CONTEXTUALIZING OUTCOMES: A GEOSPATIAL ANALYSIS OF SMOKING DURING PREGNANCY IN NORTH CAROLINA, UNITED STATES
Author(s)
Liane M. Ventura, MPH1, Qian Huang, PhD, MA, MPA2, Kate Beatty, PhD3, Adeola Omolade Ayo, DrPH4.
1The University of North Carolina at Charlotte, Charlotte, NC, USA, 2Department of Biostatistics and Epidemiology, East Tennessee State University, Johnson City, TN, USA, 3Department of Health Services Management & Policy, East Tennessee State University, Johnson City, TN, USA, 4Department of Health Services Management & Policy, East Tennessee State University, Johnson, TN, USA.
1The University of North Carolina at Charlotte, Charlotte, NC, USA, 2Department of Biostatistics and Epidemiology, East Tennessee State University, Johnson City, TN, USA, 3Department of Health Services Management & Policy, East Tennessee State University, Johnson City, TN, USA, 4Department of Health Services Management & Policy, East Tennessee State University, Johnson, TN, USA.
OBJECTIVES: Smoking during pregnancy is a preventable contributor to short- and long-term adverse health outcomes for pregnant women and infants. This study examined the association between geographic context, healthcare access, and smoking during pregnancy across North Carolina, a geographically diverse state within the Southeastern United States.
METHODS: An ecological cross-sectional study was conducted with publicly available secondary data from 2018-2022. Geographic variables were rurality and managed care regions, which guide the administration of state-sponsored health insurance. Measures of healthcare access included county-level rates of obstetrician-gynecologists (OB-GYN) and median household income. An ordinary least squares (OLS) regression model was conducted at baseline, followed by Global Moran’s I to examine spatial distribution and dependence. A spatial lag regression model was applied to adjust for spatial autocorrelation.
RESULTS: Across North Carolina’s 100 counties, the state-level rate of smoking during pregnancy was 10.3%. County-level rates ranged between 1.6% and 21.6%. In the OLS model, rurality (β=-1.86, p=0.022), managed care regions (β=-1.08, p<0.001), OB-GYN rate (β=-0.13, p=0.007), and median household income (β=-20.50, p=0.002) were significantly associated with smoking during pregnancy. Statistically significant clusters in smoking during pregnancy were evident (Moran’s I=0.55, z=7.50, p<0.01), suggesting neighboring counties have similar rates. Three high-high clusters were identified. After accounting for spatial autocorrelation, rurality (β=-1.57, p=0.018), managed care regions (β=-0.63, p<0.001), and OB-GYN rates (β=-0.13, p<0.001) remained statistically significant predictors.
CONCLUSIONS: Rurality, managed care regions, and access to OB-GYN providers significantly influence rates of smoking during pregnancy in North Carolina. Findings emphasize the importance of geographic context to assess adverse maternal and infant health outcomes and to evaluate the organizational influence of managed care regions. The association between lower OB-GYN availability and higher rates of smoking during pregnancy suggests an opportunity to expand access to reimbursable prenatal care and tobacco cessation services, particularly within high-rate clusters.
METHODS: An ecological cross-sectional study was conducted with publicly available secondary data from 2018-2022. Geographic variables were rurality and managed care regions, which guide the administration of state-sponsored health insurance. Measures of healthcare access included county-level rates of obstetrician-gynecologists (OB-GYN) and median household income. An ordinary least squares (OLS) regression model was conducted at baseline, followed by Global Moran’s I to examine spatial distribution and dependence. A spatial lag regression model was applied to adjust for spatial autocorrelation.
RESULTS: Across North Carolina’s 100 counties, the state-level rate of smoking during pregnancy was 10.3%. County-level rates ranged between 1.6% and 21.6%. In the OLS model, rurality (β=-1.86, p=0.022), managed care regions (β=-1.08, p<0.001), OB-GYN rate (β=-0.13, p=0.007), and median household income (β=-20.50, p=0.002) were significantly associated with smoking during pregnancy. Statistically significant clusters in smoking during pregnancy were evident (Moran’s I=0.55, z=7.50, p<0.01), suggesting neighboring counties have similar rates. Three high-high clusters were identified. After accounting for spatial autocorrelation, rurality (β=-1.57, p=0.018), managed care regions (β=-0.63, p<0.001), and OB-GYN rates (β=-0.13, p<0.001) remained statistically significant predictors.
CONCLUSIONS: Rurality, managed care regions, and access to OB-GYN providers significantly influence rates of smoking during pregnancy in North Carolina. Findings emphasize the importance of geographic context to assess adverse maternal and infant health outcomes and to evaluate the organizational influence of managed care regions. The association between lower OB-GYN availability and higher rates of smoking during pregnancy suggests an opportunity to expand access to reimbursable prenatal care and tobacco cessation services, particularly within high-rate clusters.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
EPH35
Topic
Epidemiology & Public Health, Health Policy & Regulatory, Study Approaches
Topic Subcategory
Public Health
Disease
Mental Health (including addiction), Reproductive & Sexual Health