Papers & Presentations


Prescription Drug Histories among Drug Overdose Decedents in Delaware

Authors: Tammy L. Anderson, Jascha Wagner, Guanwen Qi, Steve S. MartinDaniel J. O’Connell, and Ellen Donnelly

Background: The US opioid epidemic largely featured deaths from prescribed medications during Wave 1 (1990–2010), but its progression since then has resulted more so from deaths to illegal opioids—such as heroin (Wave 2 – 2010–2013) and fentanyl (Wave 3 – 2013-present). As deaths to illegally manufactured fentanyl have increased, attention to the role of prescribed opioids may be waning. However, the shifting nature of today’s opioid epidemic demands we monitor how both legal and illegal drugs are involved in overdose deaths.

Objectives: The purpose of our studyis to investigate the prescription drug (Rx) records of overdose death decedents to illuminate the continued role of prescribed medications in Wave 3 deaths.

Methods: We matched drug overdose death data and prescription drug monitoring data to investigate the prescription drug records (i.e.types of opioids and other medications) of Delaware, USA, decedents who died from a drug over-dose death between January 1, 2013, and March 31, 2015 (27 months).

Results: Fentanyl decedents differed significantly from other decedents in prescribed medications, including the amount and proximity of opioid and Rx fentanyl prescriptions before death. These relationships held while controlling for demographic characteristics and contributing health conditions.

Conclusions: Our findings show a continued presence of Rx opioids in overdose deaths and that those dying from fentanyl had different Rx records than those who died from other drugs. Continued monitoring of Rx drugs, improved toxicology testing and greater data access for more research should follow to inform effective interventions.

Click here for the full article.

Is the Gender Gap in Overdose Deaths (Still) Decreasing? An Examination of Opioid Deaths in Delaware, 2013-2017.

Authors: Mieke C.W. EckhautJascha Wagner, Logan Neitzke-Spruill, Rebecca Walker, Tammy L. Anderson

Objective: Drug overdoses among men have historically outnumbered those among women by a large margin. Yet, U.S. research on the first wave of the opioid epidemic involving prescription opioids has found women to be at increased risk. The current study considers if the narrowing gender gap in overdose deaths, as observed during the first wave, has continued into the most recent third wave, dominated by synthetic opioid deaths. This requires consideration of interactions between gender, age, and type of drug implicated.

Method: Drawing on 2013-2017 Delaware toxicology reports for a total of 890 overdose deaths involving opioids, we distinguished between four gender/age groups–women 15-44, women 45-64, men 15-44, and men 45-64–to calculate crude death rates, male-to-female death rate ratios, and younger-to-older death rate ratios by type of opioid.

Results: Opioid overdose death rates during the third wave increased among both men (+102%) and women (+46%), but the larger increase among men resulted in an increase in the male-to-female death rate ratio (from 1.9 to 2.6). This trend was driven by the growing contribution of fentanyl (from 16% to 76%) and heroin overdose deaths (from 27% to 50%) compared with other opioid overdose deaths, which disproportionately affected men and younger individuals. Higher male-to-female death rate ratios were observed among older, compared with younger, individuals.

Conclusions: Overdose deaths seem to have returned to a historically familiar pattern of dominance by younger males. Our findings suggest the gender-age distribution in deaths to specific opioid types must be considered for effective intervention.

Click here for the full article

Opioids, Race, and Drug Enforcement: Exploring Local Relationships Between Neighborhood Context and Black–White Opioid-Related Possession Arrests (2019)

Authors:  Ellen A. DonnellyJascha WagnerMadeline Stenger, Hannah G. Cortina, Daniel J. O’Connell, Tammy L. Anderson

Abstract: Opioid abuse has redefined drug problems in communities and shifted police activities to redress substance use. Changing neighborhood context around opioid issues may affect arrests and racial disparities in their imposition. This study presents a spatial analysis of arrests involving Blacks and Whites for possession of heroin, synthetic narcotics, and opium offenses. We identify the ecological conditions associated with opioid-related arrests using geographically weighted regression (GWR) methods that illuminate local patterns by allowing coefficients to vary across space. GWR models reveal spatial and racial differences in opioid-related possession arrest rates. Calls for police service for overdoses increase White arrests in more advantaged, rural communities. Economic disadvantage and racial diversity in neighborhoods more strongly elevate possession arrest rates among Blacks relative to Whites. Overdose calls predict Black arrests in poorer urban areas. Findings underscore police responsiveness to opioid problems and Black–White differences in how opioid users interact with the criminal justice system.

