Prediction of Suicide Attempts Using Clinician Assessment, Patient Self-report, and Electronic Health Records

Enhancing Suicide Risk Prediction With Polygenic Scores in Psychiatric Emergency Settings: Prospective Study

BACKGROUND

Despite growing interest in the clinical translation of polygenic risk scores (PRSs), it remains uncertain to what extent genomic information can enhance the prediction of psychiatric outcomes beyond the data collected during clinical visits alone.

OBJECTIVE

This study aimed to assess the clinical utility of incorporating PRSs into a suicide risk prediction model trained on electronic health records (EHRs) and patient-reported surveys among patients admitted to the emergency department.

METHODS

Study participants were recruited from the psychiatric emergency department at Massachusetts General Hospital. There were 333 adult patients of European ancestry who had high-quality genotype data available through their participation in the Mass General Brigham Biobank. Multiple neuropsychiatric PRSs were added to a previously validated suicide prediction model in a prospective cohort enrolled between February 4, 2015, and March 13, 2017. Data analysis was performed from July 11, 2022, to August 31, 2023. Suicide attempt was defined using diagnostic codes from longitudinal EHRs combined with 6-month follow-up surveys. The clinical risk score for suicide attempt was calculated from an ensemble model trained using an EHR-based suicide risk score and a brief survey, and it was subsequently used to define the baseline model. We generated PRSs for depression, bipolar disorder, schizophrenia, suicide attempt, and externalizing traits using a Bayesian polygenic scoring method for European ancestry participants. Model performance was evaluated using area under the receiver operator curve (AUC), area under the precision-recall curve, and positive predictive values.

RESULTS

Of the 333 patients (n=178, 53.5% male; mean age 36.8, SD 13.6 years; n=333, 100% non-Hispanic and n=324, 97.3% self-reported White), 28 (8.4%) had a suicide attempt within 6 months. Adding either the schizophrenia PRS or all PRSs to the baseline model resulted in the numerically highest discrimination (AUC 0.86, 95% CI 0.73-0.99) compared to the baseline model (AUC 0.84, 95% Cl 0.70-0.98). However, the improvement in model performance was not statistically significant.

CONCLUSIONS

In this study, incorporating genomic information into clinical prediction models for suicide attempt did not improve patient risk stratification. Larger studies that include more diverse participants are required to validate whether the inclusion of psychiatric PRSs in clinical prediction models can enhance the stratification of patients at risk of suicide attempts.

Developing a suicide risk model for use in the Indian Health Service

We developed and evaluated an electronic health record (EHR)-based model for suicide risk specific to an American Indian patient population. Using EHR data for all patients over 18 with a visit between 1/1/2017 and 10/2/2021, we developed a model for the risk of a suicide attempt or death in the 90 days following a visit. Features included demographics, medications, diagnoses, and scores from relevant screening tools. We compared the predictive performance of logistic regression and random forest models against existing suicide screening, which was augmented to include the history of previous attempts or ideation. During the study, 16,835 patients had 331,588 visits, with 490 attempts and 37 deaths by suicide. The logistic regression and random forest models (area under the ROC (AUROC) 0.83 [0.80-0.86]; both models) performed better than enhanced screening (AUROC 0.64 [0.61-0.67]). These results suggest that an EHR-based suicide risk model can add value to existing practices at Indian Health Service clinics.

Identifying pathways from childhood adversity to suicidal thoughts and behaviors among sexual minority adults: An exploratory mediation analysis

BACKGROUND

The current study uses a nationally representative longitudinal dataset of sexual minority adults in the US to investigate the pathways from adverse childhood experiences (ACEs) to adulthood suicidal thoughts and behaviors.

METHODS

ACEs were measured at year one, potential mediators at year two, and suicidal thoughts and behaviors (suicidal ideation, intent, plan, and attempt) at year three. We conducted an exploratory mediation analysis to identify potential mediating factors linking ACEs to suicidal thoughts and behaviors. Ten candidate mediators were examined: social well-being, felt stigma, experiences of everyday discrimination, social support, psychological distress, alcohol and drug use, importance of sexual identity, community connection, and internalized homophobia.

RESULTS

Participants were 1518 adults who identified as lesbian or gay (n = 833; 55 %), bisexual (n = 493; 33 %), or with other sexual minority identities (n = 181; 12 %) and were on average 36.48 years (SD = 14.7) of age. Psychological distress served as a common mediator between ACEs and suicidal ideation, intent, plan, and attempt. Additionally, experiences of everyday discrimination emerged as a specific mediator leading to suicidal intent, whereas social support uniquely mediated the relation between ACEs and suicide plan.

