Skip to main content
NCBI home page
Maternal & Child Nutrition logoLink to Maternal & Child Nutrition
. 2014 Dec 17;12(3):632–638. doi: 10.1111/mcn.12160

Validity of birth certificate‐derived maternal weight data in twin pregnancies

Lisa M Bodnar 1,2,3,✉, Barbara Abrams 4, Lara Siminerio 1, Timothy L Lash 5
PMCID: PMC4470895  NIHMSID: NIHMS652692  PMID: 25522306

Abstract

Birth certificates are an important source of pre‐pregnancy body mass index (BMI) and gestational weight gain (GWG) data for surveillance and aetiologic studies, but little is known about their validity in twin pregnancies. Twins experience high rates of adverse perinatal outcomes that have been associated with BMI and GWG in singletons. Our objective was to evaluate the accuracy of birth certificate‐derived pre‐pregnancy BMI and GWG compared with medical record‐derived data in a sample of 186 twin pregnancies at a teaching hospital in Pennsylvania (2003–2010). Twelve strata were created by simultaneous stratification on pre‐pregnancy BMI (underweight, normal weight/overweight, obese class 1, obese classes 2 and 3) and GWG (<20th, 20–80th, >80th percentile). The agreement of birth certificate‐derived pre‐pregnancy BMI category with medical record BMI category was lowest among underweight mothers [75% (95% confidence interval 51–91%) ] and highest among normal/overweight [97% (90–99%) ] and obese classes 2 and 3 mothers [97% (85–99%) ]. Agreement for GWG category from the birth certificate varied from 57% (41–70%) for GWG >80th percentile to 80% (65–91%) and 82% (72–89%) for GWG <20th and 20th–80th percentiles, respectively. The misclassification of BMI and GWG was primarily due to error in pre‐pregnancy weight rather than weight at delivery or height. Agreement proportions for twins were not meaningfully different from the proportions in a comparable sample of singleton pregnancies. These data suggest that birth certificate‐based BMI and GWG data are prone to error in twin pregnancies. Those who use these data should conduct internal validation studies and adjust their results using bias analyses.

Keywords: birth certificate, obesity, pregnancy, twins, validation study, weight

Introduction

In 2009, the National Academy of Science/Institute of Medicine (IOM) Committee to Reevaluate Pregnancy Weight Gain Guidelines published revised gestational weight gain (GWG) recommendations for singleton pregnancies that are tailored according to the mother's pre‐pregnancy body mass index (BMI) category (IOM 2009). For twin pregnancies, the IOM Committee provided only provisional GWG guidelines because the evidence base to inform weight gain recommendations for twin pregnancies was limited (IOM 2009). The lack of guidance is problematic because twin births have risen 75% in the last 30 years (Martin et al. 2012) and are at high risk for poor health outcomes that have been linked with maternal weight gain in singletons (IOM 2009; Martin et al. 2011). To expand knowledge of optimal GWG guidelines for twin pregnancies, the committee called for ‘studies among women carrying multiple fetuses that link GWG to relevant health outcomes among both mothers and infants’ (IOM 2009).

Birth certificates are an important source of data on GWG and perinatal outcomes. Birth certificate‐derived GWG and BMI data have been widely used for research among singleton births and their use among twins is on the rise (Salihu et al. 2008, 2010; Sauber‐Schatz et al. 2012). Nevertheless, there is concern that information on maternal height, pre‐pregnancy weight, and weight at delivery from vital records is inaccurate because it is based on self‐report (IOM 2009), and that this misclassification may lead to biased associations with adverse perinatal outcomes (Bodnar et al. 2010). We recently undertook a large validation study of birth certificate‐derived data on pre‐pregnancy BMI and GWG compared with medical record information among singleton births in a Pennsylvania hospital and found poor agreement, particularly for extremes of maternal weight (Bodnar et al. 2014). In this paper, we expand this analysis to determine whether vital records BMI and GWG data among twin pregnancies are accurate enough to be used without major concern about misclassification bias, or whether measurement error requires that conventional associations between maternal weight and adverse outcomes be adjusted for the bias (Fox et al. 2005; Lash et al. 2009, 2014; MacLehose & Gustafson 2012).

Key messages.

  • Pre‐pregnancy body mass index and gestational weight gain data from birth certificates are prone to error in twin pregnancies.

  • The magnitude of the error is not different from results found in singletons.

