2006 HSR&D National Meeting Abstract
1043 — Randomized Controlled Trial of a Low-Intensity Collaborative Intervention for Depression
Dobscha SK (Portland VA Medical Center)
Corson K (Portland VA Medical Center)
Hickam DH (Portland VA Medical Center)
Gerrity MS (Portland VA Medical Center)
Collaborative interventions improve depression outcomes, but typically require additional clinical resources that can be difficult to sustain. The purpose of this study was to determine if a low-intensity care management intervention improves depression outcomes in a VA primary care setting.
Forty-one clinicians and 375 patients from five clinics of one VA Medical Center participated in this randomized controlled trial. Clinicians received depression education, then were randomized to receive Depression Decision Support (DDS) versus usual care; patient assignment followed clinician status. Patients due for primary care appointments were contacted for phone-screening. Those with Patient Health Questionnaire (PHQ-9) scores >=10 were invited for initial interviews. Patients with PHQ-9>=10 or Hopkins Symptom Checklist-20 (SCL-20)>=1.0 at initial interviews were enrolled. DDS included one early patient educational contact, depression monitoring, and clinician feedback. Further intervention targeted patients not improving, and could include additional recommendations, psychiatric consultation, or referral to mental health. Depression severity, health-related quality-of-life (SF-36V), and satisfaction were assessed at 6 and 12 months.
Comorbid conditions were prevalent, and included PTSD (39%), alcohol use disorder (16%), >=3 general medical conditions (69%), and moderate to severe pain (68%). Intervention patients were more likely to receive antidepressants for 90 (76% vs. 62%, p=.003) and 180 days (63% vs. 53%, p=.04), attend mental health appointments (42% vs. 27%, p=.005), and report greater satisfaction at 12 months (3.59 vs. 3.15, p<.001). Intervention clinicians were more likely to perform initial depression assessments (94% vs. 78%, p<.001) and follow-up depression-related actions over 12-months (2.08 vs. 1.14, p<.001). There was no effect of the intervention compared to usual care on SCL-20 or SF-36V scores.
This low-intensity approach improved process of care, but not depression outcomes. Intervention effectiveness may have been limited by the high degree of comorbidity. Because recruitment did not rely on clinician referral or patient self-referral, patient motivation may have varied considerably. That usual care clinicians received depression education and had access to on-site mental health support may also have mitigated the intervention effect.
Our findings suggest that more intensive care management and/or addition of specialty treatment modules are needed to improve depression-related outcomes in this population.