Frequent police contact and nonfatal overdose among people who use drugs in rural Appalachia

Javier A. Cepeda; Talia Loeb; Zach R. Salazar; Brittney Williams; Brooke Hiltz; Brandon del Pozo; Eileen V. Pitpitan; Leo Beletsky; April M. Young; Jennifer R. Havens

doi:10.1016/j.drugalcdep.2026.113188

Frequent police contact and nonfatal overdose among people who use drugs in rural Appalachia

Javier A. Cepeda
, Talia Loeb
, Zach R. Salazar
, Brittney Williams
, Brooke Hiltz
, Brandon del Pozo
, Eileen V. Pitpitan
, Leo Beletsky
, April M. Young
, Jennifer R. Havens

Research output: Contribution to journal › Article › peer-review

Abstract

Introduction: Police contact shapes subsequent health practices and outcomes among people who use drugs (PWUD). However, the relationship between police encounters with overdose remains under-researched. We evaluated how police stop-and-search was associated with nonfatal overdose and identified correlates of frequent police contact among PWUD in rural Kentucky. Methods: From May 2023 – April 2024, 774 PWUD were enrolled in the Social Networks of Appalachian People and KyOSK studies. The exposure of interest was self-reported police stop-and-search in the past 6 months (0, 1–3, ≥4 times). The primary outcome was self-reported nonfatal overdose in the past 6 months. We evaluated the association between police stops and nonfatal overdose using multivariable logistic regression, adjusting for sociodemographic, housing, polysubstance use, and mental health covariates. Results: Nearly all participants were non-Hispanic White (99%), 52% were female, and 45% used both opioids and methamphetamine in the past 6 months. Thirty-seven percent reported being stopped and searched; 17% experienced a nonfatal overdose. PWUD who were stopped and searched by police ≥ 4 times in the past six months had more than double the odds of nonfatal overdose (adjusted odds ratio: 3.19, 95% CI: 1.60, 6.32) compared to PWUD with no police contact, controlling for other factors. PWUD reporting frequent police contact were significantly more likely to be unstably housed and less likely to be on medications for opioid use disorder. Conclusion: Police stop-and-search was associated with elevated subsequent overdose risk among PWUD. Efforts to improve PWUD health should recast police encounters as opportunities for harm reduction.

Original language	English
Article number	113188
Journal	Drug and Alcohol Dependence
Volume	284
DOIs	https://doi.org/10.1016/j.drugalcdep.2026.113188
State	Published - Jul 1 2026

Bibliographical note

Publisher Copyright:
© 2026 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license. http://creativecommons.org/licenses/by-nc-nd/4.0/

Funding

This work was supported by the National Institutes of Health through the National Institute on Drug Abuse, which funds KyOSK ( R01DA055872 to AMY as principal investigator) and SNAP ( R01DA033862 and R01DA024598 to JRH as principal investigator). Funding from the National Institute of Allergy and Infectious Diseases and National Institute on Drug Abuse also supported this analysis ( R61DA060622 and R33DA060622 to JAC, EP, and LB as principal investigators and T32AI102623 which supported TL). The San Diego Center for AIDS Research provided support for the preparation of this manuscript (to EP; P30AI036214 ). We thank the KyOSK and SNAP study staff, study participants, and board members for their valuable contributions to the study. This work was supported by the National Institutes of Health through the National Institute on Drug Abuse, which funds KyOSK (R01DA055872 to AMY as principal investigator) and SNAP (R01DA033862 and R01DA024598 to JRH as principal investigator). Funding from the National Institute of Allergy and Infectious Diseases and National Institute on Drug Abuse also supported this analysis (R61DA060622 to JAC, EP, and LB as principal investigators and T32AI102623 which supported TL). The San Diego Center for AIDS Research provided support for the preparation of this manuscript (to EP; P30AI036214). This work was supported by the National Institutes of Health through the National Institute on Drug Abuse, which funds KyOSK (R01DA055872 to AMY as principal investigator) and SNAP (R01DA033862 and R01DA024598 to JRH as principal investigator). Funding from the National Institute of Allergy and Infectious Diseases and National Institute on Drug Abuse also supported this analysis (R61DA060622 and R33DA060622 to JAC, EP, and LB as principal investigators and T32AI102623 which supported TL). The San Diego Center for AIDS Research provided support for the preparation of this manuscript (to EP; P30AI036214). We thank the KyOSK and SNAP study staff, study participants, and board members for their valuable contributions to the study. This work was supported by the National Institutes of Health through the National Institute on Drug Abuse, which funds KyOSK (R01DA055872 to AMY as principal investigator) and SNAP (R01DA033862 and R01DA024598 to JRH as principal investigator). Funding from the National Institute of Allergy and Infectious Diseases and National Institute on Drug Abuse also supported this analysis (R61DA060622 to JAC, EP, and LB as principal investigators and T32AI102623 which supported TL). The San Diego Center for AIDS Research provided support for the preparation of this manuscript (to EP; P30AI036214).

