Garen J. Wintemute’s research while affiliated with University of California System and other places

Background In 2022, a nationally representative longitudinal survey in the USA found concerningly high prevalences of support for and personal willingness to engage in political violence, but those prevalences decreased in 2023. This study examines changes in those prevalences from 2023 to 2024, an election year in the USA. Methods Participants were members of Ipsos KnowledgePanel. Wave 3 of the survey was conducted May 23-June 14, 2024; invitations to participate were sent to all respondents to prior waves who remained in KnowledgePanel. Outcome measures concern justification for the use of violence to advance any of 17 specified political objectives, personal willingness to engage in political violence at 4 levels of severity and against 9 target populations, and expectation of firearm use in political violence. Outcomes are expressed as weighted proportions. Year-to-year change is based on the means of aggregated individual change scores, which have a potential range from 0 (no change) to ± 2. Results The 2024 completion rates were 88.4% (8896 respondents/10,064 invitees) overall, 91.6% (8185 respondents/8932 invitees) for invitees in 2024 who had responded in 2023, and 62.8% (711 respondents/1132 invitees) for invitees in 2024 who had responded in 2022 but not in 2023. After weighting, 50.9% (95% confidence interval (CI) 49.5%, 52.3%) were female; weighted mean (SD) age was 48.5 (24.9) years. From 2023 to 2024, the prevalence of the view that violence was usually or always justified to advance at least 1 political objective did not change (2024: 26.2%, 95% CI 25.0%, 27.5%; 2023: 25.3%, 95% CI 24.1%, 26.5%). There were no changes from 2023 to 2024 in willingness to damage property, threaten a person, injure a person, or kill a person in an act of political violence, and no changes in expectations of firearm use in situations where respondents considered political violence justifiable. Changes on other measures were infrequent (17 of 58 comparisons in the main analysis) and small where they occurred (with 2 exceptions, change < 0.05). Conclusions Contrary to expectation, support for and willingness to participate in political violence in this cohort showed little to no change from 2023 to 2024, an election year in the USA. These findings can help guide prevention efforts.

Background In recent years, the United States (US) has witnessed a rise in political violence. Prior research has found that an individual’s social network is associated with their likelihood of engaging in various forms of violence, but research on social networks and political violence in the US context is limited. This study examined associations between social network size and endorsement of political violence in a recent nationally representative survey and explored how the relationship varied by use of social media as a major news source, perceptions of the government as an enemy, and membership in a marginalized or privileged racial or ethnic group. Methods This was a nationally representative cross-sectional survey study of adults aged 18 and older in the US, administered from May 13-June 2, 2022. The exposure was social network size, defined by the number of strong social connections. We examined three violence-related outcomes: support for non-political violence, support for political violence, and personal willingness to engage in political violence. We estimated prevalence ratios for associations using survey-weighted Poisson regression with robust standard errors, adjusting for hypothesized confounders and including interaction terms to examine effect measure modification. Results The sample included 8,620 respondents. Median age was 48.4 years (95% CI = 47.9–48.8), 51.5% were female (95% CI = 50.4–52.7%), and 62.7% self-identified as non-Hispanic White (95% CI = 61.4–65.9%). In adjusted models, those with zero strong connections were more likely than those with 1–4 strong social connections to consider political violence usually/always justified in general (PR = 2.43, 95% CI = 1.47–4.01). Those with 50 + strong connections were more likely than those with 1–4 strong social connections to consider political violence usually/always justified in at least one situation (PR = 1.19, 95% CI = 1.03–1.37) and were more likely to report being willing to personally use political violence (PR = 1.52, 95% CI = 1.13–2.04). Associations varied somewhat by social media use, perceptions of the government as an enemy, and racialized identity. Conclusions Individuals who reported very few and very many strong social connections were more likely than others to support political violence or be personally willing to engage in it in one form or another. Findings point toward potential intervention and prevention opportunities.

