UC Davis

Background

The cost-effectiveness of screening mammography beyond age 75 years remains unclear.

Objective

To estimate benefits, harms, and cost-effectiveness of extending mammography to age 80, 85, or 90 years according to comorbidity burden.

Design

Markov microsimulation model.

Data sources

SEER (Surveillance, Epidemiology, and End Results) program and Breast Cancer Surveillance Consortium.

Target population

U.S. women aged 65 to 90 years in groups defined by Charlson comorbidity score (CCS).

Time horizon

Lifetime.

Perspective

National health payer.

Intervention

Screening mammography to age 75, 80, 85, or 90 years.

Outcome measures

Breast cancer death, survival, and costs.

Results of base-case analysis

Extending biennial mammography from age 75 to 80 years averted 1.7, 1.4, and 1.0 breast cancer deaths and increased days of life gained by 5.8, 4.2, and 2.7 days per 1000 women for comorbidity scores of 0, 1, and 2, respectively. Annual mammography beyond age 75 years was not cost-effective, but extending biennial mammography to age 80 years was ($54 000, $65 000, and $85 000 per quality-adjusted life-year [QALY] gained for women with CCSs of 0, 1, and ≥2, respectively). Overdiagnosis cases were double the number of deaths averted from breast cancer.

Results of sensitivity analysis

Costs per QALY gained were sensitive to changes in invasive cancer incidence and shift of breast cancer stage with screening mammography.

Limitation

No randomized controlled trials of screening mammography beyond age 75 years are available to provide model parameter inputs.

Conclusion

Although annual mammography is not cost-effective, biennial screening mammography to age 80 years is; however, the absolute number of deaths averted is small, especially for women with comorbidities. Women considering screening beyond age 75 years should weigh the potential harms of overdiagnosis versus the potential benefit of averting death from breast cancer.

Primary funding source

National Cancer Institute and National Institutes of Health.