Pregnancy and Lactation

1F-02 - Prevalence of Major Congenital Malformations Among the Insured Population in the United States Using Seven Algorithms

Sunday, August 24, 2025

10:45 AM - 11:00 AM ET

Presenting Author(s)

SA

Sruthi Adimadhyam, PhD, MS

Senior Research Scientist
Harvard Medical School & Harvard Pilgrim Health Care Institute, United States

Co-Author(s)

TT

Thuy N. Thai, PhD

Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, Massachusetts, USA
Boston, Massachusetts, United States

Co-Author(s)

CC

Catherine Callahan, PhD, MA (she/her/hers)

Senior Epidemiologist
Office of Surveillance and Epidemiology, Center for Drug Evaluation and Research, United States Food and Drug Administration, United States
LS

Leyla Sahin, MD

Deputy Director for Safety, Division of Pediatrics and Maternal Health
FDA, United States
MD

Miriam Dinatale

Office of New Drugs, Center for Drug Evaluation and Research, United States Food and Drug Administration, United States
Jummai Apata, MBBS, DrPH

Epidemiologist
Office of Surveillance and Epidemiology, Center for Drug Evaluation and Research, U. S. Food and Drug Administration, Maryland, United States

Co-Author(s)

MW

Megan Wiley

Harvard Pilgrim Health Care Institute, United States
AS

Andrew L. Simon, ScM

Harvard Pilgrim Health Care Institute, United States
JL

Jennifer G. Lyons, PhD

Harvard Medical School and Harvard Pilgrim Health Care Institute, United States

Co-Author(s)

WH

Wei Hua

United States Food and Drug Administration, United States

Co-Author(s)

JM

Judith C. Maro

Associate Professor
Harvard Medical School & Harvard Pilgrim Health Care Institute, United States

Background: Major congenital malformations (MCM) are important drug safety outcomes. The definition of MCM can be heterogenous owing to differences in source data and interpretation of “major”, which may limit comparability between post-marketing safety studies.

Objectives: To compare and identify key drivers in estimates of MCM prevalence derived using seven published or established algorithms.

Methods: We conducted a review of the literature to identify various algorithms used to define MCM, prioritizing those compatible with U.S. data in the International Classification of Disease, Tenth Revision (ICD-10) era: three published algorithms (Kharbanda et al., Berard et al., Chomistek et al.), two algorithms used by the National Birth Defects Prevention Network to capture malformations of public health importance using the Metropolitan Atlanta Congenital Defects Program (MACDP) and ICD-10 vocabularies (MACDP-29 and MACDP-47), and two others that were developed in the Covid-19 infectiOn aNd medicineS In preGnaNcy (CONSIGN) and Big data apprOaches fOr Safe Therapeutic in Healthy Pregnancy (BOOST-HP) projects. We estimated MCM prevalence in a cohort of mother-infant linked live births using the Merative™ MarketScan® Research Databases from 2015-2024. The codes used in the seven algorithms were implemented by four approaches: ≥ 1 or ≥ 2 diagnosis code(s) in either maternal or infant claims within 90 or 180 days since birth. Our primary analysis required the presence of ≥ 2 codes within 180 days since birth.

Results: We included 719,281 linked live births. MCM prevalence observed in our primary analysis was 0.9% by MACDP-29, 1.9% by Kharbanda, 3.2% by MACDP-47, 3.8% by BOOST-HP, 3.9% by Berard, 16.4% by CONSIGN, and 17.0% by Chomistek. Inclusion of malformations such as ankyloglossia and plagiocephaly were drivers of the five-fold increase in prevalence with the CONSIGN and Chomistek algorithms. Sensitivity analyses using other implementation approaches resulted in a similar pattern across 7 algorithms with varying magnitudes. Prevalence was lower when requiring ≥2 codes during 90 days (range: 0.8%-13.4%) and higher when requiring ≥1 code (range: 1.4%-29.6% and 1.5%-33.6% within 90 and 180 days, respectively).

Conclusions: 3 of 7 algorithms (MACDP-47, BOOST-HP and Berard) yielded an estimate of MCM prevalence comparable to that reported by the U.S. Centers for Disease Control and Prevention (3-4%). Variation in estimates of MCM prevalence across algorithms was driven by the inclusion of structural malformations sometimes deemed minor. With an existing crosswalk between the vocabularies commonly used in registries (MACDP) and administrative databases (ICD), the use of MACDP-47 algorithm can be one way to harmonize MCM definitions in post-marketing safety studies.