Relapse prediction in schizophrenia through digital phenotyping: a pilot study

Abstract

Among individuals diagnosed, hospitalized, and treated for schizophrenia, up to 40% of those discharged may relapse within 1 year even with appropriate treatment. Passively collected smartphone behavioral data present a scalable and at present underutilized opportunity to monitor patients in order to identify possible warning signs of relapse. Seventeen patients with schizophrenia in active treatment at a state mental health clinic in Boston used the Beiwe app on their personal smartphone for up to 3 months. By testing for changes in mobility patterns and social behavior over time as measured through smartphone use, we were able to identify statistically significant anomalies in patient behavior in the days prior to relapse. We found that the rate of behavioral anomalies detected in the 2 weeks prior to relapse was 71% higher than the rate of anomalies during other time periods. Our findings show how passive smartphone data, data collected in the background during regular phone use without active input from the subjects, can provide an unprecedented and detailed view into patient behavior outside the clinic. Real-time detection of behavioral anomalies could signal the need for an intervention before an escalation of symptoms and relapse occur, therefore reducing patient suffering and reducing the cost of care.

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1

Monthly rate of anomalies detected in the cohort of 15 patients with schizophrenia. After performing anomaly detection for each day of data collection for each patient across the sample, the frequency of anomalies detected at the 0.05 significance level per 30 days are calculated separately for each patient. These numbers should be compared to the expected number of anomalies under the null hypothesis, or 30*0.05 = 1.5 per 30 days. The distribution of anomaly rates in all data streams match tend to be centered near this null rate of 1.5 anomalies per 30 days. As a point of comparison, in the patients who relapsed the rate of anomalies within 2 weeks of hospitalization was 2.5 anomalies per 30 days

Figure 2

Daily p values for anomaly detection in mobility, sociability, and self-report of clinical outcomes in a patient leading up to relapse and hospitalization. The horizontal dotted line represents the significance level 0.05 after adjusting for multiple comparisons. Each point represents the p value for the test of anomaly detection on that day, with feature category indicated by color. Nine days prior to hospitalization there are significant anomalies in all data streams. These anomalies correspond with an escalation in self-reported symptoms in the days leading up to hospitalization and relapse

Figure 3

A demonstration of the multivariate time-series anomaly detection method. A bivariate example with toy data is presented to demonstrate how anomaly detection is performed at the daily level. The blue lines represent the sum of the overall trend and weekly components for each time series, with the remaining ε is regarded as the error. Red vertical lines show the specific error for day 19 for each univariate time series. After the rank(·) function ranks the errors, and the probability integral transforms these ranks to normally distributed adjusted errors, the multivariate time series across all features becomes multivariate normal, and the Hotelling’s T2 test is used to test them simultaneously for anomalies. Here, m represents the number of days in the time series and Φ is the standard normal cumulative distribution function