RESUMO
Ecological momentary assessment (EMA) for mental disorders, using application-based (app) technology capable of passive and ambient data collection, has been insufficiently evaluated and validated with rigorous, adequately-powered, high-quality studies. Herein, we sought to validate the mind.me application for the assessment of depressive symptoms in adults. Adults (ages 18-65) who self-identified as having clinically significant depressive symptoms [i.e. Patient Health Questionnaire 9 (PHQ-9) ≥ 5] utilized the mind.me app-a mobile phone technology that collects data passively and continuously, and is capable of integrating broad multimodal data [e.g., location variance (e.g. GPS), behavioural (e.g. social network activity), and communication data (e.g. SMS texting, phone calls)]. The primary outcome was predictive accuracy (i.e. convergent validity with depressive symptom measurement, as captured by the PHQ-9). 200 subjects were enrolled in the study (mean age 46 ± 12.71). The average PHQ-9 score was 12.8 ± 6.9. The predictive accuracy of the mind.me app was 0.91 ± 0.06. The sensitivity was 0.98 and the specificity was 0.93. The mind.me app was rated by 200 users as highly usable and informative to their illness. The mind.me app exhibits robust predictive accuracy in detecting depressive symptoms in adults with clinically relevant depressive symptoms. The mind.me app more specifically demonstrates convergence with the PHQ-9.