Objective: To examine the effectiveness of integrating generalist and specialist care for veterans with depression.
Method: We conducted a randomized trial of patients screening positive for depression at two Veterans Affairs Medical Center general medicine clinic firms. Control firm physicians were notified prior to the encounter when eligible patients had PRIME-MD depression diagnoses. In the intervention firm, a mental health clinical nurse specialist (CNS) was to: design a treatment plan; implement that plan with the primary care physician; and monitor patients via telephone or visits at two weeks, one month and two months. Primary outcomes (depressive symptoms, patient satisfaction with health care) were collected at 3 and 12 months.
Results: Of 268 randomized patients, 246 (92%) and 222 (83%) completed 3- and 12-month follow-up interviews. There were no between-group differences in depressive symptoms or satisfaction at 3 or 12 months. The intervention group had greater chart documentation of depression at baseline (63% versus 33%, p = 0.003) and a higher referral rate to mental health services at 3 months (27% versus 9%, p = 0.019). There was no difference in the rate of new prescriptions for, or adequate dosing of, anti-depressant medications. In 40% of patients, CNSs disagreed with the PRIME-MD depression diagnosis, and their rates of watchful waiting were correspondingly high.
Conclusions: Implementing an integrated care model did not occur as intended. Experienced CNSs often did not see the need for treatment in many primary care patients identified by the PRIME-MD. Integrating integrated care models in actual practice may prove challenging.