Suicide Risk Research Articles

Identifying momentary suicidal ideation using machine learning in patients at high-risk for suicide.

Strategies to detect the presence of suicidal ideation (SI) or characteristics of ideation that indicate marked suicide risk are critically needed to guide interventions and improve care during care transition periods. Some studies indicate that machine learning can be applied to momentary data to improve classification of SI. This study examined whether the classification accuracy of these models varies as a function of type of training data or characteristics of ideation. A total of 257 psychiatric inpatients completed a 3-week battery of ecological momentary assessment and measures of suicide risk factors. The accuracy of machine learning models in classifying the presence, duration, or intensity of ideation was compared across models trained on baseline and/or momentary suicide risk data. Relative feature importance metrics were examined to identify the risk factors that were most important for outcome classification. Models including both baseline and momentary features outperformed models with only one feature type, providing important information in both correctly classifying and differentiating individual characteristics of SI. Models classifying SI presence, duration, and intensity performed similarly. Results of this study may not generalize beyond a high-risk, psychiatric inpatient sample, and additional work is needed to examine temporal ordering of the relationships identified. Our results support using machine learning approaches for accurate identification of SI characteristics and underscore the importance of understanding the factors that differentiate and drive different characteristics of SI. Expansion of this work can support use of these models to guide intervention strategies.

Variables Associated with Abandonment, Permanence, or Clinical Discharge from a Suicide Risk Attention Program

Variables Associated with Abandonment, Permanence, or Clinical Discharge from a Suicide Risk Attention Program

The relationship between mindfulness and childhood trauma with suicide: The mediating role of cognitive bias

The purpose of this study is to investigate the relationship between mindfulness, childhood trauma, and suicide, specifically focusing on the mediating role of cognitive bias. Suicide is a pressing issue that poses a significant threat to public health in society, and understanding the factors that contribute to it is crucial. The research design employed in this study is a correlation-based structural equation model. A total of 370 students from the second level of secondary school in Iran city were selected as participants using an available sampling method during the academic year of 2023. To gather data, several questionnaires were utilized, including the Kentucky Mindfulness Questionnaire, Bernstein et al.'s Childhood Trauma Questionnaire (1994), Hammachi and Ozturk's Cognitive Bias Questionnaire (2004), and the Multi-attitude Suicidal Tendency Questionnaire (MAST). Data analysis was conducted using SPSS-23 and Amos software, encompassing both descriptive and inferential levels. The findings of this research revealed several significant relationships. Firstly, there is a positive and significant direct path from childhood trauma to cognitive bias. Secondly, mindfulness has a negative and significant direct path to cognitive bias. Thirdly, childhood trauma has a direct and significant path to suicide. Additionally, cognitive bias has a positive and significant path to suicidality, while mindfulness has a negative and significant path to suicide. Moreover, there are positive and significant indirect paths from suicide to childhood trauma and negative and significant indirect paths from suicide to mindfulness. In conclusion, this study highlights the importance of mindfulness and cognitive bias as variables related to adolescent suicide. Adolescents who have experienced childhood trauma are more likely to consider suicide as a solution when faced with life problems. These findings emphasize the need for interventions and support systems that address mindfulness, cognitive bias, and childhood trauma to prevent and mitigate the risk of suicide among adolescents.

Suicide in young and middle-aged individuals: lessons from a psychosocial autopsy study

