To Determine If Maternal Adverse Outcomes Predicted by Obstetric Comorbidity Index (OBCMI) Varies According to Race [32C]. (May 2020)

• Record Type: Journal Article
• Authors: Ranjit, Anju; Olufajo, Olubode; Zogg, Cheryl; Robinson, Julian N.; Luo, Guoyang
• Is Part Of: Obstetrics and gynecology. Volume 135(2020)Supplement 1
• Publication Date: 2020-05
• Subjects: Obstetrics -- Periodicals; Gynecology -- Periodicals; 618
• DOI: 10.1097/01.AOG.0000663520.38197.38
• Languages: English
• ISSNs: 0029-7844

Abstract

Abstract : INTRODUCTION: To use nationwide data to determine if including race/ethnicity to the Obstetric Co-morbidity Index (OBCMI) improves its ability to predict severe maternal morbidity/mortality. METHODS: The National Inpatient Database (2010–2014) was used to identify all admissions for deliveries. Patient demographics including race/ethnicity (white, black, Hispanic) and co-morbidities were extracted. Primary outcome was defined as presence of severe maternal morbidity/mortality using ICD-9 codes. Using a stepwise logistic regression model, as previously described, maternal comorbid conditions predicting the primary outcome were determined. With separate models constructed for each race/ethnicity, weights assigned to each condition were used to determine the race/ethnicity-specific OBCMI. For each group, models predicting the primary outcome by their race specific OBCMI and other races' OBCMI was compared using c-statistic. RESULTS: A total of 3, 830, 958 deliveries were identified, among which 1, 867, 534 (48.7%) were white, 513, 412 (13.4%) were to black, and 759, 906 (19.8%) were Hispanic mothers. Severe maternal morbidity/mortality was present in 79, 310 (2.07%) women. For each point increase in OBCMI derived for white women, the odds ratios for the outcome were 1.30 (CI 1.29–1.30) for white, 1.31 (CI 1.30–1.31) for black and 1.34 (CI 1.33–1.35) for Hispanic women. The c-statistic for these models were 0.652, 0.651 and 0.650 for white, black and Hispanic women, respectively. The odds ratios and c-statistics for models using black and Hispanic specific OBCMI were similar when compared between races/ethnicities. CONCLUSION: The obstetric comorbidity index provides a simple predictive tool to determine likelihood of severe maternal morbidity at the time of delivery, and race/ethnicity does not seem to play a large role in its predictive ability.