Sage Journals: Discover world-class research

Abstract

The opioid crisis is a pressing issue with implications for the treatment of chronic pain, addiction, and criminal justice responses to substance use. This paper examines associations between prescription opioid dispensing rates and drug-related arrest rates using data from the Centers for Disease Control and the Uniform Crime Reporting System for the years 2009–2020. Fixed effects Poisson regression models are estimated to assess arrest-related outcomes. Findings suggest a positive association between opioid dispensing rates and arrest rates for possession and sale of synthetic narcotics, but a negative association with arrest rates for possession of heroin and cocaine. Future research should continue to assess measures of crime with consideration of opioid prescribing practices and policies. Policy implications include evaluation of criminal justice responses to substance use, and how diversion might be used in place of sanctions to offer treatment to those struggling with substance use disorders.

Keywords

opioids dispensing arrests drug crime

Introduction

Opioids are a class of drugs that include legal substances, such as pharmaceutical opioids like oxycodone and hydrocodone, as well as illicit substances like heroin and fentanyl (National Institute on Drug Abuse, 2024). In 2021, over 75% of fatal overdoses were opioid-related, and the number of overdose deaths attributed to opioids was ten times the number in 1999 (Hedegaard et al., 2021; Spencer et al., 2024). Recently, there has been a sharp increase in deaths attributed to fentanyl – a highly potent and lethal synthetic opioid (Volkow, 2021). While heroin and fentanyl have been especially deadly in recent years, the role of pharmaceutical opioids and the impacts of both overprescribing and limiting access to these substances is still a crucial area of study. County-level opioid dispensing rates are associated with higher rates of harmful opioid use, dependance, and frequency of use, which demonstrates the importance of continued study of the dispensing of opioid prescriptions (El Ibrahimi et al., 2023; Vuolo & Kelly, 2022). Prior research has focused strongly on associations between prescribing rates and overdose mortality, and while overdose mortality is an important and informative measure, there are other phenomena, such as arrests, that are indicative of problematic substance use that ought to be studied. These measures are important as they provide findings about the experiences of those living with substance use dependency, which mortality rates do not adequately capture.

Arrest rates can demonstrate increasing visibility of substance use and distribution of these substances, and how law enforcement responds to this (Deiana & Giua, 2021; Jewett et al., 2023; Ray et al., 2023). Arrest is typically the first stage of criminal justice system involvement and prompts interaction between police and vulnerable populations (Jewett et al., 2023; Ray et al., 2023; Shachar et al., 2020). By understanding associations between opioid dispensing rates and drug-related arrest rates, we can better understand how police respond to changes in opioid access and patterns of use. In doing so, more facets of the opioid crisis besides overdose mortality can be elucidated. Winkelman et al. (2018) found that those who use opioids are significantly more likely to have had contact with the criminal justice system within the past year, and intensity of opioid use and criminal justice involvement were significantly related. Opioid dispensing contributes to the available supply of addictive substances, which could be related to higher arrest rates for both sale and possession of these substances. Alternatively, the availability of pharmaceutical opioids might decrease demand within illicit markets because those who are dependent on opioids are able to access them through licit rather than illicit markets (Mallatt, 2022; Martin et al., 2018). This decreased activity within illicit markets may lead to lower drug-related arrest rates. Examining opioid dispensing and arrest rates provides important insight into the experiences of those who use opioids that goes beyond fatal overdose rates.

The current study examines two research questions. (1) Are opioid dispensing rates associated with arrest rates for drug possession? (2) Are opioid dispensing rates associated with arrest rates for drug sale/manufacturing? Three categories of drug arrests are assessed for both possession and sale/manufacturing - any drug, heroin/cocaine, and synthetic narcotics. I assess these questions at the county-level between 2009 and 2020 using data from the Center for Disease Control (CDC), the Uniform Crime Reporting System (UCR), the U.S. Department of Agriculture (USDA) and the American Community Survey (ACS) and estimate Poisson fixed effects regression models. This study captures an important time period of the opioid epidemic that was characterized by restrictive prescribing policies, decreases in prescription dispensing rates, and upticks in fentanyl-related mortality (Deiana & Giua, 2021; Hadland & Beletsky, 2018; Kolodny et al., 2015). The results have important implications for research on connections between opioids and crime and speak to important links between the opioid epidemic, public health practices, and drug-related crime.

Methods

Measures

Dependent Variable

Drug-Related Arrest Rates

The outcome variable of interest was county-level arrest rates for drug possession and sale/manufacturing between 2009 and 2020. These data came from the Uniform Crime Reporting System (UCR), which compiles data from over 18,000 law enforcement agencies (FBI, 2021; Kaplan, 2021). Due to concerns regarding the accuracy of county-level data (Maltz & Targonski, 2002), yearly county-level estimates of arrest rates were calculated by aggregating agency-level data at the county-level and converting these estimates to per capita rates. The UCR provides measures specific to arrests for the sale and possession of heroin/cocaine (both drugs are included in a single measure) as well as synthetic narcotics. They were selected because heroin is a type of opioid, and prior research has suggested spillover effects on heroin use when prescription opioids are restricted (Lee et al., 2021). The UCR defines synthetic narcotics as manufactured narcotics, such as methadone, Demerol, or fentanyl, that have similar addictive properties as traditional narcotics. Their list includes examples of highly potent synthetic narcotics like fentanyl as well as pharmaceutical opioids that are commonly prescribed, such as Oxycontin and Vicodin. The synthetic narcotics measures were selected because of the proliferation of fentanyl-related overdose mortality, particularly for those suffering from opioid use disorder (Volkow, 2021). In addition to the two categories of heroin/cocaine and synthetic narcotics, the “all drugs” category for both sale and possession arrests was included as well. The “all drugs” arrest measures were included to examine broad impacts on drug-related arrests, which includes any illicit substance that police can makes arrests for the sale and possession of.

