Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Aug 4.
Published in final edited form as: J Rural Health. 2026 Jan;42(1):e70130. doi: 10.1111/jrh.70130

County Health Rankings, Provider Shortages, and the Health of Incarcerated Women with Opioid Use Disorder

Meghan Urhahn 1, Martha Tillson 2, Brianna Gagen 1, Michele Staton 2,3
PMCID: PMC13430536  NIHMSID: NIHMS2195423  PMID: 41728810

Abstract

Purpose:

Incarcerated women in rural Kentucky face significant barriers to healthcare, including provider shortages, geographic isolation, and limited access to preventive services. This study examines how county health rankings and primary care provider (PCP) availability relate to women’s self-reported health conditions and healthcare utilization prior to incarceration.

Methods:

A total of 900 incarcerated women across nine Kentucky jails were screened for opioid use disorder and enrolled as part of a larger clinical trial. County-level data, including 2023 County Health Rankings and 2022 PCP-to-population ratios, were merged with self-reported data on health conditions and service use prior to incarceration. Logistic regression models assessed associations between county-level indicators and pre-incarceration healthcare utilization and health status.

Findings:

On average, women were 37.3 years old, 92.6% non-Hispanic White, and 67.1% lived in a rural county before jail. Women from counties with better health rankings were more likely to access substance use treatment (AOR=0.993, p=0.028) before incarceration. Those from counties with poorer rankings were more likely to report hepatitis C (AOR=1.008, p=0.002) but less likely to report other health concerns, possibly reflecting reduced access or awareness. Fewer available PCPs were associated with higher rates of chronic conditions such as issues with blood pressure (e.g., hypertension; AOR=0.782, p=0.010).

Conclusion:

County-level disparities in healthcare infrastructure may significantly affect the health of criminal legal system-involved women. Strengthening provider networks, expanding telehealth, and investing in rural health systems are critical to improving access and outcomes for this vulnerable population and may reduce recidivism and adverse post-release health events.

Keywords: Incarcerated women, Rural health disparities, Healthcare access, Primary care provider shortages, Substance use disorder

1. Introduction

County-level health indicators, such as those compiled by the University of Wisconsin Population Health Institute’s County Health Rankings, offer a valuable lens through which to examine disparities in population health. These rankings combine clinical, behavioral, environmental, and social factors into a composite score that reflects the overall health status and healthcare infrastructure of U.S. counties. Such rankings provide a more comprehensive understanding of local conditions than binary urban-rural classifications alone.

Counties with less favorable health rankings often share characteristics of rurality, including geographic isolation, provider shortages, and high levels of socioeconomic disadvantage.1,2 Many Appalachian and rural Kentucky counties are federally designated Health Professional Shortage Areas for primary care, mental health, and substance use treatment, reflecting longstanding deficits in provider availability, including limited access to buprenorphine-waivered clinicians and methadone programs.3,4 Using 2023 Health Resources and Services Administration (HRSA) data, approximately 105 of Kentucky’s 120 counties (87.5%) contained whole- or partial-county primary care Health Professional Shortage Area (HPSA) designations, with over 90% of the state’s population residing in HPSA-designated areas.4 Rural residents consistently experience higher rates of mortality from chronic diseases and face structural barriers to care, including limited provider availability and transportation challenges.5,6 These challenges are particularly acute in regions such as Appalachia, where individuals face elevated risks of opioid use disorder (OUD), mental health issues, and reproductive health concerns.7,8

Compared to men, women often face unique health issues and inequities, and women living in rural areas are often underserved in health research and healthcare delivery.9,10 They report greater unmet needs, lower access to specialty care, and poorer outcomes for chronic and mental health conditions.11,12 These disparities are compounded by stigma and structural inequities, leading many women to delay or avoid care, especially for substance use and mental health services.13,14 Women with OUD and prior justice involvement often experience fragmented care, stigma in clinical encounters, and limited continuity of services, all of which can reduce recognition of health issues and discourage disclosure of symptoms.15–17

Among rural women, those involved in the criminal justice system are even more vulnerable. Incarcerated women have disproportionately high rates of chronic disease, substance use disorders, and mental health conditions.18,19. Individuals with substance use disorders often underreport health conditions due to stigma, normalization of chronic illness, and prior negative healthcare experiences, and these patterns may be intensified within the criminal justice system where disclosure can feel risky or unhelpful.20,21 These conditions often exist before arrest and are exacerbated by incarceration and reentry into underserved communities (Knittel, 2019; Willging et al., 2013). Post-incarceration, rural women face compounded barriers: few providers, long travel distances, and limited continuity of care, all of which increase risks of relapse, overdose, and recidivism.22,23 The lack of adequate healthcare infrastructure in rural areas is a critical driver of poor health outcomes.24,25 Prior research demonstrates that county-level health system capacity, including behavioral health workforce availability and SUD treatment infrastructure, shapes both health system engagement and criminal legal system outcomes.26 Taken together, these patterns highlight why examining the intersection of incarceration and place of residence is essential for addressing disparities and informing targeted interventions.

