Effects of Hospital Procedure Volume and Resident Training on Clinical Outcomes and Resource Use in Radical Retropubic Prostatectomy Surgery in the Department of Veterans Affairs

PurposeIn this retrospective cohort study we used data from the National Surgical Quality Improvement Program of the Veterans Health Administration to examine the effect of case volume and resident participation on radical retropubic prostatectomy outcomes.Materials and MethodsThe study population included 5,736 patients who underwent radical retropubic prostatectomy in Veterans Administration hospitals between October 1, 2001 and September 30, 2004. Resource related outcomes included operative times and length of stay. Clinical outcomes included blood transfusion, complications, readmissions and reoperations. Hierarchical multivariate regression models were developed to predict outcomes. Risk adjustment was performed using patient chronic health factors and results of preoperative laboratory testing.ResultsA total of 5,070 radical retropubic prostatectomy surgeries met inclusion criteria. After adjustment for case mix, academic training institutions had longer operative times (3.2 vs 2.4 hours, p <0.01) but shorter length of stay (3.4 vs 4.2 days, p <0.01). Surgery at academic institutions was not associated with greater risk of transfusion (p = 0.36), reoperation (p = 0.93), complications (p = 0.53) or readmissions (p = 0.97). However, among the academic institutions low vs high hospital radical retropubic prostatectomy volume was associated with longer length of stay (3.7 vs 3.1 days, p = 0.02) and higher transfusion rate (29.6% vs 18.2%, p = 0.02). Substantial clustering of outcomes at the hospital level was observed.ConclusionsWithin the Veterans Administration system academic training institutions have longer operative times for radical retropubic prostatectomy, but shorter length of stay. Among the same institutions, high volume hospitals tend to have lower transfusion rates and shorter length of stay. Clustering of outcomes at the hospital level suggests that unmeasured institutional factors are key determinants of clinical and resource related outcomes.