Interrupted Time-Series Analysis: A Robust Method for Observational Outcome Analysis in Surgical Emergency Research (Motivated by the Study on Emergency General Surgery Utilization During COVID-19 by Greenberg et al.)

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Interrupted time-series analysis is an increasingly important method in observational healthcare research, offering a structured and rigorous approach for evaluating the effects of interventions when randomized controlled trials are not feasible. This report illustrates its application through an analysis of emergency general surgery utilization during the COVID-19 pandemic, drawing on insights from a study conducted at a major tertiary-care hospital in San Francisco. Using segmented linear regression, the study examined weekly volumes of surgical emergency presentations before and after the city’s shelter-in-place order. Findings showed a modest immediate rise in presentations, followed by a gradual decline over time, with notable reductions in conditions such as gallbladder disease and peritonitis. Patients presenting during the pandemic were younger, more likely to be privately insured, and generally exhibited less severe illness. Fewer patients underwent surgery or presented with sepsis, suggesting shifts in both access to care and clinical decision-making. The analysis was supported by residual and autocorrelation checks, which confirmed the reliability of the model. Interrupted time-series analysis offers a transparent and practical framework for assessing temporal effects in real-world healthcare data. Future research should focus on integrating advanced modeling techniques, improving data visualization, and enhancing reproducibility to broaden its impact in public health and clinical decision-making.

Article activity feed