Funders	Funder number
National Institutes of Health (NIH)
Author National Institute on Drug Abuse DA031791 Mark J Ferris National Institute on Drug Abuse DA006634 Mark J Ferris National Institute on Alcohol Abuse and Alcoholism AA026117 Mark J Ferris National Institute on Alcohol Abuse and Alcoholism AA028162 Elizabeth G Pitts National Institute of General Medical Sciences GM102773 Elizabeth G Pitts Peter McManus Charitable Trust Mark J Ferris National Institute on Drug Abuse	R01DA055872
National Institute of Allergy and Infectious F32-AI286447 Cydney N. Johnson Diseases National Institute of Allergy and Infectious R01AI168214 Jason W. Rosch Diseases National Institute of Allergy and Infectious P30 Cydney N. Johnson Diseases National Institute of Allergy and Infectious R00-AI166116 Christopher D. Radka Diseases National Institute of Allergy and Infectious T32-AI106700 Cydney N. Johnson Diseases National Institute of Allergy and Infectious R01AI192221 Jason W. Rosch Diseases National Inst...	R61DA060622, P30AI036214, T32AI102623
Snap Inc.	R01DA033862, R01DA024598

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Law enforcement
Overdose
Rural health

ASJC Scopus subject areas

Toxicology
Pharmacology
Psychiatry and Mental health
Pharmacology (medical)

Access to Document

10.1016/j.drugalcdep.2026.113188

Cite this

@article{edd1c0a3e54a46be851919f578abf033,

title = "Frequent police contact and nonfatal overdose among people who use drugs in rural Appalachia",

abstract = "Introduction: Police contact shapes subsequent health practices and outcomes among people who use drugs (PWUD). However, the relationship between police encounters with overdose remains under-researched. We evaluated how police stop-and-search was associated with nonfatal overdose and identified correlates of frequent police contact among PWUD in rural Kentucky. Methods: From May 2023 – April 2024, 774 PWUD were enrolled in the Social Networks of Appalachian People and KyOSK studies. The exposure of interest was self-reported police stop-and-search in the past 6 months (0, 1–3, ≥4 times). The primary outcome was self-reported nonfatal overdose in the past 6 months. We evaluated the association between police stops and nonfatal overdose using multivariable logistic regression, adjusting for sociodemographic, housing, polysubstance use, and mental health covariates. Results: Nearly all participants were non-Hispanic White (99\%), 52\% were female, and 45\% used both opioids and methamphetamine in the past 6 months. Thirty-seven percent reported being stopped and searched; 17\% experienced a nonfatal overdose. PWUD who were stopped and searched by police ≥ 4 times in the past six months had more than double the odds of nonfatal overdose (adjusted odds ratio: 3.19, 95\% CI: 1.60, 6.32) compared to PWUD with no police contact, controlling for other factors. PWUD reporting frequent police contact were significantly more likely to be unstably housed and less likely to be on medications for opioid use disorder. Conclusion: Police stop-and-search was associated with elevated subsequent overdose risk among PWUD. Efforts to improve PWUD health should recast police encounters as opportunities for harm reduction.",

keywords = "Law enforcement, Overdose, Rural health",

author = "Cepeda, \{Javier A.\} and Talia Loeb and Salazar, \{Zach R.\} and Brittney Williams and Brooke Hiltz and \{del Pozo\}, Brandon and Pitpitan, \{Eileen V.\} and Leo Beletsky and Young, \{April M.\} and Havens, \{Jennifer R.\}",

note = "Publisher Copyright: {\textcopyright} 2026 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license. http://creativecommons.org/licenses/by-nc-nd/4.0/",

year = "2026",

month = jul,

day = "1",

doi = "10.1016/j.drugalcdep.2026.113188",

language = "English",

volume = "284",

}

TY - JOUR

T1 - Frequent police contact and nonfatal overdose among people who use drugs in rural Appalachia

AU - Cepeda, Javier A.

