Anouk Meijs did her internship in epidemiology at the department of Geriatric medicine of the Radboud University Medical Center and is now working as an epidemiologist at the National Institute of Public Health and the Environment in the Netherlands. She has recently been published in Age and Ageing journal.
Although Alzheimer’s disease makes up the bulk of dementia diagnoses, many other forms of dementia exist as well. Establishing the correct dementia diagnosis is important for treatment but can be difficult. Therefore, we use additional diagnostic tests such as brain scans and lumbar punctures, but it is not really known what benefits these tests – that always come with costs and patient burden – have to offer in daily practice.
The benefits of diagnostic tests are uncertain largely because there are no ‘gold standards’ for the dementia subtypes to serve as a benchmark against which to compare the final diagnosis. We therefore investigated whether and how additional diagnostics actually change the clinical diagnoses established in every day practice. If the additional tests don’t change the diagnosis that is made after a basic clinical workup (history taking, laboratory tests, cognitive screening and physical and neurological examination) then what is the point? We studied this research question by evaluating the diagnostic process in our memory clinic.
The diagnostic process of persons with suspected dementia in our hospital starts with a basic clinical evaluation leading to a clinical diagnosis set by a multidisciplinary team. If the clinicians involved remain uncertain about the diagnosis, additional diagnostic tools (including MRI scan of the brain, cerebrospinal fluid analysis by lumbar puncture and extensive neuropsychological assessment) are requested.
For this study we evaluated the diagnostic process of all patients who visited our memory clinic between 2007 and 2010. During this period 767 patients were examined in our clinic, of whom 70% underwent additional diagnostics.We categorized the initial clinical diagnosis and the final diagnosis after additional testing into three groups: Alzheimer’s dementia, non-Alzheimer’s dementia and no dementia. Our main finding was that additional diagnostics changed the diagnosis in 17% of the patients. In patients with an initial clinical diagnosis of non-Alzheimer’s dementia the diagnosis changed in 54% of the cases, whereas the diagnosis changed far less frequently in the patients with a clinical Alzheimer’s disease or no dementia diagnosis (11% and 14%, respectively).
From these results it can be concluded that additional diagnostic tools should especially be considered in patients suspected of non-Alzheimer’s disease, whereas in the large group of patients with an uncomplicated picture of Alzheimer’s disease, or no dementia, additional tests should be be used with more restraint.
British Geriatrics Society 
This is a really helpful and careful analysis of the utility of diagnostic investigation of suspected dementia. I would observe that the one extra dimension that might have been helpful to include is that of time – sometimes the most invaluable of diagnostic tools.
The impact of knowing the clinical trajectory over time, in my experience, is often underestimated – even though increasingly patients (or more often, in this setting, their advocates!) want a result now, rather than in a few weeks or months, even if this will have little effect on the overall outcome.
Please don’t undervalue, or worse still ignore, this key component to the diagnostic process!