scholarly journals Detection of Suicidal Ideation on Social Media: Multimodal, Relational, and Behavioral Analysis (Preprint)

2020 ◽  
Author(s):  
Diana Ramírez-Cifuentes ◽  
Ana Freire ◽  
Ricardo Baeza-Yates ◽  
Joaquim Puntí ◽  
Pilar Medina-Bravo ◽  
...  
Keyword(s):  
At Risk ◽  
Social Media ◽  
Suicidal Ideation ◽  
Mental Disorders ◽  
Suicide Risk ◽  
Risk Group ◽  
Control Group ◽  
Suicide Risk Assessment ◽  
Control Groups ◽  
Multimodal Data

BACKGROUND Suicide risk assessment usually involves an interaction between doctors and patients. However, a significant number of people with mental disorders receive no treatment for their condition due to the limited access to mental health care facilities; the reduced availability of clinicians; the lack of awareness; and stigma, neglect, and discrimination surrounding mental disorders. In contrast, internet access and social media usage have increased significantly, providing experts and patients with a means of communication that may contribute to the development of methods to detect mental health issues among social media users. OBJECTIVE This paper aimed to describe an approach for the suicide risk assessment of Spanish-speaking users on social media. We aimed to explore behavioral, relational, and multimodal data extracted from multiple social platforms and develop machine learning models to detect users at risk. METHODS We characterized users based on their writings, posting patterns, relations with other users, and images posted. We also evaluated statistical and deep learning approaches to handle multimodal data for the detection of users with signs of suicidal ideation (suicidal ideation risk group). Our methods were evaluated over a dataset of 252 users annotated by clinicians. To evaluate the performance of our models, we distinguished 2 control groups: users who make use of suicide-related vocabulary (focused control group) and generic random users (generic control group). RESULTS We identified significant statistical differences between the textual and behavioral attributes of each of the control groups compared with the suicidal ideation risk group. At a 95% CI, when comparing the suicidal ideation risk group and the focused control group, the number of friends (<i>P</i>=.04) and median tweet length (<i>P</i>=.04) were significantly different. The median number of friends for a focused control user (median 578.5) was higher than that for a user at risk (median 372.0). Similarly, the median tweet length was higher for focused control users, with 16 words against 13 words of suicidal ideation risk users. Our findings also show that the combination of textual, visual, relational, and behavioral data outperforms the accuracy of using each modality separately. We defined text-based baseline models based on bag of words and word embeddings, which were outperformed by our models, obtaining an increase in accuracy of up to 8% when distinguishing users at risk from both types of control users. CONCLUSIONS The types of attributes analyzed are significant for detecting users at risk, and their combination outperforms the results provided by generic, exclusively text-based baseline models. After evaluating the contribution of image-based predictive models, we believe that our results can be improved by enhancing the models based on textual and relational features. These methods can be extended and applied to different use cases related to other mental disorders.

scholarly journals Detection of Suicidal Ideation on Social Media: Multimodal, Relational, and Behavioral Analysis

10.2196/17758 ◽  
2020 ◽  
Vol 22 (7) ◽  
pp. e17758 ◽  
Author(s):  
Diana Ramírez-Cifuentes ◽  
Ana Freire ◽  
Ricardo Baeza-Yates ◽  
Joaquim Puntí ◽  
Pilar Medina-Bravo ◽  
...  
Keyword(s):  
At Risk ◽  
Social Media ◽  
Suicidal Ideation ◽  
Mental Disorders ◽  
Suicide Risk ◽  
Risk Group ◽  
Control Group ◽  
Suicide Risk Assessment ◽  
Control Groups ◽  
Multimodal Data

