Rate of Colonoscopies Performed by Family Physician Decreased in United States

A substantial proportion of colonoscopies have been performed by family physicians (FPs) in the US. However, this rate has decreased between 2016 and 2021, according to a study published in The Journal of Rural Health.¹ This decrease was especially pronounced in patients living in rural areas.

Screening rates for colorectal cancer (CRC) have improved in the United States from 65% in 2012 to 70% in 2020 due to different ways to screen for CRC.² However there are approximately 30% of adults aged 50 to 75 years in the US who have average risk and are not up to date with their screening, with most of these patients coming from rural areas or identifying as Black, Hispanic, or American Indian.¹ To compensate for lack of access, FPs could potentially perform colonoscopies to increase patient access. There is not a lot of research on how effective this could be in the long run, however. This study aimed to use national insurance claims data to assess whether there was any benefit to FPs performing colonoscopies for their patients.

Family physicians perform a substantial percentage of colonoscopies in the United States | Image credit: romaset - stock.adobe.com

The Merative’s Marketscan database was used to conduct the retrospective study, which is a claims database for the US. The data from inpatient, outpatient, and drug claims is drawn from 300 employers and health plans across the country. US Census data and National Center for Health Statistics’ (NCHS) Urban-Rural classification scheme were used to determine the metropolitan statistical area of each beneficiary.

Colonoscopies performed between January 2016 and December 2021 were the main focus of the study. All colonoscopies performed on patients aged 49 and younger were excluded from this study through the May 2021 data, after which patients aged 45 years and older were included to compensate for the updated screening guidelines for CRC. Physicians who performed the colonoscopy were classified as either FP or another medical specialty by using the Marketscan’s Provider Type variable.

The NCHS was used to create classification codes for the primary beneficiary to assess where they lived and the urbanicity vs rurality of the patients. These categories were large central metro, large fringe metro, medium metro, small metro, micropolitan, and non-core, with the first 4 regarded as urban and the others regarded as rural.

There were 2,943,900 colonoscopies in the database during the study period that were eligible for this study. The number of colonoscopies changed from year to year, with 2016 having the highest number at 564,255 colonoscopies and 2020 having the lowest number at 365,172. The mean patient age spanned from 57.4 to 58.5 years. Patients in rural areas had an improvement of colonoscopies performed from 14.1% to 16.9% from 2016 to 2021.

There was a downward trend in colonoscopies performed by FPs between 2016 (11.32%; 95% CI, 11.23%-11.4%) and 2021 (6.73%; 95% CI, 6.66-6.81). This decrease mainly came from patients in the non-core areas (16.1% in 2016 vs 10.0% in 2021). Slightly older patients were more likely to get a colonoscopy from a FP (60.3 years vs 57.8 years) and men were more likely to receive a colonoscopy from a FP (49.2% female in colonoscopies performed by FP vs 52.3% females in groups that had colonoscopies outside of FPs).

Patients were more likely to have coinsurance or a deductible charge when they received colonoscopies from FPs (24.8% vs 22.3% for coinsurance, 28.6% vs 22.2% for deductible). Patients who had colonoscopies performed by an FP had lower mean coinsurance ($140 vs $159), copay ($73 vs $81), and deductible amounts ($318 vs $395). FPs performed more colonoscopies on patients living in a rural area, specifically noncore, small metro, and medium metro, compared with other medical specialties (20.6% vs 14.2%). Patients in the Midwest also had their colonoscopies performed by FPs more often than other medical specialties (32.2% vs 21.9%).

Patients who were older (OR, 1.048; 95% CI, 1.048-1.049), had a copay (OR, 1.16; 95% CI, 1.14-1.18), had any deductible (OR, 1.30; 95% Ci, 1.29-1.32), or had a capitated health plan (OR, 1.12; 95% CI, 1.08-1.16) were more likely to receive colonoscopies performed by a FP. Patients who were female were less likely to receive a colonoscopy from a FP (OR, 0.91; 95% CI, 0.90-0.92). Patients were significantly less likely to receive a colonoscopy from a FP in 2021 (OR, 0.54; 95% CI, 0.53-0.55) compared with 2016.

There were some limitations to this study. Marketscan databases do not cover Medicaid, uninsured patients, or patients with Medicare without employer-sponsored Medicare supplemental insurance, which could affect the results as these populations may have a higher rate of colonoscopies performed by FPs. The broad definition of primary care clinician may not have been accounted for in this database. A multidiscipline approach was not used for this analysis. Random coding errors are possible in claims data.

The researchers concluded that FPs performed a significant percentage of colonoscopies for patients, even if the percentage declined over the course of 5 years. FPs should continue to be a stakeholder in colonoscopy workforces and future studies should look into other potential clinicians who can help perform colonoscopies in different patients to increase screening rates.

"[Demand] for colonoscopy will likely continue to outpace supply for the near future. A multipronged strategy will need to be developed and sustained by practitioners and policymakers to address this deficit," they wrote, adding, "Although advancements in other lower intensity CRC screening modalities, such as FIT and other stool-DNA screening tests have been made, these interventions appear to have lower neoplastic detection rate than colonoscopy, particularly for right-sided neoplasia."

References

1. Edwardson N, van der Goes D, Pankratz VS, et al. Trends in and factors associated with family physician-performed screening colonoscopies in the United States: 2016-2021. J Rural Health. Published online June 26, 2024. doi:10.1111/jrh.12858
2. 80% in every community. National Colorectal Cancer Roundtable. 2024. Accessed July 1, 2024. https://nccrt.org/our-impact/80-in-every-community/