(B-085) Estimating Travel Time and Spatial Access to Insulin in China Before and After National Volume-Based Procurement: A Geospatial Analysis

Background: Insulin is essential for diabetes management, yet geographic and economic disparities hinder its accessibility. In 2021, China implemented the National Volume-Based Procurement (NVBP) policy to improve insulin affordability by reducing prices. However, its impact on spatial access to insulin remains unclear.

Objectives: This study evaluates changes in travel time and spatial accessibility to insulin before and after NVBP implementation, identifying regional disparities that may affect insulin availability.

Methods: We conducted a repeated cross-sectional geospatial analysis of insulin procurement data from 2017 to 2023 across 31 provinces and 360 cities in mainland China. Travel times from each census tract to the nearest insulin-providing facility were estimated using Geographic Information Systems (GIS) and the Dijkstra algorithm, incorporating pharmacy distribution and regional population density. Spatial inequality was assessed using the Gini coefficient and Theil index. The primary outcome was the change in travel time, while secondary outcomes included improvements in spatial equity and the proportion of the population with insulin access within specific travel time thresholds (e.g., ≤60 minutes). A mixed-effects model was applied to adjust for potential confounders such as population density, regional GDP, and healthcare infrastructure.

Results: NVBP implementation improved insulin accessibility, increasing the proportion of the population with ≤5-minute travel time from 10.9% to 24.3%, with notable gains for analogue insulin (0.0% to 6.3%). Travel time reductions varied regionally: Chongqing (-23.3 min), Guangxi (-19.6 min), and Hubei (-17.8 min) saw the largest reductions, whereas remote regions like Tibet (-1.7 min) and Qinghai (-3.5 min) experienced minimal changes. Spatial inequality improved slightly, with the Gini coefficient decreasing from 0.2229 to 0.2195. Mixed-effects models confirmed NVBP’s association with a modest yet statistically significant reduction in travel time (-0.46 min, 95% CI: -0.48 to -0.44, p < 0.001) and improved equity (ΔGini = -0.0016, 95% CI: -0.0017 to -0.0016, p < 0.001).

Conclusions: Geospatial analysis provides a feasible approach to quantifying disparities and evaluating policy impacts on insulin accessibility. While NVBP has improved affordability and accessibility, significant regional disparities persist. Targeted interventions are needed to further enhance insulin access, particularly in underserved and remote areas.