Background
Pregnant women in the United States encounter several challenges when accessing adequate prenatal care (PNC), with transportation being a significant hurdle. Nevertheless, assessing the proximity to the nearest provider may overstate the extent of the access inequality. Additionally, no prior research has explored the association between perceived travel burden and PNC utilization. This study aims to investigate whether travel distances to the initial visited and the predominant PNC provider and patients’ perceived travel burden, measured by the travel disadvantage index (TDI), are associated with PNC utilization.
Methods
A retrospective cohort of pregnant women with live births were identified from South Carolina Medicaid claims files in 2015–2018. Travel distances were calculated using Google Maps. The estimated TDI within the zip-code tabulate area was derived from local pilot survey data. PNC utilization was measured by PNC initiation and frequency during each pregnancy. Comparisons were drawn between pregnancies with long (> 24 miles) or short (≤ 24 miles) distance to the predominant PNC provider. Repeated measure logistic regression test was employed for categorical variables and one-way repeated measures ANOVA for continuous variables. Unadjusted and adjusted ordinal logistic regressions with repeated measure were utilised to examine the association of travel burdens with PNC usage.
Results
Among 25,801 pregnancies continuously enrolled in Medicaid, pregnant women traveled an average of 24.9/24.2 miles to their initial/predominant provider, with an average TDI of -11.4. Of these pregnancies, 60% initiated PNC in the first trimester, with an average of 8 total visits. Compared to women with long distance, those travelling shorter distance were less likely to initiate PNC in the first trimester (61% vs. 59%, p < 0.01) and had fewer total visits (8.2 vs. 8.1, p = 0.04). Multiple regression analysis revealed that women with doubled travel distance or TDI were less likely to initiate timely PNC (odds ratio: 0.95, p < 0.001; 0.94, p = 0.04; travel distance & TDI, respectively) and had a lower visit frequency (odds ratio: 0.85, p < 0.001, travel distance).
Conclusions
Findings suggest that the association between travel burden and PNC utilization was statistically significant but of limited practical significance.