Abstract Background Sociodemographic characteristics of suicides are available from Statistics Netherlands. To improve suicide prevention, it is not only necessary to know who dies but also why people die. A psychosocial autopsy was conducted in the Netherlands to investigate differences in psychosocial characteristics and precipitating factors of suicides of young (<30) and middle-aged (40-70) suicide decedents. Methods Data were collected from bereaved individuals. A questionnaire detailed sociodemographic and psychosocial characteristics of the decedents. We reconstructed the suicidal process in the months preceding the suicide in an interview. Findings from the questionnaire and interview were integrated. An independent expert committee formulated recommendations for suicide prevention. Results 162 suicides were investigated. Young decedents showed high levels of substance use and academic pressure. Young adults with developmental disorders were vulnerable in the transition to adulthood. Young males reached out less to support and care. By contrast, suicides of young females were distinguished by comorbidity of psychiatric disorders, frequent suicidal behaviours, high levels of care use, wait lists and a loss of (treatment) perspective, harmful social media use and bullying. Work and financial problems were common in middle-aged decedents. Physical disability and pain were frequently mentioned for middle-aged females. Suicide in middle-aged males was characterized by financial distress, interpersonal conflicts such as divorce, high levels of binge drinking and addiction. Conclusions Our data provide a topology of the needs of individuals at risk for suicide. Suicide prevention strategies can be improved by better transition support and explicit suicide monitoring in young adults with a developmental disorder, active waitlist policies in mental healthcare, and improved collaboration between debt counsellors and mental care professionals based on shared accountability. Key messages • Bereaved individuals elucidated a wide range of psychosocial problems in suicide decedents. • Suicide prevention should adhere to a needs-based approach.

Prediction of Suicidal Thoughts and Suicide Attempts in People Who Gamble Based on Biological-Psychological-Social Variables: A Machine Learning Study.

Recent research has shown that people who gamble are more likely to have suicidal thoughts and attempts compared to the general population. Despite the advancements made, no study to date has predicted suicide risk factors in people who gamble using machine learning algorithms. Therefore, current study aimed to identify the most critical predictors of suicidal ideation and suicidal attempts among people who gamble using a machine learning approach. An online survey conducted a cross-sectional analysis of 741 people who gamble (mean age: 25.9 ± 5.56). To predict the risk of suicide attempts and ideation, we employed a comprehensive set of 40 biological, psychological, social, and socio-demographic variables. The predictive models were developed using Logistic Regression, Random Forest (RF), robust eXtreme Gradient Boosting (XGBoost), and ensemble machine learning algorithms. Data analysis was performed using R-Studio software. Random Forest emerged as the top-performing algorithm for predicting suicidal ideation, with an impressive AUC of 0.934, sensitivity of 0.7514, specificity of 0.9885, PPV of 0.9473, and NPV of 0.9347. Across all models, dissociation, depression, and anxiety symptoms consistently emerged as crucial predictors of suicidal ideation. However, for suicide attempt prediction, all models exhibited weaker performance. XGBoost showed the best performance in this regard, with an AUC of 0.663, sensitivity of 0.78, specificity of 0.8990, PPV of 0.34, NPV of 0.984, and accuracy of 0.8918. Depressive symptoms and rumination severity were highlighted as the most important predictors of suicide attempts according to this model. These findings have important implications for clinical practice and public health interventions. Machine learning could help detect individuals prone to suicidal ideation and suicide attempts among people who gamble, assisting in creating tailored prevention programs to address future suicide risks more effectively.

Provider Perspectives on Healthcare System Barriers to Engaging Caregivers of Rural US Veterans in Firearm Suicide Prevention.

Firearms account for the majority of US Veteran suicides. Studies demonstrate that secure firearm storage can decrease risk of suicide. Veterans prefer to have discussions about secure firearm storage with familiar individuals, such as their caregivers. Providers in the Department of Veterans Affairs (VA) are well-positioned to engage caregivers of Veterans in firearm suicide prevention; however, challenges remain. Elucidate healthcare system barriers to caregiver engagement in firearm suicide prevention, as seen by Suicide Prevention and Caregiver Support Program staff, to inform development of an intervention to increase involvement of caregivers in efforts to reduce Veteran firearm suicides. An Advisory Board consisting of key stakeholders provided input regarding study methods and interpretation of findings. Participants included 16 Caregiver Support Program and 14 Suicide Prevention staff recruited from 8 geographically diverse rural VA medical centers. Interview topics and data analyses were guided by the Consolidated Framework for Implementation Research (CFIR). Rapid analysis and a team-based approach were employed to identify key barriers; emergent themes were mapped onto relevant constructs of the CFIR inner setting domain. Healthcare system barriers included structural characteristics (e.g., understaffing and heavy workloads), relational connections and communications (e.g., poor channels of communication between Caregiver Support Program and Suicide Prevention staff), culture (e.g., lack of recipient-centered care), and access to knowledge and information (e.g., lack of training in how to incorporate caregivers in Veteran care). The present study provides much needed perspective from VA staff in Suicide Prevention and Caregiver Support programs regarding barriers to involving caregivers in firearm suicide prevention efforts within the VA healthcare setting. Identifying these barriers and potential solutions is a critical step towards increasing engagement of caregivers in reducing firearm suicide among Veterans.

