The rate of maternal morbidity in the United States has risen precipitously over the last few decades. Patients of color are at increased risk of morbidity compared with White patients, but reasons for these disparities have not been sufficiently explained. Although recognition of the impact of structural racism and healthcare system factors on maternal outcomes is growing, most research continues to focus on individual characteristics such as education, income, and comorbidities rather than other possible factors such as provider bias or healthcare system-level inequities. Some studies suggest that clinical comorbidities that are more common in patients of color, such as obesity, may help explain these disparities, but associations between race/ethnicity and adverse outcomes often remain significant after adjustment for such factors.
A recent retrospective cohort study was conducted to explore the extent to which pre-pregnancy obesity explained the association between race and ethnicity and severe maternal morbidity in 409,021 patients with 591,455 live births in New York City. The primary composite outcome of severe maternal morbidity was defined as condition-specific ICD-9-CM billing codes and hospital length of stay, including any blood transfusion.
Severe maternal morbidity occurred in 15,158 live births (2.5%). Compared with patients with normal weight, increased risk of severe maternal morbidity was found in patients with overweight and obesity. Similarly, increased risk of severe maternal morbidity was seen in Black, Latina, and Asian patients as well as those of unspecified race/ethnicity compared with White patients. Having established binary relationships between severe maternal morbidity and both pre-pregnancy obesity and race/ethnicity, the authors then analyzed the extent to which the relationship between race/ethnicity and severe maternal morbidity was mediated (or explained) by pre-pregnancy obesity. This mediation analysis (the details of which are outside the scope of this summary) demonstrated that pre-pregnancy obesity explained only a very small portion of the relationship between race/ethnicity and severe maternal morbidity in Black patients (3%, or aOR 1.03, 95% CI 1.02-1.05), Latina patients (2%, or aOR 1.02, 95% CI 1.01-1.03), and patients of unspecified race/ethnicity (1%, or aOR 1.01, 95% CI 1.01-1.02). In contrast, pre-pregnancy overweight did not mediate the relationship between race/ethnicity and severe maternal morbidity for any patient group, and pre-pregnancy obesity did not mediate the relationship between race and severe maternal morbidity in Asian patients.
Mediation analysis is a complicated but useful way to determine to what extent two variables are associated by way of a third. In this study, the association found between race/ethnicity and severe maternal morbidity was only very slightly explained by pre-pregnancy obesity. Although pre-pregnancy obesity was somewhat more explanatory of severe maternal morbidity other than blood transfusion, these results still suggest that interventions aiming to reduce racial and ethnic disparities in adverse maternal outcomes by targeting maternal obesity may only be met with minor success.
For more information, see the topics Racial and Ethnic Disparities in Obstetric and Gynecologic Care and Obesity in Pregnancy in DynaMed.
