Improving Suicide Prevention Through Evidence-Based Strategies: A Systematic Review

Mental and Physical Health-Related Risk Factors Among Females Who Died by Firearm Suicide

Importance

Firearm suicide among females has increased in the past decade; routes for prevention remain unclear.

Objective

To identify classes of firearm suicide decedents among females based on preceding mental and physical health-related risk factors for suicide.

Design, Setting, and Participants

This cross-sectional retrospective study of females who died by firearm suicide between January 2014 and December 2018 used state-based surveillance data from the National Violent Death Reporting System Restricted Access Database representing 1 or more years of data from all 50 US states, the District of Columbia, and Puerto Rico; data were analyzed from March 2022 to September 2023. A latent class analysis was used to identify and characterize categories of mental and physical health-related factors among females (ie, classes).

Exposure

Known mental and physical health-related risk factors for suicide.

Main Outcome and Measure

The numbers (percentages) of females who died by firearm suicide across various factors.

Results

Of the 8318 female decedents in our sample (mean [SD] age, 47.2 [17.0] years), more than half (4816 of 8318 [57.9%]) had more than 1 risk factor endorsed. Among this subgroup, a 4-class model was identified: (1) alcohol use disorder or substance use disorder (1273 [26.4%]); (2) depression and suicidal thoughts (2289 [47.5%]); (3) physical health problems and pain (1054 [21.9%]); and (4) all conditions (multimorbid; 200 [4.2%]); 3502 (42.1%) could not be classified. Black females made up a higher proportion (244 of 3502 [7.0%]) of the females with a 0 or 1 risk factor endorsed than of the females with multiple risk factors (184 of 4816 [3.8%]). Mental health problems were apparent in 51.7% of the full sample (4303 of 8318), with 28.6% (n = 2376) demonstrating evidence of mental health treatment. More than one-fifth (1766 of 8318 [21.2%]) had physical health problems. More than one-fourth of the full sample (2239 of 8318 [26.9%]) and nearly 40% (505 of 1273 [39.7%]) in class 1 had intimate partner problems before their death.

Conclusions and Relevance

In this cross-sectional study of females who died by firearm suicide, meaningfully distinct groups of female firearm suicide decedents were identified. Given the high proportion of females without mental health and physical health problems, it is plausible that many female firearm suicide decedents did not frequently intersect with the health care system before their death. Results suggest opportunities for preventing firearm suicide among females through nonmedical care settings.

Preventing Suicide Through Lethal Means Restriction in Pediatric Care

Suicidal ideation in the pediatric population has increased by 50% over the last decade. Suicide is the primary contributor to violent deaths and ranks as the second leading cause of mortality within the 10-24-year-old age group. The lethality of the methods selected notably elevates the likelihood of children and adolescents dying by suicide. Awareness of the commonly chosen means of suicide at each developmental stage and understanding their respective lethality are vital in developing prevention strategies. Lethal means safety counseling is a specific evidence-based suicide prevention strategy that pediatric providers can use during patient encounters to help mitigate the risk of self-inflicted injuries and suicide. Providers are well-positioned to integrate discussions of lethal means restriction with patients and their families into their protocols for suicide prevention, screening, and assessment. This article explores the data and practice evidence for lethal means restriction and counseling, focusing on clinical strategies and policy changes to reduce access to lethal means. Emphasis is given to how pediatric providers speak with parents about the importance of restricting access when developing a safety plan.

Suicidal Thoughts and Behaviors Among Health Care Trainees, Staff and Faculty at an Academic Medical Center

Background/Objectives: Health care workers are at greater risk for death by suicide compared to the general population and are less likely to seek assistance. More information about correlates of suicidality and treatment-seeking behavior are needed to reduce risk. Methods: The American Foundation for Suicide Prevention developed an Interactive Screening Program to identify and engage at-risk staff and trainees in health care settings. The study reports on the prevalence and demographic and clinical predictors of current suicidal thoughts, behaviors and mental health treatment at a single site (n = 5898) from 2009 to 2024. Results: The study found that 18.2% of respondents reported current suicidal thoughts and behaviors. These were more common among respondents who were younger, male, and who identified as a race/ethnicity other than non-Hispanic White. Suicidal thoughts and behaviors were more likely among those with higher PHQ-8 scores (OR = 1.23, p < 0.01) and those who endorsed maladaptive coping behaviors, hopelessness, loneliness, stress and nervousness (ORs 1.36-3.04, ps < 0.01). Current mental health treatment was more likely among women, non-Hispanic White respondents compared with Asian or Pacific Islander respondents, and nurses relative to physicians. Mental health treatment was also associated with higher PHQ-8 scores, lifetime suicide attempts, difficulty controlling eating and alcohol consumption, and recent feelings of anxiety, stress and nervousness. Conclusions: Findings suggest a continued need to identify and engage health care trainees and staff who are at risk for suicide and to establish new approaches to linking these individuals to resources or interventions aimed at reducing risk. The study identified male and/or Asian/Pacific Islander-identifying health care workers who reported intense loneliness and/or hopelessness, use of non-prescription drugs and recent suicidal thoughts and/or behaviors as high-risk individuals who may require enhanced methods of outreach, identification, acceptance and accessibility of treatment.

