Related Experiment Video
Updated: May 15, 2026
07:31
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Validation of an algorithm to estimate gestational age in electronic health plan databases
Qian Li1, Susan E Andrade, William O Cooper
1Department of Epidemiology, Harvard School of Public Health, Boston, MA, USA.
Pharmacoepidemiology and Drug Safety
|January 22, 2013
Summary
This study validates an algorithm using electronic health data to determine gestational age at birth. The algorithm accurately identifies medication exposure during pregnancy, crucial for reproductive health research.
Area of Science:
- Reproductive Health
- Health Informatics
- Pharmacovigilance
Background:
- Electronic health records (EHRs) are valuable for research.
- Accurate gestational age (GA) is critical for pregnancy studies.
- Validating EHR algorithms is essential for reliable data.
Purpose of the Study:
- To validate an algorithm using delivery date and diagnosis codes to determine gestational age at birth in EHR databases.
- To assess the algorithm's accuracy in classifying prenatal medication exposure.
Main Methods:
- Utilized data from 225,384 live births (2001-2007) across eight health plans.
- Compared algorithm-derived GA with gold-standard GA from birth certificates.
- Evaluated prenatal exposure classification for antidepressants and antibiotics.
Main Results:
- Algorithm-derived GA was slightly lower for singleton births but similar for multiple-gestation births compared to gold-standard.
- High sensitivity (≥95%) and specificity (>99%) for classifying antidepressant and antibiotic exposure.
- Positive predictive values were ≥90% and negative predictive values >99% for antibiotics.
Conclusions:
- The EHR-based gestational age algorithm accurately classifies medication exposure in most deliveries.
- Potential for trimester-specific misclassification exists for short-duration medications.
- Algorithm shows promise for large-scale pregnancy research using EHR data.