Impact of medical exclusion data sources on population-based colorectal cancer screening round participation in France: a counterfactual analysis

The conventional participation rate (participation-rate) in the French population-based colorectal cancer screening program remains below the European target. One underexplored factor is the heterogeneity in access to data sources used to exclude medically ineligible individuals, potentially affecting the accuracy of participation rate estimates. We conducted a retrospective ecological study across 60 French departments using aggregated screening-round data from Santé publique France and information from a 2021 national survey on departmental exclusion practices. We assessed the association between duration of access to three external data sources—primary health insurance data, hospital morbidity data, and cancer registry data—and crude medical exclusion rates (exclusion-rate). Because participation-rate is mechanically influenced by exclusion ascertainment, we analyzed both the participation-rate and the screened fraction of the target population (screened-fraction) as a complementary outcome. Mixed-effects models and model-based counterfactual projections were used to estimate how screening indicators might have evolved under two hypothetical scenarios of more sustained access to provider data: Scenarios-1: uninterrupted Provider data use until 2023 in departments that had used them at any point between 2010 and 2023; Scenarios-2: continuous Provider data use from 2010 onward in all departments. Among the sixty departments, 21 used Provider data for an average of 6.8 years. Longer use of primary health insurance data (β = 0.35[0.24; 0.47]) and hospital morbidity data (β = 0.14[0.04; 0.24]) was associated with higher exclusion rates, whereas cancer registry data showed no independent association. Departments with provider-data access also showed higher reported participation-rates, but part of this difference reflected more complete refinement of the eligible population. The complementary analysis of the screened-fraction suggested a more limited change in actual screening uptake. Globally, model-based projections suggested that more sustained access to provider data would have increased the reported exclusion-rate by 6% points and the reported participation-rate by 2% points under Scenario-2. Unequal access to exclusion data sources likely contributed to under-ascertainment of medical exclusions and reduced comparability of participation indicators across departments. Equitable access to Provider data and standardized exclusion algorithms using the SNDS (“Système national des données de santé”) data would improve program evaluation and policy planning. What question did the study address? This study examined how heterogeneous access to external medical data sources—such as primary health insurance data, hospital morbidity data, and cancer registries—may influence medical exclusion rates and reported participation indicators in France’s population-based colorectal cancer screening program between 2010 and 2023. What are the main findings? Departments with more consistent access to and use of primary health insurance and hospital morbidity data had higher medical exclusion rates. They also showed higher reported participation rates, although part of this difference reflected more complete refinement of the eligible population rather than only greater screening uptake. Model-based projections suggested that more sustained use of these data sources could have improved the comparability and interpretation of screening indicators. Why are these findings important? These findings highlight the importance of high-quality medico-administrative data for accurate evaluation of population-based screening programs. Better data linkage and more standardized exclusion algorithms could improve the accuracy, comparability, and operational relevance of colorectal cancer screening indicators in France.