Appendix C1. Contextual Question: What is the cost-effectiveness of screening?

A total of 298 abstracts relevant to costs of breast cancer screening were identified by searches and 29 full text articles were retrieved for further review. Studies focused on costs and cost savings of screening, comparisons of screening strategies or programs, and costs for older women.

Data from 10,048 women screened at an integrated cancer center in the United States were used to estimate the financial impact of a screening mammography program, including costs for mammography, diagnostic procedures, and therapeutic procedures.¹ Overall results showed that screening mammography operated at a loss, and payer reimbursement was not sufficient to cover overhead costs. The screening mammography program was not financially viable without clear criteria to increase the yield of diagnostic and therapeutic procedures.

A retrospective cohort study of 566 Finnish women diagnosed with invasive cancer determined mortality rates and costs for screened and unscreened women.² Women were age 40–74 years at time of diagnosis. Twenty-five percent of unscreened women died of breast cancer versus 12% of screened (p<0.001). The non-discounted mean treatment costs were 2.8-fold for those dying of breast cancer compared to survivors (26,222 euros [$36,283 USD] versus 9,434 euros [$13,053.8 USD]; mean difference 16,788 euros; 95% confidence interval (CI), 14,915, 18,660; p<0.001). Approximately one third of costs for fatal breast cancer were avoided through mammography screening, accounting for 72–81% of the estimated total treatment cost savings achieved by screening. It was also estimated that approximately 31–35% of the screening costs for 1987–1993 were offset by savings in treatment costs.²

A recent retrospective cost-effectiveness analysis in the United States compared costs when using actual patterns of screening mammography for women age 40–80 years, no screening, and other screening strategies.³ Usual screening practices in the model were informed by data from the National Health Interview Survey and the Breast Cancer Surveillance Consortium (BCSC) using a combination of frequent and infrequent screening patterns including no screening. Screening patterns from 1990–2000 accrued 947.5 million quality-adjust life years (QALYs) and cost $166 billion over the lifetimes of the screened women. This represents a gain of 1.7 million QALYs for an additional cost of $62.5 billion compared with no screening. The actual population screening scenario presumed that in the year 2000, 25% of the population had no screening, women being screened every 1 or 2 years increased to 50%, and overall screening participation rose to nearly 70%.³ The incremental cost per QALY accrued was estimated at $37,000 for actual screening patterns compared to no screening, well within the accepted level of $50,000 per QALY for health services in general. The most expensive option was annual screening of all women age 40–80 years, consistent with current guidelines. Many alternative screening strategies generated more QALYs for less cost compared to current guidelines. However, results differed depending on the level of participation in the program and when considering adverse effects of screening.

An analysis of Japanese data compared the cost-effectiveness of 3 screening strategies in a hypothetical cohort of 100,000 women age 30–79 years. These included annual clinical breast exam (CBE), annual CBE combined with mammography, and biennial CBE combined with mammography.⁴ The number of expected survival years was highest for annual CBE combined with mammography, implying the most effective treatment. Biennial CBE combined with mammography had a higher cost-effectiveness ratio compared with annual CBE combined with mammography, followed by annual CBE in all age groups. Annual CBE did not confer any advantages in terms of effectiveness or cost-effectiveness.⁴

An evaluation of the cost-effectiveness of a quality controlled mammographic screening program compared to an opportunistic screening program used cancer registry and clinical data from Switzerland.⁵ Results showed that the discounted incremental cost-effectiveness ratio comparing quality controlled mammographic screening programs verses established opportunistic screening programs ranged from $73,018 ($61,545.8 USD) at age 40 years to $118,193 ($99,623.2 USD) at age 70 years per life-year gained.

Many cost-effectiveness decision modeling studies focus on mammography screening for older women to consider the appropriate age to discontinue screening. A decision analysis model suggested that screening saves lives at all ages, even among older women.⁶ For women age 65–69 years or age 85 years or older with screen-detected breast cancer, screening increased life expectancy by 311 and 126 days, respectively. An analysis utilizing measurement of bone mineral density to predict higher breast cancer risk among elderly women found that continuing biennial mammography from ages 65–79 years among women in the top 3 quartiles of bone density would avert 9.4 deaths per 10,000 women screened.⁷ As treatment for chronic diseases improves and life expectancy increases, screening for breast cancer among older women may yield greater benefit.

Using a $50,000 (USD) per life-year saved acceptability threshold, a recent cost-effectiveness and computer modeling study suggested screening was equitable when starting at age 35 and ending at age 85.⁸ Also, two reviews in this area focused on the costs, benefits, and harms of screening mammography in older women. One systematic review and cost-analysis showed that the estimated cost of extending biennial screening mammography to 75 or 80 years was $34,000–$88,000 (2002 USD) per life-year gained, compared with stopping screening at 65 years.⁹ In a similar review done in Australia, cost-effectiveness estimates for extending the upper age limit for screening from 69 to 79 years ranged from $8,119 to $27,751 [6,746.88 to 23,061 USD] per QALY saved.¹⁰

Appendix Figure C2Statistical Tests for Meta-analysis of Screening Trials of Women Age 39 to 49 Years