BackgroundDepression and suicide are critical social problems worldwide, but tools to objectively diagnose them are lacking. Therefore, this study aimed to diagnose depression through machine learning and determine whether it is possible to identify groups at high risk of suicide through words spoken by the participants in a semi-structured interview.MethodsA total of 83 healthy and 83 depressed patients were recruited. All participants were recorded during the Mini-International Neuropsychiatric Interview. Through the suicide risk assessment from the interview items, participants with depression were classified into high-suicide-risk (31 participants) and low-suicide-risk (52 participants) groups. The recording was transcribed into text after only the words uttered by the participant were extracted. In addition, all participants were evaluated for depression, anxiety, suicidal ideation, and impulsivity. The chi-square test and student’s T-test were used to compare clinical variables, and the Naive Bayes classifier was used for the machine learning text model.ResultsA total of 21,376 words were extracted from all participants and the model for diagnosing patients with depression based on this text confirmed an area under the curve (AUC) of 0.905, a sensitivity of 0.699, and a specificity of 0.964. In the model that distinguished the two groups using statistically significant demographic variables, the AUC was only 0.761. The DeLong test result (p-value 0.001) confirmed that the text-based classification was superior to the demographic model. When predicting the high-suicide-risk group, the demographics-based AUC was 0.499, while the text-based one was 0.632. However, the AUC of the ensemble model incorporating demographic variables was 0.800.ConclusionThe possibility of diagnosing depression using interview text was confirmed; regarding suicide risk, the diagnosis accuracy increased when demographic variables were incorporated. Therefore, participants’ words during an interview show significant potential as an objective and diagnostic marker through machine learning.
ObjectiveThis study examined the psychometric properties of the French–Canadian version of the Stress and Anxiety to Viral Epidemics-6 items (SAVE-6) scale for assessing the anxiety response to the viral epidemic among the general population in Quebec, Canada.MethodsA total of 590 participants responded to a confidential online survey between September 28 and October 18, 2020. Confirmatory Factor Analysis (CFA) was conducted to explore the factor structure of the scale. Psychometric properties were assessed using the Item Response Theory (IRT) approach. To explore the convergent validity, a Pearson correlation analysis between the SAVE-6 scale and the depression (Patient Health Questionnaire-2, PHQ-2) or anxiety subscale (Generalized Anxiety Disorder-2, GAD-2) of the Patient Health Questionnaire-4 items scale was conducted.FindingsThe French–Canadian version of the SAVE-6 scale was clustered into a single factor. The CFA of the SAVE-6 scale showed a good model fit (CFI = 0.985, TLI = 0.976, RMSEA = 0.051, RSMR = 0.048), and the multi-group CFA revealed that the SAVE-6 scale can measure anxiety response in the same way across gender or the presence of elevated depressive and anxiety symptoms. It showed good internal consistency (Cronbach's alpha = 0.76, McDonald's Omega = 0.77) and significant correlation with the PHQ-2 score and GAD-2 score. The IRT model suggested the efficiency in discrimination among individuals in this latent trait.ConclusionThe French–Canadian version of the SAVE-6 scale is a valid and reliable rating scale, which can measure the general population's anxiety response to the viral epidemic.
Purpose: Attitudes toward suicide are essential in suicide prevention, as suicide is socio-culturally nuanced. Although the relationship between individual attitudes and suicidal behavior has been extensively studied, the role of community attitudes—aggregated by region—on suicide mortality remains ambiguous. This study explored the association between community attitudes and real-world suicide mortality. Methods: Data on attitudes toward suicide from the 2018 Korea National Suicide Survey (N=1500) and individual mortality data from the MicroData Integrated System were obtained. Confirmatory factor analysis supported a factor structure with three factors: "Permissiveness," "Unjustified behavior," and "Readiness to help/Preventability." Thirty regional units in South Korea aggregated the data for ecological analysis. We used negative binomial models to examine the association at the regional level, and stratified analysis by gender and age group was conducted. Results: "Permissiveness" was associated with reduced suicide rates in a univariate model (P<0.001). Adjusting for gender, age, and additional sociodemographics did not alter the association. Additionally, this relationship was observed in males and individuals under 60 years of age after stratification. However, "Unjustified Behavior" and "Readiness to help/Preventability " exhibited no significant association with suicide in any model or stratum. Conclusion: The observed inverse association between permissive community attitudes and suicide contradicts the findings of previous research that links permissive individual attitudes to increased suicidal behavior. Instead, our findings indicate that attitudes may behave differently at the individual and group levels, and this could potentially guide future research and novel approaches to suicide prevention.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.