Validation of algorithms to estimate gestational age at birth in the Medicaid Analytic eXtract—Quantifying the misclassification of maternal drug exposure during pregnancy

Abstract

Purpose: Accurate ascertainment of gestational age (GA) has been a challenge in perinatal epidemiologic research. To date, no study has validated GA algorithms in Medicaid Analytic eXtract (MAX). Methods: We linked livebirths of mothers enrolled in Medicaid ≥30 days after delivery in 1999-2010 MAX to state birth certificates. We used clinical/obstetric estimate of gestation on the birth certificates as gold standard to validate claims-based GA algorithms. We calculated the proportions of deliveries with algorithm-estimated GA within 1−/2-weeks of the gold standard, the sensitivity, specificity, and positive/negative predictive value (PPV/NPV) of exposure to select medications during specific gestation windows, and quantified the impact of exposure misclassification on hypothetical relative risk (RR) estimates. Results: We linked 1 336 495 eligible deliveries. Within 1-week agreement was 77%-80% overall and 47%-56% for preterm deliveries. The trimester-specific drug exposure status had high sensitivities and PPVs (88.5%-98.5%), and specificities and NPVs (>99.0%). Assuming a hypothetical RR of 2.0, bias associated with exposure misclassification during first trimester ranged from 10% to 40% under non-differential/differential misclassification assumptions. Conclusions: Claims-based GA algorithms had good agreement with the gold standard overall, but lower agreement among preterm deliveries, potentially resulting in biased risk estimated for pregnancy exposure evaluations.

A Pharmacoepidemiologic Approach to Evaluate Real-world Effectiveness of Hormonal Contraceptives in the Presence of Drug–drug Interactions

“…Based on previous literature, estimation of conception will have varying degrees of accuracy for each type of pregnancy endpoint. 6 , 22 In our pregnancy identification algorithm, we assigned gestational age for live-birth episodes based on an algorithm of ICD codes indicating gestational age or preterm status, and the overall agreement against birth certificates is reported to be >93% in the Medicaid database. 22 For ectopic pregnancy, induced/spontaneous abortion, and stillbirth, we used fixed values for gestational age (8, 10, and 28 weeks, respectively) similar to previous literature.…”

“… 6 , 22 In our pregnancy identification algorithm, we assigned gestational age for live-birth episodes based on an algorithm of ICD codes indicating gestational age or preterm status, and the overall agreement against birth certificates is reported to be >93% in the Medicaid database. 22 For ectopic pregnancy, induced/spontaneous abortion, and stillbirth, we used fixed values for gestational age (8, 10, and 28 weeks, respectively) similar to previous literature. 6 , 7 This approach has shown a moderate agreement both against medical charts (70% for ectopic pregnancy and 67% for spontaneous abortions within four weeks) 7 , and a cohort of in-vitro fertilization patients (77% agreement for ectopic pregnancy, 47% for stillbirth, and 36% for abortions cases).…”

“… 6 Validation studies on pregnancy identification algorithms typically report the agreement between the estimated LMP and the gold standard (e.g., the clinical estimate of gestational age on the birth certificate) within a prespecified number of weeks (e.g., 2 weeks) as the margin of error. 4 , 6 , 22 It should be noted; however, that such a margin might need to be varied according to the pregnancy endpoint that was used to estimate LMP. For example, previous studies have shown that the capture of preterm status among deliveries improves with decreasing gestational age.…”

“…Conception was estimated via the pregnancy identification algorithm that uses medical encounters with ICD-9/10-CM, Current Procedural Terminology, and Healthcare Common Procedure Coding System codes to identify specific pregnancy endpoints, including live birth, ectopic pregnancies, stillbirth, terminations, and prenatal screening visits. 6 , 7 , 21 , 22 Once pregnancy episodes were identified, the algorithm estimated gestational age to calculate the last menstrual period (LMP). The conception date was assumed to be 14 days after the estimated LMP date.…”

Real-World Fetal Exposure to Acne Treatments in the United States: A Retrospective Analysis from 2006 to 2015

“…The study outcome was conception during the exposure period. To estimate conception date, we used an algorithm based on previous validation studies that identified pregnancy episodes among WCBP [13][14][15][16].…”