  • Error is due to poor reporting of pre‐pregnancy weight as opposed to weight at delivery or height.

  • Researchers and public health professionals who use birth certificate data should conduct internal validation studies and adjust their results using bias analyses.

Methods

There were 34 628 twin births in Pennsylvania from 2003 to 2010 to non‐Hispanic black or non‐Hispanic white mothers with complete data on gestational age, pre‐pregnancy BMI, and maternal weight at delivery. From this eligible cohort, we extracted all twin pregnancies at Magee‐Womens Hospital in Pittsburgh, Pennsylvania (n = 550) to serve as the source population for the validation study. We limited the study to non‐Hispanic black or white mothers because other racial/ethnic groups were rare. We selected twin pregnancies into the validation study using a balanced design, which allows for equally precise measurements of accuracy across strata of interest (Holcroft & Spiegelman 1999). We created 12 strata by simultaneously stratifying on pre‐pregnancy BMI [underweight (<18.5 kg m−2), normal weight/overweight (18.5–29.9 kg m−2), obese class 1 (30–34.9 kg m−2), obese classes 2 and 3 (≥35 kg m−2) ] and GWG (<20th percentile, 20–80th percentile, >80th percentile). We collapsed normal weight and overweight because we expected the accuracy of their self‐report to be similar (Bodnar et al. 2010). We aimed to have each stratum contain 30 records.

The study design, setting, and methods of data collection were the same for the twin validation as for the singleton validation, and have been presented in detail previously (Bodnar et al. 2014). The U.S. birth certificate collects self‐reported pre‐pregnancy height and weight via interview before discharge (CDC National Center for Health Statistics 2003), which we validated against self‐reported pre‐pregnancy weight and height at first prenatal visit [median 9 (interquartile range 7–11) weeks gestation]. We used this alloyed gold standard because medical records did not contain measured preconception height and weight. Pre‐pregnancy BMI [pre‐pregnancy weight (kg)/height (m)2] was classified as earlier.

The birth certificate gathers maternal weight at delivery using either prenatal records or the labour and delivery admission history and physical (CDC National Center for Health Statistics 2003). We validated maternal weight at delivery using a measured weight at ≤4 weeks before delivery gathered from the prenatal record. GWG was defined as the difference between self‐reported pre‐pregnancy weight and weight at delivery. We divided GWG by gestational age at delivery to calculate a rate of GWG, which was categorized using percentiles of the distribution shown earlier.

Per cent agreement was defined as the proportion of births in a birth certificate BMI or GWG category that were correctly classified based on medical record review. We calculated exact binomial 95% confidence intervals for the agreement proportions. To determine whether the agreement proportions for birth certificate‐derived BMI and GWG categories among twins were similar to singletons, we compared results from the singleton validation sample (Bodnar et al. 2014) after weighting them to reflect the race‐ethnicity and preterm birth distribution of the twins validation sample. A Fisher's exact test was used to compare proportions.

Results

Three validation strata (all in normal/overweight strata) met the goal of including 30 pregnancies, although later, two pregnancies were found to be ineligible and were removed (Table 1). Eight of the 12 strata included at least 10 pregnancies. The final validation study sample size was 186 twin pregnancies.

Table 1.

Pre‐pregnancy BMI and GWG of twin births in the Pennsylvania birth certificate cohort (2003–2010, n = 34 628), twin pregnancies that served as the source population for the validation study (n = 550), and the subset of twins in the validation study (n = 186)

Pre‐pregnancy BMI category GWG category Twins birth certificate cohort Twin pregnancies in the birth certificate cohort that served as the source population for the validation study Twins validation substudy sample
n (%) n (%) n (%)
Underweight <20th percentile 225 (0.7) 5 (0.9) 5 (2.7)
Underweight 20th–80th percentile 649 (1.9) 16 (2.9) 13 (7.0)
Underweight >80th percentile 219 (0.6) 2 (0.4) 2 (1.1)
Normal weight or overweight <20th percentile 5 090 (15) 62 (11) 28 (15)
Normal weight or overweight 20th–80th percentile 15 039 (43) 277 (51) 30 (16)
Normal weight or overweight >80th percentile 4 995 (14) 95 (17) 30 (16)
Obese class 1 <20th percentile 852 (2.5) 6 (0.9) 6 (3.2)
Obese class 1 20th–80th percentile 2 558 (7.4) 29 (5.3) 27 (14)
Obese class 1 >80th percentile 853 (2.5) 14 (2.6) 10 (5.4)
Obese classes 2 and 3 <20th percentile 829 (2.4) 6 (1.1) 2 (1.1)
Obese classes 2 and 3 20th–80th percentile 2 491 (7.2) 28 (5.1) 22 (12)
Obese classes 2 and 3 >80th percentile 828 (2.4) 11 (1.8) 11 (5.9)
Open in a new tab