While prior research has highlighted rural-urban disparities and the heightened health burdens among justice-involved women, little is known about how specific structural county-level characteristics, beyond binary rural classifications, shape pre-incarceration health and healthcare utilization in this population.2,18,27 To address this gap, our study uses two county-level indicators: primary care provider (PCP)–to–population ratios and overall county health rankings. Together, these measures reflect both healthcare system capacity and broader community-level determinants of health that influence women’s ability to access and engage with care prior to incarceration.1,28–30 Such structural indicators may be particularly informative for populations experiencing chronic instability and fragmented healthcare engagement. Prior work demonstrates that county-level healthcare capacity and organizational resources influence participation in health system innovations and are associated with variation in jail populations, underscoring the importance of examining structural context when studying justice-involved populations.31,32 By employing these more nuanced metrics, our study seeks to uncover localized patterns in health status and service use that may be obscured by traditional rural-urban comparisons. This approach offers a more detailed understanding of how local health system capacity influences outcomes for one of the most underserved populations.12,33

Our study investigates the relationship between health conditions, healthcare utilization, and county-level health indicators among incarcerated women with opioid use disorder (OUD) in Kentucky. Specifically, it examines how county health rankings and PCP-to-population ratios are associated with self-reported health status and service use prior to incarceration. We hypothesize that women reporting more health conditions and greater healthcare use before incarceration are more likely to come from counties with lower health rankings and fewer PCPs, although limited access to diagnostic services and under recognition of symptoms may also contribute to lower reporting in resource-poor counties.2,34 Transportation barriers and the geographic dispersion of providers in rural areas may contribute to underutilization of essential health services.5,14 Understanding how these structural factors influence pre-incarceration healthcare utilization can help inform more targeted community-level and policy interventions.23,25

2. Methods

2.1. Participants

This project was conducted within the framework of the NIDA-funded Kentucky Justice Community Opioid Innovation Network (JCOIN) cooperative initiative (UG1DA050069). From nine Kentucky jails, 900 incarcerated women aged 18 years or older were randomly identified using jail census reports and assessed for opioid use disorder (OUD). Those meeting eligibility were invited to join a clinical trial aimed at expanding treatment availability and initiation during and after incarceration.

Five of the participating jails served as experimental sites, selected because they had sufficient administrative and medical infrastructure to support treatment delivery but did not already provide OUD services for women. The remaining four jails served as comparison sites, as they offered state-supported substance use disorder (SUD) programming in the form of a six-month, jail-based modified therapeutic community that included access to extended-release naltrexone prior to release (see Staton et al., 202135 for detailed methods).

Eligibility criteria required women to (1) report opioid use during the 12 months preceding incarceration, (2) meet thresholds on standardized screening tools (≥2 on the DSM-5 OUD Checklist36 or ≥4 on the NM-ASSIST opioid scales37), (3) anticipate release within 7–60 days of screening, (4) demonstrate capacity for informed consent without evidence of severe mental illness, active psychosis, or significant cognitive impairment, and (5) agree to study participation.

2.2. Procedures

Women expressing interest in the study underwent a screening process conducted by trained female research personnel. This process took place either in-person at the jail or through teleconference, contingent upon jails’ COVID-19 safety protocols. All screening and data collection procedures were held in a confidential environment, ensuring the absence of any jail staff during the process.

During the screening, participants were provided with comprehensive information regarding the study. Those who showed interest were then evaluated for their eligibility to participate. Following this assessment, research staff obtained informed consent from eligible participants and invited them to engage in a baseline interview. This interview encompassed various demographic factors, including age, race, housing situation, educational background, and employment status, as well as inquiries into health conditions and health service utilization.

Research staff read interview questions aloud and recorded participants’ responses using Research Electronic Data Capture (REDCap), a secure platform designed for data collection and management.38 The average duration of the interviews was approximately 90 minutes, and participants received a compensation of $45 for their involvement in the screening and baseline data collection. All procedures associated with the study were safeguarded by a federal Certificate of Confidentiality and received approval from the Institutional Review Board (IRB) of the University.

2.3. Measures

2.3.1. Participant-level Measures.

Participant-level measures were collected during interviews with incarcerated women enrolled in the study. All data were self-reported.

2.3.1.1. Demographics.

Demographic variables included age, race/ethnicity (other race/ethnicity=0, non-Hispanic white=1), years of education (coded as 0=less than high school diploma, 1=high school diploma/GED or higher education), and employment prior to incarceration (PTI; coded as 0=unemployed, 1=employed full-time, part-time, or doing day labor). Participants also shared their health insurance status (currently or prior to jail), which was coded as 0=uninsured or don’t know insurance status and 1=insured (active or suspended). Women were asked the number of pregnancies they had experienced in their lifetime, which was included as a relevant reproductive health indicator that may be associated with healthcare utilization, substance use patterns, and overall physical and mental health status. Lastly, women’s scores on the DSM-5 OUD Checklist36 were included for the 30 days PTI as an indicator of substance use severity.

2.3.1.2. Health Service Utilization.

Women were asked to report, during the 90 days PTI, whether they had visited a hospital or emergency room for any reason, stayed overnight in a treatment facility (for physical, mental, or behavioral health), or received any form of outpatient treatment. Follow-up probes to these initial questions provided detail about the types of services received. A new variable was calculated indicating whether the participant had received services related to substance use or SUD (including residential or inpatient, outpatient or intensive outpatient, and services from a primary care provider). Participants also reported whether they had received prescribed medications for the treatment of OUD (MOUD) in the 90 days PTI. All health services variables were coded as 0=no, 1=yes.

2.3.1.3. Self-reported Health Issues

Participants were asked whether they had ever in their lifetime been diagnosed with hepatitis C virus (HCV), any sexually transmitted infections or diseases (STI/STD), and COVID-19 (all coded as 0=no, 1=yes). These infections were selected based on their known association with opioid use disorder (OUD), elevated prevalence in incarcerated populations, and relevance to public health39,40. Lifetime diagnoses were included to capture chronic conditions that may have long-term implications for health and healthcare needs. Participants also reported whether they had been bothered by any health problems in the 90 days PTI, and if so, to specify what medical problem(s) they had been experiencing (open response). These open-text responses allowed for a broader assessment of acute and chronic health conditions relevant to recent healthcare utilization. Responses were coded by two of the paper’s authors (MU and BG) to indicate which category or categories of health issues were indicated.

