Mobile Phone Sensor Correlates of Depressive Symptom Severity in Daily-Life Behavior: An Exploratory Study.
Saeb S, Zhang M, Karr CJ, Karr CJ, Schueller SM, Corden ME, Kording KP, Mohr DC. · 2015
Read Mobile Phone Sensor Correlates of Depressive Symptom Severity in Daily-Life Behavior: An Exploratory Study. by Saeb S, Zhang M, Karr CJ, Karr CJ, Schueller SM, Corden ME, Kording KP, Mohr DC. (2015) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc4526997.txt.
Introduction
Sohrab Saeb leads a ten-author team testing whether phone sensors can flag depression, in the Journal of Medical Internet Research in 2015. Depression is common and often missed. Phones are carried everywhere and packed with sensors, so their data might reveal daily patterns that track mood. The team set out to explore that link.
Forty adults carried a phone running a data-gathering app for two weeks; 28 gave enough data to analyze. Each person also filled in a standard depression survey at the start. Location patterns, such as how regular a person's daily movement rhythm was, lined up with symptom severity, as did some phone-use measures. The authors frame the work as an early step toward passive mood screening.
Catalog record
| CREATOR | Saeb S, Zhang M, Karr CJ, Karr CJ, Schueller SM, Corden ME, Kording KP, Mohr DC. |
|---|---|
| YEAR | 2015 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Mobile Phone Sensor Correlates of Depressive Symptom Severity in Daily-Life Behavior: An Exploratory Study By Gunther Eysenbach, Jeroen Ruwaard, Jakob Bardram, Sohrab Saeb, Mi Zhang, Christopher J Karr, Stephen M Schueller, Marya E Corden, Konrad P Kording, David C Mohr Published in: Journal of Medical Internet Research DOI: 10.2196/jmir.4273 ABSTRACT Background Depression is a common, burdensome, often recurring mental health disorder that frequently goes undetected and untreated. Mobile phones are ubiquitous and have an increasingly large complement of sensors that can potentially be useful in monitoring behavioral patterns that might be indicative of depressive symptoms. Objective The objective of this study was to explore the detection of daily-life behavioral markers using mobile phone global positioning systems (GPS) and usage sensors, and their use in identifying depressive symptom severity. Methods A total of 40 adult participants were recruited from the general community to carry a mobile phone with a sensor data acquisition app (Purple Robot) for 2 weeks. Of these participants, 28 had sufficient sensor data received to conduct analysis. At the beginning of the 2-week period, participants completed a self-reported depression survey (PHQ-9). Behavioral features were developed and extracted from GPS location and phone usage data. Results A number of features from GPS data were related to depressive symptom severity, including circadian movement (regularity in 24-hour rhythm; r=-.63, P=.005), normalized entropy (mobility between favorite locations; r=-.58, P=.012), and location variance (GPS mobility independent of location; r=-.58, P=.012). Phone usage features, usage duration, and usage frequency were also correlated (r=.54, P=.011, and r=.52, P=.015, respectively). Using the normalized entropy feature and a classifier that distinguished participants with depressive symptoms (PHQ-9 score ≥5) from those without (PHQ-9 score <5), we achieved an accuracy of 86.5%. Furthermore, a regression model that used the same feature to estimate the participants’ PHQ-9 scores obtained an average error of 23.5%. Conclusions Features extracted from mobile phone sensor data, including GPS and phone usage, provided behavioral markers that were strongly related to depressive symptom severity. While these findings must be replicated in a larger study among participants with confirmed clinical symptoms, they suggest that phone sensors offer numerous clinical opportunities, including continuous monitoring of at-risk populations with little patient burden and interventions that can provide just-in-time outreach. INTRODUCTION Depression is a common mental disorder. Estimates of the 12-month prevalence rate for major depressive disorder range from 6.6-10.3%, while lifetime risk has been estimated at 16.6-17.1% [1-3]. Subthreshold depressive symptoms are far more common [4], causing significant impairment in people’s lives and putting them at risk for future mental health concerns [5]. Depression, at both diagnosable and subthreshold levels, imposes a very high societal burden in terms of cost, lost productivity, morbidity, suffering, and mortality [6-8] and is a leading cause of disability and disease burden worldwide [9]. By the year 2020, the World Health Organization estimates that depression will be the second largest cause of “lost years of healthy life” worldwide [10]. Depression is treatable using a variety of methods, including antidepressants and psychotherapy; however, very few people who need treatment receive it [11]. It often takes months or years for depression to be identified and treated in our health care system, when it is treated at all [12]. One of the most common settings where depression is managed is primary care [13,14]; however, primary care physicians might fail to identify most patients with depressive symptoms [15,16]. Thus, more efficient methods of monitoring could significantly improve the delivery of services to those in need. Mobile phones are becoming the most ubiquitous consumer device in our world. Equipped with powerful sensing, computation, and communication capabilities, mobile phones—specifically smartphones—can continuously monitor an individual’s context including physical activity, location, and environment. Depression is associated with several behavioral components (eg, reduction in activity, psychomotor retardation, changes in sleep) and motivational states (eg, anhedonia), some of which may be detectable using mobile phone sensors [17,18]. Thus, mobile phones hold significant promise as a platform to monitor behavioral and environmental indicators of risk and resilience and to improve long-term management and treatment delivery to people suffering from depression. Indeed, some work has shown promise in this area. A first study in our group found that phone sensor data could detect social patterns among depressed patients, but this was a small study with only 8 participants [19]. Other groups have found that phone sensors were effective at detecting social and sleep behaviors among patients with depression [20,21], and such features correlated significantly with severity of depressive symptoms [22]. The aim of this study was to extend previous work by focusing specifically on behavioral markers related to movement through geographic space, which we hypothesized would be related to depressive symptom severity, given depression results in decreased motivation, withdrawal, and activity. In addition, excessive use of mobile phones is considered compulsive behavior and has been linked to some symptoms of depression [23,24]. Thus, we also explored the relationship of depression symptom severity to the use of the phone that was used to collect the sensor data. To achieve these aims, we used our mobile phone app, Purple Robot [25,26], to collect global positioning system (GPS) location and phone usage data from participants with varying levels of depression severity. We defined a number of behavioral features based on these data and built classification and regression models to examine their relationship to depression symptom severity. METHODS Participants We recruited 40 adult participants from April-July 2013 using craigslist advertisements. Participants were eligible if they had an email account, computer, and broadband access to the Internet, were within a cellular network range the majority of the day, were able to speak and read English, were at least 19 years of age, and lived within the United States of America. Participants signed an online consent form, and research staff reviewed the consent over the phone. The study was approved by the Northwestern University Institutional Review Board. At the beginning of the study, participants were asked to complete an online assessment consisting of a demographics questionnaire and the Patient Health Questionnaire-9 (PHQ-9), a commonly used measure of self-reported depressive symptom severity [27], which produces scores ranging from 0-27. Scores of less than 5 indicate no depression, 5-9 mild depression, 10-14 moderate depression, 15-19 moderately severe, and over 20 severe depression [28]. If participants owned and used an Android device with operating system 2.3 (Gingerbread) or higher, research staff assisted with the download, installation, and configuration of Purple Robot (see the Purple Robot section below). Participants who did not own a compatible phone were given an Android Nexus 4 with Purple Robot installed and configured. Phones were either picked up from the study’s office or mailed directly to the participant. Participants were instructed to keep the phone with them and charged throughout the day for 2 weeks. In addition, research staff explained that Purple Robot would be collecting GPS location and phone usage data. As part of the purpose of this study was to test Purple Robot’s functionality in the field, research staff checked in periodically, via phone and email, with the participants to ensure the app was working properly and to answer questions. Purple Robot Purple Robot is an open-source Android app that we developed to collect mobile phone sensor data [25,26].