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

Child maltreatment mortality in Canada: An analysis of coroner and medical examiner data

BACKGROUND

Fatalities are the most severe consequence of child maltreatment, but there are gaps in what is known about the epidemiology of such deaths in Canada.

OBJECTIVES

The objectives of this study were to: (1) identify child maltreatment deaths among those classified as homicide or undetermined manner; (2) estimate rates of child maltreatment mortality by sex, age, geography, and year; and (3) measure differences between rates of child maltreatment mortality and homicide mortality.

METHODS

We used a cross-sectional design and analyzed mortality data from the Canadian Coroner and Medical Examiner Database for a ten-year period (2007 to 2016). To identify child maltreatment deaths, we reviewed narrative data about children aged 0 to 17 years old whose deaths were classified as homicide or undetermined. Descriptive statistics, mortality rates, and incidence rate ratios were calculated.

RESULTS

Among the 1758 child deaths due to homicide or an undetermined manner, maltreatment was the probable cause in 20.4 % (n = 359) of deaths; most child maltreatment deaths (72.1 %, n = 259) were among children younger than 5 years old. The national child maltreatment mortality rate was 0.55 deaths per 100,000; the rate was highest among infants (3.43 per 100,000) and decreased at older ages.

CONCLUSION

This study provides evidence about the epidemiology of child maltreatment mortality in Canada.

Nutrition facts, drug facts, and model facts: putting AI ethics into practice in gun violence research

OBJECTIVE

Firearm injury research necessitates using data from often-exploited vulnerable populations of Black and Brown Americans. In order to reduce bias against protected attributes, this study provides a theoretical framework for establishing trust and transparency in the use of AI with the general population.

METHODS

We propose a Model Facts template that is easily extendable and decomposes accuracy and demographics into standardized and minimally complex values. This framework allows general users to assess the validity and biases of a model without diving into technical model documentation.

EXAMPLES

We apply the Model Facts template on 2 previously published models, a violence risk identification model and a suicide risk prediction model. We demonstrate the ease of accessing the appropriate information when the data are structured appropriately.

DISCUSSION

The Model Facts template is limited in its current form to human based data and biases. Like nutrition facts, it will require educational programs for users to grasp its full utility. Human computer interaction experiments should be conducted to ensure model information is communicated accurately and in a manner that improves user decisions.

CONCLUSION

The Model Facts label is the first framework dedicated to establishing trust with end users and general population consumers. Implementation of Model Facts into firearm injury research will provide public health practitioners and those impacted by firearm injury greater faith in the tools the research provides.

Deaths Related to Domestic Violence in Washington State

Importance

Domestic violence (DV; including intimate partner and family violence) is associated with heightened lethality risks, yet limited research has comprehensively assessed the connection between DV and fatal violence considering both homicides and suicides. Understanding the fatal consequences of DV can point to missed opportunities to support individuals and their families.

Objective

To assess the proportion of violent deaths that were connected to DV and describe contacts with the legal system or social services prior to each DV-related fatality.

Design, Setting, and Participants

This cross-sectional study used quantitative and qualitative data from the National Violent Death Reporting System (NVDRS) for all individuals who died by homicide or suicide in Washington from January 1, 2015, to December 31, 2020. Analyses were conducted from August 1, 2022, to September 30, 2023.

Main Outcomes and Measures

A multipronged approach was used to assess DV history using existing NVDRS variables, leveraging data from prior review of NVDRS death narratives, applying a validated natural language processing tool, and linking related deaths. Domestic violence was recorded as yes or no, but the decedent's role in the abusive relationship (ie, experiencing or enacting DV) could not be differentiated. To describe system involvement prior to each death, keyword searching and hand review of NVDRS death narratives were used.

Results

A total of 7352 intentional violent deaths (1192 homicides [16.2%]; 6160 suicides [83.8%]) with known circumstances were recorded in Washington during the study period. Of these, 948 deaths (12.9%) were connected to DV (624 [65.8%] among males; mean [SD] age at death, 45.3 [19.2] years), including 588 suicides (62.0%) and 360 homicides (38.0%). For 420 DV-related deaths (44.3%), there was evidence to suggest that the person who died or their intimate partner(s), family, or cohabitants had prior contacts with the legal system or social services. Specifically, 318 records (33.5%) mentioned prior contacts with law enforcement or the criminal legal system (eg, prior 9-1-1 calls, criminal convictions), and 225 (23.7%) described engagement with social services or the civil legal system (eg, civil protection order, divorce, or child custody problems).

