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.

System dynamics models of depression at the population level: a scoping review.

AIMS: Depression is a disease driven by dynamic processes both at the individual- and system-level. System dynamics (SD) models are a useful tool to capture this complexity, project the future prevalence of depression and understand the potential impact of interventions and policies. SD models have been used to model infectious and chronic disease, but rarely applied to mental health. This scoping review aimed to identify population-based SD models of depression and report on their modelling strategies and applications to policy and decision-making to inform research in this emergent field. METHODS: We searched articles in MEDLINE, Embase, PsychInfo, Scopus, MedXriv, and abstracts from the System Dynamics Society from inception to October 20, 2021 for studies of population-level SD models of depression. We extracted data on model purpose, elements of SD models, results, and interventions, and assessed the quality of reporting. RESULTS: We identified 1899 records and found four studies that met the inclusion criteria. Studies used SD models to assess various system-level processes and interventions, including the impact of antidepressant use on population-level depression in Canada; the impact of recall error on lifetime estimates of depression in the USA; smoking-related outcomes among adults with and without depression in the USA; and the impact of increasing depression incidence and counselling rates on depression in Zimbabwe. Studies included diverse stocks and flows for depression severity, recurrence, and remittance, but all models included flows for incidence and recurrence of depression. Feedback loops were also present in all models. Three studies provided sufficient information for replicability. CONCLUSIONS: The review highlights the usefulness of SD models to model the dynamics of population-level depression and inform policy and decision-making. These results can help guide future applications of SD models to depression at the population-level.

Social isolation, loneliness and positive mental health among older adults in Canada during the COVID-19 pandemic. / Isolement social, solitude et santé mentale positive chez les aînés au Canada pendant la pandémie de COVID-19.

INTRODUCTION: Social isolation and loneliness are associated with poorer mental health among older adults. However, less is known about how these experiences are independently associated with positive mental health (PMH) during the COVID-19 pandemic. METHODS: We analyzed data from the 2020 and 2021 cycles of the Survey on COVID-19 and Mental Health to provide estimates of social isolation (i.e. living alone), loneliness and PMH outcomes (i.e. high self-rated mental health, high community belonging, mean life satisfaction) in the overall older adult population (i.e. 65+ years) and across sociodemographic groups. We also conducted logistic and linear regressions to separately and simultaneously examine how social isolation and loneliness are associated with PMH. RESULTS: Nearly 3 in 10 older adults reported living alone, and over a third reported feelings of loneliness due to the pandemic. When examined separately, living alone and loneliness were each associated with lower PMH. When assessed simultaneously, loneliness remained a significant independent factor associated with all three PMH outcomes (overall and across all sociodemographic groups), but living alone was only a significant factor for high community belonging in the overall population, for males and for those aged 65 to 74 years. CONCLUSION: Overall, social isolation and loneliness were associated with poorer wellbeing among older adults in Canada during the pandemic. Loneliness remained a significant factor related to all PMH outcomes after adjusting for social isolation, but not vice versa. The findings highlight the need to appropriately identify and support lonely older adults during (and beyond) the pandemic.

The association between social media use and physical activity among Canadian adolescents: a Health Behaviour in School-aged Children (HBSC) study.

OBJECTIVE: To determine the association between social media use (SMU) and physical activity (PA) among Canadian adolescents. METHODS: We used data from 12,358 participants in grades 6 to 10 who responded to the Canadian component of the 2017/2018 Health Behaviour in School-aged Children (HBSC) survey. Social media intensity and problematic SMU were assessed using a 4-point mutually exclusive scale that contained three categories based on intensity (non-active, active, and intense SMU) and one category based on the presence of addiction-like symptoms irrespective of intensity (problematic SMU). PA was assessed for five domains (i.e., school curriculum, organized sport, exercise, outdoor play, and active transport) and dichotomized using the first quartile to represent high PA engagement in each domain. Meeting PA recommendation of 60 min per day of moderate-to-vigorous PA was calculated using the sum of the five domains. Logistic regression models were used to assess the association between SMU and PA, with active SMU used as the reference group for all models. RESULTS: Non-active SMU was associated with lower odds of meeting the daily PA recommendations and of high engagement in all five domains of PA when compared to active SMU. Intense SMU was associated with higher odds of meeting the daily PA recommendations. Problematic SMU was not associated with meeting daily PA recommendations, but it was significantly associated with lower odds of high PA engagement in the exercise domain. CONCLUSION: The findings of this study suggest that non-active SMU was significantly associated with lower PA levels. Problematic SMU was only significantly associated with lower PA levels in the exercise domain. Intense SMU was associated with higher odds of meeting the PA recommendation.

