Southwark. The ‘field test’ will also provide a supply of research grade images, which has important implications for the development of the next generation of drugs for dementia and individualised treatments.

 

There are 750,000 people with dementia in the UK. The financial cost of dementia to the UK is over £20 billion a year. According to the Alzheimer’s Society, in just 15 years a million people will be living with dementia. This will soar to 1.7 million people by 2051.

 

Provided by King's College London

 

Aha!

 

So I find the 85% figure, and the authors of the Press Release (Kings College) and the fact that the scans are being field tested i.e. seeing if they actually produce meaningful clinical information that assists in managing the treatment and care of people with Alzheimer’s.

 

It’s pretty much as written on the King’s media web page here.

 

How did King’s manage to achieve no mention in any news coverage despite doing the Press Release? Poor them.

 

And I still don’t know what anyone thinks about the 15% who get an inaccurate diagnosis – especially the false positives.