Long-term Recall of Pregnancy-related Events

Abstract

Background Early-life factors can be associated with future health outcomes and are often measured by maternal recall. Methods We used data from the North Carolina Early Pregnancy Study and Follow-up to characterize long-term maternal recall. We used data from the Early Pregnancy Study as the gold standard to evaluate the accuracy of pre-pregnancy weight, early pregnancy behaviors, symptoms and duration of pregnancy, and child’s birthweight reported at follow-up, for 109 women whose study pregnancies had resulted in a live birth. Results Most (81%) participants reported a pre-pregnancy weight at follow-up that correctly classified them by BMI category. Women reported experiencing pregnancy symptoms later at follow-up than what they reported in the Early Pregnancy Study. Accuracy of reporting of early pregnancy behaviors varied based on exposure. Overall, women who had abstained from a behavior were more likely to be classified correctly. Sensitivity of reporting was 0.14 for antibiotics, 0.30 for wine, 0.71 for brewed coffee, and 0.82 for vitamins. Most misclassification at follow-up was due to false-negative reporting. Among women who gave birth to singletons 94% could report their child’s correct birthweight within ½ pound and 86% could report duration of pregnancy within 7 days at follow-up. Conclusions Self-report of pre-pregnancy weight, duration of pregnancy, and child’s birthweight after almost 30 years was good whereas self-reported pregnancy-related exposures resulted in higher levels of reporting error. Social desirability appeared to influence women’s report of their behaviors at follow-up. Self-reported assessment of confidence in the recalled information was unrelated to accuracy.

Breastfeeding duration is associated with offspring’s adherence to prudent dietary pattern in adulthood: results from the Nutritionist’s Health Study

“…However, previous studies pointed out that these variables, as well as birthweight and gestational age, were reported accurately 20 years after birth. 58,59 The FFQ used to assess dietary intake of the previous year could also involve recall bias although this tool has been used widely in nutrition epidemiology, allowing classification according to diet exposure level. [60][61][62] The dichotomization of the dietary pattern scores and the consequent reduction of statistical power should be another limitation.…”

Renal Complications in Pregnancy Preceding Glomerulonephropathy Diagnosis

“…We found that among women with GN, those who experienced worsening kidney function, increasing proteinuria, worsening blood pressure control, and/or preeclampsia during prior pregnancy were diagnosed with GN sooner than their counterparts without these pregnancy complications. Although we considered recall bias as a potential explanation for this difference (i.e., women newly diagnosed with GN recall recent pregnancy complications more readily), literature suggests that maternal recall of even distant pregnancy events is generally accurate,6, 7 and recent validation of the question “Did you have a hypertensive disorder in pregnancy?” in a Dutch population of women was found to be 84% sensitive and 94% specific when compared to medical chart review 8 . In our study, pregnancies with reported renal-relevant complications were also objectively more likely to deliver prematurely with a lower mean birth weight, disputing recall bias as the sole explanation for the findings.…”

Pre-pregnancy BMI is associated with offspring body composition in adulthood before adiposity-related disorders: a retrospective cohort

“…Although there is possibility that women could selfreport pre-pregnancy weight with reasonable accuracy almost 30 years later (24) , the recall of maternal prepregnancy BMI implies recall bias that should be considered a limitation of the present study. Additionally, our small and homogenous sample, composed of young, apparently healthy and educated women (and their mothers), implies limitation regarding external validity; that is, generalisation of our results to different samples with different characteristics is not appropriate.…”