JMIR Publications

ABSTRACT

Background:

Among the areas that has challenged the progress of dementia care has been the assessment of change in symptoms over time. Digital biomarkers provide an interesting approach as they allow objective, ecologically valid, and long term follow-up with continuous assessment. There is a multitude of sensors and devices that can be applied but there are no agreed upon standards for digital biomarkers, nor is there comprehensive evidence-based results for which digital biomarkers may be most effective.

Objective:

In this review, we try to answer the following questions: (i) what is the evidence for real life in-home use of technologies for early detection and follow-up of MCI or dementia? And, (ii) what transformation might clinicians expect in their everyday practices?

Methods:

A systematic search was conducted in PubMed, Cochrane and Scopus databases from inception to July 2018 for studies examining the implementation of digital biomarker technologies for MCI to mild Alzheimer’s disease follow-up and detection in real life setting. All studies including community dwelling older adults (65+), cognitively healthy or presenting cognitive decline (from subjective cognitive complaints to early Alzheimer disease), focused on technology-based in-home evaluation for non-interventional follow-up and remote diagnosis of cognitive deterioration were examined.

Results:

An initial sample of 4811 English-language papers were retrieved. After screening and review, 27studies were eligible for inclusion in the review. These studies ranged from 12 to 279 participants for durations between three days to 3.6 years. Most common reasons for exclusion were: inappropriate setting (e.g. hospital setting), intervention (e.g. drugs, rehabilitation) or population (e.g. psychiatry, Parkinson disease). We generally summarized these studies into four groups (although there is overlap) based on the proposed technological solutions to extract relevant data: (i) data from dedicated embedded or passive sensors, (ii) data from dedicated wearable sensors, (iii) data from dedicated technological solutions (e.g. games or surveys), and (iv) data derived from non-dedicated technological solutions use (e.g. computer mouse movements).

Conclusions:

The first key observation is that few papers dealt with in-home, real life evaluations. Most technologies were far removed from everyday life experiences and are not mature enough for use under non optimal conditions. Nevertheless, the evidence base from embedded sensors appear to represent the relatively most mature research area and we can reasonably state that several indicators turn green: some of these solutions could be proposed to a large population in the coming decade. It’s time for the clinicians to get interested in this field.