David C. Pyrooz

Restrictive housing reforms have spread across U.S. prisons, but evidence evaluating their effectiveness remains limited. The studies that do exist have focused largely on institutional outcomes, rather than investigating whether reforms change the experience of confinement. Using data from a randomized controlled trial of Oregon’s restrictive housing step-down program (SUP) that randomly assigned incarcerated men to either transfer to the SUP or remain in the business-as-usual Intensive Management Unit, we integrate intent-to-treat, local average treatment effects, and qualitative analyses to compare daily life in restrictive housing, perceptions of correctional officers, and participants’ recommendations for restrictive housing reform. Those assigned to the SUP reported more time out-of-cell and greater interaction with other prisoners, both of which were described as meaningful improvements to their daily life. Perceptions of correctional officers depended more on interpersonal interactions than housing assignment. Participants also identified greater opportunities for movement, social interaction, and meaningful programming as priorities for reform. The findings suggest that step-down programs can improve the experience of confinement without fundamentally changing the function of restrictive housing, while also underscoring the importance of evaluating reforms through a combination of institutional outcomes and prisoner lived experiences.

Background: Firearm violence in the United States is highly concentrated within specific demographic, economic, geographic, and social population groups. Prior research indicates elevated violence exposure among gang-involved individuals, but the extent and mental health implications of firearm victimization at the national level remain poorly understood. Methods: We analyzed data from a national online survey of 10,000 U.S. adults fielded in 2024. Respondents self-reported lifetime gang membership and lifetime exposure to five forms of firearm victimization: presence at a mass shooting, gun threats, being shot at without injury, accidental gunshot injury, and intentional gunshot injury. Weighted descriptive statistics compared prevalence by lifetime gang status. Multivariable logistic regression estimated adjusted odds ratios controlling for several covariates. Among victims of firearm violence, self-reported psychological impacts, including anxiety, fear, depression, panic attacks, and post-traumatic stress symptoms, were assessed. Results: People with a history of gang involvement reported substantially higher lifetime exposure to all forms of firearm victimization. After adjustment, gang involvement was associated with 2–4 times greater odds of firearm exposure across outcomes. Psychological impacts following firearm victimization were prevalent in both gang and non-gang groups. Differences in reported mental health impacts by gang status were generally modest, with relatively few statistically significant differences in adjusted models. Conclusions: Firearm victimization is highly concentrated among people with a history of gang involvement, and such exposure is associated with substantial psychological distress. Mental health impacts were broadly similar across gang-involved and non-gang victims. These findings highlight the importance of recognizing gang-involved adults as a population with disproportionate exposure to firearm violence and significant trauma-related needs within a public health framework.

Community violence intervention and prevention (CVIP) initiatives aim to develop local infrastructures that inoculate communities from violence. The purpose of this mixed methods study was to evaluate an intervention designed to facilitate disengagement from gangs and desistance from crime. An impact evaluation, based on a preregistered randomized controlled trial, was paired with a process evaluation, based on field observations and interviews, of the Gang Reduction Initiative of Denver's centerpiece intervention that uses multidisciplinary teams (MDTs) and street outreach workers to construct and implement coordinated case plans for high‐risk gang members. The results of the impact evaluation revealed that clients assigned to the intervention were less likely to self‐report perpetrating violence, yet more likely to self‐identify as gang members; judicial records revealed no differences in arrests, charges, or dispositions. The results of the process evaluation indicated that MDTs were effective in securing services for their clients. Despite their limited training and heterogeneous strategies, street outreach workers were viewed positively by their clients and were observed to prioritize behavioral over identity change. These findings are situated within long‐standing tensions and disparate approaches in using outreach workers and targeting of gang behavior and/or involvement in violence reduction strategies, along with large‐scale federal investment in CVIP.

Criminologists maintain a vested interest in hard-to-reach populations, such as active offenders, former prisoners, and affiliates of criminal enterprises. For five decades, policymakers and researchers have sought national estimates of gang activity. Traditional methods, such as surveys sampling law enforcement agencies or youth populations, have provided valuable insights into gang activity in the United States and beyond. However, these approaches face limitations, including bias, obsolescence, high costs, outdated data, and restricted scope. This study examines the potential of online opt-in panels to study gang populations. Contracting with YouGov, we administered four surveys to 13,148 respondents between January 2023 and January 2024, measuring lifetime gang membership and its correlates. Lifetime gang membership estimates ranged from 2.0% to 2.4% in custom surveys and 5.3–6.2% in omnibus surveys. While demographic and socioeconomic correlates showed mixed results, ecological and legal correlates consistently aligned with expectations based on prior research. Our findings suggest cautious optimism about the utility of online opt-in panels for studying hard-to-reach populations like gang members and advancing cross-national, comparative research in line with the Eurogang Program of Research. These panels offer advantages, such as cost-efficiency, accessibility, and timeliness, but further validation is necessary to establish their reliability and validity for population-level estimates.

Many U.S. cities witnessed both de-policing and increased crime in 2020, yet it remains unclear whether the former contributed to the latter. Indeed, much of what is known about the effects of proactive policing on crime comes from studies that evaluate highly focused interventions atypical of day-to-day policing, use cities as the unit of analysis, or cannot rule out endogeneity. This study addresses each of these issues, thereby advancing the evidence base concerning the effects of policing on crime. Leveraging two exogenous shocks presented by the onset of COVID-19 and social unrest following the murder of George Floyd, we evaluated the effects of sudden and sustained reductions in high-discretion policing on crime at the neighborhood level in Denver. Multilevel models accounting for trends in prior police activity, neighborhood structure, seasonality, and population mobility revealed mixed results. On one hand, large-scale reductions in stops and drug-related arrests were associated with significant increases in violent and property crimes, respectively. On the other hand, fewer disorder arrests did not affect crime. These results were not universal across neighborhoods. We discuss the implications of these findings in light of debates concerning the appropriate role of policing in the 21st century.