Click here for the full article

Revisiting neighborhood context and racial disparities in drug arrests under the opioid epidemic (2019)

Authors: Ellen Donnelly, Jascha Wagner, Tammy L. Anderson, and Daniel J. O’Connell

Abstract: As opioid addiction has risen in recent years, racial disparities in drug arrests may be changing in their size and sources. Neighborhood conditions, like economic disadvantage and racial composition, are powerful determinants of racial differences in arrests. Overdoses and police responses to these incidents may, however, alter the neighborhood context of drug arrests, especially those tied to heroin, synthetic narcotics, and related opium derivatives offenses. This study revisits the environmental correlates of arrest disparities by conducting a neighborhood-level analysis of Black–White differences in drug possession and selling arrests by substance type across the State of Delaware. Spatial model estimates suggest economic disadvantage and racial diversity in neighborhoods substantially increase Black arrest rates. Conversely, White arrest rates grow with more calls for service for overdose incidents, racial homogeneity, and to a lesser extent, economic disadvantage within a community. Disparities in arrest also vary by substance type, as heroin arrests for Whites are most correlated with higher overdose service calls relative to White arrests for marijuana, cocaine, and other substances or Black arrests for any substance. Results underscore the need to reexamine neighborhood conditions and arrest disparities due to emerging shifts in drug use and drug law enforcement.

Click here for the full article.

Understanding Geographic and Neighborhood Variations in Overdose Death Rates (2018)

Authors: Jascha Wagner, Logan Neitzke-Spruill, Daniel O’Connell, James Highberger, Steve S. Martin,  Rebecca Walker, & Tammy Anderson

Abstract: The current opioid epidemic continues to challenge us in new and potentially troubling ways. For example, research today finds more overdose deaths occurring in rural, rather than urban, geographic areas. Yet, studies have often ignored heterogeneities within these spaces and the neighborhood variations therein. Using geodemographic classification, we investigate neighborhood differences in overdose death rates by geographical areas to further understand where and among what groups the problem might be most concentrated. For deaths between 2013 and 2016, we find significant variation in rates among neighborhoods, defined by their socio-economic and demographic characteristics. For example, overdose death rates vary up to 13-fold among neighborhoods within geographic areas. Our results overall show that while the rural or urban classification of a geographic area is important in understanding the current overdose problem, a more segmented analysis by neighborhood’s socio-economic and demographic makeup is also necessary.

Click here for the full paper.

​Understanding Differences in Types of Opioid Prescriptions Across Time and Space: A Community-Level Analysis (2018)

Authors: Tammy Anderson, Xiaoke Zhang, Steve S. Martin, Yiqian Fang, and Jiamin Li

Abstract: For the better part of the 21st century, opioid abuse and related consequences have beleaguered the United States. Effectively fighting the crisis may require a better understanding of potential differences among the types of opioids available as treating them as one homogeneous group may mask emerging trends and conflate more benign ones with those more troubling. The purpose of our study is to investigate changes in prescribing patterns of four groups of opioids (hydrocodone, oxycodone, fentanyl, and other) and how community-level factors explain their variation over time. We use a census tract–level data set with population, concentrated disadvantage, and prescription drug monitoring payment variables to address our goals. Findings show disparate prescribing patterns among the four types of opioids and considerable differences in the community factors that predict their change. Implications for future research and interventions follow.

