Development and external validation of a logistic and a penalized logistic model using machine-learning techniques to predict suicide attempts: A multicenter prospective cohort study in Korea.

Despite previous efforts to build statistical models for predicting the risk of suicidal behavior using machine-learning analysis, a high-accuracy model can lead to overfitting. Furthermore, internal validation cannot completely address this problem. In this study, we created models for predicting the occurrence of suicide attempts among Koreans at high risk of suicide, and we verified these models in an independent cohort. We performed logistic and penalized regression for suicide attempts within 6 months among suicidal ideators and attempters in The Korean Cohort for the Model Predicting a Suicide and Suicide-related Behavior (K-COMPASS). We then validated the models in a test cohort. Our findings indicated that several factors significantly predicted suicide attempts in the models, including young age, suicidal ideation, previous suicidal attempts, anxiety, alcohol abuse, stress, and impulsivity. The area under the curve and positive predictive values were 0.941 and 0.484 after variable selection and 0.751 and 0.084 in the test cohort. The corresponding values for the penalized regression model were 0.943 and 0.524 in the original training cohort and 0.794 and 0.115 in the test cohort. The prediction model constructed through a prospective cohort study of the suicide high-risk group showed satisfactory accuracy even in the test cohort. The accuracy with penalized regression was greater than that with the "classical" logistic model.

Risk factors of reattempt among suicide attempters in South Korea: A nationwide retrospective cohort study.

This study aimed to identify underlying demographic and clinical characteristics among individuals who had previously attempted suicide, utilizing the comprehensive Health Insurance Review and Assessment Service (HIRA) database. Data of patients aged 18 and above who had attempted suicide between January 1 and December 31, 2014, recorded in HIRA, were extracted. The index date was identified when a suicide attempt was made within the year 2014. The medical history of the three years before the index date and seven years of follow-up data after the index date were analyzed. Kaplan-Meier estimate was used to infer reattempt of the suicide attempters, and Cox-proportional hazard analysis was used to investigate risk factors associated with suicide reattempts. A total of 17,026 suicide attempters were identified, of which 1,853 (10.9%) reattempted suicide; 4,925 (28.9%) patients had been diagnosed with depressive disorder. Of the reattempters, 391 (21.1%) demonstrated a history of suicide attempts in the three years before the index date, and the mean number of prior attempts was higher compared to that of the non-reattempters (1.7 vs.1.3, p-value < 0.01). Prior psychiatric medication, polypharmacy, and an increase in the number of psychotropics were associated with suicide reattempt in overall suicide attempters. (Hazard ratio (HR) = 3.20, 95% confidence interval [CI] = 2.56-4.00; HR = 2.42, 95% CI = 1.87-3.14; HR = 19.66, 95% CI = 15.22-25.39 respectively). The risk of reattempt decreased in individuals receiving antidepressant prescriptions compared to those unmedicated, showing a reduction of 78% when prescribed by non-psychiatrists and 89% when prescribed by psychiatrists. Similar risk factors for suicide reattempts were observed in the depressive disorder subgroup, but the median time to reattempt was shorter (556.5 days) for this group compared to that for the overall attempters (578 days). Various risk factors including demographics, clinical characteristics, and medications should be considered to prevent suicide reattempts among suicide attempters, and patients with depressive disorder should be monitored more closely.

Shifting attitudes toward suicide over time: A latent profile analysis using the Korea National Suicide Survey.

Introduction: South Korea has a high suicide rate, and changes in sociodemographic factors can further increase the rate. This study aims to (1) classify participants using the Attitudes toward Suicide Scale (ATTS) through latent profile analysis (LPA), (2) identify and compare the associations between sociodemographic factors with the ATTS in two survey years (2013, 2018), and (3) determine the moderating effect of survey year. Methods: Six sub-factors of the ATTS were used for LPA with a total of 2,973 participants. Sociodemographic characteristics were compared between groups, and multinomial logistic regression was conducted for each survey year. A moderation analysis was conducted with the survey year as moderator. Results: LPA identified three groups of attitudes toward suicide: incomprehensible (10.3%), mixed (52.8%), and permissive (36.9%). The proportion of permissive attitudes increased from 2013 (32.3%) to 2018 (41.7%). Participants reporting suicidal behavior were more likely to be in the mixed and permissive groups than the incomprehensible group in both years. People reporting no religious beliefs were associated with the permissive group in the two survey years. The influence of education and income levels on groups differed by survey year. Discussion: There were significant changes between 2013 and 2018 in attitudes toward suicide in the Korean population.

Acoustic Analysis of Speech for Screening for Suicide Risk: Machine Learning Classifiers for Between- and Within-Person Evaluation of Suicidality.

