Measuring interpersonal firearm violence: natural language processing methods to address limitations in criminal charge data.

OBJECTIVE: Firearm violence constitutes a public health crisis in the United States, but comprehensive data infrastructure is lacking to study this problem. To address this challenge, we used natural language processing (NLP) to classify court record documents from alleged violent crimes as firearm-related or non-firearm-related. MATERIALS AND METHODS: We accessed and digitized court records from the state of Washington (n = 1472). Human review established a gold standard label for firearm involvement (yes/no). We developed a key term search and trained supervised machine learning classifiers for this labeling task. Results were evaluated in a held-out test set. RESULTS: The decision tree performed best (F1 score: 0.82). The key term list had perfect recall (1.0) and a modest F1 score (0.65). DISCUSSION AND CONCLUSION: This case report highlights the accuracy, feasibility, and potential time-saved by using NLP to identify firearm involvement in alleged violent crimes based on digitized narratives from court documents.

Research needs related to firearm rights restoration.

BACKGROUND: In the USA, firearms are commonly involved in many incidents of serious interpersonal harm. Federal law prohibits the purchase and possession of firearms by certain high-risk groups including those with prior felony or domestic violence misdemeanor convictions. Evidence supports the effectiveness of these prohibitions, but little is known about how often prohibited persons later seek to have their firearm rights restored. MAIN BODY: For this commentary, we systematically searched the empirical literature for information about who requests firearm rights restoration in the USA, how often it is granted, and what its consequences are. We found a dearth of empirical literature on this topic. CONCLUSION: We call for attention to this gap in the research. There is a need to build an evidence base that can help inform state policy and courtroom practices regarding the eligibility, appropriateness, and risk for subsequent harm following firearm rights restoration among persons who are prohibited based on a criminal conviction history.

Intimate partner violence and suicide mortality: a cross-sectional study using machine learning and natural language processing of suicide data from 43 states.

INTRODUCTION: Intimate partner violence (IPV) is associated with suicidal ideation, yet it remains unclear how often IPV precipitates suicide mortality. To overcome limitations with national data, we applied novel methods to: (1) document the prevalence of IPV-related suicide in the USA and (2) identify correlates for IPV-related suicide. METHODS: Using National Violent Death Reporting System data (NVDRS, 2015-2019, n=1 30 550), we recorded IPV circumstances (yes/no) by leveraging prior textual reviews of death narratives and applying a validated natural language processing tool. We could not systematically differentiate IPV perpetration versus victimisation given limited details in NVDRS. Logistic regression compared IPV-related suicides with referent group suicides (no evidence of IPV), stratified by sex. RESULTS: 7.1% of suicides were IPV related (n=9210), most were isolated suicide events (82.8%, n=7625; ie, not homicide suicide). There were higher odds of IPV circumstances when the decedent had civil legal problems (aOR for men: 3.6 (3.3 to 3.9), aOR for women: 2.6 (2.2 to 3.2)), criminal legal problems (aOR men: 2.3 (2.2 to 2.5), aOR for women: 1.7 (1.4 to 2.1)), or used a firearm (aOR men: 1.9 (1.8 to 2.0), aOR for women: 1.9 (1.7 to 2.1)). There were lower odds of IPV circumstances when the decedent had a current mental health problem (aOR men: 0.7 (0.7 to 0.8), aOR for women: 0.7 (0.6 to 0.8)). CONCLUSIONS: IPV circumstances contribute to a notable proportion of suicides. IPV-related suicides are distinct from other suicide deaths. Targeted suicide screening and intervention in IPV settings may be beneficial for prevention.

Intimate Partner Violence Circumstances for Fatal Violence in the US.

This cross-sectional study investigates intimate partner violence circumstances associated with violent deaths in the US from 2015 to 2019.

Examining Intimate Partner Violence-Related Fatalities: Past Lessons and Future Directions Using U.S. National Data.