Real-time exposure to negative news media and suicidal ideation intensity among LGBTQ young adults in a high-stigma US state

Abstract Background With a recent surge in anti-LGBTQ policies and associated news/media coverage worldwide, there is a crucial need to study the role of LGBTQ negative news/media on proximal risk for suicide among LGBTQ youth, especially in high-stigma contexts. Methods Using a smartphone-based protocol, participants responded to brief self-report surveys 3x per day for 28 consecutive days. LGBTQ young adults (ages 18-24 years old) residing in Tennessee with recent suicidal ideation were recruited. At each assessment, participants reported real-time exposure to negative news/media, whether the news/media was related to LGBTQ topics, expectations of anti-LGBTQ rejection, and current intensity of passive suicidal ideation, active suicidal ideation, and self-harm ideation. Statistical analyses employed multilevel path analyses with restricted maximum likelihood estimation. Results 31 participants completed 2189 assessments (90.5% median compliance). Within-person effects showed that real-time exposure to LGBTQ - but not general - negative news/media was positively associated with suicidal ideation intensity. Significant indirect effects were present from exposure to LGBTQ negative news/media to higher suicidal ideation intensity through expectations of rejection. Mediation through expectations of rejection accounted for 25% and 39% of the total effect for active and passive suicidal ideation, respectively. Conclusions Findings have important public health implications related to media reporting, policy, and clinical intervention. Interventions targeting media organizations should promote responsible reporting practices and increase awareness of the potential suicidogenic impact of negative LGBTQ news coverage. This study joins numerous others in documenting the potential mental health harms of policies that restrict LGBTQ visibility and rights. Mental health professionals play a vital role in promoting coping in the face of exposure to LGBTQ negative news/media content.

Ketamine and chronic treatment-resistant depression: real-world practice and after relapse

BackgroundChronic treatment-resistant depression (TRD) poses a major challenge for clinicians. Ketamine has shown a rapid but short-lived antidepressant effect in several studies involving TRD patients with different demographic and clinical profiles. Our study aimed to assess the antidepressant effect of serial infusion sessions of ketamine in patients with chronic TRD and evaluate the severity of symptoms after relapse and the general psychiatric health of the responding patients.MethodsIn this single arm open-label study, six infusions of ketamine at 0.5 mg/kg were administered to chronic TRD patients for approximately two weeks. Response and remission rates, side effects, adverse events and after-relapse symptoms were evaluated, and patients were followed for three months.Results23 patients underwent at least one infusion session, and 18 patients completed the six sessions. Twelve (66.67%) patients responded to the treatment at some point, and 11 (61.11%) patients maintained response after the end of the treatment protocol. One infusion was not sufficient to achieve a response (P > 0.9999, z = 1.81), and more than half of the responders met the response criteria after the third infusion. Only one patient (5.56%) achieved remission at the end of the infusion phase. All but one ketamine responders relapsed within one month after the end of the treatment. There was no statistical difference between baseline and after-relapse MADRS scores (P = 0.7886, 95% CI=-5.512-4.312, R2 = 0,008411). However, a high incidence of serious adverse events related to suicidality was evident; one of the non-responding patients attempted suicide and several attempts to sedate this patient with benzodiazepines failed. Two responding patients ended up with a suicidal attempt or severe suicidal thoughts.ConclusionsIntroducing rapid-acting antidepressant to manage TRD patients in clinical practice demands further investigation, and the benefit-to-harm ratio should be assessed in the light of the increased risk of suicidality.

