3104 — Quality of VA depression care, health service use, and mortality over a 6-year period

Khan MM (Houston Center for Quality of Care and Utilization Studies, Michael E. DeBakey VA Medical Center, and Baylor College of Medicine), Zimmer MP (Houston Center for Quality of Care and Utilization Studies, Michael E. DeBakey VA Medical Center), Petersen LA (Houston Center for Quality of Care and Utilization Studies, Michael E. DeBakey VA Medical Center, and Baylor College of Medicine), Cully JA (Houston Center for Quality of Care and Utilization Studies, Michael E. DeBakey VA Medical Center, and Baylor College of Medicine)

Objectives:
Little is known about the effects of depression care quality on health service use and mortality. The Department of Veterans Affairs (VA) has focused management and policy resources on improving the quality of depression care for veterans. We assessed the level of depression care quality within the VA over a six-year period and determined if depression care quality was related to patient mortality.

Methods:
We identified 205,156 veterans nationwide, with a new onset depressive disorder and at least one filled prescription for an antidepressant medication between 10/1/99-9/30/05 using VA administrative databases. Depression care quality was assessed by calculating antidepressant medication possession ratios (MPR)and adequacy of depression care follow-up during the 12 weeks after the index depression diagnosis. Adequate follow-up was defined as three or more follow-up mental health encounters during the 84 days (acute period) after the index depression diagnosis. Logistic regression modeling procedures were utilized to predict 12-month all-cause mortality.

Results:
Forty-eight percent of the cohort received an adequate supply of antidepressant medications and 31 percent received three or more follow-up encounters during the 84-day period following their index depression diagnosis. Rates of adequate medication supply did not change over the six year time period (p=NS), while the proportion of the cohort with adequate follow-up care showed improvements over time (p=0.033). Adequate depression follow-up was significantly related to decreased 12-month likelihood of all-cause mortality (OR 0.855, CI 0.784-0.993, p < 0.001).

Implications:
Less than half of patients with newly diagnosed depression received high quality care according to current quality indicators. Despite focused clinical and policy efforts within VA, few changes have occurred in the percentage of patients receiving high quality depression care. Notably, the provision of three or more encounters during the acute period following a new onset case of depression was significantly related to reduction in odds of patient mortality at 12-month follow-up.

Impacts:
Our findings suggest that depression care quality may significantly impact patient outcomes like mortality, and that further efforts to improve quality appear warranted.