LIMITATIONS

Potential recall bias due to retrospective reporting of ACEs may be a limitation. Future studies should broaden the measurement scope of ACEs and implement intersectional methods.

CONCLUSION

The current findings underscore the urgent need for targeted interventions that address the specific mental health needs of sexual minority individuals, particularly focusing on mitigating psychological distress, combating systemic discrimination, and enhancing social support.

Performance of Machine Learning Suicide Risk Models in an American Indian Population

Importance

Few suicide risk identification tools have been developed specifically for American Indian and Alaska Native populations, even though these populations face the starkest suicide-related inequities.

Objective

To examine the accuracy of existing machine learning models in a majority American Indian population.

Design, Setting, and Participants

This prognostic study used secondary data analysis of electronic health record data collected from January 1, 2017, to December 31, 2021. Existing models from the Mental Health Research Network (MHRN) and Vanderbilt University (VU) were fitted. Models were compared with an augmented screening indicator that included any previous attempt, recent suicidal ideation, or a recent positive suicide risk screen result. The comparison was based on the area under the receiver operating characteristic curve (AUROC). The study was performed in partnership with a tribe and local Indian Health Service (IHS) in the Southwest. All patients were 18 years or older with at least 1 encounter with the IHS unit during the study period. Data were analyzed between October 6, 2022, and July 29, 2024.

Exposures

Suicide attempts or deaths within 90 days.

Main Outcomes and Measures

Model performance was compared based on the ability to distinguish between those with a suicide attempt or death within 90 days of their last IHS visit with those without this outcome.

Results

Of 16 835 patients (mean [SD] age, 40.0 [17.5] years; 8660 [51.4%] female; 14 251 [84.7%] American Indian), 324 patients (1.9%) had at least 1 suicide attempt, and 37 patients (0.2%) died by suicide. The MHRN model had an AUROC value of 0.81 (95% CI, 0.77-0.85) for 90-day suicide attempts, whereas the VU model had an AUROC value of 0.68 (95% CI, 0.64-0.72), and the augmented screening indicator had an AUROC value of 0.66 (95% CI, 0.63-0.70). Calibration was poor for both models but improved after recalibration.

Conclusion and Relevance

This prognostic study found that existing risk identification models for suicide prevention held promise when applied to new contexts and performed better than relying on a combined indictor of a positive suicide risk screen result, history of attempt, and recent suicidal ideation.

Using natural language processing to evaluate temporal patterns in suicide risk variation among high-risk Veterans

Measuring suicide risk fluctuation remains difficult, especially for high-suicide risk patients. Our study addressed this issue by leveraging Dynamic Topic Modeling, a natural language processing method that evaluates topic changes over time, to analyze high-suicide risk Veterans Affairs patients' unstructured electronic health records. Our sample included all high-risk patients that died (cases) or did not (controls) by suicide in 2017 and 2018. Cases and controls shared the same risk, location, and treatment intervals and received nine months of mental health care during the year before the relevant end date. Each case was matched with five controls. We analyzed case records from diagnosis until death and control records from diagnosis until matched case's death date. Our final sample included 218 cases and 943 controls. We analyzed the corpus using a Python-based Dynamic Topic Modeling algorithm. We identified five distinct topics, "Medication," "Intervention," "Treatment Goals," "Suicide," and "Treatment Focus." We observed divergent change patterns over time, with pathology-focused care increasing for cases and supportive care increasing for controls. The case topics tended to fluctuate more than the control topics, suggesting the importance of monitoring lability. Our study provides a method for monitoring risk fluctuation and strengthens the groundwork for time-sensitive risk measurement.

Perception of social inclusion/exclusion and response inhibition in adolescents with past suicide attempt: a multidomain task-based fMRI study

The occurrence of suicidal behaviors increases during adolescence. Hypersensitivity to negative social signals and deficits in cognitive control are putative mechanisms of suicidal behaviors, which necessitate confirmation in youths. Multidomain functional neuroimaging could enhance the identification of patients at suicidal risk beyond standard clinical measures. Three groups of adolescents (N = 96; 78% females, age = 11.6-18.1) were included: patients with depressive disorders and previous suicide attempts (SA, n = 29); patient controls with depressive disorders but without any suicide attempt history (PC, n = 35); and healthy controls (HC, n = 32). We scanned participants with 3T-MRI during social inclusion/exclusion (Cyberball Game) and response inhibition (Go-NoGo) tasks. Neural activation was indexed by the blood-oxygenation-level dependent (BOLD) of the hemodynamic response during three conditions in the Cyberball Game ("Control condition", "Social Inclusion", and "Social Exclusion"), and two conditions in Go-NoGo task ("Go" and "NoGo" blocks). ANCOVA-style analysis identified group effects across three whole-brain contrasts: 1) NoGo vs. Go, 2) Social inclusion vs. control condition, 3) Social exclusion vs. control condition. We found that SA had lower activation in the left insula during social inclusion vs. control condition compared to PC and HC. Moreover, SA compared to PC had higher activity in the right middle prefrontal gyrus during social exclusion vs. control condition, and in bilateral precentral gyri during NoGo vs. Go conditions. Task-related behavioral and self-report measures (Self-reported emotional reactivity in the Cyberball Game, response times and number of errors in the Go-NoGo Task) did not discriminate groups. In conclusion, adolescent suicidal behaviors are likely associated with neural alterations related to the processing of social perception and response inhibition. Further research, involving prospective designs and diverse cohorts of patients, is necessary to explore the potential of neuroimaging as a tool in understanding the emergence and progression of suicidal behaviors.

Validation of a Multivariable Model to Predict Suicide Attempt in a Mental Health Intake Sample

Importance

Given that suicide rates have been increasing over the past decade and the demand for mental health care is at an all-time high, targeted prevention efforts are needed to identify individuals seeking to initiate mental health outpatient services who are at high risk for suicide. Suicide prediction models have been developed using outpatient mental health encounters, but their performance among intake appointments has not been directly examined.

Objective

To assess the performance of a predictive model of suicide attempts among individuals seeking to initiate an episode of outpatient mental health care.

Design, Setting, and Participants

This prognostic study tested the performance of a previously developed machine learning model designed to predict suicide attempts within 90 days of any mental health outpatient visit. All mental health intake appointments scheduled between January 1, 2012, and April 1, 2022, at Kaiser Permanente Northern California, a large integrated health care delivery system serving over 4.5 million patients, were included. Data were extracted and analyzed from August 9, 2022, to July 31, 2023.

Main Outcome and Measures

Suicide attempts (including completed suicides) within 90 days of the appointment, determined by diagnostic codes and government databases. All predictors were extracted from electronic health records.

Results

The study included 1 623 232 scheduled appointments from 835 616 unique patients. There were 2800 scheduled appointments (0.17%) followed by a suicide attempt within 90 days. The mean (SD) age across appointments was 39.7 (15.8) years, and most appointments were for women (1 103 184 [68.0%]). The model had an area under the receiver operating characteristic curve of 0.77 (95% CI, 0.76-0.78), an area under the precision-recall curve of 0.02 (95% CI, 0.02-0.02), an expected calibration error of 0.0012 (95% CI, 0.0011-0.0013), and sensitivities of 37.2% (95% CI, 35.5%-38.9%) and 18.8% (95% CI, 17.3%-20.2%) at specificities of 95% and 99%, respectively. The 10% of appointments at the highest risk level accounted for 48.8% (95% CI, 47.0%-50.6%) of the appointments followed by a suicide attempt.

Conclusions and Relevance

In this prognostic study involving mental health intakes, a previously developed machine learning model of suicide attempts showed good overall classification performance. Implementation research is needed to determine appropriate thresholds and interventions for applying the model in an intake setting to target high-risk cases in a manner that is acceptable to patients and clinicians.

Differentiation between suicide attempt and suicidal ideation in patients with major depressive disorder using cortical functional network

Studies exploring the neurophysiology of suicide are scarce and the neuropathology of related disorders is poorly understood. This study investigated source-level cortical functional networks using resting-state electroencephalography (EEG) in drug-naïve depressed patients with suicide attempt (SA) and suicidal ideation (SI). EEG was recorded in 55 patients with SA and in 54 patients with SI. Particularly, all patients with SA were evaluated using EEG immediately after their SA (within 7 days). Graph-theory-based source-level weighted functional networks were assessed using strength, clustering coefficient (CC), and path length (PL) in seven frequency bands. Finally, we applied machine learning to differentiate between the two groups using source-level network features. At the global level, patients with SA showed lower strength and CC and higher PL in the high alpha band than those with SI. At the nodal level, compared with patients with SI, patients with SA showed lower high alpha band nodal CCs in most brain regions. The best classification performances for SA and SI showed an accuracy of 73.39%, a sensitivity of 76.36%, and a specificity of 70.37% based on high alpha band network features. Our findings suggest that abnormal high alpha band functional network may reflect the pathophysiological characteristics of suicide and serve as a clinical biomarker for suicide.