Purpose: Among homicides in the United States, intimate partners kill almost 50% of female and 10% of male victims. Intimate partner violence (IPV) also contributes to an estimated 6% of suicides. These trends suggest that opportunities for IPV interventions prior to the fatalities may have been missed. Thus, researchers must investigate the context and circumstances of IPV-related fatalities to inform effective prevention strategy development. There are two primary national fatality databases that can be used to examine such factors: the National Violent Death Reporting System (NVDRS, homicide and suicides); and the Uniform Crime Reporting-Supplementary Homicide Reports (UCR-SHR, homicides). These datasets include data on many IPV-related violent deaths but are limited by variations in data quality. Method: This critical review summarizes opportunities and challenges when examining IPV-related fatalities using these national datasets. To document how the current literature is conceptualizing IPV, a rapid review on IPV-related homicide and suicide articles was performed (2019-2022). Missingness analyses were conducted to describe limitations in key dataset variables. Results: These datasets enable tracking IPV-related fatalities nationally over time. However, issues with the operationalization of variables that record IPV circumstances, particularly in the UCR-SHR, and high levels of missingness represent significant barriers to research. Novel methodologies can optimize the use of these datasets. Conclusion: National-level datasets enable researchers to examine IPV-related fatalities, evaluate policy differences between states, and monitor trends and disparities. This research can inform key recommendations for interventions to prevent IPV-related fatalities. Supplementary Information: The online version contains supplementary material available at 10.1007/s10896-022-00487-2.

Detecting intimate partner violence circumstance for suicide: development and validation of a tool using natural language processing and supervised machine learning in the National Violent Death Reporting System.

BACKGROUND: Intimate partner violence (IPV) victims and perpetrators often report suicidal ideation, yet there is no comprehensive national dataset that allows for an assessment of the connection between IPV and suicide. The National Violent Death Reporting System (NVDRS) captures IPV circumstances for homicide-suicides (<2% of suicides), but not single suicides (suicide unconnected to other violent deaths; >98% of suicides). OBJECTIVE: To facilitate a more comprehensive understanding of the co-occurrence of IPV and suicide, we developed and validated a tool that detects mentions of IPV circumstances (yes/no) for single suicides in NVDRS death narratives. METHODS: We used 10 000 hand-labelled single suicide cases from NVDRS (2010-2018) to train (n=8500) and validate (n=1500) a classification model using supervised machine learning. We used natural language processing to extract relevant information from the death narratives within a concept normalisation framework. We tested numerous models and present performance metrics for the best approach. RESULTS: Our final model had robust sensitivity (0.70), specificity (0.98), precision (0.72) and kappa values (0.69). False positives mostly described other family violence. False negatives used vague and heterogeneous language to describe IPV, and often included abusive suicide threats. IMPLICATIONS: It is possible to detect IPV circumstances among singles suicides in NVDRS, although vague language in death narratives limited our tool's sensitivity. More attention to the role of IPV in suicide is merited both during the initial death investigation processes and subsequent NVDRS reporting. This tool can support future research to inform targeted prevention.

Co-occurring Epidemic Conditions Among Southern U.S. Black Men Who Have Sex with Men in an Online eHealth Intervention.

Black men who have sex with men (BMSM) face disproportionately higher risks for adverse sexual health outcomes compared to their non-Hispanic White counterparts. This disparity can be attributable to overlapping and intersecting risk factors at the individual and structural levels and can be understood through syndemic theory. Using longitudinal data from the HealthMPowerment trial (n = 363), six conditions related to stigma syndemics were indexed as a cumulative risk score: high alcohol use, polydrug use, depression and anxiety symptomology, and experiences of racism and sexual minority stigma. Using Poisson regression, we found a positive association between baseline risk scores and sexual risk behavior (b: 0.32, SE: 0.03, p < 0.001). Using a Generalized Estimating Equation, we also found a 0.23 decrease in the within-participant risk scores at 3-month follow-up (SE: 0.10, p < 0.020). Future work examining how care and prevention trials improve health outcomes in this population is needed.

A Critical Review of Text Mining Applications for Suicide Research.

Purpose of Review: Applying text mining to suicide research holds a great deal of promise. In this manuscript, literature from 2019 to 2021 is critically reviewed for text mining projects that use electronic health records, social media data, and death records. Recent Findings: Text mining has helped identify risk factors for suicide in general and specific populations (e.g., older adults), has been combined with structured variables in EHRs to predict suicide risk, and has been used to track trends in social media suicidal discourse following population level events (e.g., COVID-19, celebrity suicides). Summary: Future research should utilize text mining along with data linkage methods to capture more complete information on risk factors and outcomes across data sources (e.g., combining death records and EHRs), evaluate effectiveness of NLP-based intervention programs that use suicide risk prediction, establish standards for reporting accuracy of text mining programs to enable comparison across studies, and incorporate implementation science to understand feasibility, acceptability, and technical considerations.

