Inter-physician heterogeneity in colorectal cancer screening participation in France: a study based on the French National Health Data System

Abstract Background Participation in the colorectal cancer screening program (CRC-SP) based on biennial faecal occult blood testing is low in France as in many countries. The aim was to evaluate the inter-physician heterogeneity in CRC-SP participation, and its determinants. Methods The study used the French National Health Data System. All beneficiaries eligible to CRC-SP (50–74 years old, no history of CRC or inflammatory bowel disease, nor colonoscopy within 5 years) and affiliated to 5,000 general practitioners (GPs) randomly selected from 41,589 eligible liberal GPs (main selection criteria: less than 75 years-old, caseload eligible to CRC-SP > 50, pay-for-performance indicators available) were included. Associations between beneficiaries’ and GPs’ characteristics and 2016–2017 CRC-SP participation, and its inter-physician heterogeneity (intraclass correlation coefficient, ICC) were evaluated using logistic mixed models. Results Of the 1.5 million beneficiaries included, 32.2% participated in the CRC-SP. Younger age, male gender, social deprivation and comorbidity score of beneficiaries were associated with lower participation, as well as the absence of GP consultation, or the absence of any healthcare reimbursement in the year before the screening campaign. GPs’ characteristics positively associated with participation were younger age, statutory sector with no extra billing, and high pay-for-performance prevention indicators. The volume of activity was negatively associated with participation. Adjustment on beneficiaries’ characteristics had little effect on the ICC (4.9% vs. 5.2% in the empty model). The ICC decreased after the addition of GPs’ characteristics (4.2%, including age, gender, extra billing, caseload characteristics and volume of activity) and pay-for-performance indicators (2.9%). Conclusions GPs and healthcare engagement play a key role in CRC screening participation. Improving uptake requires reducing structural barriers, enhancing GP support for prevention, and tailoring strategies to reach vulnerable populations. These findings support multifaceted interventions targeting both patient-level disparities and provider-level capacities to foster better and equitable participation. Trial registration not applicable.