Quantifying digital behavior on smartphones in patients with stroke - QuantStroke:A longitudinal cohort study - Protocol

Abstract

Question Unobtrusive, convenient and innovative approaches for long term monitoring after disease are needed to detect functional decline. Modern digital behavior, as reflected in smart phone use, can be applied to monitor health in different groups of users with and without health issues. People with stroke are among the most relevant target groups for unobtrusive monitoring, since many have to live with lasting disabilities and are prone to declining function, cognitive impairment and depression. In this study, we want to examine the linkage between smart phone use after stroke and health related quality of life. Methods We want to record continuously the interaction of 50 stroke survivors with their smartphones during the first 3 months after discharge. Patients with confirmed stroke who can provide informed consent are eligible. Touchscreen data gathering is done in the background. Shortly before discharge from inpatient rehabilitation, participants will be asked to install the app on their smart phones or tablet computers (Android users only due to system restrictions) and continue to use those as usual. We shall collect the touchscreen data at a millisecond resolution using the off-the-shelf data gathering tool available through QuantActions, Lausanne. Raw smartphone taps will be converted into candidate measures for health, such as (i) Tapping speed (ii) Typing speed (iii) The speed of entering a passcode (iv) The speed of finding an App icon on the home screen (v) The number of interactions No images or texts messages or other user generated content will be collected and cannot be inferred from the raw data. The linkage between smart phone use and health related quality of life will be assessed with standardized outcome measures such as EQ-5D-5L (weekly) and Stroke Specific Quality of Life and SF 36 (monthly). Participants can opt-in for a few additional memory tests. Outlook This is an ongoing study. With the acquired data, we want to contribute to a database where digital behavior is analyzed with advanced computational methods. Different groups of users can be compared, with the objective to identify potential indicators of functional decline. The study is funded by Carlsberg Foundation.

Authors 2

1. Iris Charlotte Brunner
   Affiliation as printed

   Department of Clinical Medicine - Hammel Neurocenter, Universitetsklinik for Neurorehabilitering

2. Arko Ghosh Aachen

   Leiden University

   Affiliation as printed

   Leiden University