Background Suicide risk assessment usually involves an interaction between doctors and patients. However, a significant number of people with mental disorders receive no treatment for their condition due to the limited access to mental health care facilities; the reduced availability of clinicians; the lack of awareness; and stigma, neglect, and discrimination surrounding mental disorders. In contrast, internet access and social media usage have increased significantly, providing experts and patients with a means of communication that may contribute to the development of methods to detect mental health issues among social media users. Objective This paper aimed to describe an approach for the suicide risk assessment of Spanish-speaking users on social media. We aimed to explore behavioral, relational, and multimodal data extracted from multiple social platforms and develop machine learning models to detect users at risk. Methods We characterized users based on their writings, posting patterns, relations with other users, and images posted. We also evaluated statistical and deep learning approaches to handle multimodal data for the detection of users with signs of suicidal ideation (suicidal ideation risk group). Our methods were evaluated over a dataset of 252 users annotated by clinicians. To evaluate the performance of our models, we distinguished 2 control groups: users who make use of suicide-related vocabulary (focused control group) and generic random users (generic control group). Results We identified significant statistical differences between the textual and behavioral attributes of each of the control groups compared with the suicidal ideation risk group. At a 95% CI, when comparing the suicidal ideation risk group and the focused control group, the number of friends (P=.04) and median tweet length (P=.04) were significantly different. The median number of friends for a focused control user (median 578.5) was higher than that for a user at risk (median 372.0). Similarly, the median tweet length was higher for focused control users, with 16 words against 13 words of suicidal ideation risk users. Our findings also show that the combination of textual, visual, relational, and behavioral data outperforms the accuracy of using each modality separately. We defined text-based baseline models based on bag of words and word embeddings, which were outperformed by our models, obtaining an increase in accuracy of up to 8% when distinguishing users at risk from both types of control users. Conclusions The types of attributes analyzed are significant for detecting users at risk, and their combination outperforms the results provided by generic, exclusively text-based baseline models. After evaluating the contribution of image-based predictive models, we believe that our results can be improved by enhancing the models based on textual and relational features. These methods can be extended and applied to different use cases related to other mental disorders.

scholarly journals What Has Clinical Research in Suicide Prevention Done for You Lately?

CNS Spectrums ◽  
2006 ◽  
Vol 11 (6) ◽  
pp. 440-441 ◽  
Author(s):  
Jan Fawcett
Keyword(s):  
Suicidal Ideation ◽  
Suicide Risk ◽  
Clinical Status ◽  
Panic Attacks ◽  
Current Evidence ◽  
Suicide Risk Assessment ◽  
Short Term ◽  
The Past ◽  
Standard Risk ◽  
Suicide Plan

What have you heard or read over the past 10 years that has improved you ability to assess and manage suicide risk in your patients?There has been a paucity of data. What little data there is reviewed in this month's articles.They highlight findings that you should know about. Clinicians seem to cling to the familiar, unless some intense marketing is done.For instance, are you aware that the current evidence shows that a denial of suicide thoughts, plans, or intent—even a contract for safety—means absolutely nothing in the absence of a full suicide risk assessment?Yet clinicians seem to rely on these ’reassurances“ from their patients and are shocked when the patient later commits suicide. Why should a patient who is deciding that life is too painful to live tell you the truth? Robert I. Simon, MD, and Daniel W. Shuman, JD, review these facts.Are you aware that severe psychic anxiety, panic attacks, agitation, and severe insomnia often precede suicide within hours, days, or weeks and can be rapidly modified with treatment?On the other hand, standard risk factors for suicide such as suicidal ideation, hopelessness, and past suicidal attempts are not good predictors of suicide in the short term. A suicide plan, recent high intent attempt, or refusal to contract for safety may well indicate immediate risk, but a denial of suicidal ideation or intent and a contract for no harm mean absolutely nothing without a full suicide assessment that takes current clinical status, past suicidal tendencies, social support, and willingness to accept help into account.

Ethical Issues in the Treatment of Suicidal Military Personnel and Veterans

2017 ◽  
pp. 121-130
Author(s):  
W. Brad Johnson ◽  
Gerald P. Koocher
Keyword(s):  
Risk Assessment ◽  
Suicidal Ideation ◽  
Military Personnel ◽  
Suicide Risk ◽  
Ethical Issues ◽  
Military Veterans ◽  
Active Duty ◽  
Suicide Risk Assessment ◽  
Active Duty Military ◽  
The Military

This chapter reviews the key ethical issues involved in treating active duty military personnel and military veterans who present with suicidal ideation. The primary issue is striving to help while minimizing harm. Central ethical issues involve competence in suicide risk assessment and intervention, confidentiality, and multiple role situations. All of these play out differently from civilian situations because of contextual demands associated with military settings and rules governing patients and providers in the military context. Critical strategies for addressing these issues include using appropriate consent processes and understanding how best to respect the needs of the patient within the constraints of the military context.

