Podcast: How accurate are cognitive assessment tools, that are completed by people themselves, at detecting dementia?

There are more than 100 Cochrane reviews relevant to dementia, and these were added to in December 2024 with a new review of the accuracy of self-administered cognitive assessment tools for its detection.

- Read transcript

Mike: Hello, I'm Mike Clarke, podcast editor for the Cochrane Library. There are more than 100 Cochrane reviews relevant to dementia, and these were added to in December 2024 with a new review of the accuracy of self-administered cognitive assessment tools for its detection.

Today, I’m joined by one of the authors, Fariha Naeem from Glasgow Royal Infirmary in Scotland, to talk about the review and its findings. Hi Fariha, thanks for telling us about this important new review. Perhaps you could begin by saying a little about self-administered cognitive assessment tools?

Fariha: Hi Mike, it’s good to be here. Briefly, these are tools designed to assess various aspects of cognition, such as memory and language, which are completed by the person themself without help from a healthcare professional. They may require them to answer some questions or complete some tasks.

Mike: So how might the tools be used in the diagnosis of dementia?

Fariha: Firstly, it’s important to say that these tools are not designed to diagnose dementia on their own but, rather, they would be used within the diagnostic process, which usually involves a clinical assessment, blood tests and imaging. Their role in this process is not clear at the moment but they have the potential to be used in both research and clinical situations. From a research perspective, they could be used to screen participants for studies. While, from a clinical perspective, they could be used where face-to-face cognitive assessment might be difficult. This was highlighted recently in the COVID-19 pandemic, but other potential uses are when cognitive assessments are done in rural areas or resource-poor settings, where in-person assessment with a clinician may be challenging.

Mike: With that range of potential uses, it’s clearly important to know if they actually work. So, how much evidence did you find about their accuracy in diagnosing dementia?

Fariha: We identified 11 studies involving 2300 participants that had compared self-administered cognitive assessment tools to a standard clinical assessment for dementia diagnosis. Five of these studies looked at a tool called ‘Test Your Memory’ in five different languages. The other six studies looked at five other self-administered assessment tools. Most of the tools were in print format but two were electronic.

Mike: How good were the 11 studies?

Fariha: Most were at some risk of bias, especially with regard to patient selection, and there was significant heterogeneity between the studies. For example, they took place across a range of settings, including community and residential settings, as well as secondary healthcare and outpatient clinics. They also provided a wide range of accuracy results, with no one test being clearly superior to the others. Some of the studies used the same cognitive assessment tools, there was no consensus amongst the studies as to the test threshold or cut off score for indicating an ‘abnormal’ or ‘normal’ score.  The standard clinical assessment used to diagnose dementia was also different between the studies.

Mike: It sounds as though it was difficult to reach a consistent conclusion from the currently available research, but is there a take home message about the accuracy of self-administered cognitive assessment tools?

Fariha: Yes, I would say that self-administered cognitive assessment tools have the potential to be useful in the face of increased demand for cognitive assessments with limited healthcare resources. Their test accuracy scores were similar to the range seen with standard pencil and paper cognitive testing, which is conducted by clinicians. However, we don’t have enough evidence to tell us how accurate they are or to recommend the use of any single tool. We need more research to establish the optimal setting, the test threshold and the role that these tests might play in the diagnosis of dementia.

Mike: Thanks very much for joining us today Fariha and telling us about this really important review. If listeners would like to read the full version of the review, they can go to CochraneLibrary.com and search “self‐administered assessment for dementia” to see a link to it.

Close transcript