Strategies for delivery of faecal occult blood test kits and participation to colorectal cancer screening in the Emilia-Romagna Region of Italy.

There is a lack of agreement about which routine invitation strategy should be adopted in colorectal cancer screening. We conducted an observational study to assess the impact of three invitation strategies on participation. Invitation records for the years 2005-2009 were evaluated. There were 2,234,276 invitations from 1,230,683 individuals. Among first invitations, participation associated with direct mailing of the faecal occult blood test kits was slightly lower (relative risk, RR 0.985; 95% confidence interval 0.979-0.990) than that of the reference invitation strategy, that is, the distribution of the test kits by pharmacies. In repeated invitations/previous non-responders, the participation associated with the direct mailing of the test kits was even lower (RR 0.914; 95% confidence interval 0.895-0.933) and this was also the case for the distribution of the test kits by primary care centres (RR 0.983; 95% confidence interval 0.971-0.995). In contrast, in repeated invitations/previous responders, the impact of primary care centres and direct mailing of the test kits was greater than the use of pharmacies, showing only modest RRs: 1.021 (95% confidence interval 1.019-1.023) and 1.029 (95% confidence interval 1.025-1.033) respectively. The faecal occult blood test mailing strategy modestly increased participation in previous responders.