RéSUMé: OBJECTIF: Établir la relation entre l'utilisation des médias sociaux et l'activité physique chez les adolescents canadiens. MéTHODES: Nous avons utilisé les données de 12 358 participants de la 6e à la 10e année qui ont répondu au volet canadien de l'Enquête sur les comportements de santé des jeunes d'âge scolaire de 2017­2018 (HBSC). L'intensité des médias sociaux et leur utilisation problématique ont été évaluées à l'aide d'une échelle à quatre points mutuellement exclusifs contenant trois catégories basées sur l'intensité (utilisation non active, active et intense des médias sociaux) et une catégorie basée sur la présence de symptômes de dépendance indépendamment de l'intensité (utilisation problématique des médias sociaux). L'activité physique a été évaluée pour cinq domaines (c'est-à-dire le programme scolaire, le sport organisé, l'exercice, le jeu en plein air et le transport actif) et dichotomisée en utilisant le premier quartile pour représenter la participation à une activité physique élevée dans chaque domaine. Le respect de la recommandation de 60 min par jour d'activité physique modérée à vigoureuse a été calculé en utilisant la somme des cinq domaines. Des modèles de régression logistique ont été utilisés pour évaluer la relation entre l'utilisation des médias sociaux et l'activité physique, l'utilisation active des médias sociaux étant utilisée comme groupe de référence pour tous les modèles. RéSULTATS: L'utilisation non active des médias sociaux était associée à une probabilité plus faible de respecter les recommandations en matière d'activité physique quotidienne et à une participation élevée dans les cinq domaines de l'activité physique, par rapport à l'utilisation active des médias sociaux. Une utilisation intense des médias sociaux était associée à une probabilité plus élevée de respecter les recommandations en matière d'activité physique quotidienne. Une utilisation problématique des médias sociaux n'était pas associée au respect des recommandations en matière d'activité physique quotidienne, mais elle était fortement associée à une probabilité moindre de participation élevée à une activité physique dans le domaine de l'exercice. CONCLUSION: Les résultats de cette étude suggèrent que l'utilisation non active des médias sociaux est fortement associée à des niveaux d'activité physique plus faibles. Une utilisation problématique des médias sociaux n'était fortement associée à des niveaux d'activité physique plus faibles que dans le domaine de l'exercice. Une utilisation intense des médias sociaux était associée à une probabilité plus élevée de respecter la recommandation en matière d'activité physique.

Differential Trajectories of Delusional Content and Severity Over 2 Years of Early Intervention for Psychosis: Comparison Between Chennai, India, and Montréal, Canada.

BACKGROUND: There exist few direct studies of delusional content in psychosis across geo-cultural contexts, especially those in which treatment protocols and measures are comparable. To directly examine an illness outcome that is potentially culturally mediated, this study investigated the baseline presentation and longitudinal trajectory of delusions in first-episode psychosis (FEP) across 2 similar treatment settings in Montréal (Canada) and Chennai (India). STUDY DESIGN: Patients entering an early intervention program for FEP in Chennai (N = 168) and Montréal (N = 165) were compared on site-level differences in the presentation of delusions across specific time points over 2 years of treatment. Delusions were measured using the Scale for Assessment of Positive Symptoms. Chi-square and regression analyses were conducted. STUDY RESULTS: At baseline, delusions were more frequent in Montréal than in Chennai (93% vs 80%, respectively; X2(1) = 12.36, P < .001). Thematically, delusions of grandiosity, religiosity, and mind reading were more common in Montréal than in Chennai (all P < .001); however, these baseline differences did not persist over time. Regression revealed a significant time-by-site interaction in the longitudinal course of delusions, which differs from the trajectory of other FEP-positive symptom domains. CONCLUSIONS: To the best of our knowledge, this is the first direct comparison of delusions in similar programs for FEP across 2 different geo-cultural contexts. Our findings support the notion that delusion themes follow consistent ordinal patterns across continents. Future work is needed to unpack the differences in severity that present at baseline and minor differences in content.

