Promoting participation in a population screening program for breast and cervical cancer: a randomized trial of different invitation strategies

Abstract

Aims and background: Attendance level has been identified as a major determinant of cost-effectiveness of organized screening programs. We tested the effectiveness of 4 different invitation systems in the context of an organized population screening program for cervical and breast cancer.

Methods: Women eligible for invitation--8385 for cervical and 8069 for breast cancer screening--listed in the rosters of 43 and 105 general practitioners (GP), respectively, who had accepted to collaborate in the program, were randomized to 4 invitation groups: Group A--letter signed by the GP, with a prefixed appointment; Group B--open-ended invitation, signed by the GP, prompting women to contact the screening center to arrange an appointment; Group C--letter (same as for group A), signed by the program coordinator, with a prefixed appointment; Group D--extended letter (highlighting the benefits of early cancer detection) signed by the GP, with a prefixed appointment. Assignment to the interventions was based on a randomized block design (block=GP).

Results: Assuming Group A as the reference, the overall compliance with cervical cancer screening was reduced by 39% in Group B (RR=0.61; 95% CI, 0.56-0.68) and by 14% in Group C (RR=0.86; 95% CI, 0.78-0.93); no difference was observed for Group D (RR=1.03; 95% CI, 0.95-1.1). The response pattern was similar for breast screening (Group B: RR=0.71; 95% CI, 0.65-0.76; Group C: RR=0.87; 95% CI, 0.81-0.94; Group D: RR=1.01; 95% CI, 0.94-1.08).

Conclusions: Personal invitation letters signed by the woman's GP, with preallocated appointments, induce a significant increase in compliance with screening. Efficiency can be ensured through the adoption of overbooking, provided that attendance levels are regularly monitored.