2.3.2. County-level Measures.

County-level measures were obtained from publicly available data sources noted below and provide information for individual counties relevant to health outcomes. For purposes of the present study, the specific county where each participant reported living PTI was coded with a value for each of the three variables described below (rural-urban residence, county health ranking, and physician to population ratio).

2.3.2.1. Rural-urban Residence.

Each participant reported their county of residence PTI. Counties were coded using USDA ERS Rural-Urban Continuum Codes with codes 1–3 designated as urban (=0) and 4–9 designated as rural (=1).41

2.3.2.2. Health Rankings.

Health rankings were sourced from the County Health Rankings, a program by the University of Wisconsin Population Health Institute in collaboration with the Robert Wood Johnson Foundation.30 The rankings are based on a conceptual model that includes two primary categories: health outcomes (mortality and morbidity) and health factors (encompassing health behaviors, clinical care, social and economic factors, and physical environments). Health factors are considered modifiable determinants that influence the future health of a county. Rankings are calculated using an average of multiple years of data with mean imputation to rank all 3,143 counties or county equivalents in the United States.

Data for each of the 30 ranking components are collected from national sources, such as the National Center for Health Statistics, the Behavioral Risk Factor Surveillance System, and the American Community Survey (University of Wisconsin Population Health Institute, 2023).30 For each measure, Z-scores are calculated, weighted, and summed to create composite scores.30 These scores are then ranked within each state, from best to worst health.

2.3.2.3. Primary Care Physician to Population Ratio.

According to the Kentucky Physician Report 2022 provided by the University of Kentucky (UK) Center of Excellence in Rural Health, specific criteria and statistical methods were used to analyze physician and primary care physician (PCP) ratios across Kentucky.42 Data were initially obtained from the Kentucky Board of Medical Licensure and processed to ensure accuracy by eliminating inactive, retired, semi-retired, out-of-state, unlisted physicians, and residents.

Physicians were categorized by specialty, specifically identifying PCPs in Family Medicine, Internal Medicine, Pediatrics, or Geriatric Medicine. To prevent duplicate counts, physicians listing multiple specialties were classified under one primary specialty, typically the one listed as their primary area of practice. PCPs were only included if they practiced in private settings, hospitals, or as employed outpatient physicians, excluding those in emergency medicine.

The UK Center of Excellence then used IBM SPSS Statistics software to calculate frequency distributions for physician counts and ratios. Population data was sourced from U.S. Census data to determine the provider-to-population ratios for each county.

2.4. Analytic plan.

Data were excluded for all participants who did not report a county of residence PTI (n=9) or who were not living in Kentucky (n=35), resulting in a final sample of N=856. Descriptive statistics (means, standard deviations, and proportions) were calculated for all variables of interest. Next, preliminary analyses were performed using t-tests to assess bivariate relationships between dichotomous participant-level health measures (self-reported health service utilization and health issues) and continuous county-level variables (county health rankings and PCP population ratios). Any significant bivariate relationships were then further analyzed using logistic regression models, with dichotomous participant-level health measures as dependent variables and county-level variables as independent variables. Altogether, participants reported residence pre-incarceration in a total of 93 unique counties, over half of which were represented by fewer than five participants. Despite the large number of clusters, logistic regression was deemed the most appropriate analytic approach given small sample sizes within clusters, which would have been inappropriate for hierarchical methods.43,44 All demographic variables were included in regression models to allow for examination of the unique contribution of county-level measures, controlling for individual differences. All analyses were conducted using SPSS 28.

3. Results

As shown in Table 1, participants (N=856) were 37.3 years old on average, 92.6% non-Hispanic White, and about two-thirds (67.1%) reported living in a rural county PTI. Most participants indicated that they had a high school diploma or higher education (73.5%) although fewer reported being employed PTI (23.2%). Health insurance (active or suspended) was reported by 84.5% of women. Lastly, participants reported a lifetime average of 3.2 pregnancies, scored on average 9.9 out of 11 on DSM OUD severity for the 30 days PTI, and reported having been incarcerated for an average of 38.6 total months in their lifetime (range 0–300).

Table 1.

Descriptive profile of the sample (N=856)

% or M (SD)
Demographics
 Age (range 18–62) 37.3 (8.7)
 Race/ethnicity (non-Hispanic White) 92.6%
 Living in rural county PTI 67.1%
 aHS diploma or higher ed 73.5%
 Employed PTI (part-time, full-time, or day labor) 23.2%
 Had health insurance (active or suspended) 84.5%
 bNumber of lifetime pregnancies (range 0–20) 3.2 (2.2)
 DSM OUD severity during 30 days PTI (range 0-11) 9.9 (2.5)
 Lifetime months spent incarcerated (range 0-300) 38.6 (40.1)
Health services utilization during 90 days PTI
 Visited a hospital or emergency room for any reason 21.3%
 Stayed overnight in a treatment facility 9.5%
 Received any form of outpatient treatment 22.9%
SUD treatment services received during 90 days PTI
 Received any form of SUD treatment 20.8%
 Prescribed MOUD 14.8%
Self-reported health issues – lifetime, ever diagnosed with…
 cHepatitis C virus (HCV) 38.1%
 dSTI/STD (e.g., gonorrhea, chlamydia) 18.8%
 eCOVID-19 33.1%
Self-reported – 90 days PTI, bothered by health issues related to…
 Chronic musculoskeletal 13.4%
 Mental health 11.3%
 Neurological 8.3%
 Pulmonary 4.9%
 Renal 3.0%
 Blood pressure 2.5%
 Gastrointestinal 2.3%
 Cardiac 1.8%
 Reproductive 1.5%
 Dental 1.3%
 Endocrine 0.9%
 Diabetic 0.8%
 Cancer 0.8%
 No health issues 57.0%
Open in a new tab
a

Education missing n=8.

b

Lifetime pregnancies missing n=3.

c

HCV missing n=8.

d

STI/STD missing n=1.

e

COVID-19 missing n=5.

Before incarceration, participants most commonly reported receiving health services from outpatient providers (22.9%), followed by hospitals or emergency rooms (21.3%) and inpatient settings (9.5%). About one in five participants (20.8%) had received any form of SUD treatment during the 90 days PTI, including 14.8% who had been prescribed MOUD.

Participants also reported a history of health issues, including lifetime diagnoses of HCV (38.1%), STIs or STDs (18.8%), and COVID-19 (33.1%). When asked which health problems they had been bothered by during the 90 days PTI, participants provided a variety of responses, but most commonly cited chronic musculoskeletal issues (13.4%), mental health (11.3%), or neurological problems (8.3%). Over half of women (57.0%) said there were no specific health problems that had bothered them in the 90 days before jail.

3.1. Bivariate relationships

Results from t-tests indicated several significant relationships between county health rankings and participant-level specific health issues and service utilization. Participants who reported visiting a hospital or emergency room (t[854] = 2.1, p = .040), receiving outpatient treatment (t[854] = 2.0, p = .043), and receiving SUD treatment specifically (t[854] = 2.7, p = .008) also reported residence PTI in counties with better county health rankings. Prior diagnosis of HCV was associated with PTI residence in a county with a worse health ranking (t[628.1] = −3.0, p = .003). However, women who reported a prior COVID-19 diagnosis (t[639.3] = 2.8, p = .005) and feeling bothered by mental health problems (t[122.1] = 2.8, p = .006) also reported living in a county with a better health ranking. Participants who reported not being bothered by any health issues were more likely to live in a county with a worse health ranking (t[831.2] = −2.6, p = .010).

PCP population ratios were also associated with differences in health concerns at the bivariate level. Participants reporting a prior HCV diagnosis reported living in counties with higher PCP ratios (t[619.1] = −3.1, p = .002). However, feeling bothered by health issues related to blood pressure (t[840] = 2.5, p = .011) or endocrine concerns (t[7.7] = 2.9, p = .022) were both associated with residence in counties with lower PCP ratios.

3.2. Logistic regression models examining health services utilization and health issues

Extending significant bivariate findings, a series of logistic regression models examined likelihood of reporting specific types of health issues and use of specific types of health services PTI, with county health rankings (Models 1–7) and PCP population ratios (Models 8–10) as independent variables of interest, controlling for demographic variables.

Regarding use of health services, residence in a county with a better health ranking (i.e., lower numeric value) was independently associated with greater odds of receiving SUD treatment during the 90 days PTI (Model 3; AOR=0.994, p=.043). No significant relationship for county health outcomes ranking was found in either Model 1 (hospital or emergency room use) or Model 2 (outpatient treatment). For previous health issues, having been diagnosed with COVID-19 (Model 5; AOR=0.991, p<.001) was significantly associated with living in a county with a better health ranking. However, residence in a county with a worse health ranking (i.e., higher numeric value) was associated with higher odds of having been diagnosed with HCV (Model 4; AOR=1.005, p=.006.

For models including primary care physician (PCP) population ratios, odds of a lifetime HCV diagnosis (Model 8) were higher among participants who lived in counties with higher PCP population ratios (AOR=1.105, p=.002). Blood pressure issues (Model 9) were more common among women who lived in counties with lower PCP population ratios (AOR=0.781, p=.011). Lastly, in Model 10, PCP population ratios were unrelated to odds of participants reporting being bothered by endocrine issues.

4. Discussion

By using county-level indicators rather than binary rural–urban classifications, this study offers a more nuanced view of structural health disparities. Such insights are critical for shaping public health, criminal justice, and policy responses aimed at improving outcomes for justice-involved women with OUD and reducing preventable health crises post-release, particularly during periods of incarceration and community reentry.

Although not identical, health rankings and PCP ratios reflect overlapping but distinct characteristics: the former includes broader social determinants, such as economic stability and education, while the latter captures direct access to medical care. Understanding this distinction is critical for identifying the root causes of unmet health needs in underserved populations, especially in geographically isolated communities where systemic barriers are compounded.1,2

Bivariate results showed that women from counties with better health rankings were more likely to have used outpatient care, emergency services, and received treatment for substance use disorder (SUD) prior to incarceration. These differences likely reflect improved availability of services and greater integration of behavioral health within primary care in better-ranked counties.45,46 Conversely, participants from worse-ranked counties were more likely to report hepatitis C diagnoses, commonly associated with a history of injection drug use.47,48 Interestingly, participants from worse-ranked counties were less likely to report current health concerns despite elevated risks, potentially due to underdiagnosis, stigma around disclosure, or normalization of chronic health issues in these settings.49,50 This disconnect highlights the complexity of health self-reporting and raises concerns about hidden morbidity in lower-resourced areas. Among justice-involved women, underreporting may be further influenced by fear of judgment, concerns about how disclosed information may be used within the carceral system, and prior negative experiences with healthcare providers.51,52

Bivariate differences in PCP ratios revealed that provider shortages were independently associated with a higher prevalence of chronic disease, such as hypertension and endocrine disorders. These conditions are typically managed in outpatient settings, and their presence suggests a lack of sustained access to preventive and chronic care.53,54 Given that these health issues require long-term follow-up and medication adherence, provider scarcity may directly contribute to worsening disease burden and greater reliance on emergency care.