A novel multi-model feature generation technique for suicide detection

Automated expert systems (AES) analyzing depression-related content on social media have piqued the interest of researchers. Depression, often linked to suicide, requires early prediction for potential life-saving interventions. In the conventional approach, psychologists conduct patient interviews or administer questionnaires to assess depression levels. However, this traditional method is plagued by limitations. Patients might not feel comfortable disclosing their true feelings to psychologists, and counselors may struggle to accurately predict situations due to limited data. In this context, social media emerges as a potentially valuable resource. Given the widespread use of social media in daily life, individuals often express their nature and mental state through their online posts. AES can efficiently analyze vast amounts of social media content to predict depression levels in individuals at an early stage. This study contributes to this endeavor by proposing an innovative approach for predicting suicide risks using social media content and machine learning techniques. A novel multi-model feature generation technique is employed to enhance the performance of machine learning models. This technique involves the use of a feature extraction method known as term frequency-inverse document frequency (TF-IDF), combined with two machine learning models: logistic regression (LR) and support vector machine (SVM). The proposed technique calculates probabilities for each sample in the dataset, resulting in a new feature set referred to as the probability-based feature set (ProBFS). This ProBFS is compact yet highly correlated with the target classes in the dataset. The utilization of concise and correlated features yields significant outcomes. The SVM model achieves an impressive accuracy score of 0.96 using ProBFS while maintaining a low computational time of 5.63 seconds even when dealing with extensive datasets. Furthermore, a comparison with state-of-the-art approaches is conducted to demonstrate the significance of the proposed method.

Regional deprivation and suicide mortality in Germany

Abstract Background Regional deprivation has been shown to be an important determinant of suicide risk. However, little is known about regional differences in suicide mortality in Germany. Methods Data from cause of death statistics (2015 to 2021) were analyzed. Regression analysis for data with negative binomial distribution was used to examine the association between the German Index of Multiple Deprivation 2015, covering 401 districts and district-free cities in Germany, and suicide mortality. Results The adjusted rate ratio for suicide in the most deprived deprivation area compared to the least deprived area was 1.18 (95% CI [1.72, 2.00]). A sensitivity analysis shows similar rate ratios for men (1.99 (95% CI [1.80, 2.18])) and women (1.69 (95% CI [1.49, 1.92])), but a higher rate ratio for rural areas (2.29 (95% CI [2.04, 2.57])) compared with urban areas (1.51 (95% CI [1.36, 1.68])). Conclusions The findings suggest that regional deprivation is associated with suicide mortality, and call for a more targeted approach to suicide prevention in deprived regions. Key messages • Study reveals higher suicide rates in deprived regions in Germany. • Study calls for tailored prevention strategies to address this disparity.

Disability Status and Early Suicide Risk Among Children With and Without a History of Suspensions/Expulsions

Suicidal behaviors are relatively common among children, yet literature concerning the associated risks remains limited. Using a cumulative risk model, this study examined the effects of disability status, suspensions/expulsions, school disconnectedness, and negative peer relationships on suicidal behaviors. Data used were from The Future of Families and Child Wellbeing Study, which yielded an analytic sample ( N = 2,463) that was majority male (52%), Black and Hispanic (81%), and with an average age of 9.3 years. Using multivariate logistic regression with subgroup analysis, negative peer relationships and being male were associated with suicidal ideation in children without a history of suspension/expulsion. On the contrary, children with a history of suspension/expulsion and a disability status had three times higher odds of suicidal ideation in comparison to children with no disability status. These findings have important research and practical implications, which are discussed.

Adaptation of an evidence-based, preventive intervention to promote mental health in Hispanic adolescents: eHealth Familias Unidas Mental Health.

