Evaluation of gestational age estimate method on the calculation of preterm birth rates

Abstract

The objectives of this study is to evaluate the impact of vital record gestational age estimation method on resulting preterm birth (PTB) rate calculations. This retrospective analysis reviewed three methods of gestational age estimation using all Ohio live birth records from 2006 to 2009. PTB rates were calculated using each gestational age representation and agreement between classifications of PTB was evaluated with respect to maternal age and race. For each of 608,530 births, gestational age estimates based on last menstrual period (LMP) were compared to clinically-based obstetric estimates. When gestational age estimates did not perfectly agree, differences in the consequential classification of PTB status were evaluated with respect to a third reconciliatory combined gestational age estimate. Mean birth weight at each week of gestation was calculated and compared for all three estimate methods. Substantial agreement was found in PTB classification among gestational age estimates (kappa: 0.748; 95% Confidence Interval: 0.745-0.750); agreement was weakest among black mothers and among mothers less than 20 years of age. LMP-based gestational age estimates did not perfectly agree with obstetric estimates in 238,262 records (39.2%). Disagreement in gestational age led to disagreement in PTB status in 32,033 records (5.3% of total cases) resulting in a 1.8 percentage point difference in PTB rate calculations (11.0% using obstetric and 12.8% using combined estimates). Researchers and policy makers need consistency in selecting which gestational age estimate method to use when calculating or comparing PTB rates.