A multi-method spatial examination of factors associated with changes in geographic accessibility to buprenorphine providers in HEALing communities study states Kentucky, Massachusetts, and Ohio

Shikhar Shrestha, Olivia Lewis, Daniel R. Harris, Peter Rock, Anita Silwal, Sumeeta Srinivasan, Thomas J. StopkaPreventive 2025 · DOI

TL;DR

This study examines changes in geographic access to buprenorphine providers across three US states between 2020 and 2022, using multiple spatial analysis methods. While overall access improved, particularly in urban areas, rural regions continue to face significant barriers, underscoring the need for targeted interventions.

What Problem Does This Paper Address?

Opioid overdose remains a critical public health crisis in the US, with medications for opioid use disorder (MOUD) like buprenorphine being crucial treatment options. However, access to these treatments is often limited by geographic and structural barriers. Previous studies have often relied on single measures of accessibility, which can lead to imprecise estimates. This paper addresses the need for a more comprehensive understanding of how geographic accessibility to buprenorphine providers has changed over time and whether these changes have been equitable across different regions and sociodemographic groups. Specifically, it investigates changes within the HEALing Communities Study (HCS) states of Kentucky, Massachusetts, and Ohio, aiming to identify factors associated with these changes and any existing disparities.

Methods

The researchers employed a multi-method spatial approach to analyze changes in geographic accessibility to buprenorphine providers between January 2020 and June 2022. This comprehensive strategy aimed to overcome the limitations of single-method analyses. The four key methods used were:

  • Kernel Density Analysis: This technique was used to visualize and quantify the concentration of buprenorphine providers across the study areas, allowing for the identification of "hotspots" of provider density and changes in these densities over time.
  • Change in Rates Analysis: This method calculated the number of buprenorphine providers per 100,000 population at the county or municipality level. By comparing these rates between the two time points, the study assessed changes in provider availability relative to population size.
  • Minkowski Distance (Drive-Time Approximation): To quantify travel time to the nearest provider, the researchers used Minkowski distance, a generalized metric for distance. This method, with a specific parameter (p=1.31) chosen to approximate drive-time, allowed for a consistent and computationally feasible estimation of travel burdens for individuals seeking treatment.
  • Enhanced 2-Step Floating Catchment Area (E2SFCA): This advanced spatial analysis method was used to assess geographic accessibility by considering both the supply of providers and the demand from the population. It accounts for travel time and provider capacity, offering a more nuanced measure of access.

These spatial analyses were complemented by multivariable regression modeling to examine the factors associated with changes in E2SFCA accessibility, controlling for HCS community status (Wave 1 intervention vs. Wave 2 waitlist control vs. non-HCS) and various sociodemographic indicators.

Datasets

The study utilized the following datasets:

  • Buprenorphine Provider Data: Obtained from the U.S. Drug Enforcement Administration (DEA) database, specifically identifying providers with Drug Addiction Treatment Act (DATA) of 2000 waivers (X-waiver). Data was collected for January 2020 (reference point) and June 2022.
  • Sociodemographic Data: Compiled from the U.S. Census Bureau's American Community Survey (ACS) 5-year estimates (2016-2020) at the census tract level. This included data on population demographics, education levels, employment, income, housing vacancy, and health insurance status.
  • Geographic Data: County-level and census tract-level shapefiles were obtained from the U.S. Census Bureau. For Massachusetts, municipality-level shapefiles were acquired from MassGIS.