Logistic regression models controlling for demographics revealed persistent associations between county health rankings and healthcare utilization. Women from better-ranked counties were more likely to have received SUD treatment and received a COVID-19 diagnosis prior to incarceration. These findings likely reflect both improved provider availability and broader systemic engagement with health services.11,13 In contrast, the underreporting of health problems among women in worse-ranked counties may suggest low detection rates, cultural stigma around disclosing health issues, or mistrust in medical systems12,14

PCP-to-population ratios were also independently associated with chronic health outcomes – specifically, HCV diagnosis and blood pressure issues – supporting prior evidence that provider availability influences not just access but also ongoing disease management.22,25 The significance of structural indicators even after adjustment underscores the nature of rural disadvantage, substance use, and incarceration.33,55 These findings reinforce that healthcare utilization is shaped not only by individual behavior but also by the structural characteristics of one’s environment. Structural constraints such as insurance instability, limited Medicaid acceptance, and fragmented funding for behavioral health services likely further mediate these associations, particularly for women cycling between incarceration and underserved communities.17,56 Counties with inadequate healthcare infrastructure leave women with fewer opportunities for early detection, treatment, and continuity of care. This is especially consequential for justice-involved women with OUD, who face heightened health risks during incarceration and reentrys.18,23 Addressing these gaps requires coordinated interventions across health and criminal justice systems, including expansion of telehealth services, peer navigation and recovery support programs, provider incentives for rural practice, and policies that ensure Medicaid continuity during incarceration and reentry.7,19

While this study provides important insight into the structural factors shaping pre-incarceration healthcare experiences, several limitations should be noted. The analysis did not include measures of variation in jail intake screening procedures, which may influence selection into the research study as well as self-reporting of health conditions. Additionally, county-level counts of specific substance use treatment providers, such as buprenorphine-waivered clinicians or methadone programs, were not available. Instead, county health rankings and PCP-to-population ratios were used to capture broader health system capacity and access constraints relevant to justice-involved women’s healthcare experiences prior to incarceration.29,57

Conclusions

The study’s findings emphasize critical policy and clinical actions needed to improve healthcare access for incarcerated women in underserved areas, including rural communities. A major issue is the shortage of PCPs and mental health professionals, which limits preventive care and chronic disease management.58,59 Policymakers should expand loan forgiveness programs and financial incentives to encourage healthcare workers to serve in rural correctional and reentry settings.60 Additionally, integrating telehealth into jails and post-release programs can improve access to mental health and substance use treatment61,62

Future studies should examine the long-term health outcomes of formerly incarcerated women in rural communities, particularly the impact of Medicaid expansion and telehealth accessibility.58 Research comparing rural and urban reentry healthcare access could further inform policies to address geographic disparities.63,64

Table 2.

Logistic regression models examining health conditions and health service utilization by county health outcomes ranking

Model number, dependent variable County health outcomes ranking AOR p-value Nagelkerke Pseudo-R2
Model 1: Hospital or ER (n=832) 0.999 0.846 0.019
Model 2: Outpatient Treatment (n=832) 0.995 0.126 0.057
Model 3: Any SUD Treatment (n=832) 0.994* 0.043 0.062
Model 4: HCV Diagnosis (n=825) 1.008** 0.003 0.062
Model 5: COVID Diagnosis (n=827) 0.991*** <0.001 0.069
Model 6: Mental Health Problems (n=832) 0.992 0.052 0.032
Model 7: No Health Problems (n=832) 1.005 0.066 0.074
Open in a new tab

Note:

*

p ≤ 0.05

**

p ≤ 0.01

***

p ≤ 0.001. All models included covariates of age, race, education, employment, rural residence, health insurance, number of lifetime pregnancies.

Table 3.

Logistic regression models examining health conditions and health service utilization by physician-to-population ratio

Model number, dependent variable Physician-to-population ratio AOR p-value Nagelkerke Pseudo-R2
Model 8: HCV Diagnosis (n=812) 1.105** 0.002 0.065
Model 9: Blood Pressure Issues (n=818) 0.781* 0.011 0.107
Model 10: Endocrine Issues (n=818) 0.823 0.160 0.121
Open in a new tab

Note:

*

p ≤ 0.0

**

p ≤ 0.01. All models included covariates of age, race, education, employment, rural residence, health insurance, number of lifetime pregnancies.

Acknowledgement

The authors take full responsibility for the content of this publication, which does not necessarily reflect the official perspectives of the NIH, the NIH HEAL Initiative, or the participating institutions. We also wish to express our gratitude to our collaborators at the Kentucky Department of Corrections and the Kentucky Department of Behavioral Health for their contributions.

Funding

This research was supported by the JCOIN cooperative, funded by the National Institute on Drug Abuse, National Institutes of Health, through the NIH HEAL Initiative under award number UG1DA050069.

Footnotes

Disclosure statement

The authors report no conflicts of interest in this work.

Data availability

County-level data that were used for analyses in this article are publicly available in the University of Wisconsin Population Health Institute's County Health Rankings repository (www.countyhealthrankings.org) and in the University of Kentucky Center of Excellence’s Rural Health Physician Report (https://medicine.uky.edu/). Participant-level data from the clinical trial may be made available upon reasonable request to the study principal investigator, Dr. Michele Staton (mstaton@uky.edu).