BACKGROUND: Assessing a patient's suicide risk is challenging for health professionals because it depends on voluntary disclosure by the patient and often has limited resources. The application of novel machine learning approaches to determine suicide risk has clinical utility. OBJECTIVE: This study aimed to investigate cross-sectional and longitudinal approaches to assess suicidality based on acoustic voice features of psychiatric patients using artificial intelligence. METHODS: We collected 348 voice recordings during clinical interviews of 104 patients diagnosed with mood disorders at baseline and 2, 4, 8, and 12 months after recruitment. Suicidality was assessed using the Beck Scale for Suicidal Ideation and suicidal behavior using the Columbia Suicide Severity Rating Scale. The acoustic features of the voice, including temporal, formal, and spectral features, were extracted from the recordings. A between-person classification model that examines the vocal characteristics of individuals cross sectionally to detect individuals at high risk for suicide and a within-person classification model that detects considerable worsening of suicidality based on changes in acoustic features within an individual were developed and compared. Internal validation was performed using 10-fold cross validation of audio data from baseline to 2-month and external validation was performed using data from 2 to 4 months. RESULTS: A combined set of 12 acoustic features and 3 demographic variables (age, sex, and past suicide attempts) were included in the single-layer artificial neural network for the between-person classification model. Furthermore, 13 acoustic features were included in the extreme gradient boosting machine learning algorithm for the within-person model. The between-person classifier was able to detect high suicidality with 69% accuracy (sensitivity 74%, specificity 62%, area under the receiver operating characteristic curve 0.62), whereas the within-person model was able to predict worsening suicidality over 2 months with 79% accuracy (sensitivity 68%, specificity 84%, area under receiver operating characteristic curve 0.67). The second model showed 62% accuracy in predicting increased suicidality in external sets. CONCLUSIONS: Within-person analysis using changes in acoustic features within an individual is a promising approach to detect increased suicidality. Automated analysis of voice can be used to support the real-time assessment of suicide risk in primary care or telemedicine.

Self-Transcendence Mediates the Relationship between Early Trauma and Fatal Methods of Suicide Attempts.

BACKGROUND: Early trauma is known to be a risk factor of suicide-related behavior. On the other hand, people who attempt suicide using a fatal method are reported to be more likely to complete suicide. In this study, we assumed that early trauma affects an individual's temperament and character and thereby increases the risk of a fatal method of suicide attempts. METHODS: We analyzed 92 people with a history of previous suicide attempts. We compared the Temperament and Character Inventory-Revised scores between the groups with and without early trauma, and between the groups with and without a history of suicide attempt using fatal methods through an analysis of covariance with age, sex, and presence of a psychiatric history as covariates. A mediation analysis was conducted of the relationship between early trauma and fatal methods of suicide attempt with self-transcendence as a mediator. RESULTS: Higher self-transcendence was reported in the fatal group (27.71 ± 13.78 vs. 20.97 ± 12.27, P = 0.010) and the early trauma group (28.05 ± 14.30 vs. 19.43 ± 10.73, P = 0.001), respectively. The mediation model showed that self-transcendence mediates the relationship between early trauma and fatal methods of suicide attempt. The 95% confidence intervals for the direct and indirect effect were (-0.559, 1.390) and (0.026, 0.947), respectively. CONCLUSION: Self-transcendence may mediate the relationship between early trauma and fatal methods of suicide attempt. Self-transcendence may be associated with unhealthy defenses and suicidal behavior for self-punishment and may constitute a marker of higher suicide risk.

Association of heart rate variability with REM sleep without atonia in idiopathic REM sleep behavior disorder.

STUDY OBJECTIVES: Idiopathic rapid eye movement sleep behavior disorder (iRBD), characterized by rapid eye movement sleep without atonia (RSWA) and dream-enactment behavior, has been suggested to be a predictor of α-synucleinopathies. Autonomic instability, represented by heart rate variability, is a common characteristic of both iRBD and α-synucleinopathies. Previous studies reported that RSWA was associated with autonomic dysfunction and was a possible predictor of phenoconversion. Therefore, we sought to compare heart rate variability between iRBD and control groups and explore the relationship between heart rate variability and RSWA in patients with iRBD. METHODS: Nocturnal polysomnographic data on 47 patients (28 men, 19 women) diagnosed with iRBD based on video-polysomnography and 26 age-matched and sex-matched controls were reviewed. The first 5-minute epoch with a stable electrocardiogram lead II on video-polysomnography was selected from stage N2, wake, and rapid eye movement. For quantification of RSWA, tonic activity was analyzed from the submentalis electromyogram and phasic activity from the submentalis and bilateral anterior tibialis electromyogram channels. RESULTS: Compared to the control group, the iRBD group showed significant reductions in the standard deviation of the R-R intervals, the root mean square of successive R-R interval differences, and high-frequency values. Quantified tonic activity was inversely correlated with normalized low-frequency values and low-frequency/high-frequency ratios and positively correlated with normalized high-frequency values. CONCLUSIONS: This study implied decreased cardiac autonomic function in patients with iRBD, which showed parasympathetic predominance. Heart rate variability of the patients with iRBD in this study was associated with quantified tonic RSWA, which was previously reported to be a possible predictor of phenoconversion.

Electron density profile measurements from hydrogen line intensity ratio method in Versatile Experiment Spherical Torus.

Electron density profiles of versatile experiment spherical torus plasmas are measured by using a hydrogen line intensity ratio method. A fast-frame visible camera with appropriate bandpass filters is used to detect images of Balmer line intensities. The unique optical system makes it possible to take images of Hα and Hß radiation simultaneously, with only one camera. The frame rate is 1000 fps and the spatial resolution of the system is about 0.5 cm. One-dimensional local emissivity profiles have been obtained from the toroidal line of sight with viewing dumps. An initial result for the electron density profile is presented and is in reasonable agreement with values measured by a triple Langmuir probe.