Variation in utilization of multivessel percutaneous coronary intervention: Influence of hospital volume

Background The purpose of this study was to investigate the contemporary trends in the utilization of multivessel percutaneous coronary interventions (MVPCIs) in the USA. Methods We queried the Healthcare Cost and Utilization Project's Nationwide Inpatient Sample between 2006 and 2011 using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) procedure codes 00.40 (single stent), 00.46, 00.47, and 00.48 (single vessel and multiple stents) and 00.41, 00.42 and 00.43 (MVPCI). We built a hierarchical three-level model adjusted for multiple confounding factors. Results A total of 543 434 (weighted: 2 683 206) procedures were identified. Independent predictors of increased MVPCI utilization (odds ratio, 95% confidence interval, P-value) were found to be age (1.05, 1.04-1.07, P<0.001) and comorbid conditions on using Deyo's modification of Charlson's comorbidity index of at least 2 (1.13, 1.09-1.16, P<0.001). Female sex (0.88, 0.87-0.90, P<0.001), myocardial infarction (0.86, 0.83-0.89, P<0.001), weekend admissions (0.94, 0.91-0.96, P<0.001), and urgent admissions (0.88, 0.83-0.93, P<0.001) predicted decreased utilization. Highest quartile of hospital (1.34, 1.16-1.54, P<0.001) predicted higher utilization. Betweenhospital variation of 7.7% (interclass correlation coefficient) was observed, which was minimally affected by patient or hospital mix. A randomly selected patient was ~1.6 (median odds ratio) times more likely to receive an MVPCI from a given hospital compared with another identical patient being treated at a different random hospital. Conclusion The utilization rate of MVPCI varied considerably among hospitals. Higher annual hospital volume was associated with a higher utilization rate of MVPCI.