Validity of self-reported number of pregnancies and maternal age at first birth among females attending organized mammographic screening

Abstract

Aims: Reproductive history conveys information about potential health risks later in adulthood. This study aimed to examine the validity of self-reported number of pregnancies and maternal age at first birth (AFB) among females attending BreastScreen Norway. Methods: Participants were identified through the Janus Serum Bank cohort in Norway and were eligible for this cross-sectional validation study if they participated in a health survey issued by BreastScreen Norway between 2006 and 2015. Retrospective self-reported survey information on number of pregnancies and AFB in years was validated against prospectively collected information from the Medical Birth Registry of Norway (MBRN) using the Spearman rank (rs) and intraclass correlation coefficients (ICC) with 95% confidence intervals (CI). Results: After exclusions, 51,598 subjects were included in the analysis on number of pregnancies and 46,919 in the analysis on AFB. On average, study subjects were 59–60 years old when completing the health survey and had become first-time mothers roughly 36 years earlier. Survey-based information about number of pregnancies was highly correlated and demonstrated high agreement with the registry data (rs=0.967, 95% CI 0.964–0.969; ICC=0.884, 95% CI 0.882–0.885). Survey-based information about AFB demonstrated even higher correlation and very high agreement with the registry data (rs=0.975, 95% CI 0.973–0.976; ICC=0.974, 95% CI 0.974–0.975). Conclusions: Retrospectively recalled survey-based information about number of pregnancies and AFB was highly accurate when validated against prospectively collected information in the MBRN. These survey-based data are valuable for future epidemiological research, and linkage to the MBRN may not be required when these data are available.