scholarly journals Natural Language Processing of Social Media as Screening for Suicide Risk

2018 ◽  
Vol 10 ◽  
pp. 117822261879286 ◽  
Author(s):  
Glen Coppersmith ◽  
Ryan Leary ◽  
Patrick Crutchley ◽  
Alex Fine
Keyword(s):  
At Risk ◽  
Social Media ◽  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Suicide Risk ◽  
Automated System ◽  
World Health ◽  
Primary Care Doctor ◽  
Trade Off

Suicide is among the 10 most common causes of death, as assessed by the World Health Organization. For every death by suicide, an estimated 138 people’s lives are meaningfully affected, and almost any other statistic around suicide deaths is equally alarming. The pervasiveness of social media—and the near-ubiquity of mobile devices used to access social media networks—offers new types of data for understanding the behavior of those who (attempt to) take their own lives and suggests new possibilities for preventive intervention. We demonstrate the feasibility of using social media data to detect those at risk for suicide. Specifically, we use natural language processing and machine learning (specifically deep learning) techniques to detect quantifiable signals around suicide attempts, and describe designs for an automated system for estimating suicide risk, usable by those without specialized mental health training (eg, a primary care doctor). We also discuss the ethical use of such technology and examine privacy implications. Currently, this technology is only used for intervention for individuals who have “opted in” for the analysis and intervention, but the technology enables scalable screening for suicide risk, potentially identifying many people who are at risk preventively and prior to any engagement with a health care system. This raises a significant cultural question about the trade-off between privacy and prevention—we have potentially life-saving technology that is currently reaching only a fraction of the possible people at risk because of respect for their privacy. Is the current trade-off between privacy and prevention the right one?

Suicide Risk in Bipolar Disorder

2018 ◽  
Author(s):  
Rachel Katz ◽  
Robert Beech
Keyword(s):  
At Risk ◽  
Bipolar Disorder ◽  
Study Design ◽  
Suicide Risk ◽  
Risk Group ◽  
Clinical Scenario ◽  
Relevant Case ◽  
Study Intervention ◽  
Landmark Study

This chapter provides a summary of a landmark study on bipolar disorder, tackling the serious issue of suicidality in this at-risk group of patients. Is there a difference in suicide risk for patients with bipolar disorder who are treated with lithium, divalproex, or carbamazepine? Starting with that question, it describes the basics of the study, including funding, study location, who was studied, how many patients, study design, study intervention, follow-up, endpoints, results, and criticism and limitations. The chapter briefly reviews other relevant studies and information, discusses implications, and concludes with a relevant case that applies the findings from the study to a clinical scenario.

Effects of suicide awareness materials on individuals with recent suicidal ideation or attempt: online randomised controlled trial

2019 ◽  
Vol 217 (6) ◽  
pp. 693-700 ◽  
Author(s):  
Thomas Niederkrotenthaler ◽  
Benedikt Till
Keyword(s):  
Suicidal Ideation ◽  
Randomised Controlled Trial ◽  
Personal Experience ◽  
Suicide Risk ◽  
Controlled Trial ◽  
Control Group ◽  
End Point ◽  
Randomised Controlled ◽  
Suicide Awareness ◽  
Group 1

BackgroundAwareness materials featuring ways of coping with suicidal ideation can reduce suicidal ideation, the so-called Papageno effect. All of the previous experimental studies on this subject have been conducted with individuals not at risk of suicide.AimsTo assess effects of suicide awareness materials in a sample of individuals with recent suicidal ideation. Trial registration: German Clinical Trial Registry ID number DRKS00013613.MethodAdults (n = 266) with recent self-reported suicidal ideation or attempt were randomised to read an educative article featuring a lay individual with personal experience of suicidality (n = 86), a similar article featuring a mental health expert (n = 90), or an unrelated article (n = 90) in a double-blind online randomised controlled trial. Questionnaire data were collected before (T1) and immediately after exposure (T2) as well as 1 week later (study end-point, T3) and analysed with linear mixed models. The primary outcome was suicide risk as assessed using the Survival and Coping Beliefs subscale of the Reasons for Living Inventory (RFLI); secondary outcomes were suicide-prevention knowledge and mood.ResultsThere was an immediate beneficial effect on suicide risk in the intervention group exposed to the message delivered by the individual with personal experience (group 1) as compared with the control group that was maintained until the study end-point (study end-point: RFLI score mean difference from baseline within group 1 MD = −0.36 (95% CI −0.66 to −0.06), mean difference compared with control group MD = −0.71 (95% CI −1.27 to −0.14); d = −0.18). The effect was particularly pronounced for individuals with recent suicide attempt (RFLI score at T3, compared with control group: MD = −1.55 (95% CI −2.52 to −0.57); d = −0.23). Participants in this group also showed increased prevention-related knowledge compared with the control group.ConclusionsIndividuals with a recent suicide attempt appear to benefit from a printed narrative of positive coping with suicidal ideation. The intervention materials do not increase short-term suicide risk.