Adolescent use of social media and associations with sleep patterns across 18 European and North American countries.

OBJECTIVE: Over the past decade, concurrent with increasing social media use (SMU), there has been a shift toward poorer sleep among adolescents in many countries. The purpose of this study was to examine the cross-national associations between adolescent SMU and sleep patterns, by comparing 4 different categories of SMU (nonactive, active, intense, and problematic use). DESIGN, SETTING, AND PARTICIPANTS: Data were from 86,542 adolescents in 18 European and North American countries that participated in the 2017/18 Health Behaviour in School-aged study. MEASUREMENTS: Mixed-effects linear regression models were used to examine cross-national associations between 4 SMU categories and adolescent sleep duration, bedtime and social jetlag derived from self-reported data. RESULTS: For all countries combined, nonactive SMU was associated with longer sleep, earlier bedtimes, and less social jetlag, compared to active SMU, although the differences were minor. By comparison, intense and problematic SMU were associated with less sleep and later bedtimes on both school and nonschool days, and greater social jetlag, compared to active SMU. While findings were relatively consistent between countries, some differences were observed, suggesting that the national and cultural context may be important in interpreting results. CONCLUSIONS: These findings suggest that both intense and problematic SMU are associated with poorer sleep patterns in adolescents across most countries. Further research is needed to identify effective policies, programs, and messaging to promote the healthy use of social media and prevent potential negative impacts on adolescent sleep.

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.

Increased alcohol use, heavy episodic drinking, and suicide ideation during the COVID-19 pandemic in Canada.

OBJECTIVE: Alcohol use is a known risk factor for suicidality, yet this relationship has not been explored during the pandemic in Canada. As a growing body of evidence demonstrates the negative impact of COVID-19 on alcohol consumption and associated harms in Canada, there is a need to examine this more closely. METHODS: Using the Survey on COVID-19 and Mental Health 2020, we compared the prevalence of suicide ideation among: (1) individuals who reported an increase in alcohol consumption vs those who reported a decrease/no change, and (2) individuals who reported past month heavy episodic drinking vs those who did not. We compared overall unadjusted odds ratios and across a number of sociodemographic and mental health variables. All estimates were weighted to ensure they were nationally representative. RESULTS: The prevalence and likelihood of suicide ideation were significantly higher among people who reported increased alcohol consumption during the pandemic (4.9% vs 2.0%; OR = 2.6, 95% CI: 1.8, 3.7) and people who reported past month heavy episodic drinking (3.4% vs 2.1%; OR = 1.7, 95% CI: 1.2, 2.3). Males and middle-aged and older-aged individuals had the highest odds ratios for increased alcohol consumption and past month heavy episodic drinking with suicide ideation. CONCLUSION: In the Canadian general population during the COVID-19 pandemic, there were significant associations between suicide ideation and increased alcohol use as well as past month heavy episodic drinking across specific sociodemographic subgroups. Future research could explore these associations while adjusting for social determinants of health such as income security, employment, education, social support, stress, and mental health.

RéSUMé: CONTEXTE: La consommation d'alcool est un facteur de risque connu de suicidabilité, mais cette relation n'a pas été explorée pendant la pandémie au Canada. Comme un nombre croissant de preuves démontre les répercussions négatives de la COVID-19 sur la consommation d'alcool et les méfaits associés au Canada, il est nécessaire d'examiner cette question de plus près. MéTHODES: En utilisant l'enquête de 2020 sur la COVID-19 et la santé mentale, nous avons comparé la prévalence des idées suicidaires chez : (1) les individus qui ont déclaré une augmentation de leur consommation d'alcool par rapport à ceux qui ont déclaré une diminution ou aucun changement, et (2) les individus qui ont déclaré une consommation épisodique excessive d'alcool au cours du dernier mois par rapport à ceux qui n'en ont pas déclaré. Nous avons comparé les rapports des cotes globaux non ajustés et selon un certain nombre de variables sociodémographiques et de santé mentale. Toutes les estimations ont été pondérées afin de garantir leur représentativité nationale. RéSULTATS: La prévalence et la probabilité d'idées suicidaires étaient considérablement plus élevées chez les personnes ayant déclaré une consommation d'alcool accrue pendant la pandémie (4,9 % contre 2,0 %; RC=2,6, IC à 95 % : 1,8 à 3,7) et chez les personnes ayant déclaré une consommation épisodique excessive au cours du mois écoulé (3,4 % contre 2,1 %; RC=1,7, IC à 95 % : 1,2 à 2,3). Les hommes, les personnes d'âge moyen et les personnes âgées présentaient les rapports des cotes les plus élevés pour une consommation d'alcool accrue et une consommation épisodique excessive d'alcool au cours du mois précédent avec des idées suicidaires. CONCLUSION: Dans la population générale canadienne, pendant la pandémie de COVID-19, il y avait des associations significatives entre les idées suicidaires et l'augmentation de la consommation d'alcool ainsi que la consommation épisodique excessive d'alcool au cours du dernier mois dans des sous-groupes sociodémographiques précis. Les recherches futures pourraient explorer ces associations tout en ajustant les déterminants sociaux de la santé tels que la sécurité du revenu, l'emploi, l'éducation, le soutien social, le stress et la santé mentale.

