Rolan, E.P., Robertson, O., Nonkovic, N. et al. Reliability of prospective and retrospective maternal reports of prenatal experiences. BMC Pregnancy Childbirth 22, 968 (2022). https://doi.org/10.1186/s12884-022-05286-7
Date of this Version
Maternal recall, Reliability of prenatal retrospective reports, Socioeconomic status, Prospective longitudinal, Prenatal stress
Extant perinatal research utilizes retrospective reports on the prenatal environment, but there are limited data on the validity of retrospective data compared with prospective data. The current study examined the reliability of birth mothers’ memory of prenatal stress and distress and perinatal risks at 6-months postpartum with maternal reports gathered across each trimester of pregnancy and explored whether recall varied with maternal socioeconomic status.
Surveys were collected from 34 pregnant women (M age = 29.14, SD = 5.06 years, 83% non-Hispanic White) on stress, distress, and pregnancy complications at 12(T1), 26(T2), and 38(T3) weeks of pregnancy, and at 6-month post-partum asking the same questions but specifically about the pregnancy. Cohen’s kappa and Pearson’s correlations were used to investigate maternal recall at post-partum with prospective reports at T1, T2, T3 and an average score of T1, T2, and T3. Correlations were also examined separately for those with high and relatively lower socioeconomic status.
Birth mothers’ recall was generally reliable. Retrospective reports were most strongly related to prospective reports in T1 for perceived stress, T1 and T3 for anxiety symptoms and exposure to toxins, but T3 for depressive symptoms. Recall of pregnancy complications best reflected the average score across trimesters (rather than specific trimesters). Women with higher socioeconomic status better recalled prenatal (di)stress, but women with relatively lower socioeconomic status better recalled exposure to toxins.
This study provides support for utilizing retrospective reports of maternal prenatal experiences at 6-months post-partum, with implications for interpretation of specific recalled phenotypes.