Abstract
This study characterizes driving distances to opioid treatment programs in urban vs rural counties in states with the highest rates of opioid mortality and compares them with distances to methadone-prescribing federally qualified health centers and dialysis centers.
Methadone for opioid use disorder can be dispensed only from US Substance Abuse and Mental Health Services Administration (SAMHSA)–certified opioid treatment programs (OTPs), creating access barriers in rural counties with a shortage of facilities. Canada and Australia allow primary care prescribing and pharmacy dispensing of methadone to expand access.1 Therefore, we examined drive times to the nearest OTP in urban and rural counties in 5 US states with the highest county rates of opioid-related overdose mortality.2 In addition, we compared drive times to federally qualified health centers (FQHCs) as potential primary care methadone-prescribing locations and to dialysis centers as treatment locations for a different chronic disease requiring frequent engagement.
Methods
The outcome was the minimum drive time in minutes from the county mean center of population to the nearest OTP, FQHC, and dialysis center using the Esri ArcGIS rural drive-time tool (September 2017 version), which simulates automobile movement between 2 points along a national street network based on historical average speeds.3 From the 2010 US Census, we obtained the coordinates of the county mean center of population for all counties in Indiana, Kentucky, Ohio, Virginia, and West Virginia, excluding counties with geographic changes after the census. We geocoded 2017 OTP, FQHC, and dialysis center street addresses from the SAMHSA OTP Directory and the Health Resources and Services Administration data warehouse. Addresses not matched during batch geocoding were hand reviewed. We excluded school-based FQHCs and facilities remaining unmatched after hand review.
We stratified counties by the 2013 National Center for Health Statistics urban-rural county classification scheme, dividing counties into urban (large central metros, large fringe metros, medium metros, and small metros) and rural (micropolitan and noncore) levels (Table). We assessed the association across urban-rural classification using Welch analysis of variance. We used a paired t test to compare drive times to the nearest OTP with drive times to the nearest FQHC or dialysis center, using a Bonferroni correction for multiple comparisons. Hypothesis tests were 2-sided with α=.05. We completed our analyses in Stata 15 (StataCorp).
Table. Drive Time From County Mean Center of Population to the Nearest Treatment Centers by Urban-Rural Classification, 2017 .
| Classificationa | Drive Time, Mean (95% CI), min | Difference in Drive Time, Mean (95% CI), min | |||||
|---|---|---|---|---|---|---|---|
| To OTP | To FQHC | P Valueb | To Dialysis Center | P Valuec | To OTP vs FQHC | To OTP vs Dialysis Center | |
| All counties | 37.3 (35.5 to 39.1) |
15.8 (14.8 to 16.9) |
<.001 | 15.1 (14.1 to 16.2) |
<.001 | 21.5 (19.5 to 23.4) |
22.1 (20.5 to 23.8) |
| Noncore | 49.1 (46.3 to 51.8) |
17.3 (15.4 to 19.2) |
<.001 | 22.6 (20.5 to 24.6) |
<.001 | 31.7 (28.3 to 35.2) |
26.5 (23.8 to 29.2) |
| Micropolitan | 41.1 (37.7 to 44.6) |
15.7 (13.2 to 18.2) |
<.001 | 10.1 (8.6 to 11.6) |
<.001 | 25.4 (21.6 to 29.2) |
31.0 (27.2 to 34.9) |
| Small metro | 35.0 (29.4 to 40.6) |
14.7 (11.8 to 17.6) |
<.001 | 14.9 (11.9 to 17.9) |
<.001 | 20.3 (14.3 to 26.3) |
20.1 (14.7 to 25.6) |
| Medium metro | 21.1 (17.7 to 24.5) |
13.4 (11.0 to 15.7) |
<.001 | 9.6 (7.1 to 12.2) |
<.001 | 7.8 (4.8 to 10.7) |
11.5 (8.4 to 14.6) |
| Large fringe metro | 25.2 (22.5 to 27.9) |
16.2 (13.8 to 18.6) |
<.001 | 11.3 (9.7 to 12.9) |
<.001 | 9.0 (6.1 to 12.0) |
13.9 (11.5 to 16.4) |
| Large central metro | 7.8 (5.7 to 9.9) |
6.3 (3.4 to 9.2) |
.32 | 5.4 (4.0 to 6.8) |
.06 | 1.4 (−1.7 to 4.5) |
2.4 (−0.1 to 4.8) |
Abbreviations: FQHC, Federally Qualified Health Center; OTP, opioid treatment program.
2013 National Center for Health Statistics urban-rural county classification scheme divides counties into urban (large central metros, large fringe metros, medium metros, and small metros) and rural (micropolitan and noncore) levels.
Paired t test for drive time to OTP vs drive time to FQHC.
Paired t test for drive time to OTP vs drive time to dialysis center.
Results
Of the 487 of 489 counties included, 270 (55.3%) were rural. Within the 5 states, 109 OTPs, 952 FQHCs, and 837 dialysis centers were included. Among all counties, the mean drive time to the nearest OTP was 37.3 (95% CI, 35.5-39.1) minutes and the mean drive time to the nearest OTP increased from 7.8 (95% CI, 5.7-9.9) minutes in the urban classification to 49.1 (95% CI, 46.3-51.8) minutes in the noncore rural classification (P < .001; Table). The mean drive time to the nearest FQHC was 15.8 (95% CI, 14.8-16.9) minutes (difference with OTP, 21.5 [95% CI, 19.5-23.4] minutes) and to the nearest dialysis center was 15.1 (95% CI, 14.1-16.2) minutes (difference with OTP, 22.1 [95% CI, 20.5-23.8] minutes). Longer drive times for OTPs vs FQHCs and dialysis centers were found for all urban-rural classifications (Figure) except large central metros, with the greatest difference in rural counties.
Figure. Drive Time From County Mean Center of Population to Nearest Opioid Treatment Program (OTP), Federally Qualified Health Center (FQHC), and Dialysis Center, 2017.
Discussion
Rural county classification was associated with longer drive times to the nearest OTP compared with urban counties. Drive times to OTPs were longer than to FQHCs or dialysis centers. The greater geographic availability of hemodialysis, which requires engagement 3 times a week, contrasts with methadone treatment availability, for which federal law requires engagement 6 times a week for medication dispensing. Enabling FQHC methadone provision in the United States, mirroring practices in Canada and Australia, would expand geographic access without construction of additional facilities and may further integrate opioid use disorder treatment into primary care. An alternative path to improving access would be constructing new OTPs, as was done previously with dialysis centers whose access was expanded by the 1972 extension of Medicare disability coverage,4 although this would require significantly more investment in rural health care infrastructure. Limitations include that drive times were county-level population estimates, individual drive times within counties vary, and smaller geographic units would improve drive time estimation. County estimates are presented given the importance of local government approval of OTPs. The urban geographic availability of methadone was likely overestimated because of public transportation.
Section Editor: Jody W. Zylke, MD, Deputy Editor.
References
- 1.Nosyk B, Anglin MD, Brissette S, et al. A call for evidence-based medical treatment of opioid dependence in the United States and Canada. Health Aff (Millwood). 2013;32(8):1462-1469. doi: 10.1377/hlthaff.2012.0846 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dwyer-Lindgren L, Bertozzi-Villa A, Stubbs RW, et al. Trends and patterns of geographic variation in mortality from substance use disorders and intentional injuries among US counties, 1980-2014. JAMA. 2018;319(10):1013-1023. doi: 10.1001/jama.2018.0900 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Apparicio P, Abdelmajid M, Riva M, Shearmur R. Comparing alternative approaches to measuring the geographical accessibility of urban health services: distance types and aggregation-error issues. Int J Health Geogr. 2008;7(1):7. doi: 10.1186/1476-072X-7-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Rettig RA. Special treatment—the story of Medicare’s ESRD entitlement. N Engl J Med. 2011;364(7):596-598. doi: 10.1056/NEJMp1014193 [DOI] [PubMed] [Google Scholar]