Automated Virtual Follow-up to Reduce Premature Treatment Discontinuation

Kaiser Foundation Health Plan Of Washington, Seattle WA

Investigators

Linked publications & trials

Abstract

Early dropout from depression treatment is, unfortunately, the norm. Interventions to reduce dropout from depression treatment have been repeatedly shown to improve clinical outcomes. The common ingredient of all of these collaborative care or care management programs is systematic outreach to assess outcomes and improve adherence. Furthermore, previous research conducted by our group indicates that early dropout from depression treatment is heterogeneous - and a substantial minority of those discontinuing treatment early experience good outcomes. These data suggest that more targeted dropout-reduction interventions (targeting only those in need of additional treatment) could reduce costs - and facilitate implementation - of outreach to improve adherence. Recent developments in health informatics have created to potential for more efficient and more targeted outreach programs to address dropout from depression treatment. First, electronic medical records databases allow real-time evaluation of patients who are overdue for prescription refills and follow-up visits. Second, increasing use of standardized depression severity measures (such as the PHQ9) allow efficient identification of those at risk for unfavorable outcomes. Third, increasing use of patient-provider online messaging will permit much more efficient outreach communication. We propose a pilot study of a completely automated outreach program for adult outpatients who appear to have dropped out of acute-phase depression treatment (either pharmacotherapy or psychotherapy). Using electronic medical records (EMR) data and communication tools embedded in those EMR systems, we will implement a population based program for: ? Automated identification patients who appear to have discontinued necessary depression treatment ? Automated outreach to assess treatment status and outcomes ? Algorithm-based tailored feedback to prompt re-engagement in care Data from this pilot study will be used to address the following specific aims: (1) Evaluate the feasibility and acceptability of automated outreach and feedback (2) Refine criteria for identifying patients likely to benefit from outreach (3) Examine potential outcome measures for a full-scale pragmatic trial