In sum, six arrest-related variables are included in the current study – possession of heroin/cocaine, synthetic narcotics, and all drugs, as well as sale/manufacturing of heroin/cocaine, synthetic narcotics, and all drugs. These data are provided as count variables, which were converted to per capita rates using the county population data provided by the UCR (FBI, 2021).¹

Independent Variable

Opioid Dispensing Rate

Opioid prescriptions were operationalized through data from the CDC on the number of retail opioid prescriptions dispensed per 100 people at the county and year level from 2009 to 2020 (CDC, 2021). The data are available for 87.1–98.5% of counties depending on the year. Table 1 displays the number of prescriptions dispensed per 100,000 residents in each county, which ranges from 0 to 225,800. The maximum value represents a county in which there are more prescriptions dispensed than there are residents, which is not unreasonable given that many people receive multiple prescriptions for opioids in a given year because they need a new prescription for each refill (see Table 1).

Table 1.

Descriptive Statistics.

Variable	N	Mean	Std. Dev.	Min	Max
Dependent variables: Arrest rates
Total drug possession	24,852	316.918	237.953	0	1440.197
Heroin/Cocaine possession	19,939	47.662	56.090	0	361.459
Synthetic narcotic possession	18,975	35.905	50.880	0	362.115
Total drug sale	23,013	72.567	71.533	0	449.716
Heroin/Cocaine sale	15,975	24.314	29.499	0	212.939
Synthetic narcotic sale	12,950	16.117	24.497	0	182.482
Independent variable
Dispensing rate	34,089	74,096.250	42,801.70	0	225,800.00
Control variables
Disadvantage factor	24,852	0.021	0.964	−3.940	5.455
Percent unemployed	24,852	6.566	2.972	1.100	29.400
Percent male	24,852	0.499	0.020	0.425	0.708
Percent ages 18–24	24,852	0.091	0.035	0	0.394
Percent white	24,852	0.845	0.146	0.113	0.999
Percent foreign born	24,852	0.047	0.054	0	0.540
Percent veteran	24,852	0.077	0.022	0.01	0.188
Population density (per square mile)	24,852	184.625	604.971	0.278	14,700.150
Violent and property crime rate	24,852	752.05	779.715	0	43,199.04

All rates reported per 100,000 people.

Control Variables

The majority of the control variables were obtained through the Social Explorer resource, which provides county-level demographic data that come from sources such as the American Community Survey (ACS), the United States Census Bureau, the FBI, and the UCR. Data on measures of disadvantage (median household income, median gross rent, per capita income, and unemployment), population density, and demographics (age, sex, race, veteran, and foreign-born composition) come from five-year estimates from the ACS (U.S. Census Bureau, 2022). The data on the percentage of adults in each county with less than a high school diploma and the percentage of people in poverty come from the U.S. Department of Agriculture (USDA, 2021). These data were available for 2009 and 2017, and values for each county were held constant between 2009–2016 and 2017–2020².

Disadvantage

A disadvantage index was created based on the percentage of adults in each county with less than a high school diploma, the percentage of people in poverty, the median household income, the median gross rent, and per capita income, which is the total income of an area averaged over the entire population. The disadvantage index was created using principal components factor analysis (Eigenvalue = 3.603, factor loadings > .762, alpha = .900). The inclusion of this index is motivated by disparities in access to treatment for substance use disorders across counties based on socioeconomic status (Berkowitz et al., 2015). Furthermore, policing practices vary across communities of different socioeconomic statuses (Panditharatne et al., 2021).

Unemployment

Unemployment was operationalized as the percentage of people unemployed in each county. This measure served as an additional indicator of disadvantage given that unemployed people are not necessarily in poverty and may have other income sources outside of employment, such as through other people in their household.

Sex Composition

Sex composition was operationalized as the percentage of the population that is male in each county. Harmful opioid use has historically been higher among males, but this gap is narrowing (McHugh, 2020). The likelihood of arrest and the likelihood of testing positive for specific drugs also varies between males and females (Weller et al., 2023).

Age Composition

Criminal behavior dramatically declines by early adulthood and thus the percentage of adults in each county that is between ages 18–24 was controlled for (Farrington, 1986). Research also shows there are significant age differences in harmful opioid use patterns, and the risk of arrest is much higher for the 18–24 age group compared to other age groups (Park & Wu, 2020; Prendergast et al., 2010).

Racial Composition

This measure captured the percentage of the population that is white. Longitudinal data indicates that opioid use is higher among white people compared to Black, Hispanic, and Asian people (Schuler et al., 2021). White people also have greater access to prescription opioids, which has contributed to their higher rates of opioid use disorder (Hansen & Netherland, 2016).

Veteran Composition

The percentage of people in each county who are veterans was controlled for because this population is at risk for both physical and mental health concerns, making exposure to opioids particularly concerning for this population (Leonard Westgate et al., 2015; Seal et al., 2009). Some literature also suggests a higher risk of incarceration for veterans compared to their non-veteran counterparts, which is an important consideration in research on arrest rates (Edwards et al., 2022; Greenberg & Rosenheck, 2011).

