Loneliness and Social Isolation Detection Using Passive Sensing Techniques: Scoping Review

Abstract

Background Loneliness and social isolation are associated with multiple health problems, including depression, functional impairment, and death. Mobile sensing using smartphones and wearable devices, such as fitness trackers or smartwatches, as well as ambient sensors, can be used to acquire data remotely on individuals and their daily routines and behaviors in real time. This has opened new possibilities for the early detection of health and social problems, including loneliness and social isolation. Objective This scoping review aimed to identify and synthesize recent scientific studies that used passive sensing techniques, such as the use of in-home ambient sensors, smartphones, and wearable device sensors, to collect data on device users’ daily routines and behaviors to detect loneliness or social isolation. This review also aimed to examine various aspects of these studies, especially target populations, privacy, and validation issues. Methods A scoping review was undertaken, following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews). Studies on the topic under investigation were identified through 6 databases (IEEE Xplore, Scopus, ACM, PubMed, Web of Science, and Embase). The identified studies were screened for the type of passive sensing detection methods for loneliness and social isolation, targeted population, reliability of the detection systems, challenges, and limitations of these detection systems. Results After conducting the initial search, a total of 40,071 papers were identified. After screening for inclusion and exclusion criteria, 29 (0.07%) studies were included in this scoping review. Most studies (20/29, 69%) used smartphone and wearable technology to detect loneliness or social isolation, and 72% (21/29) of the studies used a validated reference standard to assess the accuracy of passively collected data for detecting loneliness or social isolation. Conclusions Despite the growing use of passive sensing technologies for detecting loneliness and social isolation, some substantial gaps still remain in this domain. A population heterogeneity issue exists among several studies, indicating that different demographic characteristics, such as age and differences in participants’ behaviors, can affect loneliness and social isolation. In addition, despite extensive personal data collection, relatively few studies have addressed privacy and ethical issues. This review provides uncertain evidence regarding the use of passive sensing to detect loneliness and social isolation. Future research is needed using robust study designs, measures, and examinations of privacy and ethical concerns.

Sensor-based assessment of social isolation incommunity-dwelling older adults: A scopingreview

“…Due to these major differences in research question, search strategy and analysis, there are only two studies (i.e. Goonawardene et al [23] and Martinez et al [24]) that overlaps between ours and Qirtas et al [22] reviews.…”

“…However, our work is distinctly different in the following ways: (i) Our review is specifically focused on the use of different sensing modalities for detecting SI among community dwelling older adults, where the risks of isolation is predominant. The review by Qirtas et al [22] focuses on general population including younger adults, older adults and mixed groups, (ii) Qirtas et al [22] studied loneliness and SI has a unified construct, whereas in our scoping review we clearly differentiate loneliness as as a subjective feeling in comparison to social isolation that can be measured objectively. As a result, we only selected papers that conform to the objective determinant of social isolation.…”

“…Recently, a scoping review has been published on detecting loneliness and social isolation using passive sensing by Qirtas et al [22]. However, our work is distinctly different in the following ways: (i) Our review is specifically focused on the use of different sensing modalities for detecting SI among community dwelling older adults, where the risks of isolation is predominant.…”

Sensor-based assessment of social isolation in community-dwelling older adults: a scoping review

“…One recent non-systematic literature review [ 23 ] and another scoping review [ 27 ] aimed to survey and synthesize research on technology applications for the detection and monitoring of SI. Bouaziz et al [ 23 ] presented an in-depth survey of both wearable and non-wearable (ambient) sensors, software and algorithms that have been applied in the monitoring of older adults’ daily activities, with many conducted in laboratory settings.…”

“…They specifically focused on on meal-taking and mobility from which the risk of SI may be inferred at a later stage that was not detailed in the review. The scoping review by Qirtas et al [ 27 ] examined passive sensing including smartphones, wearable devices and ambient sensors to detect loneliness and SI. This review covered a general population that comprised younger adults (34% or 10 of 29 included studies), older adults (41% or 12 of 29 included studies) and mixed age (17% or 5 of 29 included studies) groups.…”