Search | Search by Center | Search by Source | Keywords in Title
Hysong SJ, Khan MM, Petersen LA. Passive monitoring versus active assessment of clinical performance: impact on measured quality of care. Medical care. 2011 Oct 1; 49(10):883-90.
CONTEXT: Measurement of hospitals'' clinical performance is becoming more ubiquitous in an effort to inform patient choices, payer reimbursement decisions, and quality improvement initiatives such as pay-for-performance. As more measures are developed, the intensity with which measures are monitored changes. Performance measures are often retired after a period of sustained performance and not monitored as actively as other measures where performance is more variable. The effect of actively versus passively monitoring performance on measured quality of care is not known. OBJECTIVE: We compared the nature and rate of change in hospital outpatient clinical performance as a function of a measure''s status (active vs. passive), and examined the mean time to stability of performance after changing status. We hypothesize that performance will be higher when measures are actively monitored than when they are passively monitored. DESIGN: Longitudinal, hierarchical retrospective analyses of outpatient clinical performance measure data from Veterans Health Administration''s External Peer Review Program from 2000 to 2008. SETTING: One hundred thirty-three Veterans Health Administration Medical Centers throughout the United States and its associated territories. MAIN OUTCOME MEASURES: Clinical performance on 17 measures covering 5 clinical areas common to ambulatory care: screening, immunization, chronic care after acute myocardial infarction, diabetes mellitus, and hypertension. RESULTS: Contrary to expectations, we found that measure status (whether active or passive) did not significantly impact performance over time; time to stability of performance varied considerably by measure, and did not seem to covary with performance at the stability point (ie, performance scores for measures with short stability times were no higher or lower than scores for measures with longer stability times). CONCLUSIONS: We found no significant "extinction" of performance after measures were retired, suggesting that other features of the health care system, such as organizational policies and procedures or other structural features, may be creating a "strong situation" and sustaining performance. Future research should aim to better understand the effects of monitoring performance using process-of-care measures and creating sustained high performance.