Response of US hospitals to elective surgical cases in the COVID-19 pandemic.

Editor—In anticipation of patients with coronavirus disease 2019 (COVID-19) overwhelming hospital resources, institutions and policymakers in the USA advocated a strategy to decrease surgical and interventional procedures rapidly in early 2020. , The downside has been a delay in treating patients and substantial revenue losses for many institutions. However, the precise timing, scale, and heterogeneity of US surgical case volume reduction and resumption of surgical activity have not yet been described. Understanding the response of US institutions to the first wave of COVID-19 will be critical to adjust hospital policies for upcoming or ongoing second waves in many places.

We did a nationwide analysis using data from the Multicenter Perioperative Outcomes Group, a registry of surgical procedures from academic and private hospitals across 21 US states. Briefly, electronic health record data for all patients undergoing surgical procedures from each participating institution are aggregated at the data coordinating centre each month after rigorous data quality validation. Between January 1, 2019 and May 31, 2020, all surgical cases at 33 health systems were totalled weekly to achieve maximum timing precision without influence by standard weekend reductions in case volume. Weekly case volumes were analysed via segmented regression and compared between 2019 and 2020.

We analysed 1 979 445 cases and observed a sharp decline in procedures during the week of March 16, 2020 (Fig 1 ), as COVID-19 diagnoses began to increase nationally. We observed a nadir in case volumes the week of April 6. During the week of April 6, we observed a 71% reduction compared with the same week in 2019. Between March 16 and May 31, the median per-week reduction in case volume relative to the same weeks in 2019 was 57% (inter-quartile range: 39–67%) (Fig 1). This reduction primarily reflected elective cases (10 237 cases per week in 2020 vs 27 122 in 2019; 62% reduction; paired Wilcoxon rank-sum test; P<0.001), whereas the volume of emergent cases decreased to a lesser extent (1248 vs 1350; 8% reduction; P=0.024), as observed for other medical conditions.

Fig 1

Weekly surgical volume at 33 US hospitals and cumulative number of COVID-19 cases in the USA from January 1, 2019 to May 31, 2020.

We also observed significant heterogeneity in case volume reductions across institutions, with per-institution median weekly reductions ranging from 33% to 72% (intra-class correlation coefficient: 0.53; 95% confidence interval: 0.45–0.61; F-test P<0.001). The decrease in case volumes was followed by a rapid increase such that by May 31, surgical case volumes were within 20% of case volumes at the same time in 2019 (Fig 1).

To summarise, an early rapid decrease in US surgical case volumes beginning mid-March 2020 was followed by a similarly rapid increase towards baseline beginning by mid-April while the pandemic was active and the numbers of COVID-19 cases were rising quickly. Case volume reductions varied significantly by institution. Important lessons can be learned from these observations. The global recommendations to cancel elective surgeries at the beginning of the pandemic regardless of the local situations of COVID-19 cases and hospitalisation should probably be more gradually implemented and adjusted based on local situations. The rapid increase in number of surgical procedures while the pandemic was very active certainly illustrates a perception of inappropriate adjustments of elective case volume by many local situations. The need for more local adjustments is further illustrated by the homogeneous timing of changes in surgical volumes across the country whereas the COVID-19 case surges were more temporally dispersed across the country.

Further analysis will be necessary to understand the specific factors that influenced the local and regional heterogeneity and the potential impact on patient outcomes to further inform public health response to future waves. We suggest a more locally and temporally adjusted response from US hospitals depending on COVID-19 hospitalisation trends to prevent avoidable cancellation of surgical cases, which might unnecessarily impact patient prognosis and hospital financial security.

Declarations of interest

The authors declare that they have no conflicts of interest.