Nov 25, 2005
Does Practice Make Perfect?
Published in: American Journal of Psychiatry, v. 161, no. 12, Dec. 2004, p. 2282-2286
Posted on RAND.org on December 31, 2003
OBJECTIVE: An extensive literature has demonstrated a relationship between hospital volume and outcomes for surgical care and other medical procedures. The authors examined whether an analogous association exists between the volume of mental health delivery and the quality of mental health care. METHOD: The study used data for the 384 health maintenance organizations participating in the Health Employer Data and Information Set (HEDIS), covering 73 million enrollees nationwide. Analyses examined the association between three measures of mental health volume (total annual ambulatory visits, inpatient discharges, and inpatient days) and the five HEDIS measures of mental health performance (two measures of follow-up after psychiatric hospitalization and three measures of outpatient antidepressant management), with adjustment for plan and enrollee characteristics. RESULTS: Plans in the lowest quartile of outpatient and inpatient mental health volume had an 8.45 (95% CI [confidence interval]=4.97-14.37) to 21.09 (95% CI=11.32-39.28) times increase in odds of poor 7- and 30-day follow-up after discharge from inpatient psychiatric hospitalization. Low-volume plans had a 3.49 (95% CI=2.15-5.67) to 5.42 (95% CI=3.21-9.15) times increase in odds of poor performance on the acute, continuation, and provider measures of antidepressant treatment. CONCLUSIONS: The large and consistent association between mental health volume and performance suggests parallels with the medical and surgical literature. As with that previous literature, further work is needed to better understand the mechanisms underlying this association and the potential implications for using volume as a criterion in plan choice.