Youth internalizing symptoms (i.e., depression and anxiety), suicide ideation and attempts have been rising in recent years, including among Hispanics. Disparities in mental healthcare are concerning and require intervention, ideally prevention or early intervention. Familias Unidas is a culturally-syntonic, family-centered intervention effective in reducing youth drug use and sexual risk, with evidence of unanticipated effects on internalizing symptoms. This paper describes the systematic process used to adapt the eHealth version of the Familias Unidas intervention to more directly address internalizing symptoms and suicide risk in preparation for an effectiveness-implementation hybrid trial for youth with elevated internalizing symptoms, a history of suicide ideation/attempts, or poor parent-youth communication. The resulting eHealth Familias Unidas Mental Health intervention is described. Guided by a 4-phase framework, the steps in the adaptation process involved: assessment of the community and intervention delivery setting (pediatric primary care clinics); integration of previous intervention research, including intervention mechanisms of action; and expert and community consultation via focus groups. Focus group analyses showed that youth and parents perceived that the intervention was helpful. Their feedback was categorized into themes that were used to directly target mental health by addressing technology use, parent mental health, and social support. Effective and scalable preventive interventions are needed to address mental health disparities. The systematic adaptation process described in this paper is an efficient approach to expanding interventions while maintaining known, empirical and theoretical mechanisms of action. Findings from the ongoing effectiveness-implementation trial will be critical.

Prevalence of Suicidal Ideation in Patients With COVID-19 Admitted to a Hospital in Qazvin, Iran

Background and Objectives: Many patients with COVID-19 show long-term physical and mental symptoms. This study aims to determine the prevalence of suicidal ideation in patients with COVID-19 hospitalized in a hospital in Qazvin, Iran. Methods: This is a cross-sectional study. The study population consists of all patients with COVID-19 admitted to Bou Ali Sina Hospital in Qazvin city. A two-part questionnaire was used to collect data; the first part surveys demographic characteristics, history of death due to COVID-19 in relatives, history of underlying disease, days of hospitalization, history of intubation, and history of hospitalization in intensive care units. The second part was the Beck Scale for Suicide Ideation. The collected data were analyzed in SPSS software, version 23 using ANOVA and logistic regression analysis. Results: Participants were 206 patients with COVID-19. The data showed that 12.8% of patients had high suicidal ideation. Younger patients were at a high risk of suicide (Mean age: 31.96±12.59 years) (P<0.001). In addition, patients who were single (P<0.001) and had a history of death due to COVID-19 in relatives or friends had higher suicidal ideation (P=0.005). Conclusion: About 12.8% of patients with Covid-19 admitted to the study hospital had a high suicidal ideation. There is no considerable difference between the prevalence of suicidal ideation reported in this study and the data reported for the general population of Iran before the pandemic.

Circumstantial Factors Among Kentucky Nurse Suicide Decedents, 2005 to 2019.

Background: Nurse suicide, a complex occupational health concern, is urgently in need of research due to the personal and occupational suicide risk factors experienced by nurses, namely mental health problems (e.g., depression), job problems, and substance misuse. Therefore, the study aims were to determine the contextual characteristics and circumstantial factors associated with nurse suicide in Kentucky. Methods: Secondary suicide data were obtained from the Kentucky Violent Death Reporting System (KYVDRS) from 2005 to 2019. Nurse suicide cases were identified using the Bureau of Labor Statistics Standard Occupational Classification. A mixed-methods analysis using descriptive statistics and qualitative evaluation was conducted to determine the distributions of demographic, injury, and weapon characteristics, followed by a qualitative analysis of the KYVDRS incident narrative text of nurse suicide decedents. Results: There were 88 decedents identified with a nursing occupation. The predominant means of death for male (59%) and female (45%) nurses involved firearms. Thematic analysis showed nurse decedents experienced a mental health problem (51%) and premeditated (50%) suicide preceding death. Incidentally, 51% of all cases experienced multiple circumstantial factors (e.g., relationship problems and premeditation) preceding their death by suicide. Discussion/Application to Practice: Nurse suicide decedents experienced multifactorial risk factors preceding their death by suicide, namely depression, premeditated suicide, and relationship problems. Occupational health practitioners and health care organizations can implement individual and organizational prevention efforts to help prevent nurse suicide.

Domestic Violence Experience, Past Depressive Disorder, Unplanned Pregnancy, and Suicide Risk in the First Year Postpartum: Mediating Effect of Postpartum Depression.

This study aimed to examine the relationship between domestic violence, past depressive disorder, unplanned pregnancy, suicide risk, and postpartum depression among Korean women using a mediation model approach. A web-based, cross-sectional survey was performed from September 21 to September 30, 2022. Participants included 1,486 women in their first year postpartum from Chungnam Province, South Korea. The generalized linear model mediation was analyzed using the R-based Jamovi 2.3.21 version program. Of all participants, 8.7% had domestic violence experience, 7.5% had been diagnosed with depressive disorder by the doctor in the past, 33.0% had reported that the last pregnancy was unplanned, 39.9% had postpartum depression, and 11.2% of the participants had a suicide risk. Among the participants with a high risk of suicide, 34.4%, 37.5%, 53.1%, and 96.9% reported experiencing domestic violence, past depressive disorder, an unplanned pregnancy, and postpartum depression, respectively. Postpartum depression partially mediated the relationship between domestic violence, past depressive disorder, and suicide risk and completely mediated the relationship between unplanned pregnancy and suicide risk. Our findings highlight the necessity for screening and early intervention for suicide risk and postpartum depression from pregnancy to the first year postpartum.

Examining the relationship between religious volunteering and mental health screeners among firefighters

ABSTRACT Firefighters experience a high level of traumatic incidents in their work. This high level of exposure puts them at risk for trauma-related symptoms. These trauma symptoms include PTSD, suicide risk, alcohol misuse, and depression, and anxiety. Furthermore, repeated exposure has been found to lead to desensitization, flashbacks, irritability, and PTSD. Exposure to natural disasters may also impact ecospirituality. A survey was designed to include the Life Events Checklist, PHQ-9, GAD-7, PCL-5, and the SBQ-R. This survey was administered to firefighters from Northwest Florida in the fall of 2020 and spring of 2021. Participants were found to have higher levels of PTSD, anxiety, depression, and suicidal risk than the general population. Participants were further asked about their volunteering for religious organization activity. A significant relationship was found between volunteering for religious organization and mental health screeners. Religious volunteering predicted lower levels of PTSD symptoms, depression, and anxiety. The effect size was small.

Co-occurrence of Intimate Partner Violence and Suicide Mortality Among Adolescents and Young Adults in the United States

PurposeSuicide is the second leading cause of death among people ages 10–24 in the United States (US). Intimate partner violence (IPV) is also prevalent, and studies suggest that IPV is associated with youth suicide risk. However, limited research has investigated such associations. We conducted an in-depth analysis of youth suicides that were preceded by or coincided with IPV (i.e., IPV-related) to characterize how decedents were involved in IPV, IPV types experienced, and prior legal systems involvement. MethodsUsing US National Violent Death Reporting System (2014–2018) data for suicide decedents aged 7–24 (n = 15,430 deaths), we manually reviewed case narratives for IPV-related suicides (n = 882) and coded details concerning IPV circumstances. We conducted bivariate analyses to compare IPV-related suicides to other (non-IPV) youth suicides and examined differences in antecedent circumstances between IPV perpetrators (n = 651) and victims (n = 81) using multivariable logistic regression. ResultsAmong IPV-related suicides, most decedents were male and were described as IPV perpetrators. Physical IPV was most frequently reported. Compared to decedents with a history of IPV perpetration, decedents with a history of IPV victimization were more often female and younger. Narratives of IPV victim decedents had higher odds of reporting physical IPV; narratives of IPV perpetrator decedents had higher odds of reporting psychological IPV. DiscussionFuture research should explore the relationship between IPV perpetration and suicide and potential impacts on surviving partners. Research and practice should seek to develop effective assessments and interventions for young people that address IPV and suicide risk simultaneously.

Suicide among post-Arabellion refugees in Germany.

