Abstract / Summary
Abstract Background Persistent disparities in cervical cancer screening (CCS) participation undermine the equity and effectiveness of screening programmes. This study examined the sociodemographic, socioeconomic, healthcare-access and health-related determinants of long-term non-attendance in the Flemish CCS programme, guided by Andersen’s Behavioural Model of Health Services Use. Methods We conducted a retrospective population-based cohort study using individually-linked population registry, cancer registry, health insurance and administrative socioeconomic data for all eligible women aged 31–64 residing in Flanders between 2018 and 2023. Long-term non-attendance was defined as having no registered CCS test during the six-year observation period. Women with no registered CCS were compared with women who had at least one registered CCS test during the same period. Hierarchical multivariable logistic regression models were used to examine associations with predisposing (age, migration, household type, education and occupation), enabling (household income, increased reimbursement status, recent childbirth, and general practitioner contact), and illness level factors (chronic illness and major invalidity status). Results Of 1,220,227 complete cases, 21.5% were long-term non-attenders. Non-attendance was strongly socially patterned. Indicators of healthcare contact showed the strongest associations: women with a recent childbirth (OR 0.37; 95% CI 0.37–0.38) and frequent GP contact (OR 0.54; 0.54–0.55) had markedly lower odds of long-term non-attendance. Educational level showed a strong gradient that persisted after full adjustment (lowest vs. highest: OR 1.79; 1.76–1.81), whereas household income showed only a modest, non-graded association. Elevated odds among most first-generation migrant groups were largely attenuated by adjustment for socioeconomic factors, while odds persisted among second-generation women of Maghrebi, Turkish, Asian, Sub-Saharan African, and Eastern-European origin. Older age, living alone, increased reimbursement status, chronic illness, and major invalidity were also independently associated with non-attendance. Conclusions Social inequalities in long-term non-attendance in the Flemish CCS programme persist despite organised screening. Population-level data reveal inequities among women with fewer socioeconomic resources, limited healthcare engagement, and ethno-racially minoritised second-generation women especially. Equity-oriented strategies should prioritise women who remain systematically unreached by the programme. Further research into the specific mechanisms and barriers underlying these inequities is needed to guide effective and tailored programme adaptations.