![[ Visit Client Website ]](images/banner.jpg)
BACKGROUND: Although birth certificate data are widely used for epidemiologic studies, the validity of individual elements is variable. Massachusetts adopted the 2003 revision of the U.S. Standard Certificate of Live Birth in 2011, but validation of birth defects reporting had not been undertaken. We compared reporting of birth defects on the Massachusetts live birth certificate (MA-BC), with data from the Massachusetts Birth Defects Registry (MBDR), using the MBDR as our gold standard.
METHODS: We compared 2011 MA birth data from revised MA-BC records (N=61,698) with MBDR surveillance records (N=1,070), which uses active case ascertainment and medical record review through the first year of life to confirm diagnosis. We examined 17 specific defects identifiable at birth: anencephaly, hydrocephaly, spina bifida, cleft lip with or without cleft palate, cleft palate without cleft lip, diaphragmatic hernia, limb reductions, rectal atresia, esophageal atresia/tracheoesophageal fistula, omphalocele/gastroschisis, renal agenesis, adactyly, polydactyly/syndactyly, club foot, trisomy 21, and hypospadias (among males). We analyzed data for birth prevalence estimates, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV).
RESULTS: Prevalence of any of these defects was 80/10,000 live births by MA-BC and 87/10,000 live births by MBDR surveillance. Among 61,698 live births, 493 infants had one or more of the birth defects indicated on MA-BC, with hypospadias (n=125) and polydactyly/syndactyly (n=124) the most common ones. Among the 539 infants in the MBDR with one or more of these 17 defects, the most common were polydactyly/syndactyly (n=103) and hypospadias (n=94). Sensitivity of the MA-BC was 48.1% for identifying any defect, ranging from 75.0% for anencephaly to 20.0% for limb reduction. Specificity of MA-BC report of defects ranged from 99.5%-100.0% for all 17 defects. While NPV was >99.8% for each of the defects, PPV ranged from 100% for spina bifida and 92.9% for esophageal atresia/tracheoesophageal fistula to 25.0% for hydrocephaly and renal agenesis. The PPV was 25.6% for hypospadias, a common defect, and 27.4% for polydactyly/syndactyly.
CONCLUSIONS: MA-BC showed moderate sensitivity in identifying cases of anencephaly, but poor sensitivity for other defects considered readily identifiable at birth. We will next examine the timing of diagnosis to compare accuracy of MA-BC reporting among MBDR cases identified prenatally or during the birth hospitalization according to medical records, to enhance improvement of MA-BC reporting. Although birth certificate data are widely used for research, caution is warranted when using them as the primary source for birth defects surveillance.