Although immigrants are considered to be vulnerable to mental illness, there is limited knowledge regarding their suicide mortality. To investigate standardised mortality ratios (SMR) for suicide among the largest immigrant populations in Germany before and after the refugee movement of 2015. Data on immigrants and the general population in Germany between 2000 and 2020 were provided by the scientific section of the Federal Statistical Office. SMR with 95% confidence intervals were calculated by indirect standardisation for gender, age and calendar year for the pre-2015 and post-2015 time interval, first for all the immigrant populations studied and second for the Syrian, Afghan and Iraqi populations separately. Immigrants from the countries studied showed a lower suicide risk compared with the German reference population (SMR = 0.38, 95% CI = 0.35-0.41). No differences in SMR were found between pre- and post-2015 time intervals, in either the aggregate data for all populations or the data for Syrian, Afghan and Iraqi populations. Post-2015, Afghan immigrants (SMR = 0.68, 95% CI = 0.54-0.83) showed a higher SMR than Syrians (SMR = 0.30, 95% CI = 0.25-0.36) or Iraqis (SMR = 0.37, 95% CI = 0.26-0.48). Despite the many and varied stresses associated with flight, comparison of the pre- and post-2015 time intervals showed that the suicide risk of the populations studied did not change and was considerably lower than that of the German reference population. We attribute this to lower suicide rates in the countries of origin but also to flight-related selection processes that favour more resilient individuals.

Development of an AI-Based Suicide Ideation Prediction Model for People with Disabilities

South Korea has one of the highest suicide rates among countries in the Organisation for Economic Co-Operation and Development, and the suicide rate among people with disabilities is more than twice that of the general population. This study aimed to develop an artificial intelligence-based suicide ideation prediction model for people with disabilities in order to provide a proactive approach for managing high-risk groups and offer evidence for establishing suicide prevention policies. The support vector machine, adaptive boost (AdaBoost), and bidirectional long short-term memory (Bi-LSTM) models were used in this study. Data from the Disability and Life Dynamics Panel for 2018–2021 were used. The performance of the models was evaluated based on the accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). All the prediction models demonstrated excellent performance, with AUC > 0.80 (0.83–0.87). The best-performing models were AdaBoost (0.87) for accuracy, Bi-LSTM (0.90) for sensitivity, and AdaBoost (0.90) for specificity. This study is the first to develop an artificial intelligence-based suicide ideation prediction model for disabled people and is significant in that it suggests ways to pre-emptively manage groups at high risk for suicide, providing evidence for the establishment of suicide prevention policies.

Evaluating suicide prevention gatekeeper training designed to identify and support people from asylum-seeking and refugee backgrounds

BackgroundSuicide-related behaviours and individual risk factors for suicide differ between ethnicities and demonstrate additional variation based on voluntary and forced migration. People forcibly displaced by violence and conflict, such as those seeking asylum and refugees, are likely to face stressors that can increase suicide risk. Research into evidenced-based suicide prevention strategies among people from asylum-seeking and refugee backgrounds is scarce. However, early, contextually-appropriate, identification and intervention may be a promising way to facilitate support for people in these groups. This research proposes that a contextually-responsive gatekeeper training is an appropriate strategy to increase the identification and support for people from asylum-seeking and refugee backgrounds.MethodsThe present article relates to the statistical findings of a larger mixed-method study used to validate and refine a contextually-responsive gatekeeper training program. The qualitative results of this research will be published separately. The outcome measures – knowledge about suicide in multicultural contexts, attitudes towards suicide and prevention, and self-efficacy to intervene were measured quantitatively, adopting a similar pre- and post-training procedure used in previous training evaluations. Using Generalised Estimating Equations, statistical comparisons were made between three identical self-report surveys completed by participants across three consecutive time points – pre-training, immediately post-training, and three months following training completion – known in this investigation as time-point zero (T0), time-point one (T1), and time-point two (T2). Lastly, during the T2 follow-up, additional open-ended questions were included to understand which areas of training they feel prepared them effectively and how the program could have better prepared them to intervene.ResultsA total of 28 participants took part in the study. Quantitative analysis indicated the program’s capacity to exert a significant favourable and lasting influence on knowledge about suicide and self-efficacy to intervene. In addition, follow-up measurements suggest that the content delivered to participants transferred effectively into real-world suicide prevention behaviours.ConclusionsFindings suggest that tailored suicide prevention training can have a significant influence on knowledge about suicide in multicultural contexts, self-efficacy to intervene in a crisis, and that course content translates effectively into real-world suicide prevention behaviour. Modifying training practices, based on feedback from contextually-experienced attendees, appears to be a pivotal factor in promoting the support of people from asylum-seeking and refugee backgrounds.