Click here for the full paper


Data Science Solutions to Combatting Opioid and Crime Problems

Authors: Tammy L. Anderson, Daniel O’Connell, Ellen Donnelly, Jascha Wagner, Logan Neitzke-Spruill

Abstract: In today’s data science and “big data” era, researchers across disciplines are developing innovative ways to understand and address the current opioid epidemic and its related consequences, including crime and negative health outcomes. The proposed panel features two projects—the Delaware Opioid Metric Intelligence project (DOMIP – funded by the National Institute of Justice) and the Florida Drug-Related Outcomes and Tracking System (FROST- funded by the Bureau of Justice Assistance) that adopt such approaches to combat substance abuse in Delaware and Florida. Researchers from DOMIP and FROST will discuss the scientific value in integrating data from multiple sources and development of surveillance and mapping tools for combatting the opioid problem. DOMIP and FROST presenters will briefly discuss the objectives, current research, and impacts from their projects and will then answer targeted questions from the discussant and audience members

This presentation was given by Tammy L. Anderson, Daniel O’Connel, and Ellen Donnelly alongside Chris Delcher, PhD at the 2019 meeting of the American Society for Criminology.

Developing and Using Metrics to Reduce Opioid-Related Problems in Delaware.

Authors: Tammy L. Anderson, Daniel O’Connell, and Ellen A. Donnelly

Abstract: The Delaware Opioid Metric Intelligence Project (DOMIP) provides community surveillance capabilities in Delaware to help reduce its prescription and illicit drug problems.  DOMIP achieves this by integrating data on overdose deaths, crime, population characteristics, and community resources. Such integration allows for statistical analysis and mapping of a wide range of opioid and crime-related metrics, leading to measurable impacts on drug market and drug diversion activities. DOMIP is supported by the National Institute of Justice. This project has also received support from the Delaware Criminal Justice Information System and the State of Delaware Division of Forensic Science.

This presentation was given by Ellen Donnelly at the Center for Evidence-Based Crime Policy at George Mason University in June, 2019. 

​Prescription Fentanyl’s role in fentanyl related drug overdoses: Matching overdose death data with prescription data

Authors: Tammy Anderson, Jascha Wagner, Guanwen Qi, Daniel O’ConnellSteve S. Martin

Abstract: In October 2017 the CDC estimated that Fentanyl was involved in almost 60% of all overdose deaths in the second half of 2016 (MMWR October 27th, 2017). Most reports suggest that illegally manufactured fentanyl (IMF) is the main culprit of current fentanyl-related harm, especially fentanyl overdose deaths. Yet, several new developments question the ‘benign’ effect of prescription fentanyl (PF) on the current opiate crisis. The abuse of PF patches, e.g. injection and ingestion, or freezing and sublingual absorbing, as well as the diversion of PF to the illegal market, may play an important role in the havoc wreaked by fentanyl. We matched fentanyl overdose deaths with prescription drug data to investigate the potential role of PF in overdose deaths. We find that 44% of all Delaware fentanyl overdose decedents (between January 2013 and March 2015) had a history of PF and 37% had a prescription within 60 days of their death.  On the contrary, only about 5% of those who died from overdoses to other drugs had a history of PF with only 1% having a prescription within 60 days of their death.  Since toxicology in Delaware cannot distinguish between IMF and PF, we explore plausible explanations for its role in overdose deaths including: (1) fentanyl prescribed in combination with other drugs may produce fatal drug interactions, (2) prescriptions for fentanyl and other drugs may exacerbate existing health problems, leading to death.  We use data to explore these explanations and offer further analysis by characteristics of decedents.

This paper was presented by Dr.. Tammy Anderson at the 2018 meeting of the American Public Health Association. See the presentation slides here.

​Uneven Recovery: Exploring Relations Between the Recovery from the Great Recession and Changes to Neighborhood Level Crime Rates

Author: Jascha Wagner

Abstract: This paper addresses two timely questions: How did neighborhoods recover from the financial crisis 2007-2008? And, how are differing recovery trajectories related to social inequalities and social problems? Combining Delaware neighborhood level data (census tracts) from the Federal Housing Finance Agency, the American Community Survey, and the Delaware Criminal Justice Information System, this paper shows first that the boom, bust, and recovery of the housing market, which we find at the core of the financial crisis, followed diverging trajectories with one cluster of Delaware neighborhoods still showing only very minor signs of recovery. Second, the paper shows that this segment disproportionately involves lower income neighborhoods with larger shares of Black and Hispanic residents. Third, the paper investigates associations between the clusters and changes in crime rates during the recovery period 2012-2016. The study finds that disparities in crime rates between neighborhood clusters increased over the crisis periods.

This paper was presented at the 2018 meeting of the American Society of Criminology.