Variability in Social Determinants of Health As Correlates of Prenatal Tobacco Use Among Native American Women

Background and purpose: While increasing attention has been paid to the social and structural contexts influencing women’s substance use, no known studies have used a person-centered analytic approach to identify previously unrecognized yet meaningful patterns of social determinants of health (SDOH) associated with prenatal substance use among Native American (NA) women. This study identified patterns across SDOH domains among NA pregnant women and examined whether these patterns are differentially associated with prenatal tobacco use.

Methods: Participants were 700 self-identified NA women from the North Dakota Pregnancy Risk Assessment Monitoring System (ND PRAMS) Phase 8 (2017–2020). PRAMS is a population-based surveillance system that collects data on maternal attitudes and experiences during the perinatal period, in collaboration with the Centers for Disease Control and Prevention (CDC) and state and tribal health departments. ND PRAMS oversamples NA women and includes data on social determinants of health (SDOH), including annual household income, maternal years of education, and adverse childhood experiences (ACEs), measured using 10 items from the CDC-Kaiser ACE Study (range: 0–10). Additional SDOH indicators included prenatal health insurance status (insured = 1), prenatal intimate partner violence (IPV exposed = 1), and timing of prenatal care initiation (first-trimester visit = 1). Prenatal tobacco use was assessed via self-report on tobacco use during the last trimester of pregnancy (yes = 1). Subgroups with distinct patterns of SDOH were identified using a two-step cluster analysis in SPSS v.21, an approach well-suited for exploratory classification with both continuous and categorical indicators. Group differences in prenatal tobacco use were examined using the chi-square (χ²) tests.

Results: Two distinct groups emerged, with the largest difference in the deduction of Bayesian Information Criterion (ΔBIC=2.214): NA women with higher SDOH resources (n = 440; 62.9%) and those with lower SDOH resources (n = 260; 37.1%). The higher SDOH-resources group was characterized by an annual household income of average $78,400, 10.8 years of education, an average of 3.3 ACEs, 65.2% having insurance during pregnancy, 67.6% not experiencing prenatal IPV, and 91.3% initiating prenatal care in the first trimester. The lower SDOH-resources group had an average annual household income of $39,000, 10.1 years of education, 3.8 ACEs on average, 34.8% having insurance during pregnancy, 32.4% not experiencing prenatal IPV, and only 8.7% initiating prenatal care in the first trimester. The relative importance index indicated that the groups primarily differed in timing of prenatal care access. Group differences in prenatal tobacco use were significant, χ² = 19.201, p < .001: 53% of the lower SDOH-resources group reported third-trimester tobacco use, compared to 36% in the higher SDOH-resources group.

Conclusions and implications: This person-centered analytic approach revealed meaningful subgroups characterized by distinct patterns of SDOH, highlighting variability among NA women in a single state. The clinical relevance of these subgroups was supported by significant differences in prenatal tobacco use. While further validation is needed, the findings suggest that prenatal care access may be a key, modifiable SDOH target for policy and intervention efforts aimed at improving prenatal behavioral health outcomes among NA women.