Mammography status using patient self-reports and computerized radiology database.

Abstract

OBJECTIVES: This study sought to compare self-reported mammography use of low-income women utilizing an inner-city public hospital with a computerized hospital database for tracking mammography use. METHODS: A survey of all age-eligible women using the hospital's internal medicine clinic was done; responses were matched with the radiology database. We examined concordance among the two data sources. RESULTS: Concordance between self-report and the database was high (82%) when using "ever had a mammogram at the hospital," but low (58%) when comparing self-reported last mammogram with the information contained in the database. CONCLUSIONS: Disagreements existed between self-reports and the database. Because we sought to ensure that women would know exactly what a mammogram entailed by including a picture of a woman having a mammogram, it is possible that women's responses were accurate, leading to concerns that discrepancies might be present in the database. Physicians and staff must ensure that they understand the full history of a woman's experience with mammography before recommending for or against the procedure.