AU - Loeb, Talia

AU - Salazar, Zach R.

AU - Williams, Brittney

AU - Hiltz, Brooke

AU - del Pozo, Brandon

AU - Pitpitan, Eileen V.

AU - Beletsky, Leo

AU - Young, April M.

AU - Havens, Jennifer R.

PY - 2026/7/1

Y1 - 2026/7/1

N2 - Introduction: Police contact shapes subsequent health practices and outcomes among people who use drugs (PWUD). However, the relationship between police encounters with overdose remains under-researched. We evaluated how police stop-and-search was associated with nonfatal overdose and identified correlates of frequent police contact among PWUD in rural Kentucky. Methods: From May 2023 – April 2024, 774 PWUD were enrolled in the Social Networks of Appalachian People and KyOSK studies. The exposure of interest was self-reported police stop-and-search in the past 6 months (0, 1–3, ≥4 times). The primary outcome was self-reported nonfatal overdose in the past 6 months. We evaluated the association between police stops and nonfatal overdose using multivariable logistic regression, adjusting for sociodemographic, housing, polysubstance use, and mental health covariates. Results: Nearly all participants were non-Hispanic White (99%), 52% were female, and 45% used both opioids and methamphetamine in the past 6 months. Thirty-seven percent reported being stopped and searched; 17% experienced a nonfatal overdose. PWUD who were stopped and searched by police ≥ 4 times in the past six months had more than double the odds of nonfatal overdose (adjusted odds ratio: 3.19, 95% CI: 1.60, 6.32) compared to PWUD with no police contact, controlling for other factors. PWUD reporting frequent police contact were significantly more likely to be unstably housed and less likely to be on medications for opioid use disorder. Conclusion: Police stop-and-search was associated with elevated subsequent overdose risk among PWUD. Efforts to improve PWUD health should recast police encounters as opportunities for harm reduction.

AB - Introduction: Police contact shapes subsequent health practices and outcomes among people who use drugs (PWUD). However, the relationship between police encounters with overdose remains under-researched. We evaluated how police stop-and-search was associated with nonfatal overdose and identified correlates of frequent police contact among PWUD in rural Kentucky. Methods: From May 2023 – April 2024, 774 PWUD were enrolled in the Social Networks of Appalachian People and KyOSK studies. The exposure of interest was self-reported police stop-and-search in the past 6 months (0, 1–3, ≥4 times). The primary outcome was self-reported nonfatal overdose in the past 6 months. We evaluated the association between police stops and nonfatal overdose using multivariable logistic regression, adjusting for sociodemographic, housing, polysubstance use, and mental health covariates. Results: Nearly all participants were non-Hispanic White (99%), 52% were female, and 45% used both opioids and methamphetamine in the past 6 months. Thirty-seven percent reported being stopped and searched; 17% experienced a nonfatal overdose. PWUD who were stopped and searched by police ≥ 4 times in the past six months had more than double the odds of nonfatal overdose (adjusted odds ratio: 3.19, 95% CI: 1.60, 6.32) compared to PWUD with no police contact, controlling for other factors. PWUD reporting frequent police contact were significantly more likely to be unstably housed and less likely to be on medications for opioid use disorder. Conclusion: Police stop-and-search was associated with elevated subsequent overdose risk among PWUD. Efforts to improve PWUD health should recast police encounters as opportunities for harm reduction.

KW - Law enforcement

KW - Overdose

KW - Rural health

UR - https://www.scopus.com/pages/publications/105037615055

UR - https://www.scopus.com/pages/publications/105037615055#tab=citedBy

U2 - 10.1016/j.drugalcdep.2026.113188

DO - 10.1016/j.drugalcdep.2026.113188

M3 - Article

C2 - 42090841

AN - SCOPUS:105037615055

SN - 0376-8716

VL - 284

JO - Drug and Alcohol Dependence

JF - Drug and Alcohol Dependence

M1 - 113188

ER -

Frequent police contact and nonfatal overdose among people who use drugs in rural Appalachia

Abstract

Bibliographical note

Funding

UN SDGs

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this