[Skip to Content]
[Skip to Content Landing]
Views 191
Citations 0
Comment & Response
April 2014

Methods to Estimate Gestational Age Can Significantly Affect Study Results

Author Affiliations
  • 1MedImmune, Gaithersburg, Maryland
JAMA Pediatr. 2014;168(4):388-390. doi:10.1001/jamapediatrics.2013.5175

To the Editor In their analysis of respiratory syncytial virus (RSV) hospitalization risk among preterm infants born at 32 to 34 weeks’ gestational age (GA) with siblings relative to full-term infants with siblings in Florida and Texas, Winterstein et al1 estimated infants’ GA by calculating weeks between the last menstrual period (LMP) date on the birth certificate and the date of birth. Although this method is still used to estimate GA in US natality statistics, it is not widely appreciated that the “clinical estimate” of GA was added to birth certificates in 1989, which was revised to an “obstetric estimate” in 2003.2 The clinical/obstetric estimate is based on multiple prenatal factors such as early prenatal ultrasonography and not solely the LMP.2 Clinical/obstetric GA better reflects the actual GA assigned in clinical care and provides improved accuracy relative to LMP-calculated GA, as evidenced by higher correlations of preterm GA with low birthweight3 and neonatal intensive care unit admission4 (even after adjusting LMP-calculated GA estimates incompatible with birthweight). Hediger and Kiely5 of the National Institutes of Health described the “extensive misclassification of gestational ages at ≤35 weeks based on menstrual dates,” and Qin et al3 of the Centers for Disease Control and Prevention concluded, “The clinical estimate measure itself, if used appropriately, is the best obstetrical estimate of gestational age.” According to the 1999-2004 US natality public use files, 31.8% and 38.6% of infants with siblings and an LMP-calculated GA of 32 to 34 weeks in Florida and Texas, respectively, were born at 37 to 41 weeks of gestation based on their clinical/obstetric GA. Thus, it appears likely that 30% to 40% of the infants whom Winterstein et al1 identified as 32 to 34 weeks’ GA may have been term infants based on the clinical/obstetric GA recorded on their birth certificates. If these infants experienced an RSV hospitalization risk similar to other term infants, the results from Winterstein et al1 would significantly underestimate the RSV risk among infants with a clinical/obstetric GA of 32 to 34 weeks. Could Winterstein et al provide their results for infants born at 32 to 34 weeks’ GA based on clinical/obstetric GA?