Medical Record Validation of Maternal Report of Prenatal Medical Conditions and Obstetric Interventions

Background:  Pregnancy and perinatal complications have been implicated as risk factors in the etiology of autism spectrum disorders (ASD).   Since acquisition and abstraction of data from medical records is both expensive and labor-intensive, most retrospective studies rely on maternal self-report of such events.

Objectives:  We compared the accuracy of maternal report of prenatal conditions (pre-pregnancy obesity, diabetes and hypertensive disorders) and obstetric interventions (induction of labor and mode of delivery) reported in a structured telephone interview with medical records in a population of women of 2 to 5 year old children with and without a neurodevelopmental disorder.

Methods: A subset of participants from the CHARGE (CHildhood Autism Risks from Genetics and the Environment) Study with a confirmed diagnosis of ASD, developmental delays (DD) without ASD, or typical development (TD) and with both prenatal/delivery medical records and telephone interviews was included in the validation study.  Kappa statistics measured agreement between self-report and medical records, while sensitivity and specificity indicated the validity of self-reported data. Bland-Altman plots depicted deviations in self-reported pre-pregnancy weight and body mass index compared with medical records.

Results:  In general, mothers of affected children reported pregnancy and perinatal complications more accurately than control mothers.  For diabetes, sensitivity ranged from 75% to 88% across diagnostic groups; specificity was 98% in all groups. Kappas ranged from 0.75 in controls to 0.92 in the DD group.  Only 4% incorrectly reported whether or not they had preeclampsia, yielding a sensitivity of 65%, a specificity of 98% and a kappa of 0.66.  For hypertension, sensitivity ranged from 33% in controls to 100% in the DD group; kappas ranged from 0.21 in controls to 0.42 in DD cases. Measurement error was small for weight and BMI and did not appear to vary by case status. Misclassification was differential by group when women were classified into obese and non-obese categories. Sensitivity ranged from 59% in controls to 86% in the DD group; kappas ranged from 0.68 in controls to 0.86 in DD cases. Low education and multiparity were associated with greater misclassification of hypertension; education level also influenced diabetes reporting. Self-reported diabetes and measurements to derive BMI appeared to have good validity. Misclassification increased when participants were classified into broad BMI categories. Self-reported hypertension had low validity and agreement, but this condition was also rare in this population.  Mothers correctly distinguished vaginal deliveries from cesarean deliveries 100% of the time.  Self-report of types of vaginal deliveries were less reliable, with 93% of women incorrectly reporting an operative vaginal delivery as a normal spontaneous delivery.

Conclusions: Studies attempting to link gestational conditions and obstetric interventions to the subsequent health and development of the offspring require careful attention to exposure assessment.   Maternal recall of the medical diagnoses and procedures included in this study was excellent, but varied by the condition under study, the health of the child, and maternal characteristics including education level and parity.  This potential for exposure misclassification must be taken into consideration when designing and evaluating studies using self-report of pregnancy events.