[Skip to Navigation]
Sign In
Research Letter
May 4, 2020

Association of Time With Reliability of Maternal Recall of Infant Birth Weight

Author Affiliations
  • 1Leadership Sinai Centre for Diabetes, Mount Sinai Hospital, Toronto, Ontario, Canada
  • 2Universidad Nacional Autónoma de México, Mexico City, Mexico
  • 3Instituto Nacional de Pediatría, Mexico City, Mexico
  • 4Division of Endocrinology, University of Toronto, Toronto, Ontario, Canada
  • 5Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, Ontario, Canada
JAMA Pediatr. 2020;174(12):1208-1210. doi:10.1001/jamapediatrics.2020.0389

Amid growing recognition of the Developmental Origins of Health and Disease, maternal recall of infant birth weight has emerged as a mainstay of clinical research associating fetal exposures with subsequent health in childhood. Since studies at various offspring ages have noted the accuracy of maternal report of birth weight,1-3 it is widely assumed that the reliability of this recall is stable over time. However, to our knowledge, this assumption, which underpins the validity of a critical variable, has not been directly evaluated by repeated assessment of maternal recall over time. Thus, our objective was to evaluate the reliability of initial maternal recall of birth weight at 3 months post partum and the subsequent stability of recall at 3 years post partum.


This analysis was performed within a prospective observational cohort study in which women were recruited in pregnancy and followed up longitudinally in the years thereafter. The protocol has been previously described.4 Women underwent metabolic characterization in pregnancy, followed by repeated assessments at 3 months, 1 year, and 3 years post partum. At both 3 months and 3 years post partum, participants reported infant birth weight on interviewer-administered questionnaires. We thus compared birth weights reported at 3 months with the values recorded at delivery. Similarly, we compared birth weights reported at 3 years with the values reported at 3 months. The protocol was approved by the Mount Sinai Hospital Research Ethics Board, and all participants provided written informed consent. Data collection occurred from October 2003 to June 2015, and data analysis occurred from September 2019 to October 2019. Statistical analyses were performed with SAS 9.4 (SAS Institute), with 2-tailed P values less than .05 considered significant.


The study population consisted of 335 women (233 white women [69.6%], 40 Asian women [11.9%], and 62 women of other races/ethnicities [18.5%]) with a mean (SD) age of 35.2 (4.3) years. At delivery, the mean (SD) length of gestation was 39.1 (1.5) weeks, and infants had a mean (SD) birth weight of 3381 (482) g.

The women reported mean (SD) birth weights of 3357 (529) g at 3 months and 3296 (575) g at 3 years. Bland-Altman plots5 revealed that, while there was limited difference between the birth weight reported at 3 months and the value recorded at delivery (Figure, A), there was much greater divergence between birth weights reported at 3 years vs 3 months (Figure, B). To quantify these associations, we calculated the concordance correlation coefficient (CCC),6 which assesses agreement between 2 measures of the same continuous variable. Between birth weights reported at 3 months and delivery, the CCC was 0.86 (95% limits of agreement, 0.83-0.88). However, between birth weights reported at 3 years and 3 months, the CCC decreased to 0.69 (95% limits of agreement, 0.62-0.74). Indeed, at 3 months, 292 of 317 women (92.1%; 18 women had missing data) reported a birth weight within 100 g of the value recorded at delivery. By 3 years, however, only 138 of 323 women (42.7%; 12 women had missing data) reported birth weights within 100 g of the value reported at 3 months.

Figure.  Bland-Altman Plots
Bland-Altman Plots

Bland-Altman plots showing the difference between birth weight reported at 3 months and the value recorded at delivery (A) and the difference between birth weight reported at 3 years and the value reported at 3 months (B). Data are shown as the mean difference (dashed line) and 95% limits of agreement (dotted lines).

