Dataset Information

Uncovering social-contextual and individual mental health factors associated with violence via computational inference.

ABSTRACT: The identification of human violence determinants has sparked multiple questions from different academic fields. Innovative methodological assessments of the weight and interaction of multiple determinants are still required. Here, we examine multiple features potentially associated with confessed acts of violence in ex-members of illegal armed groups in Colombia (N = 26,349) through deep learning and feature-derived machine learning. We assessed 162 social-contextual and individual mental health potential predictors of historical data regarding consequentialist, appetitive, retaliative, and reactive domains of violence. Deep learning yields high accuracy using the full set of determinants. Progressive feature elimination revealed that contextual factors were more important than individual factors. Combined social network adversities, membership identification, and normalization of violence were among the more accurate social-contextual factors. To a lesser extent the best individual factors were personality traits (borderline, paranoid, and antisocial) and psychiatric symptoms. The results provide a population-based computational classification regarding historical assessments of violence in vulnerable populations.

SUBMITTER: Santamaria-Garcia H

PROVIDER: S-EPMC7892360 | biostudies-literature | 2021 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Uncovering social-contextual and individual mental health factors associated with violence via computational inference.

Santamaría-García Hernando H Baez Sandra S Aponte-Canencio Diego Mauricio DM Pasciarello Guido Orlando GO Donnelly-Kehoe Patricio Andrés PA Maggiotti Gabriel G Matallana Diana D Hesse Eugenia E Neely Alejandra A Zapata José Gabriel JG Chiong Winston W Levy Jonathan J Decety Jean J Ibáñez Agustín A

Patterns (New York, N.Y.) 20210212 2

The identification of human violence determinants has sparked multiple questions from different academic fields. Innovative methodological assessments of the weight and interaction of multiple determinants are still required. Here, we examine multiple features potentially associated with confessed acts of violence in ex-members of illegal armed groups in Colombia (N = 26,349) through deep learning and feature-derived machine learning. We assessed 162 social-contextual and individual mental healt ...[more]

PMID: 33659906

Similar Datasets

Project description:Background Little is known about the psychological risks of dispatched citizen responders who have participated in resuscitation attempts. Methods and Results A cross-sectional survey study was performed with 102 citizen responders who participated in a resuscitation attempt from July 23, 2018, to August 22, 2018, in the Capital Region of Denmark. Psychological distress, defined as symptoms of posttraumatic stress disorder, was assessed 3 weeks after the resuscitation attempt and measured with the Impact of Event Scale-Revised. Perceived stress was measured with the Perceived Stress Scale. Individual differences were assessed as the personality traits of agreeableness, conscientiousness, extraversion, neuroticism, and openness to experience with the Big Five Inventory, general self-efficacy, and coping mechanisms (Brief Coping Orientation to Problems Experienced Inventory). Associations between continuous variables were examined with the Pearson correlation. The associations between psychological distress levels and contextual factors and individual differences were analyzed in multivariable linear regression models to determine factors independently associated with psychological distress levels. The mean overall posttraumatic stress disorder score was 0.65 of 12; the mean perceived stress score was 7.61 of 40. The most common coping mechanisms were acceptance and emotional support. Low perceived stress was significantly associated with high general self-efficacy, and high perceived stress was significantly associated with high scores on neuroticism and openness to experience. Non-healthcare professionals were less likely to report symptoms of posttraumatic stress disorder. Conclusions Citizen responders who participated in resuscitation reported low levels of psychological distress. Individual differences were significantly associated with levels of psychological distress and should be considered when engaging citizen responders in resuscitation.