Conclusions and Relevance

In this cross-sectional study, 12.9% of violent deaths in Washington were connected to DV. The findings suggest that more resources are needed to support law enforcement, court professionals, and social services specialists to proactively identify and refer families to wraparound supports before the situation can escalate to a fatality.

Population-Level Surveillance of Domestic Assaults in the Home Using the National Emergency Medical Services Information System (NEMSIS)

Violence in the home, including partner violence, child abuse, and elder abuse, is pervasive in the United States. An informatics approach allowing automated analysis of administrative data to identify domestic assaults and release timely and localized data would assist preventionists to identify geographic and demographic populations of need and design tailored interventions. This study examines the use of an established national dataset, the NEMSIS 2019, as a potential annual automated data source for domestic assault surveillance. An algorithm was used to identify individuals who utilized emergency medical services (EMS) for a physical assault in a private residence (N = 176,931). Descriptive analyses were conducted to define the identified population and disposition of patients. A logistic regression was performed to predict which characteristics were associated with consistent domestic assault identification by the on-scene EMS clinician and dispatcher. The sample was majority female (52.2%), White (44.7%), urban (85.5%), and 21-29 years old (24.4%). A disproportionate number of those found dead on scene were men (74.5%), and female patients more often refused treatment (57.8%) or were treated and then released against medical advice (58.4%). Domestic assaults against children and seniors had higher odds of being consistently identified by both the dispatcher and EMS clinician than those 21-49, and women had lower odds of consistent identification than men. While a more specific field to identify the type of domestic assault (e.g., intimate partner) would help inform specialized intervention planning, these data indicate an opportunity to systematically track domestic assaults in communities and describe population-specific needs.

Artificial Intelligence and the National Violent Death Reporting System: A Rapid Review

As the awareness on violent deaths from guns, drugs, and suicides emerges as a public health crisis in the United States, attempts to prevent injury and mortality through nursing research are critical. The National Violent Death Reporting System provides public health surveillance of US violent deaths; however, understanding the National Violent Death Reporting System's research utility is limited. The purpose of our rapid review of the 2019-2023 literature was to understand to what extent artificial intelligence methods are being used with the National Violent Death Reporting System. We identified 16 National Violent Death Reporting System artificial intelligence studies, with more than half published after 2020. The text-rich content of National Violent Death Reporting System enabled researchers to center their artificial intelligence approaches mostly on natural language processing (50%) or natural language processing and machine learning (37%). Significant heterogeneity in approaches, techniques, and processes was noted across the studies, with critical methods information often lacking. The aims and focus of National Violent Death Reporting System studies were homogeneous and mostly examined suicide among nurses and older adults. Our findings suggested that artificial intelligence is a promising approach to the National Violent Death Reporting System data with significant untapped potential in its use. Artificial intelligence may prove to be a powerful tool enabling nursing scholars and practitioners to reduce the number of preventable, violent deaths.

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.

Misclassification of firearm-related violent crime in criminal legal system records: challenges and opportunities

BACKGROUND

Criminal legal system data are one source for measuring some types of firearm-related harms, including those that do not necessarily result in injury or death, but measurement can be hampered by imprecise criminal code statutes. We quantified the degree of misclassification in Washington state criminal codes for measuring firearm-related crime.

FINDINGS

In this study of individuals aged 18 years and older who were convicted of a misdemeanor in Washington Superior Courts from 1/1/2015 through 12/31/2019, we compared firearm-related charges as measured with criminal codes and with manual review of probable cause documents, considered the gold standard. The sample included 5,390 criminal cases. Of these, 77 (1.4%) were firearm-related as measured with criminal codes and 437 (8.1%) were firearm-related as measured via manual record review. In the sample overall, the sensitivity of criminal codes was 17.6% (95% CI 14.2-21.5%), and negative predictive value (NPV) was 93.2% (95% CI 92.5-93.9%). Sensitivity and NPV were higher for cases with exclusively non-violent charges. For all cases and for cases with any violent crime charge, firearm-related crimes described in probable cause documents most often involved explicit verbal threats, firearm possession, and pointing a firearm at or touching a firearm to someone; almost 10% of all cases involved shooting/discharging a firearm. For cases with exclusively non-violent charges, the most common firearm-related crime was unlawful possession.

CONCLUSIONS

Criminal records can be used for large-scale policy-relevant studies of firearm-related harms, but this study suggests Washington state criminal codes substantially undercount firearm-related crime, especially firearm-related violent crime.

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.