Variability in utilization of drug eluting stents in United States: Insights from nationwide inpatient sample

Objectives We studied the trends and predictors of drug eluting stent (DES) utilization from 2006 to 2011 to further expound the inter-hospital variability in their utilization. Background We queried the Healthcare Cost and Utilization Project's Nationwide Inpatient Sample (NIS) between 2006 and 2011 using ICD-9-CM procedure code, 36.06 (bare metal stent) or 36.07 (drug eluting stents) for Percutaneous Coronary Intervention (PCI). Annual hospital volume was calculated using unique identification numbers and divided into quartiles for analysis. Methods and Results We built a hierarchical two level model adjusted for multiple confounding factors, with hospital ID incorporated as random effects in the model. About 665,804 procedures (weighted n = 3,277,884) were analyzed. Safety concerns arising in 2006 reduced utilization DES from 90% of all PCIs performed in 2006 to a nadir of 69% in 2008 followed by increase (76% of all stents in 2009) and plateau (75% in 2011). Significant between-hospital variation was noted in DES utilization irrespective of patient or hospital characteristics. Independent patient level predictors of DES were (OR, 95% CI, P-value) age (0.99, 0.98-0.99, <0.001), female(1.12, 1.09-1.15, <0.001), acute myocardial infarction(0.75, 0.71-0.79, <0.001), shock (0.53, 0.49-0.58, <0.001), Charlson Co-morbidity index (0.81,0.77-0.86, <0.001), private insurance/HMO (1.27, 1.20-1.34, <0.001), and elective admission (1.16, 1.05-1.29, <0.001). Highest quartile hospital (1.64, 1.25-2.16, <0.001) volume was associated with higher DES placement. Conclusion There is significant between-hospital variation in DES utilization and a higher annual hospital volume is associated with higher utilization rate of DES.