Hospital-, Anesthesiologist-, and Patient-level Variation in Primary Anesthesia Type for Hip Fracture Surgery A Population-based Cross-sectional Analysis

Background: Substantial variation in primary anesthesia type for hip fracture surgery exists. Previous work has demonstrated that patients cared for at hospitals using less than 20 to 25% neuraxial anesthesia have decreased survival. Therefore, the authors aimed to identify sources of variation in anesthesia type, considering patient-, anesthesiologist-, and hospital-level variables. Methods: Following protocol registration (NCT02787031), the authors conducted a cross-sectional analysis of a population-based cohort using linked administrative data in Ontario, Canada. The authors identified all people greater than 65 yr of age who had emergency hip fracture surgery from April 2002 to March 2014. Generalized linear mixed models were used to account for hierarchal data and measure the adjusted association of hospital-, anesthesiologist-, and patient- level factors with neuraxial anesthesia use. The proportion of variation attributable to each level was estimated using variance partition coefficients and the median odds ratio for receipt of neuraxial anesthesia. Results: Of 107,317 patients, 57,080 (53.2%) had a neuraxial anesthetic. The median odds ratio for receiving neuraxial anesthesia was 2.36 between randomly selected hospitals and 2.36 between randomly selected anesthesiologists. The majority (60.1%) of variation in neuraxial anesthesia use was explained by patient factors; 19.9% was attributable to the anesthesiologist providing care and 20.0% to the hospital where surgery occurred. The strongest patient- level predictors were absence of preoperative anticoagulant or antiplatelet agents, absence of obesity, and presence of pulmonary disease. Conclusions: While patient factors explain most of the variation in neuraxial anesthesia use for hip fracture surgery, 40% of variation is attributable to anesthesiologist and hospital- level practice. Efforts to change practice patterns will need to consider hospital- level processes and anesthesiologists' intentions and behaviors.