Foreign Born Composition

The percentage of people born outside of the U.S. was controlled for. Foreign-born populations often report different experiences with police compared to their native-born counterparts (Davis & Hendricks, 2007; Hummel, 2016). This population also has unique experience in the American healthcare system, which is an important consideration for research on opioid dispensing (Adigun et al., 2021).

Violent and Property Crime Rates

Data on total violent and property crime rates comes from the Social Explorer resource, which extracts data from both the UCR and FBI (FBI, 2020). This measure was included to determine whether changes in drug-related arrests were a function of changes in arrests overall. Data was available from 2010 to 2019. To address the missing data for 2009 and 2020, the total violent and property crime rate in 2009 was held constant from the 2010 data, and the rate in 2020 was held constant from the 2019 data³. Sensitivity analyses estimated results for 2010–2019, and the results were the same as those for 2009–2020.

Population Density

Population density was included and operationalized as the number of people per square mile in each county (United States Census Bureau, 2022). This measure has been linked to differences in policing, responses to substance use, and informal social controls between densely and less densely populated areas (Battin & Crowl, 2017; Keyes et al., 2014).

Analytic Strategy

The analytic strategy had three main components. First, descriptive statistics and distributions were reported in tables and examined. Table 1 presents the descriptive statistics for all independent, dependent, and control variables. None of the control variables were highly correlated, so autocorrelation was not a concern (see Appendix A).

Second, fixed effects Poisson models were estimated. Poisson regression is suitable for analyzing outcome variables that are count outcomes that can be viewed as changes in rates over time (Osgood, 2017). The fixed effects estimator accounts for relatively stable unobserved heterogeneity that may exist across counties. County-level dispensing rates vary systematically based on political party and regional variation, which makes comparing across counties difficult. For example, there is variation in state-level opioid prescribing restrictions that directly impacts dispensing rates at the county-level (Davis & Lieberman, 2021). The fixed effects estimator treats each county as its own control and estimates the relationship between the changes in dispensing rates and the changes in drug-related arrests. Arrest rates vary depending on the season, prescribing trends overtime, and major shifts that have occurred over the last several decades. To account for time trends, year indicators were included in every model. To ensure outliers did not significantly influence the results, the top 1% of values for both the independent and dependent variables were excluded from the analysis.

Third, several sensitivity analyses were conducted. Because some agencies reported less than 12 months of data, Poisson regression models were estimated with adjusted arrest rates based on the number of months reported (see Ansari & He, 2017), excluding those that reported less than 4 months from the analysis (see Appendix B).⁴ Fixed effects negative binomial models were also estimated for all outcomes to account for overdispersion, and results are similar (see Appendix C).⁵ Models were also estimated using one year lagged dispensing rates to assess any delayed relationship between dispensing rates and drug-related arrests. Results are substantively similar to the models presented in the main analyses (see Appendix D).

Results

Opioid Dispensing Rates and Possession Arrest Rates

Tables 2 and 3 display the Poisson fixed effects models. Coefficients are presented as incident rate ratios. Overall, the results indicate that changes in opioid dispensing rates are associated with increases in some types of drug-related arrest measures and decreases in others. The tables report the relationship between a one unit increase in the number of prescriptions per 100 people and the number of arrests per 100,000 people in each county (see Tables 2 and 3). Table 2 presents the results for arrests for total drug possession, heroin/cocaine possession, and synthetic narcotic possession. There were no significant associations between changes in opioid dispensing rates and changes in total drug possession (b = .000, n.s), though several control variables emerged as significant. Model 1 reveals that increases in the percentage of males in a county is associated with increases in total drug possession rates (b = 2.098, p < .05). This suggests that a one unit increase in the percentage of males in each county is associated with a 2.098 unit difference in the expected arrest rate for possession of any drug. Population density is inversely related to the total drug possession arrest rate (b = −.001, p < .01), suggesting that a one unit increase in population density is associated with a .001 decrease in arrest rates for any drug, controlling for all other variables.

Table 2.

Poisson Models Estimating Possession Arrest Rates.

	Total drugs	Heroin/Cocaine	Synthetic narcotics
	b(se)	b(se)	b(se)
	1	2	3
Opioid dispensing rate	0.000 (0.000)	−0.002*** (0.001)	0.004*** (0.001)
Disadvantage factor	−0.016 (0.044)	−0.104+ (0.061)	−0.028 (0.103)
Percent unemployed	−0.003 (0.004)	−0.007 (0.008)	−0.004 (0.010)
Percent male	2.098* (0.991)	2.558 (1.924)	2.837 (1.840)
Percent ages 18–24	0.734 (0.767)	0.628 (1.308)	−3.644+ (1.939)
Percent white	−0.456 (0.299)	−0.672 (0.452)	1.157+ (0.680)
Percent foreign born	0.126 (0.750)	1.118 (1.394)	−2.067 (1.439)
Percent veteran	−1.010 (0.874)	−1.869 (1.810)	0.512 (1.738)
Population density	−0.001** (0.000)	−0.001*** (0.000)	−0.001* (0.001)
Violent and property crime rate	0.000 (0.000)	0.000+ (0.000)	−0.000 (0.000)
N	24,801	19,776	18,822

Robust standard errors in parentheses.

+ p < .10, * p < .05, ** p < .01, *** p < .001*.