Key Insights and Ideas

  • Urban Areas See Greatest Gains: The most significant increases in buprenorphine provider density were concentrated in urban centers, suggesting that existing infrastructure and demand in these areas may have facilitated faster growth.
  • Rural Access Remains a Challenge: Despite overall improvements, the study highlights that rural areas continue to experience limited access to buprenorphine providers, often requiring longer travel times. This perpetuates existing disparities in treatment access.
  • Provider Numbers Don't Always Equal Prescriptions: While the number of buprenorphine providers has increased, this hasn't necessarily translated into a proportional increase in buprenorphine prescriptions. This suggests that factors beyond provider availability, such as prescriber training, patient stigma, or pharmacy dispensing barriers, may be limiting treatment uptake.
  • Baseline Access is a Strong Predictor: Areas that already had higher levels of geographic accessibility to buprenorphine providers at the start of the study (2020) tended to see the greatest improvements by the end of the study period (2022). This indicates that existing infrastructure plays a crucial role in facilitating further growth.
  • HCS Intervention Impact is Nuanced: The HEALing Communities Study (HCS) intervention did not show a consistent, significant association with increased geographic accessibility across all HCS communities compared to the state average. This suggests that community-level interventions may have complex and varied impacts on provider distribution.

Key Results

  • Increased Density: All three states (Kentucky, Massachusetts, and Ohio) showed an increase in the density of buprenorphine providers between 2020 and 2022, with the largest increases observed in urban areas.
  • Improved Drive-Times: The proportion of the population living within a certain drive-time to a buprenorphine provider increased across all states. For example, in Kentucky, the percentage of people within a 10-minute drive increased from 75% to 82%.
  • E2SFCA Index Improvement: The E2SFCA accessibility index generally increased across most census tracts in all three states, indicating an overall improvement in geographic accessibility.
  • Predictive Power of Baseline Access: The reference point E2SFCA index (from January 2020) was a statistically significant predictor of increased E2SFCA indices in all three states, demonstrating that areas with better initial access saw greater gains.
  • Population Density Matters: Population density was found to have a significant positive impact on increases in E2SFCA indices across all states, suggesting that more densely populated areas experienced greater improvements in access.

Limitations and Future Work

The authors acknowledge several limitations:

  • Provider Location as Proxy: The study used the location of buprenorphine providers as a proxy for access to prescriptions, as granular data on actual prescriptions was unavailable. The presence of a provider does not guarantee actual treatment uptake.
  • Aggregated Data: Reliance on aggregated data (county, municipality, census tract) may obscure finer spatial patterns of accessibility.
  • Temporal Limitations: The analysis used two distinct time points, limiting the ability to capture real-time changes and trends.
  • Urban-Rural Status: While population density was used as a proxy for urbanicity, it doesn't fully capture the nuances of urban-rural status across all states.
  • Driving Speed Assumption: The drive-time calculation assumed an average driving speed, which can vary significantly by location and urbanicity.
  • Spatial Lagged Effects: The multivariable models did not account for spatial lagged effects of geographic accessibility or underlying socioeconomic features on changes in accessibility.
  • Demand Estimation: The study used population over 18 as a proxy for demand, rather than the number of individuals with opioid use disorder.

Future work could focus on:

  • Investigating how distance to pharmacies and prescribers impacts initiation and adherence to buprenorphine treatment.
  • Exploring the impact of pharmacy dispensing barriers and prescriber biases on treatment access.
  • Developing real-time data collection methods for more dynamic accessibility analysis.
  • Incorporating more granular data on opioid use disorder prevalence and treatment demand.
  • Further research into the effectiveness of interventions specifically targeting rural areas and communities with limited baseline access.

Summary updated

Abstract

Since geographic accessibility to buprenorphine treatment is critical, additional efforts are needed to improve access to buprenorphine providers in rural areas and areas with limited baseline access.
BibTeX
@inproceedings{shrestha2025multi,
  author = {Shikhar Shrestha and Olivia Lewis and Daniel R. Harris and Peter Rock and Anita Silwal and Sumeeta Srinivasan and Thomas J. Stopka},
  title = {{A multi-method spatial examination of factors associated with changes in geographic accessibility to buprenorphine providers in HEALing communities study states Kentucky, Massachusetts, and Ohio}},
  booktitle = {Preventive},
  year = {2025},
  doi = {10.1016/j.pmedr.2025.103045},
}
Figure