Tools for App- and Web-Based Self-Testing of Cognitive Impairment: Systematic Search and Evaluation.

Background: Tools for app- and Web-based self-testing for identification of cognitive impairment are widely available but are of uncertain quality.
Objective: The objective of this study was to undertake a scoping review of app- and Web-based self-tests for cognitive impairment and determine the validity of these tests.
Methods: We conducted systematic searches in electronic databases, including Google search, Google Play Store, and iPhone Operating System App Store, using the search terms "Online OR Internet-based AND Memory OR Brain OR Dementia OR mild cognitive impairment OR MCI AND Test OR Screen OR Check."
Results: We identified 3057 tools, of which 25 were included in the review. Most tools meeting the inclusion criteria assessed multiple cognitive domains. The most frequently assessed domains were memory, attention, and executive function. We then conducted an electronic survey with the developers of the tools to identify data relating to development and validation of each tool. If no response to the survey was received, Google (to identify gray literature), Google Scholar, and Medical Literature Analysis and Retrieval System Online were searched using key terms "(name of developer, if available)" AND "(the name of the tool)" to identify any additional data. Only 7 tools had any information concerning psychometric quality, and only 1 tool reported data on performance norms, reliability, validity, sensitivity, and specificity for the detection of cognitive impairment.
Conclusions: The number of cognitive self-assessment electronic health tools for cognitive impairment is increasing, but most are of uncertain quality. There is a need for well-validated tools and guidance for users concerning which tools provide reliable information about possible cognitive impairment that could warrant further investigation.
(©Anna Pavlina Charalambous, Annie Pye, Wai Kent Yeung, Iracema Leroi, Malcolm Neil, Chryssoula Thodi, Piers Dawes. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 17.01.2020.)