ABSTRACT
Birth certificates are potentially a valuable source of information for studying gestational weight gain (GWG) during pregnancy, particularly important given new Institute of Medicine (IOM) guidelines. We examined factors associated with the accuracy of maternal GWG self-report by linking the gold standard obstetric electronic medical record (EMR) of women from a large urban practice to state birth certificates. Primary outcomes included maternal under-reporting of GWG (>10 lbs below the EMR), accurate reporting (within 10 lbs), and over-reporting (>10 lbs above EMR). Data were stratified across categories of pre-pregnancy body mass index (BMI) and the actual GWG IOM categories (inadequate, adequate, and excessive) acquired in the clinical setting and recorded in the EMR. Among 1,223 women, mean (SD) age was 27.4 (6.2) years, mean (SD) BMI was 28.2 (8.1) kg/m(2), and mean GWG was 26.0 (20.2) pounds. The majority of women with normal BMI (<25 kg/m(2)) and adequate GWG reported GWG accurately (78.8%), more so than any other group. After adjusting for age, race, insurance status, and number of prenatal visits, among women with actual adequate GWG, women with high BMI (≥25 kg/m(2)) were more likely to over-report GWG than women with normal BMI (RR 4.7, 95% CI 2.6-8.4). In patients with normal BMI, women with excessive GWG were more likely to under-report than women with adequate GWG (RR 6.0, 95% CI 3.0-12.1). Such findings raise concern for systematic bias that would limit the use of birth certificate data for studying population trends in GWG.
