Data Science Approaches in Criminal Justice and Public Health Research: Lessons Learned From Opioid Projects

• Published date: 01 May 2021
• DOI: 10.1177/1043986221999858
• Date: 01 May 2021
• Subject Matter: Articles

https://doi.org/10.1177/1043986221999858

Journal of Contemporary Criminal Justice

2021, Vol. 37(2) 175 –191

© The Author(s) 2021

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1043986221999858

journals.sagepub.com/home/ccj

Article

Data Science Approaches

in Criminal Justice and

Public Health Research:

Lessons Learned From

Opioid Projects

Tammy L. Anderson1, Ellen A. Donnelly1,

Chris Delcher2, and Yanning Wang3

Abstract

The persistence of the nation’s opioid epidemic has called on criminal justice

and public health agencies to collaborate more than ever. This epidemiological

criminology framework highlights the surveillance of public health and safety, often

using data science approaches, to inform best practices. The purpose of our article

is to delineate the main benefits and challenges of adopting data science approaches

for epidemiological criminology partnerships, research, and policy. We offer

“lessons learned” from our opioid research in Delaware and Florida to advise future

researchers, especially those working closely with policymakers and practitioners in

translating science into impactful best practices. We begin with a description of our

projects, pivot to the challenges we have faced in contributing to science and policy,

and close with recommendations for future research, public advocacy, and practice.

Keywords

opioids, data science, criminal justice, public health

Introduction

The current U.S. opioid epidemic is now entering its third decade. The Centers for

Disease Control and Prevention (2012) established its origin in the late 1990s

1University of Delaware, Newark, SE, USA

2University of Kentucky, Lexington, KY, USA

3University of Florida, Gainesville, FL, USA

Corresponding Author:

Tammy L. Anderson, Department of Sociology & Criminal Justice, University of Delaware, Newark,

DE 19716, USA.

Email: tammya@udel.edu

999858CCJXXX10.1177/1043986221999858Journal of Contemporary Criminal JusticeAnderson et al.

research-article2021

176 Journal of Contemporary Criminal Justice 37(2)

following increases in opioid prescribing, which expanded to illegal opioids in the late

2010s. The Trump administration declared opioid abuse a public health emergency in

2017 (ONDCP 2020). Drug-related treatment admissions, arrests, and incarceration

remain high today (Centers for Disease Control and Prevention, 2020; U.S. Department

of Justice, National Institute of Corrections, 2017). The exponential growth in fatal

and nonfatal drug overdoses since the 1980s has increased the pressure to find new and

effective approaches to address this crisis (Jalal et al., 2018). The popular catch phrase

among criminal justice stakeholders—“We can’t arrest our way out of this problem”—

reflects a potential shift in policy approaches that enhance collaborations between

criminal justice and public health agencies. Efforts include training police to revive

people who have overdosed (Banta-Green et al., 2013), creating diversionary pro-

grams that send people struggling with addiction to treatment instead of court (Str eisel

et al., 2019), and adding public health staff to law enforcement agencies to increase

communication and intelligence sharing across states (HIDTA Program, 2020).

This public health-criminal justice approach, otherwise called an epidemiological

criminology framework (Akers & Lanier, 2009), presents new challenges to the aca-

demic–practitioner partnership as research, policies and interventions will invariably

have to straddle multiple domains, goals, and expectations. For example, the

Department of Health and Human Services (2015) outlined three main initiatives to

reduce opioid deaths: decrease unsafe prescribing of prescription (Rx) opioids, distrib-

ute naloxone widely, and make medication-assisted treatment (MAT) more readily

available across populations, including people in criminal justice settings (U.S.

Department of Health & Human Services, Office of the Assistant Secretary for

Planning and Evaluation, 2015).

To date, improved surveillance has helped law enforcement reduce overprescribing

from fraudulent pain management clinics (i.e., pill mills) and diversion activities (U.S.

Department of Justice, Drug Enforcement Administration, 2020) and administer nal-

oxone to overdose victims. Correctional officials are making MAT available to prison-

ers, whereas probation and parole officers are monitoring a higher portion of their

clientele in treatment (Substance Abuse and Mental Health Services Administration,

2019). The National Institute of Drug Abuse established the Justice Community

Opioid Innovation Network (JCOIN) with substantial funding to community public

health projects in criminal justice settings to treat and reduce opioid addiction (U.S.

Department of Health & Human Services, National Institutes of Health, 2019b). To

support these efforts, public health and criminal justice agencies need robust surveil-

lance data and have increasingly relied on data science to gain insights and direct

resources for this epidemic (National Center for Injury Prevention and Control Board

of Scientific Counselors, 2018).

As demands for quick, effective, and data-driven solutions mount, data science

technologies are needed for epidemiological criminology studies (Hogle, 2016; Lynch,

2018; Perdue et al., 2018). A working definition of data science highlights that it is an

“interdisciplinary field of study that draws on quantitative and analytical processes

using large scale and complex data sets” (U.S. Department of Health & Human

Services, National Institutes of Health, 2019a). Data science approaches to addressing