Background: In 2022, a nationally representative longitudinal survey in the USA found concerningly high prevalences of support for and personal willingness to engage in political violence, but those prevalences decreased in 2023. This study examines changes in those prevalences from 2023 to 2024, an election year in the USA.Methods: Participants were members of Ipsos KnowledgePanel. Wave 3 of the survey was conducted May 23-June 14, 2024; invitations to participate were sent to all respondents to prior waves who remained in KnowledgePanel. Outcomes are expressed as weighted proportions. Year-to-year change is based on the means of aggregated individual change scores, which have a potential range from 0 (no change) to ±2. Results: The 2024 completion rates were 88.4% (8896 respondents/10,064 invitees) overall, 91.6% (8185 respondents/8932 invitees) for invitees in 2024 who had responded in 2023, and 62.8% (711 respondents/1132 invitees) for invitees in 2024 who had responded in 2022 but not in 2023. After weighting, 50.9% (95% confidence interval (CI) 49.5%, 52.3%) were female; weighted mean (SD) age was 48.5 (24.9) years. From 2023 to 2024, the prevalence of the view that violence was usually or always justified to advance at least 1 political objective did not change (2024: 26.2%, 95% CI 25.0%, 27.5%; 2023: 25.3%, 95% CI 24.1%, 26.5%). There were no changes from 2023 to 2024 in willingness to damage property, threaten a person, injure a person, or kill a person in an act of political violence, and no changes in expectations of firearm use in situations where respondents considered political violence justifiable. Changes on other measures were infrequent (17 of 58 comparisons in the main analysis) and small where they occurred (with 2 exceptions, change < 0.05). Conclusions: Contrary to expectation, support for and willingness to participate in political violence in this cohort showed little to no change from 2023 to 2024, an election year in the USA. These findings can help guide prevention efforts.

Background Too little is known about the effectiveness of efforts to prevent firearm violence. We evaluated California’s Armed and Prohibited Persons System (APPS), which identifies legal purchasers of firearms who have become prohibited persons and seeks to recover all firearms and ammunition to which they have access. Design and methods This cluster-randomised pragmatic trial was made possible by APPS’s expansion from a small pilot to a continuing statewide programme. We included 363 California cities, allocated to early (n=187) or later (n=176) intervention in blocks stratified by region within the state, and within region by population and violent crime rate. The study period began 1 February 2015; region-specific end dates ranged from 1 May 2015 to 1 February 2016. Analysis was on an intention-to-treat, difference-in-difference basis using generalised linear mixed models and generalised estimating equations with robust SEs. The population-level primary outcome measures were monthly city-level counts of firearm-related homicides, robberies and aggravated assaults. The primary model was adjusted for stratification variables; city-level population, population density, socioeconomic status and firearm purchasing; year; and month. Findings Allocation groups were well balanced on baseline characteristics and implementation measures. In adjusted models, allocation to early intervention was not associated with statistically significant differences in any primary outcome measure; these findings were robust to multiple sensitivity analyses. There was some heterogeneity across regions. Conclusions The APPS intervention directly affects a very small percentage of the population, limiting its potential for population-level effects. Individual-level analyses may provide a better estimate of the intervention’s effectiveness. Trial registration number NCT02318732 .

Background Extreme Risk Protection Orders (ERPOs) are civil court orders that prohibit firearm purchase and possession when someone is behaving dangerously and is at risk of harming themselves and/or others. As of June 2024, ERPOs are available in 21 states and the District of Columbia to prevent firearm violence. This paper describes the design and protocol of a six-state study of ERPO use. Methods The six states included are California, Colorado, Connecticut, Florida, Maryland, and Washington. During the 3-year project period (2020–2023), ERPO case files were obtained through public records requests or through agreements with agencies with access to these data in each state. A team of over four dozen research assistants from seven institutions coded 6628 ERPO cases, abstracting 80 variables per case under domains related to respondent characteristics, events and behaviors leading to ERPO petitions, petitioner types, and court outcomes. Research assistants received didactic training through an online learning management system that included virtual training modules, quizzes, practice coding exercises, and two virtual synchronous sessions. A protocol for gaining strong interrater reliability was used. Research assistants also learned strategies for reducing the risk of experiencing secondary trauma through the coding process, identifying its occurrence, and obtaining help. Discussion Addressing firearm violence in the U.S. is a priority. Understanding ERPO use in these six states can inform implementation planning and ERPO uptake, including promising opportunities to enhance safety and prevent firearm-related injuries and deaths. By publishing this protocol, we offer detailed insight into the methods underlying the papers published from these data, and the process of managing data abstraction from ERPO case files across the multi-state and multi-institution teams involved. Such information may also inform future analyses of this data, and future replication efforts. Registration This protocol is registered on Open Science Framework ( https://osf.io/kv4fc/ ).