Table 3.

Poisson Models Estimating Sale/Manufacturing Arrest Rates.

	Total drugs	Heroin/Cocaine	Synthetic narcotics
	b(se)	b(se)	b(se)
	1	2	3
Opioid dispensing rate	0.000 (0.000)	−0.001 (0.001)	0.003*** (0.001)
Disadvantage factor	−0.077 (0.071)	−0.071 (0.098)	−0.194 (0.145)
Percent unemployed	0.016* (0.006)	0.012 (0.009)	0.021 (0.013)
Percent male	−0.272 (1.333)	−2.198 (2.190)	−0.690 (2.319)
Percent ages 18–24	2.249+ (1.173)	0.205 (1.673)	4.080+ (2.248)
Percent white	0.308 (0.523)	0.773 (0.773)	0.960 (0.987)
Percent foreign born	0.101 (1.332)	4.516* (1.754)	−2.103 (2.288)
Percent veteran	−1.433 (1.307)	−1.859 (2.177)	−4.920* (2.505)
Population density	−0.000** (0.000)	−0.000** (0.000)	−0.000+ (0.000)
Violent and property crime rate	0.000 (0.000)	0.000 (0.000)	−0.000 (0.000)
N	22,921	15,720	12,643

Robust standard errors in parentheses.

+ p < .10, * p < .05, ** p < .01, *** p < .001.

Model 2 predicts heroin/cocaine possession rates. The relationship between the opioid dispensing rate and heroin/cocaine possession arrest rates is negative and significant (b = −.002, p < .001), suggesting that a one unit increase in the opioid dispensing rate (one additional opioid prescription per 100 people in a county) is associated with a .002 decrease in per capita arrest rates for possession of heroin/cocaine. This finding suggests a substitution effect whereby people who lose access to prescription opioids may substitute with other substances. Population density also had a negative association with arrest rates for heroin/cocaine possession (b = −.001, p < .001), suggesting that a one unit increase in population density is associated with a .001 decrease in arrest rates for heroin/cocaine.

Model 3 presents the results for arrest for possession of synthetic narcotics. Interestingly, unlike heroin/cocaine possession arrest rates, there is a positive relationship between the opioid dispensing rate and arrests for synthetic narcotic possession (b = .003, p < .001). That is, a one unit within county increase in opioid dispensing rates is related to a .004 within county increase in arrest rates for synthetic narcotic possession. Population density was also negatively associated with arrest rates for possession of synthetic narcotics, similar to the results of models 1 and 2 (b = −.001, p < .01).

In sum, results from Table 2 suggest that increases in opioid prescribing are associated with within-county decreases in arrest rates for heroin/cocaine possession and within-county increases in arrest rates for synthetic narcotic possession, controlling for disadvantage and demographic factors. For possession arrests, increases in percentage of males in a county tends to be associated with increases in arrest rates (only for total drug possession) and increases in population density were associated with reductions in arrest rates in all three groups of possession arrests.

Opioid Dispensing Rates and Sale/Manufacturing Arrest Rates

Table 3 presents the results for arrests for total drug sale, heroin/cocaine sale, and synthetic narcotic sale. Model 3 suggests that increases in dispensing rates were associated with increases in rates of arrest for sale of synthetic narcotics, indicating a potential amplifying effect of strong opioid prescribing (b = .003, p < .001). While there was no significant relationship between dispensing rates and total drug sale arrest rates, model 1 suggests that the unemployment rate is positively associated with total drug sale arrest rates (b = .016, p < .05). Similar to the results of Table 2, models 1 and 2 in Table 3 revealed a negative association between population density and arrest rates, and model 3 revealed a marginal negative association. Therefore, every model that included population density indicated that increases in population density were associated with within county decreases in arrest rates for both sale and possession of any drug, controlling for all other variables.

Discussion

The current study tested whether there is a relationship between opioid dispensing rates and drug-related arrest rates at the county-level using yearly fixed effects. This study is motivated by the lack of research on associations between opioid dispensing and specific kinds of drug-related arrest rates. These findings expand on prior research that has suggested that problematic opioid use is associated with outcomes such as opioid-related crime and emergency department visits (Chen et al., 2022; Deiana & Giua, 2021). The current study makes unique contributions to the literature because drug-related arrest rates capture phenomena related to policing responses to substance use, unintended consequences of opioid prescribing policies and practices, and the marginalization of people experiencing addiction. By examining county-level opioid dispensing rates and possession and sale arrest rates between 2009 and 2020, I capture a time frame in which the opioid crisis has seen major shifts in prescribing and overdose rates, criminal justice responses to the crisis, and the proliferation of dangerous substances like fentanyl (Guy et al., 2019; Hedegaard et al., 2021; Volkow, 2021).

The findings contribute to the literature on the opioid epidemic by expanding the understanding of connections between legal and illegal markets for opioids. On one hand, the inverse relationship between opioid dispensing and arrest rates for possession of heroin/cocaine suggests a potential substitution effect (Kolodny et al., 2015; Reuter et al., 2021). This extends findings from prior literature which finds restrictive opioid prescribing policies are associated with substitution for other substances (Hadland & Beletsky, 2018; Manchikanti et al., 2018). This is particularly concerning because withdrawal symptoms from opioids can be extremely painful and difficult to treat (Pergolizzi et al., 2020; Radcliffe & Parkes, 2013). If opioid dispensing rates in the U.S. continue to decrease, it will be important to understand potential unintended consequences (El Ibrahimi et al., 2023; Hadland & Beletsky, 2018). Prior research has investigated the impact of policies such as state-level restrictions on prescribing and prescription drug monitoring programs (PDMPs) (e.g., Deiana & Giua, 2021; Lee et al., 2021; Pacula & Powell, 2018) and future research should continue to assess these policies with specific attention on opioid dispensing rates.

On the other hand, there is some evidence of a positive relationship between opioid dispensing and arrest rates for both possession and sale/manufacturing of synthetic narcotics. These findings provide suggestive evidence of diversion of prescription opioids into illicit markets, and the positive relationships for both possession and sale/manufacturing suggest that police are focused on both the supply and demand aspects of this issue. Prior literature has found a positive association between opioid dispensing rates and property crime rates (Giles & Malcolm, 2021), and this research provides commentary on drug specific arrests. An important future direction is whether arrest rates influence opioid-related overdose rates. This would expand knowledge of connections between police drug interventions and public health, which is crucial to understanding how police contact affects individuals experiencing addiction (Deiana & Giua, 2021; Jewett et al., 2023; Ray et al., 2023). Future research should also investigate if continued decreases in opioid dispensing will lead to the proliferation of novel substances, increased prevalence of counterfeit prescription opioids, or higher overdose mortality from potent substances such as fentanyl, which is something both researchers and law enforcement officials have expressed great concern about (Hadland & Beletsky, 2018; Manchikanti et al., 2018; Pergolizzi et al., 2020).

The results of this study should be interpreted in light of several limitations. First, participation in the UCR is not nationally mandated; however, many states require law enforcement to provide data. Nonetheless, some counties are missing from the analysis and interpretation of the findings should be limited to counties in the analyses. Second, the UCR lists many substances within their categories for drug-related arrests. Heroin and cocaine are grouped together as one measure, and the synthetic narcotics category includes substances such as methadone, Demerol, fentanyl, and pharmaceutical opioids. Future research could link dispensing rates with arrests using the National Incident-Based Reporting System (NIBRS), which contains details about each single crime incident. The analysis in the current study utilized UCR data due to inconsistent coverage of counties in NIBRS, particularly in earlier years. Finally, there is a potential for spuriousness and the omitted variable bias. While the within county estimation reduces unobserved heterogeneity and a number of time varying covariates are accounted for, additional factors such as prevalence of manual labor jobs, prior prescribing trends, and presence of pain management clinics could be considered in future analyses, although data on these measures is limited. In this same vein, the data cannot speak to which drugs are prioritized by law enforcement, which could shape arrest statistics. Despite these limitations, this study provides an important assessment of associations between opioid prescribing practices, policing responses to substance use, and changes over time.

There are several policy implications of these findings. First, it is important to consider the potential distribution of prescription opioids in illicit markets when creating prescribing policies. Prescribing guidelines outlined by the CDC should be followed, and excessive prescribing should be avoided in order to ensure the well-being of both the patient and other members of the community (Howard et al., 2018; Morris et al., 2019; Weisberg et al., 2014). Second, the findings suggest a potential substitution effect in which people substitute prescription opioids for illicit substances when access to prescription opioids is restricted, and the expansion of medication-assisted treatment could potentially help those experiencing substance dependance and withdrawal (Homer & Wakeland, 2021; Meyers et al., 2025). Finally, the results suggest connections between opioid dispensing and interactions between police and people experiencing addiction, demonstrating the importance of expanding harm reduction practices in the criminal justice system, such as requiring police to carry naloxone, offering treatment as an alternative to incarceration, and offering medication assisted treatment in incarceration settings (Caulkins et al., 2021; Winkelman et al., 2018). This is especially important when considering that forced detoxification is common in correctional settings, and only .8% of correctional institutions in the U.S. offer medication-assisted treatment (Giftos & Tesema, 2018).

Conclusion

The opioid crisis is a pressing and dynamic issue that has led to hundreds of thousands of overdose-related deaths as well as incarceration of those experiencing addiction. While prescribing rates have significantly decreased in recent years, the effects of prescribing opioids for chronic pain that began in the 1990s are still apparent today. Even as prescribing rates decrease, there may still be unintended consequences regarding the supply and demand of opioids within illicit markets. When researchers uncover these unintended consequences, policymakers are better equipped to put structures and systems in place that address components of the opioid crisis that are not as visible or widely discussed as other components. Understanding as many facets of the opioid crisis as possible is the first step in attempting to implement innovative solutions that minimize harm and maximize care and treatment options for people struggling with addiction.

Supplemental Material

Supplemental Material - From Pain Management to Street Crime: A County-Level Analysis of Opioid Dispensing Rates and Drug-Related Arrest Rates

Supplemental Material for From Pain Management to Street Crime: A County-Level Analysis of Opioid Dispensing Rates and Drug-Related Arrest Rates by Anna Newell in Journal of Drug Issues

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Anna Newell

Data Availability Statement

All data are publicly available and cited in the references section. Concatenated data and code are available upon request.*

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biography

Anna Newell is a fourth year Ph.D. student in Criminology at Pennsylvania State University, University Park. She received a bachelor's degree in Criminology/Criminal Justice and Psychology from the University of Maryland, College Park and a master's degree in Criminology from Pennsylvania State University. Her research focuses on substance use, drug policy, illicit markets, and decision making.

References

Adigun

Barroso

Mixer

Myers

Anderson

(2021). Minding the gaps: Health care access for foreign-born people in the US: An integrative review. Journal of Health Care for the Poor and Underserved, 32(4), 1653–1674. https://doi.org/10.1353/hpu.2021.0158

Allison

P. D.

Waterman

R. P.

(2002). 7. Fixed-effects negative binomial regression models. Sociological Methodology, 32(1), 247–265. https://doi.org/10.1111/1467-9531.00117

Ansari

(2017). Explaining the UCR-NCVS convergence: A time series analysis. Asian Journal of Criminology, 12(1), 39–62. https://doi.org/10.1007/s11417-016-9236-3

Battin

J. R.

Crowl

J. N.

(2017). Urban sprawl, population density, and crime: An examination of contemporary migration trends and crime in suburban and rural neighborhoods. Crime Prevention and Community Safety, 19(2), 136–150. https://doi.org/10.1057/s41300-017-0020-9

Berkowitz

S. A.

Traore

C. Y.

Singer

D. E.

Atlas

S. J.

(2015). Evaluating area‐based socioeconomic status indicators for monitoring disparities within health care systems: Results from a primary care network. Health Services Research, 50(2), 398–417. https://doi.org/10.1111/1475-6773.12229

Caulkins

J. P.

Gould

Pardo

Reuter

Stein

B. D.

(2021). Opioids and the criminal justice system: New challenges posed by the modern opioid epidemic. Annual Review of Criminology, 4(1), 353–375. https://doi.org/10.1146/annurev-criminol-061020-125715

Center for Disease Control and Prevention (CDC) (2021). U.S. County Opioid Dispensing Rates [dataset]. Retrieved September 20, 2023 from. https://archive.cdc.gov/www_cdc_gov/drugoverdose/rxrate-maps/county2020.html

Chen

Sterner

Segel

Feng

(2022). Trends in opioid-related crime incidents and comparison with opioid overdose outcomes in the United States. International Journal of Drug Policy, 101, 103555. https://doi.org/10.1016/j.drugpo.2021.103555

Davis

C. S.

Lieberman

A. J.

(2021). Laws limiting prescribing and dispensing of opioids in the United States, 1989–2019. Addiction, 116(7), 1817–1827. https://doi.org/10.1111/add.15359

10.

Davis

R. C.

Hendricks

N. J.

(2007). Immigrants and law enforcement: A comparison of native-born and foreign-born Americans' opinions of the police. International Review of Victimology, 14(1), 81–94. https://doi.org/10.1177/026975800701400105

11.

Deiana

Giua

(2021). The intended and unintended effects of opioid policies on prescription opioids and crime. The B.E. Journal of Economic Analysis & Policy, 21(2), 751–792. https://doi.org/10.1515/bejeap-2020-0253

12.

Edwards

E. R.

Dichiara

Gromatsky

Tsai

Goodman

Pietrzak

(2022). Understanding risk in younger veterans: Risk and protective factors associated with suicide attempt, homelessness, and arrest in a nationally representative Veteran sample. Military Psychology: The Official Journal of the Division of Military Psychology, American Psychological Association, 34(2), 175–186. https://doi.org/10.1080/08995605.2021.1982632

13.

El Ibrahimi

Hendricks

M. A.

Little

Ritter

G. A.

Flores

Loy

Wright

Weiner

S. G.

(2023). The association between community social vulnerability and prescription opioid availability with individual opioid overdose. Drug and Alcohol Dependence, 252, 110991. https://doi.org/10.1016/j.drugalcdep.2023.110991

14.

Farrington

D. P.

(1986). Age and crime. Crime and Justice, 7, 189–250. https://doi.org/10.1086/449114

15.

Federal Bureau of Investigation (FBI) (2020). Total violent and property crimes rate [dataset]. In SocialExplorer.com. FBI crime data 2019. Retrieved November 2, 2023 from. https://www.socialexplorer.com/d762cef95c/explore

16.

Federal Bureau of Investigation (FBI) (2021). FBI Crime Data Explorer [dataset]. https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/home

17.

Giftos

Tesema

(2018). When less is more: Reforming the criminal justice response to the opioid epidemic. The Judges’ Journal, 57(1), 28.

18.

Giles

Malcolm

(2021). Prescription opioid misuse and property crime. Social Science Quarterly, 102(2), 663–682. https://doi.org/10.1111/ssqu.12945

19.

Greenberg

G. A.

Rosenheck

R. A.

(2011). Incarceration among male veterans: Relative risk of imprisonment and differences between veteran and nonveteran inmate. International Journal of Offender Therapy and Comparative Criminology, 56(4), 646–667. https://doi.org/10.1177/0306624X11406091

20.

Guy

G. P.

Zhang

Schieber

L. Z.

Young

Dowell

(2019). County-level opioid prescribing in the United States, 2015 and 2017. JAMA Internal Medicine, 179(4), 574–576. https://doi.org/10.1001/jamainternmed.2018.6989

21.

Hadland

S. E.

Beletsky

(2018). Tighter prescribing regulations drive illicit opioid sales. BMJ, 361, k2480. https://doi.org/10.1136/bmj.k2480

22.

Hansen

Netherland

(2016). Is the prescription opioid epidemic a white problem? American Journal of Public Health, 106(12), 2127–2129. https://doi.org/10.2105/AJPH.2016.303483

23.