We next evaluated factors that may be differentially associated with recall (Table). The Pearson correlation (r) between birth weights at 3 months and delivery was 0.86 (P < .001), while the r between birth weights at 3 years and 3 months was 0.69 (P < .001). At both 3 months and 3 years, these correlations were unchanged on adjustment for maternal age, race/ethnicity, parity, delivery mode, maternal body mass index, or the intervening time between the birth weight reports. Of note, both correlations were mildly decreased by adjustment for length of gestation, likely reflecting differential recall of birth weights in women with preterm delivery. Specifically, in women with preterm delivery (n = 22), the correlations (r) were 0.998 (P < .001) in the first interval and 0.77 (P < .001) in the second interval, while the analogous correlations in their peers were 0.83 and 0.62 (both P < .001), respectively.

Table.  Pearson Correlations Between Birth Weights Reported at 3 Months Post Partum vs Delivery and Between Birth Weights Reported at 3 Years vs 3 Months
Pearson Correlations Between Birth Weights Reported at 3 Months Post Partum vs Delivery and Between Birth Weights Reported at 3 Years vs 3 Months


We demonstrate that the validity of maternal recall of infant birth weights decreases on repeat assessment within less than 3 years. While an observational design is limited by the potential for unrecognized confounding, the repeated assessments herein suggest that studies involving maternal report of birth weights should account for the influence of time since delivery. Furthermore, these studies should also consider the possibility of differential validity of recall in women with vs without preterm delivery. Recognition of these considerations may enhance the integrity of this critical variable, which otherwise may compromise the interpretation of Developmental Origins of Health and Disease research findings.

Back to top
Article Information

Corresponding Author: Ravi Retnakaran, MD, Leadership Sinai Centre for Diabetes, Mount Sinai Hospital, 60 Murray St, Ste L5-025, Mailbox-21, Toronto, ON M5T 3L9, Canada (ravi.retnakaran@sinaihealthsystem.ca).

Published Online: May 4, 2020. doi:10.1001/jamapediatrics.2020.0389

Author Contributions: Dr Retnakaran had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Retnakaran.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: All authors.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Ye.

Obtained funding: Retnakaran.

Administrative, technical, or material support: Casas-Guzik, Retnakaran.

Supervision: Retnakaran.

Conflict of Interest Disclosures: Dr Retnakaran holds the Boehringer Ingelheim Chair in Beta-Cell Preservation, Function and Regeneration at Mount Sinai Hospital, and his research program is supported by the Sun Life Financial Program to Prevent Diabetes in Women. No other disclosures were reported.

Funding/Support: This study was supported by operating grants from the Canadian Institutes of Health Research (grants MOP-84206 and PJT-156286) and Diabetes Canada (grant CDA-OG-3-15-4924-RR).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Kassem  Z, Burmeister  C, Johnson  DA, Dakki  H, Joseph  CLM, Cassidy-Bushrow  AE.  Reliability of birth weight recall by parent or guardian respondents in a study of healthy adolescents.   BMC Res Notes. 2018;11(1):878. doi:10.1186/s13104-018-3977-2PubMedGoogle ScholarCrossref
Keenan  K, Hipwell  A, McAloon  R, Hoffmann  A, Mohanty  A, Magee  K.  Concordance between maternal recall of birth complications and data from obstetrical records.   Early Hum Dev. 2017;105:11-15. doi:10.1016/j.earlhumdev.2017.01.003PubMedGoogle ScholarCrossref
Bat-Erdene  U, Metcalfe  A, McDonald  SW, Tough  SC.  Validation of Canadian mothers’ recall of events in labour and delivery with electronic health records.   BMC Pregnancy Childbirth. 2013;13(suppl 1):S3. doi:10.1186/1471-2393-13-S1-S3PubMedGoogle ScholarCrossref
Pinnaduwage  L, Ye  C, Hanley  AJ,  et al.  Changes over time in hepatic markers predict changes in insulin sensitivity, β-cell function, and glycemia.   J Clin Endocrinol Metab. 2018;103(7):2651-2659. doi:10.1210/jc.2018-00306PubMedGoogle ScholarCrossref
Bland  JM, Altman  DG.  Statistical methods for assessing agreement between two methods of clinical measurement.   Lancet. 1986;1(8476):307-310. doi:10.1016/S0140-6736(86)90837-8PubMedGoogle ScholarCrossref
Lin  LI.  A concordance correlation coefficient to evaluate reproducibility.   Biometrics. 1989;45(1):255-268. doi:10.2307/2532051PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words