Background Surveys have found concerningly high levels of agreement that the United States will experience civil war soon. This study assesses variation in expectation of and perceived need for civil war with respondent sociopolitical characteristics, beliefs, firearm ownership, and willingness to engage in political violence. Methods Findings are from Wave 2 of a nationally representative annual longitudinal survey of members of the Ipsos KnowledgePanel, conducted May 18–June 8, 2023. All respondents to 2022’s Wave 1 who remained in KnowledgePanel were invited to participate. Outcomes are expressed as weighted proportions and adjusted prevalence differences, with p-values adjusted for the false discovery rate and reported as q-values. Results The completion rate was 84.2%; there were 9385 respondents. After weighting, half the sample was female (50.7%, 95% CI 49.4%, 52.1%); the weighted mean (± standard deviation) age was 48.5 (25.9) years. Approximately 1 respondent in 20 (5.7%, 95% CI 5.1%, 6.4%) agreed strongly or very strongly that “in the next few years, there will be civil war in the United States.” About 1 in 25 (3.8%, 95% CI 3.2%, 4.4%), and nearly 40% (38.4%, 95% CI 32.3%, 44.5%) of those who strongly or very strongly agreed that civil war was coming, also agreed strongly or very strongly that “the United States needs a civil war to set things right.” Expectation of and perceived need for civil war were higher among subsets of respondents who in Wave 1 were more willing than others to commit political violence, including MAGA Republicans, persons in strong agreement with racist beliefs or statements of the potential need for violence to effect social change, persons who strongly approved of specified extreme right-wing political organizations and movements, firearm owners who purchased firearms in 2020 or later, and firearm owners who carried firearms in public all or nearly all the time. Conclusions In 2023, the expectation that civil war was likely and the belief that it was needed were uncommon but were higher among subsets of the population that had previously been associated with greater willingness to commit political violence. These findings can help guide prevention efforts.

More than half of suicide deaths in the United States result from self-inflicted firearm injuries. Extreme risk protection order (ERPO) laws in 21 states and the District of Columbia temporarily limit access to firearms for individuals found in a civil court process to pose an imminent risk of harm to themselves or others. Research with large multistate study populations has been lacking to determine effectiveness of these laws. This study assembled records pertaining to 4,583 ERPO respondents in California, Connecticut, Maryland, and Washington. Matched records identified suicide decedents and self-injury method. Researchers applied case fatality rates for each suicide method to estimate nonfatal suicide attempts corresponding to observed deaths. Comparison of counterfactual to observed data patterns yielded estimates of the number of lives saved and number of ERPOs needed to avert one suicide. Estimates varied depending on the assumed probability that a gun owner who attempts suicide will use a gun. Two evidence-based approaches yielded estimates of 17 and 23 ERPOs needed to prevent one suicide. For the subset of 2,850 ERPO respondents with documented suicide concern, comparable estimates were 13 and 18, respectively. This study's findings add to growing evidence that ERPOs can be an effective and important suicide prevention tool.

