RESUMO
BACKGROUND: Variation in readmission rates, length of stay, and operative time associated with rotator cuff surgery should be understood if cost-control strategies are to be considered. We hypothesized that there would be variation in resource utilization as measured in terms of these factors and that surgeon and hospital practice patterns, rather than patient characteristics, would explain this variation. METHODS: We conducted a retrospective analysis of the effects of surgeon, hospital, and patient-related factors on the readmission rates, length of stay, and operative time associated with 1077 rotator cuff repairs performed by thirty-two surgeons in eleven group-model health maintenance organization hospitals, two satellite centers, and one contract facility in southern California. RESULTS: The initial unadjusted analysis of covariance showed moderate-to-strong associations between surgeon and hospital variation and the rate of hospital readmission within thirty days (p = 0.0919 and p = 0.0209, respectively), extended length of stay (p = 0.0016 and p = 0.0016, respectively), and operative time (p < 0.0001 and p < 0.0001, respectively). The hospital effect was no longer significant when patient-related factors (i.e., sociodemographic characteristics and comorbidities) and the surgeon effect were taken into account. The surgeon effect was still significant (except with regard to the readmission rate) after adjustment for patient and hospital-related factors, explaining 23% of the variation in length of stay and 69% of the variation in operative time. There was a significantly increased risk of an extended stay (p = 0.0010) and readmission (p = 0.0260) following procedures performed at hospitals with an orthopaedic residency program. Increased operative time was significantly associated with decreased surgeon volume (p < 0.0001) and the absence of an orthopaedic residency program (p < 0.0001). CONCLUSIONS: Variation in length of stay and operative time associated with rotator cuff surgery is largely explained by surgeon practice patterns. Our results suggest that surgeons have the ability to affect these two factors, which are often identified as drivers of cost.