References

  • 1.Douthit N, Kiv S, Dwolatzky T, Biswas S. Exposing some important barriers to health care access in the rural USA. Public Health. 2015;129(6):611–620. doi: 10.1016/j.puhe.2015.04.001 [DOI] [PubMed] [Google Scholar]
  • 2.Probst J, Eberth JM, Crouch E. Structural Urbanism Contributes To Poorer Health Outcomes For Rural America. Health Aff (Millwood). 2019;38(12):1976–1984. doi: 10.1377/hlthaff.2019.00914 [DOI] [PubMed] [Google Scholar]
  • 3.Cooper HL, Cloud DH, Freeman PR, et al. Buprenorphine dispensing in an epicenter of the U.S. opioid epidemic: A case study of the rural risk environment in Appalachian Kentucky. Int J Drug Policy. 2020;85:102701. doi: 10.1016/j.drugpo.2020.102701 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Health Resources and Services Administration. Health Professional Shortage Areas (HPSA) Data. U.S. Department of Health and Human Services; 2023. https://data.hrsa.gov/topics/health-workforce/shortage-areas [Google Scholar]
  • 5.Cheng L, Tan L, Zhang L, et al. Chronic disease mortality in rural and urban residents in Hubei Province, China, 2008–2010. BMC Public Health. 2013;13(1):713. doi: 10.1186/1471-2458-13-713 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Streeter RA, Zangaro GA, Chattopadhyay A. Perspectives: Using Results from HRSA ’s Health Workforce Simulation Model to Examine the Geography of Primary Care. Health Serv Res. 2017;52(S1):481–507. doi: 10.1111/1475-6773.12663 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Staton-Tindall M, Webster JM, Oser CB, Havens JR, Leukefeld CG. Drug Use, Hepatitis C, and Service Availability: Perspectives of Incarcerated Rural Women. Soc Work Public Health. 2015;30(4):385–396. doi: 10.1080/19371918.2015.1021024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Cohen EL, Wilson BR, Vanderpool RC, Collins T. Identifying Sociocultural Barriers to Mammography Adherence Among Appalachian Kentucky Women. Health Commun. 2016;31(1):72–82. doi: 10.1080/10410236.2014.936337 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Government College University, Faisalabad, Pakistan, Siddiqui S, Farah N, Government College University, Faisalabad, Pakistan, Malik N, Government College University, Faisalabad, Pakistan. CULTURAL VALUES ABOUT GENDER INEQUALITIES AND THEIR IMPLICATIONS FOR WOMEN’S HEALTH IN RURAL PUNJAB, PAKISTAN. Pak J Soc Res. 2021;03(03):44–53. doi: 10.52567/pjsr.v3i3.224 [DOI] [Google Scholar]
  • 10.Ntoimo LFC, Okonofua FE, Igboin B, Ekwo C, Imongan W, Yaya S. Why rural women do not use primary health centres for pregnancy care: evidence from a qualitative study in Nigeria. BMC Pregnancy Childbirth. 2019;19(1):277. doi: 10.1186/s12884-019-2433-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Chen X, Orom H, Hay JL, et al. Differences in Rural and Urban Health Information Access and Use. J Rural Health. 2019;35(3):405–417. doi: 10.1111/jrh.12335 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Zhang X, Dupre ME, Qiu L, Zhou W, Zhao Y, Gu D. Urban-rural differences in the association between access to healthcare and health outcomes among older adults in China. BMC Geriatr. 2017;17(1):151. doi: 10.1186/s12877-017-0538-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Bright V, Riddle J, Kerver J. Stigma Experienced by Rural Pregnant Women with Substance Use Disorder: A Scoping Review and Qualitative Synthesis. Int J Environ Res Public Health. 2022;19(22):15065. doi: 10.3390/ijerph192215065 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Cheesmond NE, Davies K, Inder KJ. Exploring the role of rurality and rural identity in mental health help-seeking behavior: A systematic qualitative review. J Rural Ment Health. 2019;43(1):45–59. doi: 10.1037/rmh0000109 [DOI] [Google Scholar]
  • 15.Hoadley A, Bass SB, Zisman-Ilani Y, et al. Combined self-stigma of justice system involvement, opioid use, and mental health disorder. Stigma Health. Published online April 7, 2025. doi: 10.1037/sah0000630 [DOI] [Google Scholar]
  • 16.Krawczyk N, Picher CE, Feder KA, Saloner B. Only One In Twenty Justice-Referred Adults In Specialty Treatment For Opioid Use Receive Methadone Or Buprenorphine. Health Aff (Millwood). 2017;36(12):2046–2053. doi: 10.1377/hlthaff.2017.0890 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Winkelman TNA, Kieffer EC, Goold SD, Morenoff JD, Cross K, Ayanian JZ. Health Insurance Trends and Access to Behavioral Healthcare Among Justice-Involved Individuals—United States, 2008–2014. J Gen Intern Med. 2016;31(12):1523–1529. doi: 10.1007/s11606-016-3845-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Butsang T, McLuhan A, Keown LA, Fung K, Matheson FI. Sex differences in pre-incarceration mental illness, substance use, injury and sexually transmitted infections and health service utilization: a longitudinal linkage study of people serving federal sentences in Ontario. Health Justice. 2023;11(1):19. doi: 10.1186/s40352-023-00218-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Stewart C, Bourke SL, Green JA, et al. Healthcare challenges of incarcerated women in Australia: An integrative review. Int J Healthc. 2020;7(1):10. doi: 10.5430/ijh.v7n1p10 [DOI] [Google Scholar]
  • 20.Maruschak LM, Berzofsky M, Unangst J. Medical Problems of State and Federal Prisoners and Jail Inmates, 2011–12. US Department of Justice, Office of Justice Programs, Bureau of Justice Statistics; 2015. https://static1.squarespace.com/static/5b7ea2794cde7a79e7c00582/t/633ca5450ecf9b281c544980/1664918854311/medical+problems.pdf [Google Scholar]
