The cost-effectiveness of mammographic screening strategies

Karen K Lindfors, C. J. Rosenquist

Research output: Contribution to journalArticle

119 Scopus citations

Abstract

Objective. - To compare and analyze the cost-effectiveness of different mammographic screening strategies. Design. - A computer simulation model was developed to compare mammographic screening with observation without screening. Cost-effectiveness was expressed as marginal cost per year of life saved (MCYLS) and was calculated for the following mammographic screening strategies. - (1) annual for ages 40 to 79 years; (2) annual for ages 50 to 79 years; (3) biennial for ages 50 to 79 years; (4) annual for ages 40 to 49 years with biennial for ages 50 to 79 years; (5) annual for ages 40 to 64 years with biennial for ages 65 to 79 years; (6) biennial for ages 40 to 49 years with annual for ages 50 to 79 years; and (7) annual for high-risk and biennial for normal-risk women aged 40 to 49 years with annual for ages 50 to 79 years. Data Sources. - The probability and cost of all outcomes were established from previously published data or community experience. Results. - The most cost-effective screening strategy is biennial mammography for women aged 50 to 79 years, with an MCYLS of $16 000. Adding annual mammography for women aged 40 to 49 years increases the MCYLS to $20 200, but is more cost-effective than other tested protocols that included women in their 40s; annual mammography for ages 40 to 49 years with biennial for ages 50 to 79 years is also more cost-effective than annual mammography for ages 50 to 79 years. Conclusion. - Screening programs that include women in their 40s can be as cost-effective as some that exclude such women. Choice of a screening strategy depends on financial resources and desired effectiveness.

Original languageEnglish (US)
Pages (from-to)881-884
Number of pages4
JournalJournal of the American Medical Association
Volume274
Issue number11
StatePublished - 1995

    Fingerprint

ASJC Scopus subject areas

  • Medicine(all)

Cite this