scholarly journals Waterborne outbreak of campylobacter enteritis after outdoors infantry drill in Utti, Finland

1989 ◽  
Vol 103 (1) ◽  
pp. 133-141 ◽  
Author(s):  
M. Aho ◽  
M. Kurki ◽  
H. Rautelin ◽  
T. U. Kosunen
Keyword(s):  
At Risk ◽  
Abdominal Pain ◽  
Campylobacter Jejuni ◽  
Risk Group ◽  
Entire Population ◽  
Control Group ◽  
Serological Response ◽  
Population At Risk ◽  
Waterborne Outbreak ◽  
Heat Stable

SUMMARYDiarrhoea, abdominal pain, malaise and fever affected 75 of the 88 conscript soldiers in Utti, Finland after an outdoors infantry drill.Campylobacter jejuni, heat-stable serotype 3/43/59. was isolated from 37 out of 63 men investigated. A clear serological response was evident in the risk group and negligible in the control group. The entire population at risk was interviewed.

Rappresentazioni mentali materne e qualitŕ degli scambi diadici in un gruppo di bambini a rischio di maltrattamento

2009 ◽  
pp. 51-63
Author(s):  
Giampaolo Nicolais ◽  
Anna Maria Speranza ◽  
Simona Guarino ◽  
Cristina Trentini
Keyword(s):  
At Risk ◽  
Risk Group ◽  
Dyadic Interaction ◽  
Traumatic Experiences ◽  
Control Group ◽  
Lower Sensitivity ◽  
Affective States ◽  
Relational Experiences ◽  
State Of Mind ◽  
Negative Affective States

- Problem: maltreatment impacts on the quality of interactive exchanges, precociously putting the infant at risk for maladaptive outcomes. Method: research involved 37 mother-infant dyads at risk for maltreatment and 37 non-referred dyads. AAI has been administered to the mothers; furthermore, dyadic interactive exchanges were observed and video-recorded. Results: in comparison with the control group, the at-risk group shows a prevalence of disorganized models (X= 12,2; p &lt; 0,01). Additionally, in the at-risk group an influence of maternal mental state on the interactive quality has been evidenced: dismissing mothers present lower sensitivity (F = 4,87; p &lt; 0,01), whereas disorganized mothers express more negative affective states (F = 5,65; p &lt; 0,01). Conclusions: early traumatic relational experiences influence maternal state of mind with respect to attachment, thus compromising the process of dyadic affective regulation.Key words: child maltreatment, traumatic experiences, attachment, dyadic interaction.Parole chiave: maltrattamento infantile, esperienze traumatiche, attaccamento, interazione diadica.

scholarly journals Efectos del programa CIPRES sobre el riesgo suicida en adolescentes

2019 ◽  
Vol 24 (2) ◽  
pp. 83
Author(s):  
Marly Johana Bahamón ◽  
Yolima Alarcón-Vásquez ◽  
Ana María Trejos-Herrera ◽  
Stefano Vinaccia Alpi ◽  
Andres Cabezas ◽  
...  
Keyword(s):  
Suicidal Ideation ◽  
Suicide Risk ◽  
Psychological Intervention ◽  
Control Group ◽  
Assessment Instruments ◽  
Suicidal Risk ◽  
Before And After ◽  
Quasi Experimental ◽  
Self Harm ◽  
Experimental Group