Hedegaard

Miniño

A. M.

Spencer

M. R.

Warner

(2021). Drug overdose deaths in the United States, 1999–2020. National Center for Health Statistics. [dataset]. NCHS Data Brief, no 428. https://doi.org/10.15620/cdc:112340external_icon

24.

Homer

Wakeland

(2021). A dynamic model of the opioid drug epidemic with implications for policy. The American Journal of Drug and Alcohol Abuse, 47(1), 5–15. https://doi.org/10.1080/00952990.2020.1755677

25.

Howard

Waljee

Brummett

Englesbe

Lee

(2018). Reduction in opioid prescribing through evidence-based prescribing guidelines. JAMA Surgery, 153(3), 285–287. https://doi.org/10.1001/jamasurg.2017.4436

26.

Hummel

(2016). Immigrant-friendly and unfriendly cities: Impacts on the presence of a foreign-born population and city crime. Journal of International Migration and Integration, 17(4), 1211–1230. https://doi.org/10.1007/s12134-015-0464-7

27.

Jewett

P. I.

Gangnon

R. E.

Hing

A. K.

Henning-Smith

Chantarat

Areba

E. M.

Borowsky

I. W.

(2023). Racial arrest disparities in the USA by rural-urban location and region. Journal of Racial and Ethnic Health Disparities, 1-23(4), 2355–2377. https://doi.org/10.1007/s40615-023-01703-5

28.

Kaplan

(2021). Jacob Kaplan’s concatenated files: Uniform crime reporting (UCR) program data: Arrests by age, sex, and race, 1974–2020 [dataset]. Inter-University Consortium for Political and Social Research [distributor]. https://doi.org/10.3886/E102263V14

29.

Keyes

K. M.

Cerdá

Brady

J. E.

Havens

J. R.

Galea

(2014). Understanding the rural–urban differences in nonmedical prescription opioid use and abuse in the United States. American Journal of Public Health, 104(2), e52–e59. https://doi.org/10.2105/AJPH.2013.301709

30.

Kolodny

Courtwright

D. T.

Hwang

C. S.

Kreiner

Eadie

J. L.

Clark

T. W.

Alexander

G. C.

(2015). The prescription opioid and heroin crisis: A public health approach to an epidemic of addiction. Annual Review of Public Health, 36(1), 559–574. https://doi.org/10.1146/annurev-publhealth-031914-122957

31.

Lee

Zhao

Yang

K. C.

Ahn

Y. Y.

Perry

B. L.

(2021). Systematic evaluation of state policy interventions targeting the US opioid epidemic, 2007–2018. JAMA Network Open, 4(2), e2036687. https://doi.org/10.1001/jamanetworkopen.2020.36687

32.

Leonard Westgate

Shiner

Thompson

Watts

B. V.

(2015). Evaluation of veterans’ suicide risk with the use of linguistic detection methods. Psychiatric Services, 66(10), 1051–1056. https://doi.org/10.1176/appi.ps.201400283

33.

Mallatt

(2022). Policy-induced substitution to illicit drugs and implications for law enforcement activity. American Journal of Health Economics, 8(1), 30–64. https://doi.org/10.1086/716462

34.

Maltz

M. D.

Targonski

(2002). A note on the use of county-level UCR data. Journal of Quantitative Criminology, 18(3), 297–318. https://doi.org/10.1023/A:1016060020848

35.

Manchikanti

Sanapati

Benyamin

R. M.

Atluri

Kaye

A. D.

Hirsch

J. A.

(2018). Reframing the prevention strategies of the opioid crisis: Focusing on prescription opioids, fentanyl, and heroin epidemic. Pain Physician, 21(4), 309–326. https://doi.org/10.36076/ppj.2018.4.309

36.

Martin

Cunliffe

Décary-Hétu

Aldridge

(2018). Effect of restricting the legal supply of prescription opioids on buying through online illicit marketplaces: Interrupted time series analysis. BMJ, 361. https://doi.org/10.1136/bmj.k2270

37.

McHugh

R. K.

(2020). The importance of studying sex and gender differences in opioid misuse. JAMA Network Open, 3(12), e2030676. https://doi.org/10.1001/jamanetworkopen.2020.30676

38.

Meyers

Herman

Schuler

Mun

Bresani

Payne

R. K.

(2025). The opioid epidemic in rural communities: Can telehealth increase access to medications for opioid use disorder and offset barriers to care? Drug and Alcohol Dependence, 271, 112628. https://doi.org/10.1016/j.drugalcdep.2025.112628

39.

Morris

M. C.

Marco

Bailey

Ruiz

Goodin

(2019). Opioid prescription rates and risk for substantiated child abuse and neglect: A Bayesian spatiotemporal analysis. Drug and Alcohol Dependence, 205, 107623. https://doi.org/10.1016/j.drugalcdep.2019.107623

40.

National Institute on Drug Abuse (NIDA) . (2024). Opioids. Retrieved from. https://nida.nih.gov/research-topics/opioids

41.

Osgood

D. W.

(2017). Poisson-based regression analysis of aggregate crime rates. In Quantitative methods in criminology (pp. 577–599). Routledge.

42.

Pacula

R. L.

Powell

(2018). A supply-side perspective on the opioid crisis. Journal of Policy Analysis and Management, 37(2), 438–446. https://doi.org/10.1002/pam.22049

43.

Panditharatne

Chant

Sibley

C. G.

Osborne

(2021). At the intersection of disadvantage: Socioeconomic status heightens ethnic group differences in trust in the police. Race and Justice, 11(2), 160–182. https://doi.org/10.1177/2153368718796119

44.

