Screening patterns within organized programs of Italian women with invasive cervical cancer

Background: The efficacy of organized cervical screening programs (OCSPs) in reducing, through the Pap-smear test, the burden of invasive cervical cancer (ICC) is well established. Opportunistic screening is widespread in Italy, while OCSP implementation is demanded to regional governments. In some regions, OCSPs have been implemented since the ‘90s whereas in others OSCPs do not cover the entire population yet. Objective: To describe screening patterns of women diagnosed with ICC, and to identify difficulties and limitations of OCSPs at a population level in Italy. Methods: OCSPs invite women aged 25-64 years to perform a free of charge Pap smear every 3 years. This study focused on 3268 women aged 25-65 years, diagnosed with ICC between 1995 and 2008, who lived in areas covered by population-based cancer registries (CRs) with active OCSPs. CRs and OCSPs databases were linked to classify each woman according to OCSP invitation, compliance with invitation, and Pap-smear results. Odds ratios (OR- adjusted for age at diagnosis) and 95% confidence intervals (CI) were computed using logistic regression models to assess the association between women characteristics, screening patterns, and ICC features. Among 2911 women (out of 3268) with updated vital status, hazard ratios (HR- adjusted for age and tumor stage) and 95% CI were computed using the Cox model to assess the impact of screening patterns on death risk. Results: Out of 3268 women with ICC, 657 (20.1%) were never invited for screening. Non invitation by OCSPs was inversely related with time elapsed between program activation and ICC diagnosis (OR=4.6, 95% CI: 3.7-5.9, for 3 years vs 6 years). Of the 2611 women with ICC invited by OCSPs, 1181 (45.2%) were non compliant. Non compliance was highest in southern Italy (80.6% vs 45.2% in the North -OR=4.9), and it significantly increased with ageing. 66.8% of women with advanced ICC stage at diagnosis (FIGO III-IV) were non compliant, as compared to 25.6% of those with micro-invasive stage (IA-IA1, OR=5.7). Among the 1430 compliant women, ICC was screen-detected by OCSPs in 1075 cases (75.2%) -in 794 cases at first Pap-smear and in 281 cases at a subsequent tests. Conversely, in 355 cases (24.8%) ICC was non screen-detected -all Pap-smears made within OCSPs were negative. As compared to compliant women with screen-detected ICC, the risk of death was significantly higher in never compliant women (HR=2.0; 95%CI: 1.6-2.5), in never invited ones (HR=1.8; 95% CI 1.4-2.3), and in women with non screen-detected ICC (HR=1.7, 95%CI: 1.2-2.2). Conclusions: In Italy, about 56% of ICC cases diagnosed between 1995 and 2008 occurred in women from target populations for OCSPs who were never invited for screening, or who were non compliant. Factors associated with lack of invitation and with non-adherence to OCSPs deserve attention in view of the associated high risks of death. Screening history outside OCSPs should also be considered to fully evaluate non screen-detected ICC cases.