Strategies for Identifying and Channeling Patients for Depression Care Management

Objective: To determine optimal methods of identifying enrollees with possible depression for additional depression screening in the context of a care management program for chronically ill Medicare recipients. Study Design: Observational analysis of telephone and mail survey and claims data collected for the Medicare Health Support (MHS) program. Methods: This study examines data from 14,902 participants with diabetes mellitus and/or congestive heart failure in the MHS program administered by Green Ribbon Health, LLC. Depression screening was performed by administering a 2-item screen (the Patient Health Questionnaire 2 [PHQ-2]) by telephone or by mail. Additional information about possible depression was drawn from International Classification of Diseases, Ninth Revision (ICD-9) depression diagnoses on claims and from self-reported use of antidepressant medications. We evaluated positive depression screens using the PHQ-2 administered via telephone versus mail, examined variations in screener-positive findings by care manager, and compared rates of positive screens with antidepressant use and with claims diagnoses of depression. Results: Almost 14% of participants received an ICD-9 diagnosis of depression during the year before program enrollment; 7.1% reported taking antidepressants, and 5.1% screened positive for depression on the PHQ-2. We found substantial variation in positive depression screens by care manager that could not be explained by case mix, prior depression diagnoses, or current depression treatment. After adjusting for demographic and clinical differences, the PHQ-2-positive screen rates were 6.5% by telephone and 14.1% by mail (P <.001). Conclusion: A multipronged effort composed of mail screening (using the PHQ-2), self-reported antidepressant use, and claims diagnoses of depression may capture the greatest number of enrollees with possible depression.