  • 21.Van Boekel LC, Brouwers EPM, Van Weeghel J, Garretsen HFL. Stigma among health professionals towards patients with substance use disorders and its consequences for healthcare delivery: Systematic review. Drug Alcohol Depend. 2013;131(1–2):23–35. doi: 10.1016/j.drugalcdep.2013.02.018 [DOI] [PubMed] [Google Scholar]
  • 22.Kramlich D, Kronk R, Marcellus L, Colbert A, Jakub K. Rural Postpartum Women With Substance Use Disorders. Qual Health Res. 2018;28(9):1449–1461. doi: 10.1177/1049732318765720 [DOI] [PubMed] [Google Scholar]
  • 23.Monnat SM. Factors Associated With County-Level Differences in U.S. Drug-Related Mortality Rates. Am J Prev Med. 2018;54(5):611–619. doi: 10.1016/j.amepre.2018.01.040 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Augsburger A, Neri C, Bodenmann P, Gravier B, Jaquier V, Clair C. Assessing incarcerated women’s physical and mental health status and needs in a Swiss prison: a cross-sectional study. Health Justice. 2022;10(1):8. doi: 10.1186/s40352-022-00171-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Lu Y. Rural and Urban Differences in Gender-Sentencing Patterns of Pennsylvania. Rural Sociol. 2018;83(2):402–430. doi: 10.1111/ruso.12204 [DOI] [Google Scholar]
  • 26.Monnat SM. Factors Associated With County-Level Differences in U.S. Drug-Related Mortality Rates. Am J Prev Med. 2018;54(5):611–619. doi: 10.1016/j.amepre.2018.01.040 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Wildeman C, Wang EA. Mass incarceration, public health, and widening inequality in the USA. The Lancet. 2017;389(10077):1464–1474. doi: 10.1016/S0140-6736(17)30259-3 [DOI] [PubMed] [Google Scholar]
  • 28.Braveman P, Gottlieb L. The Social Determinants of Health: It’s Time to Consider the Causes of the Causes. Public Health Reports®. 2014;129(1_suppl2):19–31. doi: 10.1177/00333549141291S206 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Hood CM, Gennuso KP, Swain GR, Catlin BB. County Health Rankings: Relationships Between Determinant Factors and Health Outcomes. Am J Prev Med. 2016;50(2):129–135. doi: 10.1016/j.amepre.2015.08.024 [DOI] [PubMed] [Google Scholar]
  • 30.University of Wisconsin Population Health Institute. County Health Rankings & Roadmaps 2023: Data, methods, and measures. Published online 2023. [Google Scholar]
  • 31.Cuellar AE, Ramezani N, Breno A, Johnson JE, Taxman FS. Drivers of County Engagement in Criminal Justice–Behavioral Health Initiatives. Psychiatr Serv. 2022;73(6):709–711. doi: 10.1176/appi.ps.202100485 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Ramezani N, Breno AJ, Mackey BJ, et al. The relationship between community public health, behavioral health service accessibility, and mass incarceration. BMC Health Serv Res. 2022;22(1):966. doi: 10.1186/s12913-022-08306-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Knittel AK. Resolving Health Disparities for Women Involved in the Criminal Justice System. N C Med J. 2019;80(6):363–366. doi: 10.18043/ncm.80.6.363 [DOI] [PubMed] [Google Scholar]
  • 34.Syed ST, Gerber BS, Sharp LK. Traveling Towards Disease: Transportation Barriers to Health Care Access. J Community Health. 2013;38(5):976–993. doi: 10.1007/s10900-013-9681-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Staton M, Webster JM, Leukefeld C, et al. Kentucky Women’s Justice Community Opioid Innovation Network (JCOIN): A type 1 effectiveness-implementation hybrid trial to increase utilization of medications for opioid use disorder among justice-involved women. J Subst Abuse Treat. 2021;128:108284. doi: 10.1016/j.jsat.2021.108284 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed.; 2013. https://doi/book/10.1176/appi.books.9780890425596 [Google Scholar]
  • 37.NIDA. NIDA-Modified ASSIST V2.0. Published online 2009. https://nida.nih.gov/sites/default/files/pdf/nmassist.pdf [Google Scholar]
  • 38.Patridge EF, Bardyn TP. Research Electronic Data Capture (REDCap). J Med Libr Assoc. 2018;106(1). doi: 10.5195/jmla.2018.319 [DOI] [Google Scholar]
  • 39.Weinbaum CM, Sabin KM, Santibanez SS. Hepatitis B, hepatitis C, and HIV in correctional populations: a review of epidemiology and prevention. AIDS. 2005;19(Suppl 3):S41–S46. doi: 10.1097/01.aids.0000192069.95819.aa [DOI] [PubMed] [Google Scholar]
  • 40.Moradi G, Alavian SM, Gholami F, et al. Prevalence of Hepatitis B and Hepatitis C Infections among Incarcerated Individuals in Iran: A Cross-Sectional National Bio-behavioral Study in 2019. Pathogens. 2021;10(11):1522. doi: 10.3390/pathogens10111522 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Department of Agriculture US, Economic Research Service. Rural-Urban Continuum Codes. Published online January 2024. [Google Scholar]
  • 42.Howard S, Boots K, Feltner F. 2022. Kentucky Physician Update. Univ Ky Cent Excell Rural Health Workforce Rep. [Google Scholar]
  • 43.Maas CJM, Hox JJ. Sufficient Sample Sizes for Multilevel Modeling. Methodology. 2005;1(3):86–92. doi: 10.1027/1614-2241.1.3.86 [DOI] [Google Scholar]
  • 44.McNeish DM, Stapleton LM. The Effect of Small Sample Size on Two-Level Model Estimates: A Review and Illustration. Educ Psychol Rev. 2016;28(2):295–314. doi: 10.1007/s10648-014-9287-x [DOI] [Google Scholar]
