Presentations to the emergency department with self-harm or suicidal behaviours: A role for digital mental health services?

Contact patterns in health care before and after first registered suicide attempt or self-harm: A nationwide register study

This study aimed to identify possible missed opportunities for suicide prevention by investigating health care utilization two years before and after the first recorded suicide attempt or self-harm (SA/SH). The SA/SH group was identified in 2010-2021 and was matched 1:10 with a reference group using risk set sampling on sex, birth year, and time of SA/SH diagnosis. Incidence rates for contacts to general practitioner, somatic hospital, or psychiatric hospital were estimated within quarters (i.e. three-month intervals). Negative binomial regression was used to determine incidence rate ratios for contact patterns in the SA/SH group compared to the two years prior to their first event. Underlying causes for health care contacts were also examined. The SA/SH group included 29,439 individuals. Contact to health care facilities changed markedly in the three months before SA/SH. The underlying causes for these encounters were often related to poisoning, injuries, abnormal or unspecific causes influencing health status, or neurotic, stress-related and somatoform disorders. This study provides health care professionals with useful insight into changing health care contact patterns among individuals with SA/SH.

Suicidal behaviors and associated factors among patients attending an emergency department: a facility-based cross-sectional study

BACKGROUND

Emergency departments (ED) are an important site for screening patients with suicidal behaviors. However, there is insufficient data in low-and middle-income countries regarding the magnitude of suicidal attempts among patients attending EDs. Therefore, the present study aimed to screen suicidal behavior and factors associated with suicide in patients attending an ED of Addis Ababa Burn, Emergency and Trauma Hospital, Ethiopia.

METHOD

A facility-based quantitative cross-sectional study was conducted between April and June 2018. A total of 398 participants were recruited using a consecutive sampling technique. The collected data collected includes structured questionnaires containing sociodemographic determinants, chronic medical illness conditions, substance use characteristics, social support level, common mental disorders (CMD) screening, suicidal behaviors assessment and suicidal attempts reason and method.

RESULTS

The prevalence of suicidal behavior and suicidal attempts were 8% and 6.3%, respectively. Suicide was attempted most frequently in the 18-24 age group. There was no overall difference in sex distribution for suicidal attempts. However, there were sex-based differences when the age group was taken into consideration. The commonest underlying reason for the attempt was social reasons (44%), while the most frequently reported attempt method was hanging (36%). No single factor was found to be significantly associated with the suicidal attempt.

CONCLUSION

Although suicidal behaviors are more common in patients attending the ED than in the general population, these facts have previously got little attention in patient attending EDs in low and middle income countries. The present findings support the need for a more detailed assessment of suicidal behaviours in patients attending ED and in patients with CMD.

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.