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.

Adolescent social media use and its association with relationships and connections: Canadian Health Behaviour in School-aged Children, 2017/2018.

Background: The development of healthy relationships and connections is of fundamental importance to adolescent well-being. The use of social media plays a vital role in the lives of young Canadians, yet the association between different types of social media use and the quality of relationships and connections remains unknown, and most existing analyses on this topic are based on modest and non-representative samples. Data and methods: Using 2017/2018 reports from the nationally representative Health Behaviour in School-aged Children study (n=17,149; ages 11 to 15 years), the strength, consistency and significance of associations was examined between intensive (frequent use to connect with other people) and problematic (use that depicts addictive qualities) social media use and available measures of adolescent relationships and connections. Results: Overall, intensive use (online communication with others almost all of the time) and problematic use (potential addiction to social media) were more common in girls than boys (38% of girls versus 30% of boys and 7.7% of girls versus 5.2% of boys, respectively), with prevalence levels that rose with age. Intensive use was associated with more positive social relationships with friends, especially among girls (relative risk [RR] = 1.40 [95% confidence interval (CI) 1.28 - 1.54]), while problematic use was consistently and negatively associated with strong relationships and connections in all groups in the study. Notably, problematic use was negatively associated with strong family relationships in boys (RR = 0.58 [95% CI 0.42 to 0.79]) and girls (RR=0.48 [95% CI 0.36 to 0.63]). Interpretation: Intensive use of social media has the potential to strengthen relationships and connections in adolescents. However, when social media use becomes addictive or "problematic", it is highly correlated with weaker relationships and a sense of social disconnection. Public health initiatives related to social media use should consider how different types of social media use have the potential to impact on different aspects of health.

National strategy on the integration of sleep and circadian rhythms into public health research and policies: Report from the Canadian Sleep and Circadian Network.

Scientists in sleep and circadian rhythms, public health experts, healthcare providers, partners, and stakeholders convened in 2020 for a 2-day meeting organized by the Canadian Sleep and Circadian Network to develop a national strategy for integrating sleep and circadian rhythms into public health and policies in Canada. The objective of this paper is to present the national strategy that emerged from this meeting of 60 participants from across Canada. The meeting focused on 4 key target priorities: (1) atypical working schedules, (2) sleep and circadian rhythms of children and adolescents, (3) insomnia, and (4) impact of sleep apnea on health. Following constructive discussions, it was decided that the following 4 strategic objectives should be prioritized to accelerate the integration of sleep and circadian rhythms into public health policies in Canada: (1) increase public health sleep and circadian rhythm research, (2) increase public health education and knowledge mobilization on sleep, (3) inform and support public health sleep interventions and policies, and (4) promote sleep health training. Participants recommended that research and public health efforts address needs along the continuum of sleep health. The committee noted that strategies and interventions could differ across contexts, settings, sectors, and jurisdictions. The national strategy also identified high-priority research questions in public health and recommended mechanisms to build research capacity, providing a path forward for the integration of sleep and circadian rhythms into public health research and policies.

Years of life gained when meeting sleep duration recommendations in Canada.

