HPV Self-Sampling in Non-Attendees of the Cervical Cancer Screening Program in Germany—A Decision-Analytic Evaluation of Benefit-Harm-Cost Trade-Offs

Résumé fourni par la source

Background/Objectives: In Germany, one third of screening-eligible women remain un- or under-screened. We evaluated the long-term benefits, risks, and cost-effectiveness of human papillomavirus self-sampling (HPV-SS) strategies for non-attendees. Methods: Using a validated Markov-state-transition model, we evaluated HPV-SS for non-attendees every five years at ages 25-65, 30-65 or 35-65, with ordering an HPV-SS test (opt-in), or the test sent with the invitation letter (send-to-all). German clinical, epidemiological, and economic data along with international test-accuracy and HPV-SS-attendance data were used. Outcomes included life-years gained (LYG), the incremental harm-benefit ratio (IHBR), and cost-effectiveness ratio (ICER), compared to the next non-dominated strategy. We adopted the German statutory health insurance perspective with 3% annual discount rate for health effects and costs. Comprehensive sensitivity analyses were performed. Results: Incremental undiscounted effectiveness per 1,000 women compared to standard screening without HPV-SS ranged from 0.90 LYG for 5-yearly HPV-SS (opt-in) at age 35-65 to 1.67 LYG for HPV-SS (send-to-all) at age 25-65. Discounted ICERs compared to next effective non-dominated strategies were 22,700 EUR/LYG for 5-yearly HPV-SS (opt-in) at age 35-65, and for HPV-SS (send-to-all) 25,900 EUR/LYG at age 35-65, 726,000 EUR/LYG at age 30-65, and 1.78 million EUR/LYG at age 25-65. IHBRs for these strategies were 31, 32, 769, and 1,774 additional positive screening tests per LYG, respectively. Results were robust over a wide range of variations in parameters. Conclusions: Offering HPV-SS (send-to-all) to non-attendees every five years at age 35-65may have a good benefit-harm balance and be cost-effective. Our results may inform decision makers and clinical guideline developers in Germany.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.