Dataset Information

Predicting Mood Disturbance Severity with Mobile Phone Keystroke Metadata: A BiAffect Digital Phenotyping Study.

ABSTRACT: BACKGROUND:Mood disorders are common and associated with significant morbidity and mortality. Better tools are needed for their diagnosis and treatment. Deeper phenotypic understanding of these disorders is integral to the development of such tools. This study is the first effort to use passively collected mobile phone keyboard activity to build deep digital phenotypes of depression and mania. OBJECTIVE:The objective of our study was to investigate the relationship between mobile phone keyboard activity and mood disturbance in subjects with bipolar disorders and to demonstrate the feasibility of using passively collected mobile phone keyboard metadata features to predict manic and depressive signs and symptoms as measured via clinician-administered rating scales. METHODS:Using a within-subject design of 8 weeks, subjects were provided a mobile phone loaded with a customized keyboard that passively collected keystroke metadata. Subjects were administered the Hamilton Depression Rating Scale (HDRS) and Young Mania Rating Scale (YMRS) weekly. Linear mixed-effects models were created to predict HDRS and YMRS scores. The total number of keystrokes was 626,641, with a weekly average of 9791 (7861), and that of accelerometer readings was 6,660,890, with a weekly average 104,076 (68,912). RESULTS:A statistically significant mixed-effects regression model for the prediction of HDRS-17 item scores was created: conditional R2=.63, P=.01. A mixed-effects regression model for YMRS scores showed the variance accounted for by random effect was zero, and so an ordinary least squares linear regression model was created: R2=.34, P=.001. Multiple significant variables were demonstrated for each measure. CONCLUSIONS:Mood states in bipolar disorder appear to correlate with specific changes in mobile phone usage. The creation of these models provides evidence for the feasibility of using passively collected keyboard metadata to detect and monitor mood disturbances.

SUBMITTER: Zulueta J

PROVIDER: S-EPMC6076371 | biostudies-literature | 2018 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting Mood Disturbance Severity with Mobile Phone Keystroke Metadata: A BiAffect Digital Phenotyping Study.

Zulueta John J Piscitello Andrea A Rasic Mladen M Easter Rebecca R Babu Pallavi P Langenecker Scott A SA McInnis Melvin M Ajilore Olusola O Nelson Peter C PC Ryan Kelly K Leow Alex A

Journal of medical Internet research 20180720 7

<h4>Background</h4>Mood disorders are common and associated with significant morbidity and mortality. Better tools are needed for their diagnosis and treatment. Deeper phenotypic understanding of these disorders is integral to the development of such tools. This study is the first effort to use passively collected mobile phone keyboard activity to build deep digital phenotypes of depression and mania.<h4>Objective</h4>The objective of our study was to investigate the relationship between mobile ...[more]

PMID: 30030209

Similar Datasets

Project description:BACKGROUND:Mobile technologies have the potential to be used as innovative tools for conducting research on the mental health and well-being of young people. In particular, they have utility for carrying out ecological momentary assessment (EMA) research by capturing data from participants in real time as they go about their daily lives. OBJECTIVE:The aim of this study was to explore the utility of a mobile phone app as a means of collecting EMA data pertaining to mood, problems, and coping efficacy in a school-based sample of Irish young people. METHODS:The study included a total of 208 participants who were aged 15-18 years, 64% female (113/208), recruited from second-level schools in Ireland, and who downloaded the CopeSmart mobile phone app as part of a randomized controlled trial. On the app, participants initially responded to 5 single-item measures of key protective factors in youth mental health (formal help-seeking, informal help-seeking, sleep, exercise, and sense of belonging). They were then encouraged to use the app daily to input data relating to mood states (happiness, sadness, anger, stress, and worry), daily problems, and coping self-efficacy. The app automatically collected data pertaining to user engagement over the course of the 28-day intervention period. Students also completed pen and paper questionnaires containing standardized measures of emotional distress (Depression, Anxiety, and Stress Scale; DASS-21), well-being (World Health Organization Well-Being Index; WHO-5), and coping (Coping Strategies Inventory; CSI). RESULTS:On average the participants completed 18% (5/28) of daily ratings, and engagement levels did not differ across gender, age, school, socioeconomic status, ethnicity, or nationality. On a scale of 1 to 10, happiness was consistently the highest rated mood state (overall mean 6.56), and anger was consistently the lowest (overall mean 2.11). Pearson correlations revealed that average daily ratings of emotional states were associated with standardized measures of emotional distress (rhappiness=-.45, rsadness=.51, ranger=.32, rstress=.41, rworry=.48) and well-being (rhappiness=.39, rsadness =-.43, ranger=-.27, rstress=-.35, rworry=-.33). Inferential statistics indicated that single-item indicators of key protective factors were related to emotional distress, well-being, and average daily mood states, as measured by EMA ratings. Hierarchical regressions revealed that greater daily problems were associated with more negative daily mood ratings (all at the P<.001 level); however, when coping efficacy was taken into account, the relationship between problems and happiness, sadness, and anger became negligible. CONCLUSIONS:While engagement with the app was low, overall the EMA data collected in this exploratory study appeared valid and provided useful insights into the relationships between daily problems, coping efficacy, and mood states. Future research should explore ways to increase engagement with EMA mobile phone apps in adolescent populations to maximize the amount of data captured by these tools. TRIAL REGISTRATION:Clinicaltrials.gov NCT02265978; http://clinicaltrials.gov/ct2/show/NCT02265978 (Archived by WebCite at http://www.webcitation.org/6mMeYqseA).

