An enhanced method for identifying obstetric deliveries: implications for estimating maternal morbidity

Area of Science:

• Maternal Health
• Healthcare Data Analysis
• Epidemiology

Background:

• Maternal morbidity estimates from hospital discharge data can be inaccurate due to incomplete delivery identification.
• The standard method relies solely on the International Classification of Diseases (ICD) V27 code, potentially missing many deliveries.

Purpose of the Study:

• To develop and evaluate an enhanced method for identifying obstetric deliveries using additional delivery-related codes.
• To compare the enhanced method with the standard V27 code in estimating deliveries and maternal morbidity.

Main Methods:

• Applied both enhanced and standard V27 delivery identification methods to the 1998-2004 Healthcare Cost and Utilization Project Nationwide Inpatient Sample.
• Utilized logistic regression to analyze predictors of deliveries missed by the V27 method.

Main Results:

• The enhanced method identified 958,868 (3.4%) more deliveries than the V27 code alone.
• Deliveries missed by the V27 code were strongly associated with severe maternal complications, including infections, hysterectomy, sepsis, and respiratory distress syndrome.
• The V27 method underestimated the prevalence of severe maternal complications, with underestimation rates ranging from 9% to 40%.

Conclusions:

• The standard V27 code is likely to miss deliveries with severe obstetric complications.
• Relying solely on V27 codes for delivery identification in hospital records can compromise the accuracy of maternal morbidity estimates.
• Researchers should consider using enhanced methods for more accurate identification of deliveries and associated maternal health outcomes.