Improving Breast Cancer Screening Uptake Through Tailored Reminder Letters

Overview

The goal of this trial is to learn whether a culturally and linguistically tailored reminder letter can increase participation in the Flemish Breast Cancer Screening Program among women who previously did not attend screening. The study also examines whether the effect differs between socioeconomic groups. The main questions it aims to answer are: * Does a tailored reminder letter increase participation among previous non-attenders? * Does the effect differ between women with lower and higher socioeconomic status? Researchers compare two groups: * Intervention group: receive the standard invitation letter plus an additional tailored reminder letter. * Control group: receive the standard invitation letter only. Participation in screening within 40 days of the invitation is measured using the program's registry. Participants are women aged 50-69 years who had previously been invited but had never attended screening. The reminder letter is sent as part of routine screening communication, and no additional medical procedures are involved.

This study aims to evaluate the effectiveness of a culturally and linguistically tailored reminder letter in increasing mammography uptake within the Flemish Breast Cancer Screening Program (BCSP) among women who previously did not attend screening. The objectives are to: * Quantify the difference in uptake between the intervention and control arms; and * Assess whether the effect differs according to socioeconomic status (SES). Study Setting: The study is embedded within the Flemish BCSP, which is coordinated by the Centre for Cancer Detection (CvKO). The BCSP provides biennial screening mammography, free of charge, to eligible individuals aged 50-69 years registered as female in the Belgian national registry. Recruitment occurs through the routine program invitation process, where eligible individuals receive a mailed invitation letter indicating the time and location of their screening appointment. Study design: This study is a two-arm, parallel-group, superiority, randomized controlled trial (RCT) with partial blinding. Study arms: * Intervention arm: Participants allocated to the intervention arm receive two mailed communications: (1) the standard BCSP invitation letter, and (2) a tailored reminder letter, specifically designed to address linguistic and cultural barriers that may affect screening participation. * Control arm: Participants allocated to the control arm receive the standard BCSP invitation letter only. Detailed descriptions of study Arms and Intervention, Study Design (Masking, Allocation), Outcome Measures and Eligibility are provided in the designated sections. Statistical analysis: The primary analysis compares BC screening uptake between the intervention and control arms in the overall study population. The effect of the intervention is quantified using odds ratios (ORs) with corresponding 95% confidence intervals (CIs). Statistical significance is assessed using the likelihood-ratio chi-square test (x²), with two-sided p-value \< 0.05. Stratified analyses are conducted by socioeconomic status (SES), age group, province of residence, and type of mammographic unit. Analyses are performed according to both intention-to-treat (ITT) and per-protocol (PP) principles, with ITT as the primary analytic approach and PP as a sensitivity analysis. The ITT population comprises all randomized participants, excluding records removed before analysis during routine quality control procedures (e.g., duplicate system entries or invalid identifiers). The PP population comprises participants who receive the allocated intervention as planned and for whom complete outcome data is available. Exclusion criteria for the PP analysis comprise withdrawal or opt-out from the BCSP prior to receipt of the assigned communication strategy and documented logistical issues affecting letters delivery in either study arm. Deaths or emigrations occurring prior to participation are not specifically addressed in the PP definition as they are assumed to be evenly distributed between study arms, in line with routine cross-checks with demographic registries. No multivariable adjusted analyses are performed. Stratified analyses are interpreted descriptively, and no formal correction for multiple comparisons is applied.