BMI, body mass index; GWG, gestational weight gain. Underweight, <18.5 kg m−2; normal weight and overweight, 18.5–29.9 kg m−2; obese class 1, 30–34.9 kg m−2; obese class 2 and class 3, ≥35 kg m−2.

A majority of the twins birth certificate cohort (43%) and in the validation study source population at our hospital (51%) was normal weight or overweight and had GWG in the 20th–80th percentile (Table 1). By design, the twin validation sample had a smaller proportion in this very common BMI–GWG stratum (16%) and greater proportions of births in the less common BMI–GWG strata. Consequently, the validation sample was also more likely than the birth certificate cohort to be primiparous (54% vs. 21%), non‐Hispanic black (22% vs. 16%), smokers (18% vs. 14%) and unmarried (36% vs. 28%).

The agreement proportion between the birth certificate‐derived pre‐pregnancy BMI and the medical record‐derived BMI was highest among obese classes 2 and 3 mothers [97% (95% confidence interval 85–99%) ] and normal/overweight mothers [97% (90–99%) ] and lowest (and most imprecise) among underweight women [75% (51–91%); Table 2 ]. When BMI category was misclassified, there was both over‐ and underestimation. Agreement proportions for twins were not significantly different from the weighted proportions among singletons (all P > 0.05). The misclassification of pre‐pregnancy BMI among twins was due to error in self‐reported pregravid weight rather than height. In 98% of twin births, birth certificate‐based height agreed within 1 in. (2.54 cm) of medical record height.

Table 2.

Agreement of birth certificate‐derived pre‐pregnancy body mass index categories with medical record derived categories among the subcohort of twin deliveries in the validation study compared with the weighted average of the agreement proportions in singletons

Birth certificate n Medical records
Underweight Normal weight and overweight Obese class 1 Obese class 2 and class 3
Twins
Underweight 20 75% (51%, 91%) 25% (8.7%, 49%) 0% 0%
Normal weight and overweight 88 2.3% (0.3%, 7.9%) 97% (90%, 99%) 1.1% (0.03%, 6.2%) 0%
Obese class 1 43 0% 7.0% (1.5%, 19%) 84% (69%, 93%) 9.3% (2.5%, 22%)
Obese class 2 and class 3 35 0% 0% 2.9% (0.07%, 15%) 97% (85%, 99%)
Singletons weighted average
Underweight 258 69% (63%, 75%) 31% (25%, 37%) 0% 0%
Normal weight and overweight 262 0.2% (0.01%, 2.1%) 98% (95%, 99%) 2.0% (0.6%, 4.4%) 0.2% (0.01%, 2.1%)
Obese class 1 291 0% 13% (9.4%, 18%) 73% (67%, 78%) 14% (10%, 19%)
Obese class 2 and class 3 295 0% 1.6% (0.6%, 3.9%) 6.7% (4.2%, 10%) 92% (88%, 95%)
Open in a new tab

Underweight, <18.5 kg m−2; normal weight and overweight, 18.5–29.9 kg m−2; obese class 1, 30–34.9 kg m−2; obese class 2 and class 3, ≥35 kg m−2. Agreement proportion is calculated as the probability that women in each birth certificate‐derived BMI category were underweight, normal weight or overweight, obese class 1, and obese class 2 and class 3 by the medical record. Singleton results are weighted to reflect the race‐ethnicity and preterm birth distribution in the twin validation study. Agreement proportions were not significantly different between twins and singletons (P > 0.05).

Birth certificate‐based GWG categories agreed with medical record‐based categories in 80% (65–91%) of twin births with GWG <20th percentile, 82% (72–89%) of twin births with GWG 20th–80th percentile, and 57% (42–70%) of twin births with GWG >80th percentile (Table 3). Agreement proportions for twins were not meaningfully different from the weighted proportions in the singleton data (all P > 0.05). Error in GWG in twin births was driven by errors in pre‐pregnancy weight rather than weight at delivery. Agreement between birth certificate and medical record weights within 5 lb (2.2 kg) occurred more often for delivery weight (74%) than pre‐pregnancy weight (55%).