Project description:OBJECTIVE:To assess the relationship between periodontal treatment needs by elderly Brazilians and contextual as well as individual variables. METHODS:A cross-sectional study was carried out to assess the need for clinical periodontal treatment, based on National Oral Health Survey (SB Brasil 2010) data on the presence of dental calculus, shallow (3-5 mm) and deep (? 6 mm) periodontal pockets, and gingival bleeding in elderly people (n = 7,619). The contextual variables included the Municipal Human Development Index (MHDI), income inequality (Gini Index) and coverage of the municipal population by the Family Health Strategy (FHS) program oral health teams.<0} The individual variables were sex, income, education level and self-reported skin color. Multilevel logistic regression models were used to calculate the odds ratio (OR) and 95% confidence intervals (CI95%) between periodontal treatment needs and the contextual as well as individual variables. RESULTS:Gingival bleeding was found in 20.7% of the elderly analyzed (n = 1,577), dental calculus in 34% (n = 2,590), shallow periodontal pockets in 15.6% (n = 1,189), and deep periodontal pockets in 4.2% (n = 320). Individual factors were correlated with all the outcomes assessed. Sex was a protective factor in regard to gingival bleeding (OR = 0.87; CI95% 0.76-1.00), dental calculus (OR = 0.86; CI95% 0.75-0.99), shallow periodontal pockets (OR = 0.69; CI95% 0.60-0.80) and deep periodontal pockets (OR = 0.58; CI95% 0.45-0.74). It was found that fewer women needed treatment. Elderly people who self-reported having nonwhite skin had higher chances of needing periodontal treatment. Skin color was a risk factor for gingival bleeding (OR = 1.32; CI95% 1.14-1.53), dental calculus (OR = 1.32; CI95%1.14-1.54) and shallow periodontal pockets (OR = 1.27; CI95% 1.09-1.49). Education level was associated with the presence of dental calculus (OR = 0.77; CI95% 0.66-0.89), shallow periodontal pockets (OR = 0.86; CI95% 0.73-1.00) and deep periodontal pockets (OR = 0.74; CI95% 0.57-0.97), thus acting as a risk factor for undereducated elderly people. There was a correlation between population coverage by the Family Health Strategy (FHS) program oral health teams and the presence of gingival bleeding (OR = 0.67; CI95% 0.52-0.88), shallow periodontal pockets (OR = 0.76; CI95% 0.58-0.98) and deep periodontal pockets (OR = 0.62; CI95% 0.44-0.89), making these teams act as a protective factor. CONCLUSIONS:This study showed evidence of the sociocontextual as well as individual sociodemographic characteristics influencing periodontal treatment needed by elderly Brazilians, based on the clinical features of periodontal disease. The results suggest the existence of inequality related to periodontal treatment needs among elderly Brazilians, especially in regard to sex and ethnicity, in addition to a potentially positive impact from the expansion of the Family Health Strategy (FHS) program oral health teams.

Project description:BackgroundIn recent decades, China has experienced dramatic changes to its social and economic environment, which has affected the distribution of wellbeing across its citizens. While several studies have investigated individual level predictors of wellbeing in the Chinese population, less research has been done looking at contextual effects. This cross-sectional study looks at the individual and contextual effects of (regional) education, unemployment and marriage (rate) on individual happiness, life satisfaction and depressive symptomatology.MethodsData were collected from over 29,000 individuals (aged 18 to 110, 51.91% female) in the China Family Panel Studies, and merged with county level census data obtained from the 2010 China Population Census and Statistical Yearbook. To explore contextual effects, we used multilevel models accounting for the hierarchical structure of the data.ResultsWe found that a one-year increase in education was associated with a 0.17% increase in happiness and a 0.16% decrease in depressive symptoms. Unemployed men were 1% less happy, 1% less satisfied with life and reported 0.84% more depressive symptoms than employed men while minimal effects were seen for women. Single, divorced and widowed individuals had worse outcomes than married individuals (ranging from 2.96 to 21% differences). We found interaction effects for education and employment. Less educated individuals had greater happiness and less depressive symptoms in counties with higher average education compared to counterparts in less educated counties. In contrast, more educated individuals were less satisfied with life in more educated counties, an effect that is possibly due to social comparison. Employed individuals had lower life satisfaction in areas of high unemployment, while levels were constant for the unemployed. A 1% increase in county marriage rate was associated with 0.33 and 0.24% increases in happiness and life satisfaction respectively, with no interactions. We speculate that this effect could be due to greater social cohesion in the neighbourhood.ConclusionsOur results show that policies designed to improve employment and marriage rates will be beneficial for all, while interventions to encourage positive social comparison strategies may help to offset the negative effects of increasing neighbourhood average education on the highly educated.

Dataset Information

Uncovering social-contextual and individual mental health factors associated with violence via computational inference.

Publications

Uncovering social-contextual and individual mental health factors associated with violence via computational inference.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets