Abstract
Background and Objectives:
Despite elevated overdose risk in rural communities, rural/urban differences in overdose risk factors are understudied among women with criminal legal system involvement. This study examines substance use and overdose among incarcerated women.
Methods:
Women (N = 900) were randomly selected from nine Kentucky jails, screened, and interviewed as part of a larger study. Bivariate analyses were used to identify group differences.
Results:
Rural women were less likely to report pre-incarceration overdose and to receive emergency services post-overdose. Substance use also varied.
Conclusions and Scientific Significance:
Results underscore the importance of Opioid Overdose Education and Naloxone Distribution in rural communities.
INTRODUCTION
The overdose crisis continues to have a widespread impact among those involved in the US criminal legal system (CLS),1 especially women, who are more likely to experience overdose fatality compared to their male counterparts.2 This risk is further amplified for those screening positive for opioid use disorder (OUD),1 those having previously experienced a nonfatal overdose,3 and those living in rural communities.4 In rural areas, overdose-related risks are exacerbated by the lack of resources and health services common to many rural areas5 and due to stigmatizing attitudes within rural communities.4
With the number of incarcerated women growing at a rate faster than men,6 a better understanding of risk factors for overdose among CLS-involved women is imperative, including overdose experiences prior to incarceration (PTI) and how these experiences may differ in rural and urban communities. Thus, the current study aims to compare pre-incarceration overdose experiences and related risks (e.g., substance use) among a sample of rural and urban women with OUD who are currently incarcerated.
METHODS
Participants and procedures
As part of the ongoing NIDA-funded Kentucky Justice Community Opioid Innovation Network (JCOIN) cooperative project, 900 women (aged 18 and older) were recruited from nine Kentucky jails, screened for eligibility, and invited to participate in a larger study. Recruitment and procedural details can be found in Staton et al.7
Randomly selected women were provided informed consent and screened for study eligibility either onsite at the jail or via teleconference (depending on COVID-19 protocols). Eligibility criteria included: (1) a positive score on the Diagnostic Statistical Manual (DSM-5) OUD checklist (2+; American Psychiatric Association, 2013) or the opioid scales of the NIDA-modified ASSIST (4+; NIDA, 2009); (2) an anticipated a release date between 7- and 60-days post-screening interview; (3) no evidence of cognitive impairment or active psychosis; and (4) willingness to participate. Eligible, interested women were then invited to complete a baseline interview, which included demographic information, substance use, and overdose histories. Screening sessions and interviews were conducted by trained female research staff in a private setting with no jail personnel present. Data were collected using Research Electronic Data Capture (REDCap), and participants were paid $45 for their time. All research activities were approved by the University of Kentucky IRB.
Measures
Demographics
Demographic information collected from participants included age (years), race/ethnicity (non-Hispanic White/other race or ethnicity), high school diploma/GED (yes/no), employment PTI (employed at least part-time/unemployed), marital status (married, living as married/not married, living as married), and sexuality (heterosexual/another sexuality). County of residence PTI was also collected and collapsed into two categories using the 2013 Rural-Urban Continuum Codes (United States Department of Agriculture Economic Research Service, 2020); codes 1 through 3 were recoded as urban and codes 4 through 9 were rural.
Substance use
Self-reported substance use information was dichotomized and included 90-day PTI injection drug use (IDU) and use of alcohol, marijuana, heroin, fentanyl, diverted buprenorphine (e.g., Suboxone®), other opioids, cocaine, methamphetamine/amphetamines, benzodiazepines, and gabapentin/neurontin (yes/no).
Overdose experiences
Women were asked about their overdose experiences, including lifetime overdose and overdose in the 90 days PTI (yes/no). Women who overdosed in the 90 days PTI were asked about substance(s) used in the 4 h preceding their overdose and services received post-overdose (e.g., administered naloxone or received emergency medical services [EMS]) was collected. Finally, if applicable, they were asked who administered naloxone (e.g., friend, family, or paramedic).
Analytic plan
After determining that continuous variables were asymmetric, χ2 and Mann–Whitney U-tests were used to identify differences across rural (n = 581; 65.4%) and urban (n = 308; 34.6%) women’s demographics, substance use, and overdose experiences. Eleven participants were excluded from the final sample due to missing data on at least one variable of interest. Analyses were conducted using IBM SPSS Statistics for Windows, Version 28 (IBM Corp.).
RESULTS
Women averaged 37 years old, and most were non-Hispanic white (92.5%), unemployed (76.3%), identified as heterosexual (72.7%), and had a high school diploma (73.7%), while 35.9% reported being married or living as married. As presented in Table 1, rural women were significantly more likely to be Non-Hispanic White (χ2[1, N = 889] = 15.59, p < 0.001) and to identify as heterosexual (χ2[1, N = 889] = 5.49, p = 0.019) but were less likely to be employed (χ2[1, N = 889] = 7.84, p = 0.005).
TABLE 1.
Group differences based on rurality of county of residence before jail incarceration (N = 889).
| Rural (n = 581) %/Mean (SD) |
Urban (n = 308) %/Mean (SD) |
p-value | |
|---|---|---|---|
| Demographics | |||
| Non-Hispanic White*** | 95.0% | 87.7% | <0.001 |
| Age (years) | 37.6 (8.8) | 36.9 (8.2) | 0.272 |
| Heterosexual* | 75.2% | 67.9% | 0.019 |
| Employed at least part-time** | 20.8% | 29.2% | 0.005 |
| Married or living as married | 36.1% | 35.4% | 0.823 |
| High school diploma/GED | 74.2% | 72.7% | 0.639 |
| Substance use (90 days PTI) | |||
| Used alcohol*** | 27.5% | 46.4% | <0.001 |
| Used marijuana | 62.5% | 61.7% | 0.817 |
| Used heroin | 64.2% | 69.2% | 0.138 |
| Used fentanyl*** | 52.7% | 64.9% | <0.001 |
| Used diverted buprenorphine (e.g., Suboxone®)*** | 40.6% | 26.3% | <0.001 |
| Used other opioids* | 77.5% | 70.8% | 0.028 |
| Used cocaine*** | 21.5% | 31.5% | 0.001 |
| Used methamphetamine/amphetamines*** | 89.2% | 78.6% | <0.001 |
| Used benzodiazepines | 41.1% | 43.8% | 0.439 |
| Used gabapentin/Neurontin*** | 45.8% | 29.9% | <0.001 |
| Injected drugs* | 59.9% | 51.3% | 0.014 |
| Received prescribed MOUD | 14.5% | 14.6% | 0.951 |
| Overdose experiences | |||
| Overdosed in lifetime | 52.8% | 56.8% | 0.257 |
| Overdosed in 90 days PTI* | 17.2% | 23.4% | 0.027 |
| Of those who overdosed in 90 days PTI: | (n = 100) | (n = 72) | |
| Number of overdoses | 2.0 (2.0) | 2.4 (1.9) | 0.111 |
| Drugs taken in the 4 h before overdose | |||
| Heroin | 80.0% | 80.6% | 0.928 |
| Fentanyl | 69.0% | 73.6% | 0.511 |
| Methamphetamine/amphetamine | 38.0% | 43.1% | 0.505 |
| Anti-anxiety meds/benzodiazepines | 15.0% | 11.1% | 0.460 |
| Other opioids | 10.0% | 8.3% | 0.710 |
| Marijuana | 6.0% | 8.3% | 0.553 |
| Powder or crack cocaine | 3.0% | 5.6% | 0.403 |
| Number of times following an overdose participant: | |||
| Was administered naloxone | 2.0 (2.7) | 2.3 (2.0) | 0.270 |
| Received emergency medical services* | 0.4 (0.7) | 0.6 (0.9) | 0.045 |
| Went to the emergency department | 0.3 (0.7) | 0.5 (0.9) | 0.269 |
| Referred to substance use treatment by first responder/hospital staff | 0.2 (0.5) | 0.2 (0.6) | 0.589 |
| Admitted to hospital | <0.1 (0.2) | 0.1 (0.3) | 0.942 |
| For those who were administered naloxone, who administered it: | (n = 95) | (n = 65) | |
| Friend | 50.5% | 52.3% | 0.825 |
| Paramedic* | 16.8% | 33.8% | 0.013 |
| Significant other | 23.2% | 16.9% | 0.338 |
| Other family member (e.g., parent, child) | 4.2% | 6.2% | 0.580 |
| Running partnera | 5.3% | 4.6% | 0.854 |
| ER doctor | 3.2% | 6.2% | 0.363 |
| Police | 4.2% | 4.6% | 0.902 |
| Stranger | 3.2% | 3.1% | 0.977 |
Note: Bold values indicate a statistically significant difference with a p-value of less than .05.
p ≤ 0.05;
p ≤ 0.01;
p ≤ 0.001.
Refers to someone participant frequently used or procured drugs with.
Substance use in the 90 days PTI also varied by rurality. In the 90 days PTI, rural women were significantly more likely to report IDU (χ2[1, N = 889] = 6.07, p = 0.014) and using diverted buprenorphine (χ2[1, N = 889] = 17.99, p < 0.001), other opioids (χ2[1, N = 889] = 4.80, p = 0.028), methamphetamine/amphetamines (χ2[1, N = 889] = 18.18, p < 0.001), and gabapentin/Neurontin (χ2[1, N = 889] = 21.19, p < 0.001). Urban women were more likely to report using alcohol (χ2[1, N = 889] = 31.97, p < 0.001), fentanyl (χ2[1, N = 889] = 12.35, p < 0.001), and cocaine (χ2[1, N = 889] = 10.70, p = 0.001) in the 90 days PTI (see Table 1).
More than half of the sample (54.2%) reported a lifetime overdose, and nearly one in five women reported overdosing in the 90 days PTI (19.3%). Among rural women, 17.2% (n = 100) reported having overdosed at least once in the 90 days PTI, though this was more commonly reported by urban women (23.4%; χ2[1, N = 889] = 4.90, p = .027). Neither lifetime overdose rates nor the number of overdose events in the 90 days PTI (among those who experienced overdose) differed by group. Among those reporting a nonfatal overdose in the 90 days PTI (N = 172), there were no differences in the types of drugs taken in the 4 h preceding an overdose, but there was variability in the services received post-overdose. Urban women more often received EMS (z = −2.0, p = 0.045) and were more likely to be administered naloxone by a paramedic (χ2[1, N = 889] = 6.16, p = 0.013). There were no differences in naloxone administration by friends, family, or significant others.
DISCUSSION
Existing literature has documented the elevated risk of overdose for CLS-involved women and those residing in rural communities.2,4 However, our knowledge of women’s overdose experiences and risk aRefers to someone participant frequently used or procured drugs with. behavior before incarceration is limited, especially regarding how these differ by geographic location. While urban women were more likely to report having overdosed in the 90 days PTI, this may be related to the higher rates of fentanyl use reported by urban women in the sample. Prior work has identified recent increases in fentanyl-related overdoses in urban areas.8 Nonetheless, study results showed that nearly 1 in 5 rural women overdosed in the 90 days PTI, and they had higher rates of IDU compared to urban women. Despite differing substance use patterns PTI, results also indicate that both groups of women may be engaging in high-risk polysubstance use. A known risk factor for overdose,2 future research and interventions should consider how unique polysubstance use patterns of rural and urban women may impact overdose risk.
Findings also suggest that rural individuals may be less likely to receive EMS post-overdose, and while the difference in EMS receipt is small, rural women are also less likely to be administered naloxone by a paramedic. This may be due to rural individuals’ reluctance to call EMS because of perceived community stigma and fear of arrest,4 even when protected by Good Samaritan Laws (as in Kentucky), or it may be linked to the dearth of services or challenges associated with rural EMS (e.g., travel distances and volunteer-based staff).5 This, coupled with the finding that women in both groups were most likely to be administered naloxone by a friend or significant other, underscores the importance of providing Opioid Overdose Education and Naloxone Distribution (OEND) throughout communities and to women reentering the community postjail release.
Finally, results provide additional evidence of the high prevalence of prior nonfatal overdose among CLS-involved women—suggesting a need for improved post-overdose response systems. Emergency and other health care services could be a critical point of referral to substance use treatment services following an overdose event, as has been found with “Quick Response Teams” and other naloxone-plus pathways of deflection.9 Initiating and ensuring access to medications to treat opioid use disorder (MOUD) among CLS-involved women during incarceration, and establishing effective links with community-based MOUD and treatment providers following release, may also help reduce overdose risk. This an important avenue for future research, particularly in rural areas where there are known barriers to MOUD.10
There are several study limitations to consider, including limited generalizability given the use of self-reported data collected from women with OUD in jail settings. Findings may not be generalizable to women in prison with different drug use patterns or those from other geographical contexts. Self-report data may be impacted by recall bias, and sensitive topics could affect participant disclosure, though study participants were ensured confidentiality and consented to participate, and interviews were conducted in a private setting by female interviewers. Additionally, information regarding the situational context of women’s overdose experiences (e.g., using alone vs. in a group) was not collected but could be important for assessing future overdose risk.
Despite limitations, this study provides a better understanding of differences in substance use and overdose experiences PTI among currently incarcerated rural and urban women. Results highlight the importance of community OEND programs and need for improved post-overdose response systems to reduce future overdose risk among CLS-involved women returning to the community post-jail release, especially in rural areas with limited substance use treatment and prevention services.
ACKNOWLEDGMENTS
Research reported in this manuscript was supported by the JCOIN cooperative, funded by the National Institute on Drug Abuse, National Institutes of Health, Bethesda, MD through the NIH HEAL Initiative. The authors gratefully acknowledge the collaborative contributions of NIDA and support from grant award UG1DA050069 (Principal Investigator: Michele Staton). The authors would also like to thank the Kentucky Department of Corrections and study participants. This research could not have been done without their collaboration. The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of NIDA, the NIH HEAL Initiative, the Kentucky Department of Corrections, or the University of Kentucky. Additionally, the principal investigator, Michele Staton, certifies that she had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Funding information
National Institute on Drug Abuse, Grant/Award Number: UG1DA050069
Footnotes
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
REFERENCES
- 1.Flam-Ross JM, Lown J, Patil P, et al. Factors associated with opioid-involved overdose among previously incarcerated people in the U.S.: a community engaged narrative review. Int J Drug Policy. 2022; 100:103534. doi: 10.1016/j.drugpo.2021.103534 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Pizzicato LN, Drake R, Domer-Shank R, Johnson CC, Viner KM. Beyond the walls: risk factors for overdose mortality following release from the Philadelphia Department of Prisons. Drug Alcohol Depend. 2018;189:108–115. doi: 10.1016/j.drugalcdep.2018.04.034 [DOI] [PubMed] [Google Scholar]
- 3.Caudarella A, Dong H, Milloy MJ, Kerr T, Wood E, Hayashi K. Nonfatal overdose as a risk factor for subsequent fatal overdose among people who inject drugs. Drug Alcohol Depend. 2016;162:51–55. doi: 10.1016/j.drugalcdep.2016.02.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Fadanelli M, Cloud DH, Ibragimov U, et al. People, places, and stigma: a qualitative study exploring the overdose risk environment in rural Kentucky. Int J Drug Policy. 2020;85:102588. doi: 10.1016/j.drugpo.2019.11.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Harrington RA, Califf RM, Balamurugan A, et al. Call to action: rural health: a presidential advisory from the American Heart Association and American Stroke Association. Circulation. 2020;141(10):e615–e644. doi: 10.1161/CIR.0000000000000753 [DOI] [PubMed] [Google Scholar]
- 6.Budd KM. Incarcerated Women and Girls. The Sentencing Project; 2024. Accessed September 26, 2024. https://www.sentencingproject.org/fact-sheet/incarcerated-women-and-girls/ [Google Scholar]
- 7.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 AbuseTreat. 2021;128:108284. doi: 10.1016/j.jsat.2021.108284 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gondré-Lewis MC, Abijo T, Gondré-Lewis TA. The opioid epidemic: a crisis disproportionately impacting Black Americans and Urban Communities. J Rac Ethnic Health Disparit. 2023;10(4):2039–2053. doi: 10.1007/s40615-022-01384-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Firesheets K, Juarez S, Kopak A, Ross J, Sperber K, Reichert J. Naloxone plus, plus some: examining Ohio’s quick response teams through the lens of deflection. J Public Health Manag Pract. 2022;28(suppl 6):S330–S338. doi: 10.1097/PHH.0000000000001570 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Staton M, Pike E, Tillson M, Lofwall MR. Facilitating factors and barriers for use of medications to treat opioid use disorder (MOUD) among justice-involved individuals in rural Appalachia. J Commun Psychol. 2024;52(8):997–1014. doi: 10.1002/jcop.23029 [DOI] [PMC free article] [PubMed] [Google Scholar]