OBJECTIVE: To estimate the years of life gained when meeting the sleep duration recommendations across the adult lifespan. METHODS: Three pieces of information were used to estimate and compare life expectancy at each age of adult life among Canadian adults who did and did not meet sleep duration recommendations: (i) the prevalence of self-reported short sleep duration, recommended sleep duration, and long sleep duration; (ii) the relative risks of all-cause mortality associated with sleep duration obtained from recent meta-analyses; and (iii) the probability of death during each year of life obtained from life tables. RESULTS: Adults who meet the sleep duration recommendations have an estimated life expectancy at age 20 years that is 1.2 years longer than short sleepers and 2.6 years longer than long sleepers. Differences between men and women were minimal. CONCLUSION: Meeting sleep duration recommendations is associated with increases in longevity in Canada.

Single-Item Happiness Measure Features Adequate Validity Among Adolescents.

Background: Happiness is becoming increasingly relevant in recent research, including adolescents. Many studies are using the single-item measure for adolescent happiness, however, its validity is not well known. We aimed to examine the validity of this measure among adolescents in three countries from distinct European regions - Eastern (Lithuania), Southern (Portugal), and Western (Scotland). Materials and Methods: The analysis included data from Health Behaviour in School-aged Children (HBSC) study from three countries and three last surveys (2009/10, 2013/14, and 2017/18). The total sample comprised 47,439 schoolchildren. For validity, the indicators reflecting subjective health, life satisfaction, quality of life, well-being, social support, health complaints, bullying, and self-directed violence were assessed. The calculations were conducted in the total sample and by gender, age, survey year, and country. Results: The different indicators of concurrent and convergent validity revealed consistent correlations with happiness, with better well-being, health, and subjective perceptions being related to higher happiness. Meanwhile, health complaints, bullying behaviors, and self-directed violence were related to lower happiness. The subgroup differences were consistent across gender, age groups, countries, and survey rounds. The extent of differences was more expressed among girls. Conclusion: The single item for adolescent happiness measurement features a consistent pattern of validity concerning indicators of concurrent and convergent validity. Higher self-reported happiness is associated with better mental and physical health and well-being, and less expressed negative factors (complaints, bullying, and self-directed violence). In addition, among girls the correlations tend to be stronger than boys.

Virtual Photovoice With Older Adults: Methodological Reflections during the COVID-19 Pandemic.

Photovoice is a participatory action research method in which participants take and narrate photographs to share their experiences and perspectives. This method is gaining in popularity among health researchers. Few studies, however, have described virtual photovoice data collection despite the growing interest among qualitative health researchers for online data collection. As such, the aim of this article is to discuss the implementation of a virtual photovoice study and presents some of the challenges of this design and potential solutions. The study examined issues of social isolation and mental health among older adults during the COVID-19 pandemic in the Canadian province of Québec. Twenty-six older adults took photographs depicting their experience of the pandemic that were then shared in virtual discussion groups. In this article, we discuss three key challenges arising from our study and how we navigated them. First, we offer insights into managing some of the technical difficulties related to using online meeting technologies. Second, we describe the adjustments we made during our study to foster and maintain positive group dynamics. Third, we share our insights into the process of building and maintaining trust between both researchers and participants, and amongst participants. Through a discussion of these challenges, we offer suggestions to guide the work of health promotion researchers wishing to conduct virtual photovoice studies, including with older adults.

Economic burden of insufficient sleep duration in Canadian adults.

OBJECTIVE: To provide estimates of the health care and productivity costs associated with insufficient sleep duration (<7 hours per night) in Canadian adults. METHODS: A prevalence-based approach was used to estimate the economic costs associated with insufficient sleep duration. Estimates relied on 3 pieces of information: (1) the relative risks of health outcomes consistently associated with insufficient sleep duration obtained from recent meta-analyses; (2) the direct (health care) and indirect (health-related losses of productivity) costs of these health outcomes obtained from the Economic Burden of Illness in Canada data; and (3) the prevalence of insufficient sleep duration in Canadian adults obtained from a representative national survey (17.2%). RESULTS: The estimated direct, indirect, and total costs of insufficient sleep duration in Canada in 2020 were $484 million, $18 million, and $502 million, respectively. These values represent 0.5% (direct), 2.7% (indirect), and 0.5% (total) of the overall burden of illness costs for Canada (estimated at $102 billion). The 2 most expensive chronic diseases attributable to insufficient sleep duration were depression ($219 million) and type 2 diabetes ($92 million). The main contributors of these health care costs were related to hospital care (for coronary heart disease and obesity), prescription drugs (for type 2 diabetes and depression), physician care (for hypertension and cognitive disorders), and mortality (for accidents/injuries). A 5% decrease in the prevalence of insufficient sleep duration (from 17.2% to 12.2%) in Canadian adults would lead to a yearly savings of $148 million. CONCLUSIONS: Insufficient sleep duration is an important contributor to health care spending and health-related losses of productivity in Canada. Studies are needed to test cost-effective sleep health interventions at the population level.

