Instantly diagnosing mild dementia; not possible/not desirable

Today GPs are no good at diagnosing dementia. Clare Gerada did good work on You and Yours today discussing a paper just out from Leicester University.

The press release says that “general practitioners (GPs) are struggling to correctly identify people in the early stages of dementia resulting in both missed cases (false negatives) and misidentifications (false positives)”. I think this is  unhelpful. It doesn’t reflect the organic pattern of managing risk and uncertainty in general practice.

This is how it normally happens. A person comes in feeling anxious. Ever since their brother died, who they cared for, they have been aware that their memory is not quite as good. The person can manage everything needed to do, but sometimes finds that a word or phrase does not come quickly to the lips. We talk. There is distress; the bereavement has been hard, there are lonely times. Here there are many reasons why a memory problem may have occured; dementia isn’t the only possibility; anxiety, depression, or a change in daily prompts may all have caused the difficulty. Would it be helpful for me to diagnose dementia straight away? No.

Here’s why. There is no lab test that we can do to give us an instant diagnosis of dementia. Sometimes it is useful to to lab tests to exclude other things, such as a urine infection or a thyroid abnormality. If we are considering the diagnosis of dementia which has major implications, we will want to know how symptoms have developed – does it flucuate? – is it progressively worsening?- can we find out what the family think (patient thinks fine, daughter notices gas left on) as well. We may want specialist psychological assessment. General practice is not just about the instant, but the longer term. Diagnosis can be excluded or made over time; time is a sharp diagnostic tool.

The study looked at patients who were being assessed for dementia. The headline on the press release was that “People presenting with memory problems and mild dementia are often not diagnosed promptly in primary care.” Yet the study showed that clinical judgement was correct three quarters of the time; yet GPs only wrote down their suspicions in the notes a third of the time. There may be reasons for this; not least the depression that often accompanies the diagnosis, the lack of effective early treatments, the implications for driving and life insurance. You want to be pretty certain before making the diagnosis.

The authors refer to another of their papers in recommending mental state examinations which can be done in people with memory problems. However, this paper gives sensitivities of 72% and specificities of 88%, which is not quite the same as the reference in the press release that “the application of a simple cognitive screening test after a clinical diagnosis would help GPs to achieve about 90% accuracy.”

Primary and secondary care have to work well together in order to look after people with memory problems and dementia. This means appreciating the manner in which memory problems come to our attention, and having an eye on the harms done by false positives that will inevitably arrive with ‘faster’ diagnosis.

One Response to “Instantly diagnosing mild dementia; not possible/not desirable”

  1. michael (@wassabeee) November 22, 2012 at 8:04 pm #

    I came home to a very upset wife who had scored 5 out of 5 on the “Do you know the signs of early dementia?” NHS / Alzheimers’ Society leaflet that the postman delivered today.

    I consoled her by telling her that I had scored 5/5 for as long as I can remember.

    I immediately visited the Alzheimer’s Society web site and was upset by the transparent opacity about their funding from the pharmaceutical industry: “We want to be as clear as possible about how we raise our money … 40%, £28.7 from donations” – but no indication about how much comes from individuals and how much comes from companies with a potential conflict of interest.