Mike: Hello, I'm Mike Clarke, podcast editor for the Cochrane Library. Alongside the dozens of Cochrane Reviews of possible treatments for dementia, are reviews looking at how to diagnose it, including the use of a phone or video call rather than an in-person visit. In April 2022, we added a new review to these, evaluating the accuracy of remote cognitive assessment and we asked Amy Elliott to speak with lead author and colleague at the University of Leicester in England, Lucy Beishon about the review for this podcast.

Amy: Hi Lucy, firstly, please would you sum up what is meant by a remote cognitive assessment?

Lucy: Hi Amy. It's a way of assessing someone's ability to think, learn, or remember things, in particular to help diagnose dementia, that is done during a virtual consultation. Usually, this is either a telephone call or over a video-link. Various assessment tools are in use, which are either purpose-built for a virtual consultation, or long-standing assessment tools that have been adapted for virtual use.

Amy: Thanks. The review looks at the accuracy of using these remote cognitive assessments in diagnosing dementia, so could you tell us why we need to have good evidence on this?

Lucy: Dementia is a common, progressive illness and it's important that it's identified accurately, so that we don't miss people who have dementia, but also so that we don't cause unnecessary distress by incorrectly diagnosing dementia when it isn't there. 
Since the beginning of the Covid-19 pandemic, an increasing number of appointments have been moved from face-to-face to virtual and although we could assume that the assessment tools would work the same whether done remotely or in-person, some assessments require visual information to be given, or use questions based on orientation to a new environment, which are hard to replicate if done remotely. 
I experienced this first hand during the pandemic when attempting to use one of these tools over the phone and trying to verify that the patient's location in a supermarket car park was correct! It's therefore important that we make sure that our assessments are as accurate when done virtually as when they're done face to face. 

Amy: So what does the evidence say about the accuracy of remote cognitive assessments? 

Lucy: We found seven studies involving just over 750 participants that compared a remote cognitive assessment to a gold standard diagnosis of dementia. By gold standard, I mean a diagnosis from an experienced clinician based on information from the patient's history, examination, and investigations such as brain scans and blood tests, as well as information from a cognitive assessment tool. Six of the studies looked at telephone assessments, and one at a video-call assessment. All the studies were at some risk of bias, and they had a wide range of accuracy results, with no test clearly being superior.
We also found 24 papers that compared a remote assessment to its equivalent in-person test, for example, a telephone version of a common test called the mini-mental state examination, or MMSE for short, compared to an in-person MMSE. There was good agreement between the two ways of doing the tests, but this was rarely perfect.

Amy: You mentioned the comparison of the remote tests against a gold standard of a dementia diagnosis, compared to an in-person test, why is that important?

Lucy: Cognitive assessments are commonly used to help with a diagnosis of dementia, but they are not enough on their own. There are common problems with them such as language barriers or the person's level of education. Even the most used cognitive assessment isn't perfect at telling us whether a person has dementia or not. So, when we look at a remote assessment, we should look at how good it is at telling us whether someone has dementia, rather than how it compares to an imperfect in-person test.

Amy: Thanks Lucy. Overall, what's your take home message about the diagnostic accuracy of remote cognitive assessments and how can people get hold of the full review?

Lucy: I'd say that although remote tests are increasingly used, there is not enough good-quality evidence right now to tell us how accurate they are. Carrying out consultations virtually is complicated, and very variable. We need more research to be done before we can recommend any particular test for remote use and, for now, although they do have some use as part of a wider assessment, an in-person version would still be required. Finally, to get the full version of the review, people need to go to Cochrane Library dot and search 'remote diagnosis of dementia'. They'll then see a link to the review.