What is known about the prevalence of household food insecurity in Canada during the COVID-19 pandemic: a systematic review. / État des connaissances au sujet de la prévalence de l'insécurité alimentaire des ménages canadiens durant la pandémie de COVID-19 : revue systématique.

INTRODUCTION: Household food insecurity (HFI) is a persistent public health issue in Canada that may have disproportionately affected certain subgroups of the population during the COVID-19 pandemic. The purpose of this systematic review is to report on the prevalence of HFI in the Canadian general population and in subpopulations after the declaration of the COVID-19 pandemic in March 2020. METHODS: Sixteen databases were searched from 1 March 2020 to 5 May 2021. Abstract and full-text screening was conducted by one reviewer and the inclusions verified by a second reviewer. Only studies that reported on the prevalence of HFI in Canadian households were included. Data extraction, risk of bias and certainty of the evidence assessments were conducted by two reviewers. RESULTS: Of 8986 studies identified in the search, four studies, three of which collected data in April and May 2020, were included. The evidence concerning the prevalence of HFI during the COVID-19 pandemic is very uncertain. The prevalence of HFI (marginal to severe) ranged from 14% to 17% in the general population. Working-age populations aged 18 to 44 years had higher HFI (range: 18%-23%) than adults aged 60+ years (5%-11%). Some of the highest HFI prevalence was observed among households with children (range: 19%-22%), those who had lost their jobs or stopped working due to COVID-19 (24%-39%) and those with job insecurity (26%). CONCLUSION: The evidence suggests that the COVID-19 pandemic may have slightly increased total household food insecurity in Canada during the COVID-19 pandemic, especially in populations that were already vulnerable to HFI. There is a need to continue to monitor HFI in Canada.

Socioeconomic inequalities in adolescent health complaints: A multilevel latent class analysis in 45 countries.

Our study evaluated the relationship between adolescent health complaints and socioeconomic position in 45 countries. Data are from the 2017/2018 international Health Behaviour in School-aged Children survey which used proportionate sampling among adolescents aged 11 to 15 years old (n=228,979). Multilevel, multinomial regression analysis assessed the association between the multilevel latent classes with socioeconomic status (SES; at the household and country level). Three distinct latent classes were identified: No Complaints, Psychological Complaints, and a Physical and Psychological Complaints class; where, low household SES was highest for the physical and psychological complaints class. The findings suggest that health promotion policies and interventions among adolescents should consider the specific needs of adolescents living with low household SES as they report more subjective health complaints.

Patterns and motivations of polysubstance use: a rapid review of the qualitative evidence. / Tendances et motivations associées à la polyconsommation de substances : une revue rapide des données qualitatives.

INTRODUCTION: Polysubstance use-the use of substances at the same time or close in time-is a common practice among people who use drugs. The recent rise in mortality and overdose associated with polysubstance use makes understanding current motivations underlying this pattern critical. The objective of this review was to synthesize current knowledge of the reasons for combining substances in a single defined episode of drug use. METHODS: We conducted a rapid review of the literature to identify empirical studies describing patterns and/or motivations for polysubstance use. Included studies were published between 2010 and 2021 and identified using MEDLINE, Embase, PsycINFO and Google Scholar. RESULTS: We included 13 qualitative or mixed-method studies in our analysis. Substances were combined sequentially to alleviate withdrawal symptoms or prolong a state of euphoria ("high"). Simultaneous use was motivated by an intention to counteract or balance the effect(s) of a substance with those of another, enhance a high or reduce overall use, and to mimic the effect of another unavailable or more expensive substance. Self-medication for a pre-existing condition was also the intention behind sequential or simultaneous use. CONCLUSION: Polysubstance use is often motivated by a desire to improve the experience based on expected effects of combinations. A better understanding of the reasons underlying substance combination are needed to mitigate the impact of the current overdose crisis.

The mental health of young people who are not in education, employment, or training: a systematic review and meta-analysis.