Abstract: Effects of the CIPRES program on suicidal risk in adolescents. The objective of the study was to evaluate the specific changes generated by the CIPRES program on suicidal risk in adolescents. The sample comprised 106 adolescents aged between 13 and 18 years old (53 in the experimental group, 53 in the control group), of which 54.7% were women (n = 58). The study used a quasi-experimental design of repeated pretest-posttest measures with a control group. Two assessment instruments were administered before and after the program: The Scale of the Suicide Risk (ERS) and the Inventory of Positive and Negative Suicidal Ideation (PANSI). The pretest-posttest ANCOVAs showed that the program significantly reduced (p < .05) the scores of suicidal ideation, planning, self-harm, isolation/social support, lack of family support, and global suicide risk. In addition, it significantly increased positive ideation scores. The results support the relevance of this program to reduce suicidal risk in adolescents.Keywords: CIPRES; suicide risk; adolescents; psychological intervention; quasi-experiment.Resumen: El estudio tuvo por objetivo evaluar los cambios específicos generados por el programa CIPRES sobre el riesgo suicida en adolescentes. La muestra se configuró con 106 adolescentes entre los 13 y 18 años (53 experimentales, 53 controles), de los cuales el 54.7% eran mujeres (n = 58). El estudio utilizó un diseño cuasi-experimental de medidas repetidas pretest-postest con grupo de control. Se administraron dos instrumentos de evaluación antes y después de la aplicación del programa: la Escala de Riesgo Suicida (ERS) y el Inventario de Ideación Suicida Positiva y Negativa (PANSI). Los ANCOVAs pretest-postest evidenciaron que el programa redujo significativamente (p < .05) las puntuaciones en ideación suicida, planificación, autolesión, aislamiento/soporte social, falta de apoyo familiar, y riesgo suicida global. Además, aumentó significativamente las puntuaciones en ideación positiva. Los resultados apuntan a la relevancia de este programa para reducir el riesgo suicida en adolescentes.Palabras clave: CIPRES; riesgo de suicidio; adolescentes; intervención psicológica; cuasiexperimento.

scholarly journals Access to Firearms Among People Assessed by Psychiatric Services in the Emergency Department

2019 ◽  
Vol 64 (11) ◽  
pp. 770-776
Author(s):  
Cara Katz ◽  
Joanna Bhaskaran ◽  
James M. Bolton
Keyword(s):  
Suicidal Ideation ◽  
Suicide Attempt ◽  
Odds Ratio ◽  
Suicide Risk ◽  
Tertiary Care ◽  
Group Differences ◽  
Emergency Setting ◽  
High Likelihood ◽  
Psychiatric Consultation ◽  
Suicide Risk Assessment

Objective: In Canada, 80% of firearm-related deaths are suicides. Access to firearms is associated with increased suicide rates. This study examines the frequency and factors that influence assessment of firearm access in an emergency setting. Methods: A total of 15,847 consecutive adults seen for psychiatric consultation in two tertiary emergency departments (EDs) in Winnipeg, Manitoba were interviewed. Data captured whether access to firearms was assessed, and whether respondents endorsed access or not. Comparisons were done to determine group differences among those with and without and with known and unknown firearm access. Results: Access to firearms was unknown in 47% ( n = 7,363) of psychiatric ED consultations, including 43% ( n = 998) of individuals who presented with a suicide attempt. Female sex was associated with decreased odds of firearm access (odds ratio [ OR] 0.28; 95% CI, 0.22 to 0.35). Being single was associated with lower odds of known firearm access ( OR 0.83; 95% CI, 0.77 to 0.89) yet higher likelihood of firearm access ( OR 1.36; 95% CI, 1.11 to 1.68). Presenting with a suicide attempt ( OR 2.45; 95% CI, 1.80 to 3.34), preparatory acts ( OR 6.40; 95% CI, 4.38 to 9.36) and suicidal ideation ( OR 2.45; 95% CI, 1.87 to 3.21) were associated with increased odds of reporting access. When clinicians felt there was a high likelihood of future suicide, firearm access remained unknown in half of cases. Conclusion: Firearm ownership and access is an essential component of a suicide risk assessment and remains unknown in half of individuals seen by psychiatry in this tertiary care ED sample. People presenting with suicidal ideation and attempts were more likely to report access to firearms.

Sign in / Sign up

Export Citation Format

Share Document