Examining Individual and Community Level Risk Factors for Severe Maternal Morbidity in Arizona's Native American Communities

Abstract

Over the past decades, national rates of maternal mortality (MM) and severe maternal morbidity (SMM) have shown a concerning upward trend, prompting a call for concerted efforts at the national, state, and local levels to enhance health outcomes for women and birthing people. SMM encompasses unexpected labor and delivery outcomes and carries lasting implications for birthing people and their families, including extended hospital stays, adverse outcomes for the fetus or infant, and sometimes death. Significant racial and ethnic health inequities exist in maternal health, with Native Americans experiencing disproportionately higher risks of pregnancy complications and SMM compared to their White counterparts. Arizona published a report in 2020 on maternal fatalities and morbidities. The report highlighted stark disparities across various demographic characteristics, including race, with Native Americans experiencing SMM 3.6 times more than Whites despite only representing 6% of the population. Public health interventions that target modifiable risk factors (RFs) can reduce SMM incidents and save lives. The extent to which these RFs contribute to the overall burden of SMM is not well known. Quantifying this burden will help identify priority areas for SMM prevention efforts in the state’s Native American communities. Aim 1 of this project comprehensively synthesizes published literature on individual and community RFs for maternal morbidity and mortality (MMM) among Native Americans in the United States. We systematically searched four databases for articles published between 2012 and 2022 using database-specific controlled vocabulary for MMM and Native American. Fifteen studies were included in the review. Identified RFs include, rural residency, overweight/obese BMI, advanced maternal age, nulliparity, and preexisting medical conditions. Aim 2 assesses the incidence of SMM across racial/ethnic groups in Arizona by conducting a pooled, cross-sectional analysis of 297,036 merged hospital and birth records from 2016 to 2019. We used binomial proportions for prevalence estimates and multivariable logistic regression with generalized estimating equations to calculate odds ratios and associated 95% confidence intervals (CIs), revealing higher odds among non-White birthing people, with Native Americans exhibiting the highest odds of SMM. Lastly aim 3 estimates the population-attributable risk of modifiable RFs for SMM among Native American hospital deliveries in the state. For this analysis, we used a retrospective cohort of 11,518 hospital deliveries between 2016 and 2019. We used multivariable Poisson regression with generalized estimating equations to estimate the relative risk and associated 95% CIs between RFs and SMM. We then calculated the PAR with associated 95% CIs and the number of preventable SMM cases for each RF. We discovered that non-vaginal/spontaneous deliveries, rural residency, gestational hypertension, and having an age 30 or greater had elevated adjusted risks and yielded the largest amounts of preventable SMM cases. The findings highlight the multifaceted nature of RFs, spanning individual characteristics to community-level influences. Recognizing this complexity is crucial for informing effective public health policies and interventions tailored to Native American communities' nuanced maternal health landscape. This dissertation highlights the importance of building trust and engaging these communities in selecting and adapting public health interventions (i.e., surveillance, monitoring, system adaptations, evidence-based programming) to improve maternal health outcomes in the state.