PURPOSE: There are increasing concerns about the intersection between NEET (not in education, employment, or training) status and youth mental ill-health and substance use. However, findings are inconsistent and differ across types of problems. This is the first systematic review and meta-analysis (PROSPERO-CRD42018087446) on the association between NEET status and youth mental health and substance use problems. METHODS: We searched Medline, EMBASE, Web of Science, ERIC, PsycINFO, and ProQuest Dissertations and Theses (1999-2020). Two reviewers extracted data and appraised study quality using a modified Newcastle-Ottawa Scale. We ran robust variance estimation random-effects models for associations between NEET and aggregate groups of mental ill-health and substance use measures; conventional random-effects models for associations with individual mental/substance use problems; and subgroup analyses to explore heterogeneity. RESULTS: We identified 24 studies from 6,120 references. NEET status was associated with aggregate groups of mental ill-health (OR 1.28, CI 1.06-1.54), substance use problems (OR 1.43, CI 1.08-1.89), and combined mental ill-health and substance use measures (OR 1.38, CI 1.15-1.64). Each disaggregated measure was associated with NEET status [mood (OR 1.43, CI 1.21-1.70), anxiety (OR 1.55, CI 1.07-2.24), behaviour problems (OR 1.49, CI 1.21-1.85), alcohol use (OR 1.28, CI 1.24-1.46), cannabis use (OR 1.62, CI 1.07-2.46), drug use (OR 1.99, CI 1.19-3.31), suicidality (OR 2.84, CI 2.04-3.95); and psychological distress (OR 1.10, CI 1.01-1.21)]. Longitudinal data indicated that aggregate measures of mental health problems and of mental health and substance use problems (combined) predicted being NEET later, while evidence for the inverse relationship was equivocal and sparse. CONCLUSION: Our review provides evidence for meaningful, significant associations between youth mental health and substance use problems and being NEET. We, therefore, advocate for mental ill-health prevention and early intervention and integrating vocational supports in youth mental healthcare.

Social inequalities in availability of health-promoting interventions in Québec elementary schools.

Availability of health-promoting interventions (HPIs) may vary across schools serving students with different socioeconomic backgrounds. Our objectives were to describe social inequalities across elementary schools in: (i) level of importance that school principals attribute to 13 common health-related issues among students in their school; (ii) availability of HPIs within their school addressing eight health topics and (iii) (mis)alignment between perceived importance and HPI availability. Data were collected in telephone interviews with school principals in a convenience sample of 171 elementary schools (59% of 291 schools contacted). Schools were categorized as serving very advantaged, moderately advantaged or disadvantaged students. Principals reported how important 13 health issues were among students in their schools and provided data on HPI availability for 8 health issues, 4 of which required school action under government mandates. Higher proportions of principals in schools serving disadvantaged students (36% of all 171 schools) perceived most health issues as important. The mean number of HPIs in the past year was 12.0, 12.1 and 11.7 in schools serving very advantaged, moderately advantaged and disadvantaged students, respectively. Only availability of mental health HPIs differed by school deprivation (60, 43 and 30% in very, moderately and disadvantaged, respectively). Although most schools offered oral health HPIs, dental problems were not perceived as important. Smoking was perceived as not important and smoking-related HPIs were relatively rare (9%). Given rapid evolution in public health priorities, (mis)alignment between perceived importance of specific health issues and HPI availability in elementary schools warrants ongoing reflection.

Interventions that promote health may not be present in all schools. Our objectives were to describe differences across elementary schools in: (i) level of importance that school principals attribute to common health-related issues among students in their own school; (ii) the presence of interventions that address these health issues and (iii) (mis)alignment between perceived importance and presence of interventions. Telephone interviews were conducted with school principals in 171 elementary schools serving very advantaged, moderately advantaged or disadvantaged students. Principals reported how important 13 health-related issues (four of which required school action under government mandates) were among students in their schools and whether interventions were present for 8 of the 13 issues. Higher proportions of principals in schools serving disadvantaged students perceived most health issues as important. Intervention availability did not differ across schools, except that higher proportions of schools serving advantaged students reported mental health interventions. Most schools offered oral health interventions, but dental problems were not perceived as important. Smoking was also not perceived as important and interventions were relatively rare. (Mis)alignment between perceived importance of health issues and intervention availability calls for ongoing reflection.