Question: In patients with depression, are either of 2 quality-improvement (QI) interventions for improving the treatment of depression in managed care more cost-effective than usual care?
Design: Cost-effectiveness analysis from a societal perspective for a cluster-randomized {allocation concealed*}†, unblinded,* controlled trial with 2-year follow-up.
Setting: 46 primary care clinics in 6 community-based managed-care organizations (MCOs) in the United States.
Patients: 1356 patients who were ≥ 18 years of age {mean age 44 y, 71% women}‡, planned to use the primary care clinic over the next 12 months, and met the Composite International Diagnostic Interview criteria for depression. Follow-up at 2 years was 85%.
Intervention: Matched clinics were allocated to 1 of 2 QI interventions or to usual care (i.e., mailing of practice guidelines) (16 clinics, 443 patients). The QI interventions consisted of training for practice leaders and nurses, enhanced educational and assessment resources, and either nurses for medication follow-up (QI-meds, 12 clinics, 424 patients) or access to trained psychotherapists (QI-therapy, 15 clinics, 489 patients).
Main cost and outcome measures: Outcomes were quality-adjusted life-years (QALYs), days with depression burden, and days of employment. Intervention costs (screening, intervention materials, and professional time) and health care costs (consultations and psychotropic medications) were assessed in 1998 U.S. dollars. Indirect costs for patient time were included.
Main results: Intention-to-treat analyses were adjusted for baseline patient characteristics and practice randomization blocks. Patients in the QI-therapy group had more QALYs (P = 0.006), fewer days of depression burden (P = 0.01), and more days of employment (P = 0.03) than did those receiving usual care (Table). QI-meds and usual care did not differ for any outcome (Table). The groups did not differ for health care costs (including patient time) (Table).
Conclusion: 1 of 2 quality-improvement interventions for depression in managed care was more effective but cost more than usual care.
2 quality-improvement (QI) interventions vs usual care for depression in primary care§
| Outcomes at 2 y | Usual care total | Incremental effect of QI-meds (95% CI) | Incremental effect of QI-therapy (CI) |
| Quality-adjusted life-years | 1.7 | 0.01 (−0.00 to 0.03) | 0.02 (0.01 to 0.04)¶ |
| Days of depression burden | 419.9 | −25.0 (−63.1 to 13.2) | −46.7 (−83.1 to −10.3)¶ |
| Days of employment | 279.2 | 17.9 (−1.6 to 37.4) | 20.9 (2.4 to 39.3)¶ |
| Health care costs‖ (U.S. $)** | 3835 | 419 (−467 to 1306) | 485 (−393 to 1363) |