Enhancing suicidal behavior detection in EHRs: A multi-label NLP framework with transformer models and semantic retrieval-based annotation

BACKGROUND

Suicide is a leading cause of death worldwide, making early identification of suicidal behaviors crucial for clinicians. Current Natural Language Processing (NLP) approaches for identifying suicidal behaviors in Electronic Health Records (EHRs) rely on keyword searches, rule-based methods, and binary classification, which may not fully capture the complexity and spectrum of suicidal behaviors. This study aims to create a multi-class labeled dataset with annotation guidelines and develop a novel NLP approach for fine-grained, multi-label classification of suicidal behaviors, improving the efficiency of the annotation process and accuracy of the NLP methods.

METHODS

We develop a multi-class labeling system based on guidelines from FDA, CDC, and WHO, distinguishing between six categories of suicidal behaviors and allowing for multiple labels per data sample. To efficiently create an annotated dataset, we use an MPNet-based semantic retrieval framework to extract relevant sentences from a large EHR dataset, reducing annotation space while capturing diverse expressions. Experts annotate the extracted sentences using the multi-class system. We then formulate the task as a multi-label classification problem and fine-tune transformer-based models on the curated dataset to accurately classify suicidal behaviors in EHRs.

RESULTS

Lexical analysis revealed key themes in assessing suicide risk, considering an individual's history, mental health, substance use, and family background. Fine-tuned transformer-based models effectively identified suicidal behaviors from EHRs, with Bio_ClinicalBERT, BioBERT, and XLNet achieving the F1 scores (0.81), outperforming BERT and RoBERTa. The proposed approach, based on a multi-label classification system, captures the complexity of suicidal behaviors effectively particularly "Suicide Attempt" and "Family History" instances. The proposed approach, using task-specific NLP models and a multi-label classification system, captures the complexity of suicidal behaviors more effectively than traditional binary classification. However, direct comparisons with existing studies are difficult due to varying metrics and label definitions.

CONCLUSION

This study presents a robust NLP framework for detecting suicidal behaviors in EHRs, leveraging task-specific fine-tuning of transformer-based models and a semi-automated pipeline. Despite limitations, the approach demonstrates the potential of advanced NLP techniques in enhancing the identification of suicidal behaviors. Future work should focus on model expansion and integration to further improve patient care and clinical decision-making.

The Impact of Statewide Enumerated Antibullying Laws and Local School Interventions on School Spaces, Physical Harm, and Suicide Attempts for LGBTQ Youth

Purpose: Our goal was to evaluate the impact of enumerated antibullying laws and local interventions on school affirmation, risk of physical threat or harm, and suicide attempts for lesbian, gay, bisexual, transgender, and questioning (LGBTQ) youth. Methods: This study used cross-sectional data from three national sources that provided outcomes and potential predictors: (1) The Trevor Project National Survey 2022: based on the LGBTQ youth report, (2) Movement Advancement Project: presence of LGBTQ enumerated antibullying state laws, and (3) School Health Profiles 2020: statewide prevalence of school-level strategies to improve student wellness reported by school personnel. Bivariate and multivariable logistic regression models calculated the relative risks (RRs) for each outcome. Results: LGBTQ youth (n = 27,697) were surveyed. Youth were less likely to identify their school as nonaffirming in states with enumerated laws (RR: 0.97, confidence interval [CI]: 0.94-0.99) and in states with higher percentages of gay/straight alliances (GSAs) (RR: 0.97, 95% CI: 0.96-0.97). Youth were less likely to report a history of physical threat or harm in states with more GSAs (RR: 0.96, 95% CI: 0.95-0.97). Enumerated state laws were associated with a lower risk of suicide attempts among LGBTQ youth (RR: 0.82, 95% CI: 0.77-0.87). Universal strategies/policies not specific to LGBTQ youth did not reduce risk of physical threat/harm or suicide attempts. Conclusions: Statewide enumerated antibullying protections and the presence of a GSA were associated with a decreased risk of poor outcomes among LGBTQ youth. Longitudinal studies are needed to assess the impact of recent legislative changes.