Table 3.

Agreement of birth certificate‐derived GWG categories with medical record derived categories among the subcohort of twin deliveries in the validation study compared with the weighted average of the agreement proportions in singletons*

Birth certificate n Medical records
GWG <20th percentile GWG 20th–80th percentile GWG >80th percentile
Twins
GWG <20th percentile* 41 80% (65%, 91%) 20% (8.8%, 35%) 0%
GWG 20th–80th percentile 92 13% (6.9%, 22%) 82% (72%, 89%) 5.4% (1.8%, 12%)
GWG >80th percentile 53 1.9% (0.05%, 10%) 42% (28%, 56%) 57% (42%, 70%)
Singletons weighted average
GWG <20th percentile 358 72% (67%, 77%) 25% (21%, 30%) 2.6% (1.2%, 4.7%)
GWG 20th–80th percentile 449 12% (9.1%, 15%) 82% (78%, 85%) 6.3% (4.2%, 8.9%)
GWG >80th percentile 397 2.0% (0.9%, 3.9%) 30% (25%, 35%) 68% (63%, 73%)
Open in a new tab

GWG, gestational weight gain. *Agreement proportion is calculated as the probability that women in each birth certificate‐derived GWG category gained <20th percentile, 20–80th percentile, or >80th percentile by the medical record. GWG percentile cut‐points are based on statewide pre‐pregnancy body mass index‐specific distributions for rate of GWG (twins) or maternal gestational weight gain for gestational age z‐score (singletons) (Hutcheon et al. 2013). Singleton results are weighted to reflect the race‐ethnicity and preterm birth distribution in the twin validation study. Agreement proportions were not significantly different between twins and singletons (P > 0.05).

Discussion

This validation study highlights the significant discrepancy between birth certificate‐based and medical record‐based BMI and GWG in twin pregnancies. The misclassification in calculated BMI and GWG was driven by errors in pre‐pregnancy weight rather than weight at delivery. This result is not surprising given that these mothers with twin gestations recalled pre‐pregnancy weight 30–40 weeks after conception, and inaccuracies in maternal weight reporting after many months have been documented (Schieve et al. 1999).

To our knowledge, this study is the first to validate maternal weight variables on birth certificates of twins. Mothers with twin gestations have higher weight gains during pregnancy, more pregnancy complications, and tend to be older and better educated than women with singleton gestations (IOM 1990; Martin et al. 2012). These factors could lead to important differences in the accuracy of maternal weight data on birth certificates. Nevertheless, our study suggests that the degree of error is similar to what we and others previously observed among singletons (Vinikoor et al. 2010; Park et al. 2011; Wright et al. 2012; Bodnar et al. 2014).

We could not validate pre‐pregnancy weight using a measured weight before conception because it was not available in the medical records, so we used pre‐pregnancy weight recalled in early pregnancy as an alloyed gold standard. This method is unlikely to lead to significant bias in our validation results because there is a strong correlation between measured pre‐pregnancy weight and pre‐pregnancy weight recalled at the first visit (Lederman & Paxton 1998; Phelan et al. 2011; Mandujano et al. 2012). Furthermore, we previously demonstrated in singletons that the agreement proportions for birth certificate‐derived BMI and GWG categories with medical record categories were similar when we used measured weight at ≤8 weeks or pre‐pregnancy weight recalled at the first visit (Bodnar et al. 2014).

We chose a balanced design to permit an assessment of predictive values of birth certificate‐derived BMI and GWG categories in groups that are rare, but are of great interest in public health (e.g. severely obese mothers; mothers with very low and very high GWG). However, even in our large maternity centre, we did not reach our intended sample size. This limitation led to a reduction in precision for some results. Future research with a larger source population of twin pregnancies will be needed to determine with certainty the accuracy of BMI and GWG in rare groups and whether it differs by other key variables, such as gestational age at delivery, maternal race‐ethnicity or education.