Project description:BACKGROUND:Objective behavioral markers of mental illness, often recorded through smartphones or wearable devices, have the potential to transform how mental health services are delivered and to help users monitor their own health. Linking objective markers to illness is commonly performed using population-level models, which assume that everyone is the same. The reality is that there are large levels of natural interindividual variability, both in terms of response to illness and in usual behavioral patterns, as well as intraindividual variability that these models do not consider. OBJECTIVE:The objective of this study was to demonstrate the utility of splitting the population into subsets of individuals that exhibit similar relationships between their objective markers and their mental states. Using these subsets, "group-personalized" models can be built for individuals based on other individuals to whom they are most similar. METHODS:We collected geolocation data from 59 participants who were part of the Automated Monitoring of Symptom Severity study at the University of Oxford. This was an observational data collection study. Participants were diagnosed with bipolar disorder (n=20); borderline personality disorder (n=17); or were healthy controls (n=22). Geolocation data were collected using a custom Android app installed on participants' smartphones, and participants weekly reported their symptoms of depression using the 16-item quick inventory of depressive symptomatology questionnaire. Population-level models were built to estimate levels of depression using features derived from the geolocation data recorded from participants, and it was hypothesized that results could be improved by splitting individuals into subgroups with similar relationships between their behavioral features and depressive symptoms. We developed a new model using a Dirichlet process prior for splitting individuals into groups, with a Bayesian Lasso model in each group to link behavioral features with mental illness. The result is a model for each individual that incorporates information from other similar individuals to augment the limited training data available. RESULTS:The new group-personalized regression model showed a significant improvement over population-level models in predicting mental health severity (P<.001). Analysis of subgroups showed that different groups were characterized by different features derived from raw geolocation data. CONCLUSIONS:This study demonstrates the importance of handling interindividual variability when developing models of mental illness. Population-level models do not capture nuances in how different individuals respond to illness, and the group-personalized model demonstrates a potential way to overcome these limitations when estimating mental state from objective behavioral features.

Project description:BackgroundMobile phones may help young people (YP) access health information and support health service engagement. However, in low-income settings there is limited knowledge on YP's phone and internet access to inform the feasibility of implementing digital health interventions.ObjectiveWe investigated access to information and communication technologies among adolescents and young adults in Zimbabwe.MethodsA cross-sectional population-based survey was conducted from October to December 2018 among YP aged 13-24 years in 5 communities in urban and peri-urban Harare and Mashonaland East, Zimbabwe. Consenting YP completed a self-completed tablet-based questionnaire on mobile phone ownership and use, and use of the internet. The primary outcome was the proportion who reported owning a mobile phone. Secondary outcomes included phone and internet access and use behavior, and ownership and use of other technological devices. Multivariable logistic regression was used to investigate factors associated with mobile phone ownership and with internet access, with adjustment for the one-stage cluster sampling design. A priori exploratory variables were age, sex, marital status, and urban/peri-urban residence.ResultsA total of 634/719 (88.2%) eligible YP, mean age 18.0 years (SD 3.3) and 62.6% (397/634) females, participated. Of the YP interviewed, 62.6% (396/633; 95% CI 58.5-66.5) reported owning a phone and a further 4.3% (27/633) reported having access to a shared phone. Phone ownership increased with age: 27.0% (43/159) of 13-15-year olds, 61.0% (72/118) of 16-17-year olds, 71.5% (103/144) of 18-19-year olds, and 84.7% (171/202) of 20-24-year olds (odds ratio [OR] 1.4, 95% CI 1.3-1.5) per year increase. Ownership was similar among females and males: 61.0% (236/387; 95% CI 55.6-66.1) versus 64.8% (153/236; 95% CI 57.8-71.2), age-adjusted OR 0.7 (95% CI 0.5-1.1); higher in those with secondary level education compared to primary or no education: 67.1% (346/516; 95% CI 62.6-71.2) versus 26% (21/82; 95% CI 16.4-37.7), age-adjusted OR 2.3 (95% CI 1.1-4.8); and similar across other sociodemographic factors. YP reported that 85.3% (361/423) of phones, either owned or shared, were smartphones. Among phone owners, the most commonly used phone app was WhatsApp (71.2%, 282/396), and 16.4% (65/396) reported having ever used their phone to track their health. A total of 407/631 (64.5%; 95% CI 60.3-68.5) currently had access to the internet (used in last 3 months on any device) with access increasing with age (OR 1.2, 95% CI 1.2-1.3 per year increase). In age-adjusted analysis, internet access was higher among males, the unmarried, those with a higher level of education, phone owners, and those who had lived in the community for more than 1 year. The aspect of the internet that YP most disliked was unwanted sexual (29.2%, 136/465) and violent (13.1%, 61/465) content.ConclusionsMobile phone-based interventions may be feasible in this population; however, such interventions could increase inequity, especially if they require access to the internet. Internet-based interventions should consider potential risks for participants and incorporate skill-building sessions on safe internet and phone use.

Dataset Information

Predicting Mood Disturbance Severity with Mobile Phone Keystroke Metadata: A BiAffect Digital Phenotyping Study.

Publications

Predicting Mood Disturbance Severity with Mobile Phone Keystroke Metadata: A BiAffect Digital Phenotyping Study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets