RESUMO
The effect of different invitation models on participation in cervical cancer screening (CCS) was investigated in a randomized population-based cohort study in Germany. Participants were randomly selected via population registries and randomized into intervention Arm A (invitation letter) and Arm B (invitation letter and information brochure) or control Arm C (no invitation). The intervention and control arms were compared with regard to 3-year participation and the two invitation models were compared between intervention arms. Of the 7,758 eligible women aged 30-65 years, living in the city of Mainz and in the rural region of Mainz-Bingen, 5,265 were included in the analysis. Differences in proportions of women attending CCS were investigated and logistic regression was performed to analyze various factors influencing participation. In the intervention group, 91.8% participated in CCS compared to 85.3% in the control group (p < 0.0001), with a 6.6 percentage point increase in participation [95% confidence interval (CI) 4.6-8.6] and an adjusted odds ratio (OR) of 2.69 (95% CI 2.15-3.37). Effect estimators increased to 21.9 percentage points (95% CI 16.7-27.1) and an OR of 3.64 (95% CI 2.74-4.82), respectively, when women who participated in screening annually were excluded from the analysis. The invitation letter was particularly effective among women with lower school education, migrant women and older women. No difference in participation was found between the intervention Arm A and Arm B. An accompanying information brochure did not motivate more women to undergo CCS. However, a written invitation statistically significantly increased participation in CCS in Germany.