Cost-effectiveness of 3 methods to enhance the sensitivity of Papanicolaou testing

Abstract

Context: ThinPrep, AutoPap, and Papnet are 3 new technologies that increase the sensitivity and cost of cervical cancer screening.

Objective: To estimate the cost-effectiveness of these technological enhancements to Papanicolaou (Pap) tests.

Design: We estimated the increase in sensitivity from using these technologies by combining results of 8 studies meeting defined criteria. We used published literature and additional sources for cost estimates. To estimate overall cost-effectiveness, we applied a 9-state time-varying transition state model to these data and information about specific populations.

Setting: A hypothetical program serving a cohort of 20- to 65-year-old women who begin screening at the same age and are representative of the US population.

Results: The new technologies increased life expectancy by 5 hours to 1.6 days, varying with the technology and the frequency of screening. All 3 technologies also increased the cost per woman screened by $30 to $257 (1996 US dollars). AutoPap dominated ThinPrep in the base case. At each screening interval, AutoPap increased survival at the lowest cost. The cost per year of life saved rose from $7777 with quadrennial screening to $166000 with annual screening. Papnet produced more life-years at a higher cost per year of life saved. However, when used with triennial screening, each of them produced more life-years at lower cost than conventional Pap testing every 2 years. The cost-effectiveness ratio of each technology improved with increases in the prevalence of disease, decreases in the sensitivity of conventional Pap testing, and increases in the improvement in sensitivity produced by the technology.

Conclusions: Technologies to increase the sensitivity of Pap testing are more cost-effective when incorporated into infrequent screening. Increases in sensitivity and decreases in cost may eventually make each technology more cost-effective.