  • 45.Stuart EA, Barry CL, Donohue JM, et al. Effects of accountable care and payment reform on substance use disorder treatment: evidence from the initial 3 years of the alternative quality contract: Effects of the AQC on SUD treatment. Addiction. 2017;112(1):124–133. doi: 10.1111/add.13555 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Dayal P, Chang CH, Benko WS, et al. Hospital Utilization Among Rural Children Served by Pediatric Neurology Telemedicine Clinics. JAMA Netw Open. 2019;2(8):e199364. doi: 10.1001/jamanetworkopen.2019.9364 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Yehia BR, Schranz AJ, Umscheid CA, Lo Re V. The Treatment Cascade for Chronic Hepatitis C Virus Infection in the United States: A Systematic Review and Meta-Analysis. Rizza SA, ed. PLoS ONE. 2014;9(7):e101554. doi: 10.1371/journal.pone.0101554 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Harris M, Rhodes T. Hepatitis C treatment access and uptake for people who inject drugs: a review mapping the role of social factors. Harm Reduct J. 2013;10(1):7. doi: 10.1186/1477-7517-10-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Brohan E, Evans-Lacko S, Henderson C, Murray J, Slade M, Thornicroft G. Disclosure of a mental health problem in the employment context: qualitative study of beliefs and experiences. Epidemiol Psychiatr Sci. 2014;23(3):289–300. doi: 10.1017/S2045796013000310 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Golembiewski EH, Gravholt DL, Torres Roldan VD, et al. Rural Patient Experiences of Accessing Care for Chronic Conditions: A Systematic Review and Thematic Synthesis of Qualitative Studies. Ann Fam Med. 2022;20(3):266–272. doi: 10.1370/afm.2798 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Brinkley-Rubinstein L Incarceration as a catalyst for worsening health. Health Justice. 2013;1(1):3. doi: 10.1186/2194-7899-1-3 [DOI] [Google Scholar]
  • 52.Van Boekel LC, Brouwers EPM, Van Weeghel J, Garretsen HFL. Stigma among health professionals towards patients with substance use disorders and its consequences for healthcare delivery: Systematic review. Drug Alcohol Depend. 2013;131(1–2):23–35. doi: 10.1016/j.drugalcdep.2013.02.018 [DOI] [PubMed] [Google Scholar]
  • 53.Feng XL, Pang M, Beard J. Health system strengthening and hypertension awareness, treatment and control: data from the China Health and Retirement Longitudinal Study. Bull World Health Organ. 2014;92(1):29–41. doi: 10.2471/BLT.13.124495 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Smith SA, Ansa B. A Systematic Review of Lifestyle Interventions for Chronic Diseases in Rural Communities. J Ga Public Health Assoc. 2016;5(4). doi: 10.21633/jgpha.5.404 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Willging CE, Malcoe LH, St. Cyr S, Zywiak WH, Lapham SC. Behavioral Health and Social Correlates of Reincarceration Among Hispanic, Native American, and White Rural Women. Psychiatr Serv. 2013;64(6):590–593. doi: 10.1176/appi.ps.201200120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Kulkarni SP, Baldwin S, Lightstone AS, Gelberg L, Diamant AL. Is Incarceration a Contributor to Health Disparities? Access to Care of Formerly Incarcerated Adults. J Community Health. 2010;35(3):268–274. doi: 10.1007/s10900-010-9234-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Binswanger IA, Nowels C, Corsi KF, et al. “From the prison door right to the sidewalk, everything went downhill,” A qualitative study of the health experiences of recently released inmates. Int J Law Psychiatry. 2011;34(4):249–255. doi: 10.1016/j.ijlp.2011.07.002 [DOI] [PubMed] [Google Scholar]
  • 58.Perera PS, Miller VE, Fitch KV, et al. Medicaid Expansion and Mortality Among Persons Who Were Formerly Incarcerated. JAMA Netw Open. 2024;7(9):e2429454. doi: 10.1001/jamanetworkopen.2024.29454 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Garcia-Grossman IR, Cenzer I, Steinman MA, Williams BA. History of Incarceration and Its Association With Geriatric and Chronic Health Outcomes in Older Adulthood. JAMA Netw Open. 2023;6(1):e2249785. doi: 10.1001/jamanetworkopen.2022.49785 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Weidner RR, Schultz J. Examining the Relationship Between Incarceration and Population Health: The Roles of Region and Urbanicity. Crim Justice Policy Rev. 2021;32(4):403–426. doi: 10.1177/0887403420916907 [DOI] [Google Scholar]
  • 61.Radfar SR, De Jong CAJ, Farhoudian A, et al. Reorganization of Substance Use Treatment and Harm Reduction Services During the COVID-19 Pandemic: A Global Survey. Front Psychiatry. 2021;12:639393. doi: 10.3389/fpsyt.2021.639393 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Green TC, Clarke J, Brinkley-Rubinstein L, et al. Postincarceration Fatal Overdoses After Implementing Medications for Addiction Treatment in a Statewide Correctional System. JAMA Psychiatry. 2018;75(4):405. doi: 10.1001/jamapsychiatry.2017.4614 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Moser M, Beiruti S. Healthcare Inaccessibility in Rural Indiana: A Scoping Review. J Stud Res. 2023;12(1). doi: 10.47611/jsrhs.v12i1.3923 [DOI] [Google Scholar]
  • 64.Kim SJ, Medina M, Chang J. Healthcare Utilization of Patients with Opioid Use Disorder in US Hospitals from 2016 to 2019: Focusing on Racial and Regional Variances. Clin Drug Investig. 2022;42(10):853–863. doi: 10.1007/s40261-022-01192-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

County-level data that were used for analyses in this article are publicly available in the University of Wisconsin Population Health Institute's County Health Rankings repository (www.countyhealthrankings.org) and in the University of Kentucky Center of Excellence’s Rural Health Physician Report (https://medicine.uky.edu/). Participant-level data from the clinical trial may be made available upon reasonable request to the study principal investigator, Dr. Michele Staton (mstaton@uky.edu).

RESOURCES