Our validation study was restricted to Magee‐Womens Hospital because we did not have the resources to expand data collection across the state. Compared with all twin pregnancies in Pennsylvania, mothers with twin pregnancies at our facility are more likely to have a college degree (52% vs. 41%) and to be normal weight (54% vs. 48%), but there are no important differences in age, race/ethnicity, smoking, parity or other characteristics. The external validity of our results may be limited by the fact that not all hospitals in Pennsylvania share common procedures for collecting weight data for birth certificates and in medical records. Furthermore, the predictive values calculated in our study are sensitive to prevalence of the validated variable and therefore can vary in different populations. Before applying our results to other data sets, analysts should consider how well our validation results might apply given differences in time and place.

Variables to calculate pre‐pregnancy BMI and GWG appeared on birth certificates of 90% of births in 2013 and will increase to 100% in 2015 (Hamilton et al. 2014). Their use for monitoring and surveillance of maternal weight and for aetiologic studies of BMI or GWG in relation to perinatal outcomes will undoubtedly increase. Our data suggest that in both twin and singleton pregnancies in our hospital (Bodnar et al. 2014), maternal BMI and GWG are prone to error. As BMI and GWG are of interest to researchers and public health professionals, those who use birth certificate data should devote study resources to designing and conducting validation studies in their own populations. If resources are not available, external validation results may be considered (Lash et al. 2009). Validation study results can be incorporated into the estimates of interest with bias analyses methods, some of which are semi‐automated for simple application (Fox et al. 2005; Lash et al. 2009, 2014; MacLehose & Gustafson 2012). We recommend that future data validation studies examining the revised birth and fetal death certificates include evaluation of maternal pre‐pregnancy weight and weight at delivery. Furthermore, the Centers for Disease Control and Prevention may wish to advise states to use the self‐reported pre‐pregnancy weight documented in the medical record at the first prenatal in lieu of self‐report after delivery.

Source of funding

This project was supported by the National Institutes of Health, grant R21 HD065807 and the Thrasher Research Fund (#9181).

Conflicts of interest

The authors declare that they have no conflicts of interest.

Contributions

LMB, BA, and TLL designed the study; LMB and LS drafted the manuscript; LMB analyzed the data; BA and TLL critically reviewed the manuscript.

Acknowledgements

We thank Sara Parisi, Sarah Pugh, Sean Rinella and Jennifer Taylor for their assistance with the data management and medical record reviews for this study.

Bodnar, L. M. , Abrams, B. , Siminerio, L. , and Lash, T. L. (2016) Validity of birth certificate‐derived maternal weight data in twin pregnancies. Maternal & Child Nutrition, 12: 632–638. doi: 10.1111/mcn.12160.

