This issue of Psychiatric Services includes an analysis of two-year follow-up data from Partners in Care, a seminal depression quality improvement study conducted during the late 1990s. The results include a finding that is, on its face, surprising—the Partners in Care intervention appears to have been comparably, or possibly even more, cost-effective for minor depression than for major depression. How are we to reconcile this result with the fact that other studies have found only limited effectiveness for treatment of subsyndromal depression?
Answering this question requires recognizing an important but often underemphasized distinction—the difference between evidence-based medicine (EBM) for clinical care and evidence-based quality improvement. A clinical perspective assumes that everyone in a program receives a relatively standardized set of services. Persons with more severe conditions stand to gain more from treatment and have lower rates of spontaneous remission than those with milder problems. Thus in clinical studies, effectiveness and cost-effectiveness are typically greater for subjects with the highest degree of symptomatology .
Partners in Care, however, was a quality improvement study, in which the intervention was directed at clinical practices rather than to individual patients. Although the project enrolled participants with both major and subthreshold depression, persons with subthreshold conditions could receive low-intensity monitoring rather than active psychotherapy or medication management. As a result, incremental health costs for the intervention group in this subpopulation were extremely modest—less than 5% of the costs for persons with major depression. Given that more than 40% of patients in the Partners in Care study had subthreshold rather than major depression, the low cost of treating this group is likely to have been an important driver of the intervention's overall cost-effectiveness.
Most disease management programs look more like Partners in Care than like interventions from clinical studies. Typically, these programs rely on referrals for identifying patients rather than on full diagnostic assessments and include participants with both full and subsyndromal conditions. Once patients are enrolled, disease managers generally provide supplemental guidance and support rather than direct care. The results of the study by Wells and colleagues provide encouraging news for this approach to improving care. By providing low-intensity services to persons with subsyndromal conditions and more intensive care to individuals with more serious disorders, these programs have the potential to provide population-based care that is both efficacious and cost-effective.
