Reliability and validity of women's recall of mammographic screening

Abstract

Objective: To assess the reliability and validity of self-reported attendance for mammographic screening. Methods: To assess reliability of recall of attendance for a screening mammogram, 100 women selected at random were interviewed twice (approximately one week apart). To assess validity, 127 women who reported having a mammogram within the national breast screening program (BreastScreen Australia) consented to having their reports verified by the national program. Results: Test-retest reliability for the question 'Have you ever had a mammogram?' was perfect (agreement 100%, kappa 1). Validity was also high. About one-quarter of women (24.4%) recalled the exact date of their last mammogram and a further third (39.4%) correctly reported the month in which the mammogram was done. Almost all (91.3%) women reported the mammogram date accurately to within 12 months of the recorded date. Conclusions: These data suggest that Australian women provide reliable and valid information in relation to mammographic screening attendance. Implications: Self-reported data about attendance for mammographic screening are likely to provide reliable and valid estimates for research and health services evaluation purposes.