References

  1. Bodnar L.M., Siega‐Riz A.M., Simhan H.N., Diesel J.C. & Abrams B. (2010) The impact of exposure misclassification on associations between prepregnancy BMI and adverse pregnancy outcomes. Obesity 18 (11), 2184–2190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bodnar L.M., Abrams B., Bertolet M., Gernand A.D., Parisi S.M., Himes K.P. et al (2014) Validity of birth certificate‐derived maternal weight data. Paediatric and Perinatal Epidemiology 28 (3), 203–212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Centers for Disease Control and Prevention (CDC) National Center for Health Statistics (2003) Birth edit specifications for the 2003 proposed revision of the U.S. standard certificate of birth. Available at: http://www.cdc.gov/nchs/data/dvs/birth_edit_specifications.pdf
  4. Fox M.P., Lash T.L. & Greenland S. (2005) A method to automate probabilistic sensitivity analyses of misclassified binary variables. International Journal of Epidemiology 34 (6), 1370–1376. [DOI] [PubMed] [Google Scholar]
  5. Hamilton B.E., Martin J.A., Osterman M. & Curtin C.S. (2014) Births: preliminary data for 2013. National Vital Statistics Reports 63 (2), 1–20. [PubMed] [Google Scholar]
  6. Holcroft C.A. & Spiegelman D. (1999) Design of validation studies for estimating the odds ratio of exposure–disease relationships when exposure is misclassified. Biometrics 55 (4), 1193–1201. [DOI] [PubMed] [Google Scholar]
  7. Hutcheon J.A., Platt R.W., Abrams B., Himes K.P., Simhan H.N. & Bodnar L.M. (2013) A weight‐gain‐for‐gestational‐age z score chart for the assessment of maternal weight gain in pregnancy. American Journal of Clinical Nutrition 97 (5), 1062–1067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Institute of Medicine (IOM) (1990) Nutrition during Pregnancy, National Academy Press: Washington, DC. [Google Scholar]
  9. Institute of Medicine (IOM) (2009) Weight Gain During Pregnancy: Reexamining the Guidelines, National Academies Press: Washington, DC. [PubMed] [Google Scholar]
  10. Lash T.L., Fox M.P. & Fink A.K. (2009) Applying Quantitative Bias Analysis to Epidemiologic Data, Springer: New York. [Google Scholar]
  11. Lash T.L., Abrams B. & Bodnar L.M. (2014) Comparison of bias analysis strategies applied to a large data set. Epidemiology (Cambridge, Mass.) 25 (4), 576–582. [Epub 13 May]. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Lederman S.A. & Paxton A. (1998) Maternal reporting of prepregnancy weight and birth outcome: consistency and completeness compared with the clinical record. Maternal Child Health Journal 2 (2), 123–126. [DOI] [PubMed] [Google Scholar]
  13. MacLehose R.F. & Gustafson P. (2012) Is probabilistic bias analysis approximately Bayesian? Epidemiology (Cambridge, Mass.) 23 (1), 151–158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Mandujano A., Huston‐Presley L., Waters T.P. & Catalano P.M. (2012) Women's reported weight: is there a discrepancy? Journal of Maternal‐Fetal & Neonatal Medicine 25 (8), 1395–1398. [DOI] [PubMed] [Google Scholar]
  15. Martin J.A., Hamilton B.E., Ventura S.J., Osterman M.J.K., Kirmeyer S., Mathews T.J. et al (2011) Births: Final Data for 2009, National Center for Health Statistics: Hyattsville, MD. [PubMed] [Google Scholar]
  16. Martin J.A., Hamilton B.E. & Osterman M.J.K. (2012) Three Decades of Twin Births in the United States, 1980–2009, National Center for Health Statistics: Hyattsville, MD. [PubMed] [Google Scholar]
  17. Park S., Sappenfield W.M., Bish C., Bensyl D.M., Goodman D. & Menges J. (2011) Reliability and validity of birth certificate prepregnancy weight and height among women enrolled in prenatal WIC program: Florida, 2005. Maternal Child Health Journal 15 (7), 851–859. [DOI] [PubMed] [Google Scholar]
  18. Phelan S., Phipps M.G., Abrams B., Darroch F., Schaffner A. & Wing R.R. (2011) Randomized trial of a behavioral intervention to prevent excessive gestational weight gain: the Fit for Delivery Study. American Journal of Clinical Nutrition 93 (4), 772–779. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Salihu H.M., Lynch O., Alio A.P. & Liu J. (2008) Obesity subtypes and risk of spontaneous versus medically indicated preterm births in singletons and twins. American Journal of Epidemiology 168 (1), 13–20. [DOI] [PubMed] [Google Scholar]
  20. Salihu H.M., Alio A.P., Belogolovkin V., Aliyu M.H., Wilson R.E., Reddy U.M. et al (2010) Prepregnancy obesity and risk of stillbirth in viable twin gestations. Obesity 18 (9), 1795–1800. [DOI] [PubMed] [Google Scholar]
  21. Sauber‐Schatz E.K., Sappenfield W., Grigorescu V., Kulkarni A., Zhang Y., Salihu H.M. et al (2012) Obesity, assisted reproductive technology, and early preterm birth – Florida, 2004–2006. American Journal of Epidemiology 176 (10), 886–896. [DOI] [PubMed] [Google Scholar]
  22. Schieve L.A., Perry G.S., Cogswell M.E., Scanion K.S., Rosenberg D., Carmichael S. et al (1999) Validity of self‐reported pregnancy delivery weight: an analysis of the 1988 National Maternal and Infant Health Survey. NMIHS Collaborative Working Group. American Journal of Epidemiology 150 (9), 947–956. [DOI] [PubMed] [Google Scholar]
  23. Vinikoor L.C., Messer L.C., Laraia B.A. & Kaufman J.S. (2010) Reliability of variables on the North Carolina birth certificate: a comparison with directly queried values from a cohort study. Paediatric and Perinatal Epidemiology 24 (1), 102–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Wright C.S., Weiner M., Localio R., Song L., Chen P. & Rubin D. (2012) Misreport of gestational weight gain (GWG) in birth certificate data. Maternal & Child Health Journal 16 (1), 197–202. [DOI] [PubMed] [Google Scholar]

Articles from Maternal & Child Nutrition are provided here courtesy of Wiley

RESOURCES