Explainable artificial intelligence models for predicting risk of suicide using health administrative data in Quebec.

Suicide is a complex, multidimensional event, and a significant challenge for prevention globally. Artificial intelligence (AI) and machine learning (ML) have emerged to harness large-scale datasets to enhance risk detection. In order to trust and act upon the predictions made with ML, more intuitive user interfaces must be validated. Thus, Interpretable AI is one of the crucial directions which could allow policy and decision makers to make reasonable and data-driven decisions that can ultimately lead to better mental health services planning and suicide prevention. This research aimed to develop sex-specific ML models for predicting the population risk of suicide and to interpret the models. Data were from the Quebec Integrated Chronic Disease Surveillance System (QICDSS), covering up to 98% of the population in the province of Quebec and containing data for over 20,000 suicides between 2002 and 2019. We employed a case-control study design. Individuals were considered cases if they were aged 15+ and had died from suicide between January 1st, 2002, and December 31st, 2019 (n = 18339). Controls were a random sample of 1% of the Quebec population aged 15+ of each year, who were alive on December 31st of each year, from 2002 to 2019 (n = 1,307,370). We included 103 features, including individual, programmatic, systemic, and community factors, measured up to five years prior to the suicide events. We trained and then validated the sex-specific predictive risk model using supervised ML algorithms, including Logistic Regression (LR), Random Forest (RF), Extreme Gradient Boosting (XGBoost) and Multilayer perceptron (MLP). We computed operating characteristics, including sensitivity, specificity, and Positive Predictive Value (PPV). We then generated receiver operating characteristic (ROC) curves to predict suicides and calibration measures. For interpretability, Shapley Additive Explanations (SHAP) was used with the global explanation to determine how much the input features contribute to the models' output and the largest absolute coefficients. The best sensitivity was 0.38 with logistic regression for males and 0.47 with MLP for females; the XGBoost Classifier with 0.25 for males and 0.19 for females had the best precision (PPV). This study demonstrated the useful potential of explainable AI models as tools for decision-making and population-level suicide prevention actions. The ML models included individual, programmatic, systemic, and community levels variables available routinely to decision makers and planners in a public managed care system. Caution shall be exercised in the interpretation of variables associated in a predictive model since they are not causal, and other designs are required to establish the value of individual treatments. The next steps are to produce an intuitive user interface for decision makers, planners and other stakeholders like clinicians or representatives of families and people with live experience of suicidal behaviors or death by suicide. For example, how variations in the quality of local area primary care programs for depression or substance use disorders or increased in regional mental health and addiction budgets would lower suicide rates.

Emergency department care experience of suicidal patients: A qualitative analysis of patients' perspectives.

INTRODUCTION: Individuals experiencing suicidal ideation or behavior frequently seek assistance at the emergency department (ED), yet the care they receive does not consistently align with their needs. This study explores the ED care experience of suicidal patients from their own perspective and offers recommendations to improve ED care for this population. METHOD: This qualitative study uses a descriptive interpretative design. Semi-structured interviews were conducted with 7 individuals who sought care in an ED due to suicidal ideation or behavior. Transcripts were analyzed using thematic analysis. FINDINGS: Participants' experiences were marked by an unsuited physical environment described as uncomfortable, noisy, and depressing. The organization of care was perceived as inadequate as patients complained about limited front-line access to mental health expertise, long waiting times, overworked staff, and inequities between patients with physical injuries and those with mental health concerns. Participants reported feelings of being trapped, left on their own and mistreated during their ED stay. Most found their care experience unhelpful or distressing, leaving them reluctant to reconsult. Specific recommendations based on patients' testimonials and literature are provided to enhance the ED care experience of suicidal patients. CONCLUSION: This study highlights several areas for improvement of the ED care experience of suicidal patients. Changes in current practices are needed to offer suicidal patients the satisfying care experience they deserve.

Suicidality and mood disorders in psychiatric emergency patients: Results from SBQ-R.

Patients with mood disorders are at high risk of suicidality, and emergency departments (ED) are essential in the management of this risk. This study aims to (1) describe the suicidal thoughts and behaviours of patients with mood disorders who come to ED; (2) assess the psychometric properties of the Suicidal Behaviours Questionnaire-Revised (SBQ-R) in a psychiatric ED; and (3) determine the best predictors of suicidality for these patients. A total of 300 participants with mood disorders recruited for the Signature Bank of the Institut universitaire en santé mentale de Montréal (IUSMM) were retained. Suicidality was assessed using the SBQ-R. Other clinical and demographic details were recorded. Bivariate analyses, correlations and multivariate regression analyses were conducted. SBQ-R's internal consistency, construct and convergent validities were also tested. In the Patient Health Questionnaire-9 (PHQ-9), 53.3% of the sample stated they had suicidal or self-harm thoughts in the last 2 weeks. The mean score obtained at the SBQ-R was 8.3. Multivariate analysis found that SBQ-R scores were associated with depressive symptoms and substance use, especially alcohol, accounting for 44.3% of the model variance. Cronbach's alpha was 0.81 [0.78, 0.84] and factor loadings for items 1-4 were 0.68, 0.88, 0.54, and 0.85, respectively. The confirmatory factor analysis indicated that the model fit the data well. The SBQ-R is a brief and valid instrument that can easily be used in busy emergency departments to assess suicide risk. Depressive symptoms and alcohol use shall also be assessed, as they are determinants of increased risk of suicidality.

A case-control study on predicting population risk of suicide using health administrative data: a research protocol.

INTRODUCTION: Suicide has a complex aetiology and is a result of the interaction among the risk and protective factors at the individual, healthcare system and population levels. Therefore, policy and decision makers and mental health service planners can play an important role in suicide prevention. Although a number of suicide risk predictive tools have been developed, these tools were designed to be used by clinicians for assessing individual risk of suicide. There have been no risk predictive models to be used by policy and decision makers for predicting population risk of suicide at the national, provincial and regional levels. This paper aimed to describe the rationale and methodology for developing risk predictive models for population risk of suicide. METHODS AND ANALYSIS: A case-control study design will be used to develop sex-specific risk predictive models for population risk of suicide, using statistical regression and machine learning techniques. Routinely collected health administrative data in Quebec, Canada, and community-level social deprivation and marginalisation data will be used. The developed models will be transformed into the models that can be readily used by policy and decision makers. Two rounds of qualitative interviews with end-users and other stakeholders were proposed to understand their views about the developed models and potential systematic, social and ethical issues for implementation; the first round of qualitative interviews has been completed. We included 9440 suicide cases (7234 males and 2206 females) and 661 780 controls for model development. Three hundred and forty-seven variables at individual, healthcare system and community levels have been identified and will be included in least absolute shrinkage and selection operator regression for feature selection. ETHICS AND DISSEMINATION: This study is approved by the Health Research Ethnics Committee of Dalhousie University, Canada. This study takes an integrated knowledge translation approach, involving knowledge users from the beginning of the process.

The 5-year longitudinal diagnostic profile and health services utilization of patients treated with electroconvulsive therapy in Quebec: a population-based study.

PURPOSE: Electroconvulsive therapy (ECT) is effective for treating several psychiatric disorders. However, only a minority of patients are treated with ECT. It is of primary importance to characterize their profile for epidemiological purposes and to inform clinical practice. We aimed to characterize the longitudinal profile of psychopathology and services utilization of patients first treated with ECT. METHODS: We conducted a population-based comparative study using data from a national administrative database in Quebec. Patients who received a first ECT between 2002 and 2016 were compared to controls who were hospitalized in psychiatry but did not receive ECT. We performed descriptive analyses to compare psychiatric diagnoses, domains of psychopathology (internalizing, externalizing and thought/psychotic disorders), medical services and medication use in the 5 years prior to the ECT or hospitalization. RESULTS: 5 080 ECT patients were compared with 179 594 controls. Depressive, anxiety, bipolar and psychotic disorders were more frequent in the ECT group. 96.2% of ECT patients had been diagnosed with depression and 53.8% with a primary psychotic disorder. In the ECT group, 1.0% had been diagnosed exclusively with depression and 47.0% had disorders from that belong to all three domains of psychopathology. Having both internalizing and thought/psychotic disorders was associated with an increased likelihood of receiving ECT vs having internalizing disorders alone (unadjusted OR = 2.93; 95% CI = 2.63, 3.26). All indicators of mental health services utilization showed higher use among ECT patients. CONCLUSION: Our results provide robust evidence of complex longitudinal psychopathology and extensive services utilization among ECT patients.

Workplace Factors, Burnout Signs, and Clinical Mental Health Symptoms among Mental Health Workers in Lombardy and Quebec during the First Wave of COVID-19.

Several research contributions have depicted the impact of the pandemic environment on healthcare and social care personnel. Even though the high prevalence of burnout depression and anxiety in healthcare settings before COVID-19 has been well documented in the research, the recent increase in psychological distress and mental health issues in healthcare and mental health workers should be attributed to the effect of the COVID-19 pandemic. The aim of the present study is to develop, evaluate, and compare a model of COVID-19 workplace stressors between two different territories, the Italian region of Lombardy and the Canadian province of Quebec. Within this model, burnout is depicted as the strongest determinant of mental health symptoms for mental health workers. In turn, the main workplace determinants of burnout are the perception of a lack of support from the organization and the fear of contracting COVID-19 at work. Findings also provide insights for designing interventions to promote and protect mental health workers in the context of the pandemic. In conclusion, it is necessary to monitor burnout and carefully analyze elements of organizational culture, in addition to offering clinical and psychological care for those in need.

Trajectories of adolescent poly-substance use and their long-term social and economic outcomes for males from low-income backgrounds.

Substance abuse is a significant public health concern that disproportionately burdens males and low-income communities. This study examined (1) longitudinal profiles of male adolescent poly-substance use and (2) their association with social and economic participation across early adulthood. Drawing on a cohort of males (n = 890) from low-income neighborhoods, we used group-based multi-trajectory modeling to identify profiles of poly-substance use (alcohol, tobacco, cannabis, illicit drugs) from age 13-17 years. Regression models were used to link substance use profiles to high school graduation, criminal convictions, personal and household earnings, welfare receipt and partnership from age 19-37 years, obtained from administrative records. Child IQ, family adversity and behavioral problems were adjusted for. Four poly-substance use profiles were identified: abstinent (n = 128, 14.4%), late-onset (n = 412, 46.5%), mid-onset (n = 249, 28.1%), and early-onset (n = 98, 11.1%). Relative to the late-onset (reference) group, participants in the early-onset profile were 3.0 times (95%CI = 1.68-5.53) more likely to have left school without a diploma, 2.7 times (95% CI = 1.56-4.68) more likely to have a criminal conviction by age 24 years, earned 10,185 USD less (95% CI = - 15,225- - 5144) per year at age 33-37 years and had 15,790 USD lower (95% CI = - 23,378- - 8218) household income at age 33-37 years, a 1.3 times (95%CI = 1.15-1.57) higher incidence of annual welfare receipt and a 24% (95% CI = 5-40) lower incidence of marriage/cohabitation from age 18-35 years. We show that adolescent-onset poly-substance use by age 13 is associated with poor social and economic outcomes. Delaying the onset of substance use and reducing exposure to additional substance classes has potential for high societal cost savings.

The Early Impact of the Covid-19 Emergency on Mental Health Workers: A Survey in Lombardy, Italy.

Lombardy was the epicenter of the Covid-19 outbreak in Italy, and in March 2020 the rapid escalation in cases prompted the Italian Government to decree a mandatory lockdown and to introduce safety practices in mental health services. The general objective of the study is to evaluate the early impact of the Covid-19 emergency and quarantine on the well-being and work practices of mental health service personnel and professionals. Data were collected through an online survey of workers and professionals working with people with mental health problems in Lombardy in several outpatient and inpatient services. Their socio-demographic characteristics, professional background, description of working conditions during lockdown and psychological distress levels were collected. All analyses were performed on a sample of 241. Approximately, 31% of the participants obtained a severe score in at least one of the burnout dimensions, 11.6% showed moderate or severe levels of anxiety, and 6.6% had a moderate or severe level of depression. Different work conditions and patterns of distress were found for outpatient service workers and inpatient service workers. The overall impact of the Covid-19 emergency on mental health workers' level of distress was mild, although a significant number of workers experienced severe levels of depersonalization and anxiety. More research is needed to assess specific predictive factors.