Intimate partner violence victimization and perpetration as precursors to suicide.

It remains unclear how often and under what circumstances intimate partner violence (IPV) precedes suicide. Available research on IPV and suicide focuses largely on homicide-suicide, which is a rare event (<2% of suicides). We focus instead on single suicides (i.e., suicides unconnected to other violent deaths), which are the most common type of fatal violence in the US. Unfortunately, information about IPV circumstances is often unavailable for suicides. To address this gap, we sought to identify the proportion of single suicides that were preceded by IPV in North Carolina (NC), to describe the prevalence of IPV victimization and perpetration as precursors to suicide, and to explore how IPV-related suicides differ from other suicides. We used data from the NC Violent Death Reporting System (2010-2017, n = 9682 single suicides) and hand-reviewed textual data for a subset of cases (n = 2440) to document IPV circumstances. We had robust inter-rater reliability (Kappa: 0.73) and identified n = 439 IPV-related suicides. Most were males who had perpetrated nonfatal IPV (n = 319, 72.7%) prior to dying by suicide. Our findings suggest that IPV was a precursor for at least 4.5% of single suicides. Next, we conducted logistic regression analyses by sex comparing IPV-related suicides to other suicides. For both men and women, IPV was more common when the person who died by suicide had recently disclosed suicidal intent, was younger, used a firearm, and was involved with the criminal legal system, even after controlling for covariates. We also found sex-specific correlates for IPV circumstances in suicide. Combined with homicide-suicide data (reported elsewhere), IPV is likely associated with 6.1% or more of suicides overall. Results suggest clear missed opportunities to intervene for this unique subpopulation, such as suicide screening and referral in IPV settings (e.g., batterer intervention programs, Family Justice Centers) that is tailored by sex.

Intimate Partner Violence-Related Fatalities Among U.S. Youth Aged 0-24 Years, 2014-2018.

INTRODUCTION: Little is known about intimate partner violence-related fatalities among young people. This study comprehensively identifies and describes intimate partner violence-related homicides, homicide-suicides, legal intervention deaths, and suicides among young people. METHODS: Data from the 2014-2018 National Violent Death Reporting System were analyzed for all decedents aged 0-24 years in 38 states, the District of Columbia, and Puerto Rico with known circumstances of death (n=29,702). Intimate partner violence-related deaths were identified using National Violent Death Reporting System variables across all manners of death and supplementary narrative review for suicides. This article reports the proportion of intimate partner violence versus non-intimate partner violence-related deaths by manner of death, descriptive statistics, and rate of intimate partner violence-related death per 100,000 person years. To examine disparities in intimate partner violence-related deaths, generalized estimating equations were used with robust standard errors to account for clustering of deaths within states and fixed effects for years. Statistical analyses were conducted May and August 2021. RESULTS: A total of 1,927 intimate partner violence-related deaths were identified, which represents 6.5% of violent deaths with known circumstances, at a rate of 0.35 per 100,000 person years. Supplementary narrative review identified 44.7% of all intimate partner violence-related deaths. There were significant differences by race/ethnicity and whether a firearm inflicted the fatal injury for male and female decedents by manner of death. CONCLUSIONS: If the National Violent Death Reporting System does not assess whether intimate partner violence was a precipitating factor across all death manners, the true magnitude of intimate partner violence's contribution to violent death will be underestimated. Future research that identifies factors associated with all manners of intimate partner violence-related deaths among young people will help inform intervention and prevention strategies to save young lives.

What is the role of firearms in nonfatal intimate partner violence? Findings from civil protective order case data.