Objective This study investigates the association between replacement thinking, status threat perceptions, and the endorsement of political violence among non-Hispanic white adults in the United States. It explores how perceived threats to social status can drive support for extreme measures aimed at preserving white hegemony, addressing a gap in research on factors contributing to political violence, a public health concern. Methods The 2022 Life in America Survey provided data for this cross-sectional study, focusing on status threat and replacement thinking among non-Hispanic white respondents. Status threat was inferred from relative income, education level, and racial segregation in residential census tracts, while replacement thinking was derived through agreement with the statement “in America, native-born white people are being replaced by immigrants.” The outcome was the endorsement of political violence. Analysis utilized a survey-weighted robust modified Poisson model. Results Among 5,976 non-Hispanic white respondents, 18.7 % supported political violence in at least one scenario. A U-shaped relationship was observed between racial segregation and political violence endorsement: respondents from more diverse communities were less likely to support political violence. Those endorsing replacement thinking were 233 %–229 % more likely to endorse political violence than those who did not, dependent on income levels. White respondents without a high school degree were 29 % more likely to endorse political violence. Conclusion The study found a positive association between replacement thinking, markers of status threat, and political violence endorsements among non-Hispanic white Americans. These findings emphasize the need for research and interventions to mitigate these perceptions and prevent political violence.

Optimism bias is common across health risk assessments, including firearm injury risk, and can have behavioral consequences. Using data from the 2018 California Safety and Wellbeing Survey, we examine whether optimism bias influences firearm injury prevention practices and policy support by comparing the characteristics, behaviors, and opinions of gun owners who believed having a gun at home is comparatively safer for themselves than for similar others ( Optimism Bias group) with (1) those who unequivocally believe guns increase safety for themselves and others ( Always Safer group), and (2) those who said they “don’t know” or “it depends” in both the self and other scenarios ( Uncertain group). Weighted multinomial logistic regression results indicated that gun owners in the Optimism Bias group were more often female, members of minoritized racial or ethnic groups, and new gun owners than the Always Safer and Uncertain groups; they also demonstrated greater support for 4 of 5 firearm injury prevention policies/interventions. Despite similar prevalence of owning a gun for protection, gun owners in the Optimism Bias group less often carried a loaded firearm or stored a gun in an unsecure way compared with the Always Safer group. Findings suggest that gun owners characterized by optimism bias, who acknowledged some risk associated with firearms, even if only or more so for others than for themselves, may represent a “movable middle” that is more receptive to firearm injury prevention efforts. Public health messages emphasizing other-oriented (vs. personal) risk and collective responsibility may be perceived as less threatening to the symbolic significance of guns to individual identity, thus enhancing effectiveness.

Objectives: To estimate prevalences of racism, hostile sexism, homonegativity, transphobia, xenophobia, antisemitism, and Islamophobia in the United States; to estimate their associations, individually and collectively, with support for and willingness to engage in political violence.Methods: Findings are from Wave 2 of a nationally representative longitudinal survey of members of the Ipsos KnowledgePanel, conducted May 18-June 8, 2023. Prevalences are estimated based on levels of agreement with items in abbreviated versions of previously validated scales. Associations with political violence are estimated based on agreement with measures of support for or willingness to engage in such violence and are expressed as weighted proportions and adjusted prevalence differences, with p-values adjusted for the false discovery rate and reported as q-values.Results: The Wave 2 completion rate was 84.2%; there were 9,385 respondents. After weighting, 50.7% (95% confidence interval (CI) 49.4%, 52.1%) of respondents were female and 62.7% (95% CI, 61.2%, 64.1%) were white, non-Hispanic; weighted mean (± standard deviation) age was 48.5 (25.9) years. Prevalences of strong agreement ranged from 3.2% (95% CI 2.6%, 3.7%) for antisemitism to 26.4% (95% CI 25.1%, 27.2%) for homonegativity and exceeded 10% in 4 of 7 cases. In all cases, significantly higher percentages of persons in strong agreement than in non-agreement supported political violence and were willing to engage in it; the largest differences were for Islamophobia and hostile sexism. Phobias with relatively lower prevalences of strong agreement tended to have stronger associations with political violence. Associations with political violence were larger for the 7 phobias combined than for individual phobias.Conclusions: Racism, hostile sexism, homonegativity, transphobia, xenophobia, antisemitism, and Islamophobia were all strongly associated with support for and willingness to engage in political violence. These findings can help focus political violence prevention efforts in the United States.