Park

J. Y.

L. T.

(2020). Sources of misused prescription opioids and their association with prescription opioid use disorder in the United States: Sex and age differences. Substance Use & Misuse, 55(6), 928–936. https://doi.org/10.1080/10826084.2020.1713818

45.

Pergolizzi

J. V.

Jr. Raffa

R. B.

Rosenblatt

M. H.

(2020). Opioid withdrawal symptoms, a consequence of chronic opioid use and opioid use disorder: Current understanding and approaches to management. Journal of Clinical Pharmacy and Therapeutics, 45(5), 892–903. https://doi.org/10.1111/jcpt.13114

46.

United States Census Bureau . (2022). Population Density (Per Sq. Mile) [Map] ACS 2009-2020 (5-year estimates). In SocialExplorer.com. [dataset]. Retrieved November 2, 2023 from. https://www.socialexplorer.com/a9676d974c/explore

47.

Prendergast

Huang

Evans

Hser

Y. I.

(2010). Are there gender differences in arrest trajectories among adult drug abuse treatment participants? Journal of Drug Issues, 40(1), 7–26. https://doi.org/10.1177/002204261004000102

48.

Radcliffe

Parkes

(2013). The politics of providing opioid pharmacotherapy. International Journal of Drug Policy, 24(6), e6–e10. https://doi.org/10.1016/j.drugpo.2013.09.009

49.

Ray

Korzeniewski

S. J.

Mohler

Carroll

J. J.

Del Pozo

Victor

Huynh

Hedden

B. J.

(2023). Spatiotemporal analysis exploring the effect of law enforcement drug market disruptions on overdose, Indianapolis, Indiana, 2020–2021. American Journal of Public Health, 113(7), 750–758. https://doi.org/10.2105/AJPH.2023.307291

50.

Reuter

Pardo

Taylor

(2021). Imagining a fentanyl future: Some consequences of synthetic opioids replacing heroin. International Journal of Drug Policy, 94, 103086. https://doi.org/10.1016/j.drugpo.2020.103086

51.

Schuler

M. S.

Schell

T. L.

Wong

E. C.

(2021). Racial/ethnic differences in prescription opioid misuse and heroin use among a national sample, 1999–2018. Drug and Alcohol Dependence, 221, 108588. https://doi.org/10.1016/j.drugalcdep.2021.108588

52.

Seal

K. H.

Metzler

T. J.

Gima

K. S.

Bertenthal

Maguen

Marmar

C. R.

(2009). Trends and risk factors for mental health diagnoses among Iraq and Afghanistan veterans using department of veterans affairs health care, 2002–2008. American Journal of Public Health, 99(9), 1651–1658. https://doi.org/10.2105/AJPH.2008.150284

53.

Shachar

Wise

Katznelson

Campbell

A. L.

(2020). Criminal justice or public health: A comparison of the representation of the crack cocaine and opioid epidemics in the media. Journal of health politics, policy and law, 45(2), 211–239. https://doi.org/10.1215/03616878-8004862

54.

Spencer

M. R.

Garnett

M. F.

Miniño

A. M.

(2024). Drug overdose deaths in the united Sates, 2002-2022. National Center for Health Statistics. [dataset]. NCHS Data Brief, no 491.

55.

United States Department of Agriculture (USDA) Economic Research Service . (2021). County-level Datasets [dataset]. Retrieved from: https://www.ers.usda.gov/data-products/county-level-data-sets

56.

U.S. Census Bureau . (2022).

Social Explorer Tables: ACS 2009-2020 (5-year estimates) [dataset]. In SocialExplorer.com

. Retrieved November 2, 2023 from. https://www.socialexplorer.com/tables/ACS2020_5yr

57.

Volkow

N. D.

(2021). The epidemic of fentanyl misuse and overdoses: Challenges and strategies. World Psychiatry: Official Journal of the World Psychiatric Association (WPA), 20(2), 195–196. https://doi.org/10.1002/wps.20846

58.

Vuolo

Kelly

B. C.

(2022). Effects of county-level opioid dispensing rates on individual-level patterns of prescription opioid and heroin consumption: Evidence from national US data. American Journal of Psychiatry, 179(4), 305–311. https://doi.org/10.1176/appi.ajp.2021.21060602

59.

Weisberg

D. F.

Becker

W. C.

Fiellin

D. A.

Stannard

(2014). Prescription opioid misuse in the United States and the United Kingdom: Cautionary lessons. International Journal of Drug Policy, 25(6), 1124–1130. https://doi.org/10.1016/j.drugpo.2014.07.009

60.

Weller

B. E.

Magura

Smith

D. R.

Saxton

M. M.

Amaratunga

(2023). Gender differences in drug use among individuals under arrest. Journal of Substance Use, 28(4), 541–544. https://doi.org/10.1080/14659891.2022.2070872

61.

Winkelman

T. N.

Chang

V. W.

Binswanger

I. A.

(2018). Health, polysubstance use, and criminal justice involvement among adults with varying levels of opioid use. JAMA Network Open, 1(3), e180558. https://doi.org/10.1001/jamanetworkopen.2018.0558

62.

Wooldridge

J. M.

(1999). Distribution-free estimation of some nonlinear panel data models. Journal of Econometrics, 90(1), 77–97. https://doi.org/10.1016/S0304-4076(98)00033-5

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.08 MB