BACKGROUND: Perpetrators of intimate partner violence (IPV) use firearms to injure, scare, and manipulate their partners. Abusers who have a firearm in their homes are more likely to threaten and/or kill their partner. To date, however, limited research documents the nature of IPV perpetrator firearm access or the prevalence of nonfatal firearm abuse behaviors. METHODS: Federal law restricts firearm access for IPV perpetrators in qualifying domestic violence protective order (DVPO) cases and information about firearms should be disclosed during the DVPO process. We used secondary data from civil DVPO cases (n = 406) in North Carolina that were collected using a representative sampling strategy. Data were from DVPO case files and structured DVPO hearing observations. We conducted a content analysis to record IPV perpetrator access to guns and reported firearm abuse behaviors. We used a linear regression analysis to determine whether IPV perpetrator gun access was associated with higher levels of reported abuse. We also examined factors associated with perpetration of nonfatal firearm abuse. RESULTS: We found evidence of perpetrator firearm access in nearly half of all cases (46%, n = 108). Controlling for covariates, gun access was significantly associated with higher levels of reported IPV (b = 0.5, p < .001). Firearm abuse was reported in nearly one out of four cases (23.1%, n = 101), and often entailed spoken threats, displaying a gun, or holding a partner at gun point. The only factors associated with firearm abuse in the multivariate models were related to English language speaking/fluency. CONCLUSIONS: Gun access should be considered an indicator for severe IPV. We must ensure that existing legal mechanisms to identify and restrict abuser access to firearms are fully implemented and enforced. Firearm abuse often manifests as non-physical coercive control which is traumatic and has the potential to escalate to homicide, even in the absence of past physical violence.

Fatalities related to intimate partner violence: towards a comprehensive perspective.

BACKGROUND: In 2015, 1350 people in the US were killed by their current or former intimate partner. Intimate partner violence (IPV) can also fatally injure family members or friends, and IPV may be a risk factor for suicide. Without accounting for all these outcomes, policymakers, funders, researchers and public health practitioners may underestimate the role that IPV plays in violent death. OBJECTIVE: We sought to enumerate the total contribution of IPV to violent death. Currently, no data holistically report on this problem. METHODS: We used Violent Death Reporting System (VDRS) data to identify all IPV-related violent deaths in North Carolina, 2010-2017. These included intimate partner homicides, corollary deaths, homicide-suicides, single suicides and legal intervention deaths. We used the existing IPV variable in VDRS, linked deaths from the same incident and manually reviewed 2440 suicide narratives where intimate partner problems or stalking were a factor in the death. RESULTS: IPV contributes to more than 1 in 10 violent deaths (10.3%). This represents an age-adjusted rate of 1.97 per 100 000 persons. Of the IPV-related violent deaths we identified, 39.3% were victims of intimate partner homicide, 17.4% corollary victims, 11.4% suicides in a homicide-suicide event, 29.8% suicides in a suicide-only event and 2.0% legal intervention deaths. IMPLICATIONS: If researchers only include intimate partner homicides, they may miss over 60% of IPV-related deaths. Our novel study shows the importance of taking a comprehensive approach to prevent IPV and decrease violent deaths. IPV is a risk factor for suicide as well as homicide.

The understudied half of undergraduates: Risky sexual behaviors among community college students.

Objective: This study examines the prevalence and risk factors associated with risky sexual behaviors in community college students. Participants: A diverse sample of 18-19-year-old community college students (N = 264). Methods: Baseline data from an online prevention program administered in 2015. Results: Community college students in this sample disproportionately experienced sexual assault and were unlikely to test for STIs. Higher intentions to engage in risky sexual behaviors were associated with gender and sexual experience, but also with having lower intentions to communicate with a sexual partner about pregnancy and STIs, and having higher gender norm endorsement. Conclusions: Older adolescents attending community colleges may be at high risk for poor sexual health outcomes, and appropriate theory-based education should be tailored to meet the needs of these underserved students.

Judging Domestic Violence From the Bench: A Narrative Analysis of Judicial Anecdotes About Domestic Violence Protective Order Cases.

Interview participants sometimes share anecdotes (stories about past events), to illustrate a point or discuss their perspectives. When sharing these stories, participants may imbue the events with their own personal meaning-making, selective memory, and biases. We conducted a narrative analysis of anecdotes shared by judges ( n = 20) who preside over Domestic Violence Protective Order (DVPO) hearings to examine how biases and misperceptions shape decisions in DVPO cases. We found that judges rely on biases to sort cases as "true domestic violence" compared with "frivolous cases." In the anecdotes they shared, judges often used gendered stereotypes to depict litigants, and many judges felt that DVPOs had limited efficacy in preventing violence. We argue that important cognitive insights are revealed by interview participants during the spontaneous act of storytelling. In the case of judges, their biases could lead to DVPOs being denied in situations when they are warranted.