Depression Screening Using Daily Mental-Health Ratings from a Smartphone Application for Breast Cancer Patients

• DOI: 10.2196/jmir.5598
• Publication Date: 2016-08-04T08:30:06Z
• Authors: Junetae Kim; Sanghee Lim; Yul Ha Min; Yong-Wook Shin; Byungtae Lee; Guiyun Sohn; Kyung Hae Jung; Jae-Ho Lee; Byung Ho Son; Sei Hyun Ahn; Soo-Yong Shin; Jong Won Lee
• Keywords: Patient Health Questionnaire; Depression

Mobile mental-health trackers are mobile phone apps that gather self-reported ratings from users. They have received great attention clinicians as tools to screen for depression in individual patients. While several ask simple questions using face emoticons been developed, there has no study examining the validity of their screening performance.In this study, we (1) evaluate potential a tracker uses three daily (sleep satisfaction, mood, and anxiety) indicators depression, (2) discuss approaches data processing (ratio, average, frequency) generating indicator variables, (3) examine impact adherence on reporting accuracy screening.We analyzed 5792 sets collected 78 breast cancer patients over 48-week period. Using Patient Health Questionnaire-9 (PHQ-9) measure true status, conducted random-effect logistic panel regression receiver operating characteristic (ROC) analysis performance tracker. In addition, classified into two subgroups based level (higher lower adherence) k-means clustering algorithm compared between groups.With ratio approach, area under ROC curve (AUC) is 0.8012, indicating gathered via comparable results PHQ-9 tests. Also, AUC significantly higher (P=.002) group (AUC=0.8524) than (AUC=0.7234). This result shows self-reporting associated with screening.Our support tool practice. provides guideline variables ratings. Furthermore, our provide empirical evidence critical role self-reporting, which represents crucial information both doctors