Emergency general surgery (EGS) conditions are defined by a group of acute, non-trauma diagnoses that require presentation to an emergency department for operative or non-operative care. In the US, 3-4 million adults are hospitalized and treated for an EGS condition each year such that EGS conditions are responsible for 7% of all hospitalizations. EGS conditions result in an estimated 800,000 operations annually and cost ≈ $28 billion. The burden of EGS on inpatient care is greater than that of a new diagnosis of diabetes or cancer, coronary heart disease, heart failure, stroke, or HIV. Older adults, who are at increased risk of adverse outcomes, comprise more than half of the adult EGS population. Operative management of EGS conditions results in an overall complication rate of 50%, with a readmission rate of 4-18%, and a mortality rate that can approach 15%, with older patients demonstrating a significantly higher risk. Non-operative management is considered a reasonable alternative yet studies show mixed results. Randomized control trials are limited by the invasive nature of operative treatment, the heterogeneity of conditions, and the acute nature of EGS. Furthermore, most trials have compared different types of operative or non-operative interventions, with relatively few comparing operative treatment to non-operative treatment. A lack of evidence on the comparative effectiveness of operative and non- operative treatment has been shown to result in unnecessary variations in treatment and inferior operative outcomes. Furthermore, given the growing interest in the development of a regionalized system of care for EGS patients since the Institute of Medicine Committee report on the Future of Emergency Care in the United States Health System, knowledge on hospital performance in EGS is needed. This proposal aims to address these gaps in the literature with the following specific aims: (1) To identify hospital-level factors associated with rankings in EGS performance, (2) To evaluate the comparative effectiveness of operative and non-operative treatment in specific EGS conditions and (3) To evaluate heterogeneity in the effectiveness of operative treatment. Conditional effects of age, dementia and race will be examined amongst others. Using nationwide data from Medicare beneficiaries, the proposed study will be the first comparative effectiveness analysis of operative and non-operative treatment in EGS. The proposal employs: (1) template matching to define hospital quality in EGS (2) an instrumental variable approach with optimal near-far matching to overcome confounding by indication and, (3) interaction models to examine the conditional relationship between treatment and outcomes by patient factors. In so doing, the findings will provide insights into hospital factors necessary for optimal EGS outcomes and the comparative effectiveness of OP of EGS conditions. Policy makers will have evidence to inform systems- level restructuring to address th...