Hospital-acquired conditions: predictors and implications for outcomes following spine tumor resection

Abstract

OBJECTIVE Hospital-acquired conditions (HACs) significantly compromise patient safety, and have been identified by the Centers for Medicare and Medicaid Services as events that will be associated with penalties for surgeons. The mitigation of HACs must be an important consideration during the postoperative management of patients undergoing spine tumor resection. The purpose of this study was to identify the risk factors for HACs and to characterize the relationship between HACs and other postoperative adverse events following spine tumor resection. METHODS The 2008-2014 American College of Surgeons' National Surgical Quality Improvement Program database was used to identify adult patients undergoing the resection of intramedullary, intradural extramedullary, and extradural spine lesions via current procedural terminology and ICD-9 codes. Demographic, comorbidity, and operative variables were evaluated via bivariate statistics before being incorporated into a multivariable logistic regression model to identify the independent risk factors for HACs. Associations between HACs and other postoperative events, including death, readmission, prolonged length of stay, and various complications were determined through multivariable analysis while controlling for other significant variables. The c-statistic was computed to evaluate the predictive capacity of the regression models. RESULTS Of the 2170 patients included in the study, 195 (9.0%) developed an HAC. Only 2 perioperative variables, functional dependency and high body mass index, were risk factors for developing HACs (area under the curve = 0.654). Hospital-acquired conditions were independent predictors of all examined outcomes and complications, including death (OR 2.26, 95% CI 1.24-4.11, p = 0.007), prolonged length of stay (OR 2.74, 95% CI 1.98-3.80, p < 0.001), and readmission (OR 9.16, 95% CI 6.27-13.37, p < 0.001). The areas under the curve for these models ranged from 0.750 to 0.917. CONCLUSIONS The comorbidities assessed in this study were not strongly predictive of HACs. Other variables, including hospital-associated factors, may play a role in the development of these conditions. The presence of an HAC was found to be an independent risk factor for a variety of adverse events. These findings highlight the need for continued development of evidence-based protocols designed to reduce the incidence and severity of HACs.