Abstract
Background/Objectives: Prior studies using administrative data have found that antenatal anemia is a risk factor for severe maternal morbidity (SMM). However, administrative definitions, including the commonly used definition from the Centers for Disease Control and Prevention (CDC), have a poor positive predictive value for some SMM components. We tested the relationship between hemoglobin level at delivery admission and SMM, as defined by gold-standard chart review. Methods: This was a retrospective case-control study of deliveries at a high-acuity hospital in Los Angeles, California, from 2016 to 2019. Administrative data were screened to identify patients with CDC SMM. Control-patients were selected at random from screen-negative individuals. Medical records for all individuals were reviewed for gold-standard SMM criteria, and clinical data were abstracted. Confirmed-positive and confirmed-negative patients were compared using bivariate analyses. Multiple logistic regression models were developed to test the relationship between admission hemoglobin level and gold-standard SMM. Results: Of 4202 eligible individuals, 275 (6.5%) screened positive for SMM. Of these, 107 (38.5%) met gold-standard SMM criteria; 285 confirmed-negative controls were retained for analysis. Case-patients were more likely than control-patients to have anemia on delivery admission (43.9% vs. 24.2%, p < 0.01) and had lower admission hemoglobin levels (11.2 ± 1.7 g/dL vs. 11.9 ± 1.3 g/dL, p < 0.01). After controlling for covariates, admission hemoglobin was independently and inversely associated with gold-standard SMM (aOR = 0.76, 95% CI 0.60-0.96, p = 0.02). Conclusions: Lower hemoglobin level at delivery admission was associated with an increased risk of developing gold-standard SMM.