Trends in and factors associated with family physician‐performed screening colonoscopies in the United States: 2016‐2021

Abstract

PurposeFamily physician (FP)‐performed screening colonoscopies can serve as 1 strategy in the multifaceted strategy necessary to improve national colorectal cancer screening rates, particularly in rural areas where specialist models can fail. However, little research exists on the performance of this strategy in the real world. In this study, we evaluated trends in and factors associated with FP‐performed screening colonoscopies in the United States between 2016 and 2021.MethodsUsing national data from Merative's Marketscan insurance claims database, we estimate the proportion of screening colonoscopies performed by FPs. We use logistic regression models to evaluate factors independently associated with FP‐performed colonoscopies.ResultsThe percentage of screening colonoscopies performed by FPs exhibited a downward trend from 11.32% in 2016 to 6.73% in 2021, with the largest decrease occurring among patients from the most rural areas. FPs were more likely to perform colonoscopies on slightly older patients, male patients, and rural patients. Patients were less likely to receive FP‐performed colonoscopies in large metropolitan areas compared to lesser populated areas. Patients were more likely to receive FP‐performed colonoscopies in the Midwest, South, and West, even after accounting for urban‐rural classification.ConclusionDespite a downward trajectory, FPs perform a substantial proportion of screening colonoscopies in the United States. Changes to the business side of health care delivery may be contributing to the observed decreasing rate. Whether through spatial or relational proximity, FPs may be better positioned to provide colonoscopy to some rural, male, and older patients who otherwise may not have been screened. Policy changes to expand the FP workforce, particularly in rural areas, are likely necessary to slow or reverse the downward trend of FP‐performed screening colonoscopies.