*2.1. Study Design*

A retrospective study was conducted of a consecutive series of shoulder arthroplasties performed at a single private practice institution. Institutional review board approval was obtained for this study. Our registry was queried to identify all patients who underwent elective primary total shoulder arthroplasty (anatomic (ATSA) or reverse (RTSA)) between March 2019 and March 2021 by a single fellowship-trained shoulder surgeon. The inclusion criteria were: (1) an ATSA or RTSA and (2) minimum follow-up of 90 days. To achieve a homogenous sample of patients at low surgical risk, an a priori decision was made to exclude patients whose indication for surgery was traumatic, and those undergoing revision surgery.

Following the 18 March 2020 recommendation by the Centers for Medicare and Medicaid Services to postpone non-essential surgeries in response to the COVID-19 virus, no elective shoulder arthroplasties were performed at our practice until 11 May 2020. As such, the study sample was divided into two cohorts: the "pre-COVID" group for surgeries performed before 18 March 2020, and the "post-COVID" group for cases performed on or after 11 May 2020. Notably, the treating surgeon had nearly 10 years of experience at the beginning of the study period. During the study period there was no change in postoperative protocols or in the design of the implants used by the primary surgeon.

#### *2.2. Outcomes Measures and Explanatory Variables*

The main outcomes of interest included LOS (measured in hours after surgery) and same-day discharge. Discharge disposition (home versus skilled nursing facility (SNF)) was also recorded. Electronic medical records linked to the local hospital were reviewed to collect data on ED visits, readmissions, and reoperations within 90 days of surgery.

Several patient characteristics that might affect the influence of the COVID-19 pandemic on resource allocation after shoulder arthroplasty were recorded. Specifically, data were collected on age, sex, body mass index (BMI), and the presence of co-morbidities including diabetes, chronic obstructive pulmonary disease (COPD), and tobacco use. Surgical location (hospital versus ambulatory surgery center) data were also collected.