Introduction: Mass shootings, defined as incidents where 4 or more people are shot with a firearm, have become a significant public health concern in the US.Objective: To estimate the prevalence of direct exposure to mass shootings among US adults and identify the sociodemographic groups most affected.Design, Setting, and Participants: A survey was administered in January 2024 to a sample of 10 000 respondents (18 years or older) designed to be representative of US adults using a multistage matched sampling design. In addition to exposure to mass shootings, the survey also collected sociodemographic information, including age, gender, self-reported race and ethnicity, and socioeconomic status.Exposure: Survey respondents were asked about their lifetime presence at the scene of a mass shooting, any physical injuries sustained (including being shot or trampled or experiencing a related physical injury), and the features of such incidents.Main Outcome and Measures: Multivariable logistic regression was used to generate adjusted odds ratios (AORs) of the associations between sociodemographic measures and being present on the scene of and injured in a mass shooting. All analyses undertaken to generate population and relational inferences were weighted to achieve representativeness of US adults.Results: Of the 10 000 respondents included in the analysis, 51.34% (95% CI, 50.27%-52.40%) were female. In terms of race and ethnicity, 3.04% (95% CI, 2.71%-3.38%) were Asian, 12.46% (95% CI, 11.81%-13.12%) were Black, 16.04% (95% CI, 15.10%-16.98%) were Hispanic, 62.78% (95% CI, 61.73%-63.84%) were White, and 5.67% (95% CI, 5.23%-6.11%) were other race or ethnicity. The findings indicated that 6.95% (95% CI, 6.39%-7.50%) of respondents were present at the scene of a mass shooting, and 2.18% (95% CI, 1.85%-2.50%) sustained physical injuries, such as being shot or trampled, during such incidents. A total of 54.89% of mass shootings to which respondents were exposed occurred in 2015 or more recently, and 76.15% took place in respondents’ local communities. Mass shootings were most likely to occur in neighborhoods. Younger individuals (eg, AOR for Baby Boomer and Silent generations vs Generation Z, 0.12; 95% CI, 0.09-0.18) and males (AOR, 1.55; 95% CI, 1.29-1.85) were more likely to report exposure compared with those from older generations or female individuals, respectively. Black respondents reported higher rates of being present at mass shootings (AOR, 1.87; 95% CI, 1.49-2.34), while Asian respondents reported lower rates (AOR, 0.36; 95% CI, 0.19-0.66), compared with White respondents, but there were no racial and ethnic differences in injuries sustained.Conclusions and Relevance: The findings from this survey study of US adults underscore the extensive and often overlooked regular exposure to mass shootings in this country. The demographic disparities in exposure highlight the need for targeted interventions and support for the most affected groups, particularly younger generations and males. Understanding these patterns is essential for addressing the broader impacts of gun violence on public health and community well-being.

Criminologists maintain a vested interest in hard-to-reach populations, such as active offenders, former prisoners, and affiliates of criminal enterprises. For five decades, policymakers and researchers have sought national estimates of gang activity. Traditional methods, such as surveys sampling law enforcement agencies or youth populations, have provided valuable insights into gang activity in the United States. However, these approaches face limitations, including bias, obsolescence, high costs, outdated data, and restricted scope. This study examines the potential of online opt-in panels to studying gang populations. Contracting with YouGov, we administered four surveys to 13,148 respondents between January 2023 and January 2024, measuring lifetime gang membership and its correlates. Lifetime gang membership estimates ranged from 2.0% to 2.4% in custom surveys and 5.3% to 6.2% in omnibus surveys. While demographic and socioeconomic correlates showed mixed results, ecological and legal correlates consistently aligned with expectations based on prior research. Our findings suggest cautious optimism about the utility of online opt-in panels for studying hard-to-reach populations like gang members and advancing cross-national, comparative research in line with the Eurogang Program of Research. These panels offer advantages such as cost-efficiency, accessibility, and timeliness, but further validation is necessary to establish their reliability and validity for population-level estimates.

During the last decade, health criminology—the study of health outcomes for justice‐involved individuals and their families—has gained traction in the field. We extend health criminology to the study of street gangs by drawing on the stress process perspective. Gang membership is conceptualized as a primary stressor that leads to secondary stressors with direct and indirect adverse effects on mental health. Leaving a gang, we hypothesize, offers relief by shrinking the stress universe to improve mental health. We test the gang disengagement–mental health link using panel data from a sample of 510 active gang members in the Northwestern Juvenile Project, longitudinal entropy balancing models, and mental health outcomes related to both clinical diagnosis and functional impairment. The results indicate that gang disengagement leads to improvements in mental health and functioning. Compared with those who stayed in gangs, those who left experienced improvements in global functioning, overall mental health diagnosis, behavior toward others functioning, substance abuse functioning, and alcohol‐related diagnoses. Secondary stressors partially, but not fully, mediated this association. Our findings extend the inventory of research on the benefits of disengagement from gangs to health outcomes and support interventions designed to promote gang disengagement.