Digital phenotyping data and anomaly detection methods to assess changes in mood and anxiety symptoms across a transdiagnostic clinical sample

INTRODUCTION

Clinical assessment of mood and anxiety change often relies on clinical assessment or self-reported scales. Using smartphone digital phenotyping data and resulting markers of behavior (e.g., sleep) to augment clinical symptom scores offers a scalable and potentially more valid method to understand changes in patients' state. This paper explores the potential of using a combination of active and passive sensors in the context of smartphone-based digital phenotyping to assess mood and anxiety changes in two distinct cohorts of patients to assess the preliminary reliability and validity of this digital phenotyping method.

METHODS

Participants from two different cohorts, each n = 76, one with diagnoses of depression/anxiety and the other schizophrenia, utilized mindLAMP to collect active data (e.g., surveys on mood/anxiety), along with passive data consisting of smartphone digital phenotyping data (geolocation, accelerometer, and screen state) for at least 1 month. Using anomaly detection algorithms, we assessed if statistical anomalies in the combination of active and passive data could predict changes in mood/anxiety scores as measured via smartphone surveys.

RESULTS

The anomaly detection model was reliably able to predict symptom change of 4 points or greater for depression as measured by the PHQ-9 and anxiety as measured for the GAD-8 for both patient populations, with an area under the ROC curve of 0.65 and 0.80 for each respectively. For both PHQ-9 and GAD-7, these AUCs were maintained when predicting significant symptom change at least 7 days in advance. Active data alone predicted around 52% and 75% of the symptom variability for the depression/anxiety and schizophrenia populations respectively.

CONCLUSION

These results indicate the feasibility of anomaly detection for predicting symptom change in transdiagnostic cohorts. These results across different patient groups, different countries, and different sites (India and the US) suggest anomaly detection of smartphone digital phenotyping data may offer a reliable and valid approach to predicting symptom change. Future work should emphasize prospective application of these statistical methods.

Predictive Validity of the K-CAT-SS in High-Risk Adolescents and Young Adults

OBJECTIVE

Suicide is a leading cause of death in adolescents and young adults and has increased substantially in the past 15 years. Accurate suicide risk stratification based on rapid screening can help reverse these trends. This study aimed to assess the ability of the Kiddie Computerized Adaptive Test Suicide Scale (K-CAT-SS), a brief computerized adaptive test of suicidality, to predict suicide attempts (SAs) in high-risk youth.

METHOD

A total of 652 participants (age range, 12-24 years), 78% of whom presented with suicidal ideation or behavior, were recruited within 1 month of mental health care contact. The K-CAT-SS, scaled from 0 to 100, was administered at baseline, and participants were assessed at about 1, 3, and 6 months after intake. Weekly incidence of SAs was assessed using the Adolescent Longitudinal Interval Follow-up Evaluation and Columbia-Suicide Severity Rating Scale. A secondary outcome was suicidal behavior, including aborted, interrupted, and actual SAs.

RESULTS

The K-CAT-SS showed a 4.91-fold increase in SAs for every 25-point increase in the baseline score (95% CI 2.83-8.52) and a 3.51-fold increase in suicidal behaviors (95% CI 2.32-5.30). These relations persisted following adjustment for prior attempts; demographic variables including age, sex, gender identity, sexual orientation, and race/ethnicity; and other measures of psychopathology. No moderating effects were identified. At 3 months, area under the receiver operating characteristic curve was 0.83 (95% CI 0.72-0.93) for 1 or more SAs.

CONCLUSION

The K-CAT-SS is an excellent tool for suicide risk stratification, particularly in higher-risk populations where other measures have shown lower predictive validity.

Clinical, behavioral, and electrophysiological profiles along a continuum of suicide risk: evidence from an implicit association task

BACKGROUND

An urgent need exists to identify neural correlates associated with differing levels of suicide risk and develop novel, rapid-acting therapeutics to modulate activity within these neural networks.

METHODS

Electrophysiological correlates of suicide were evaluated using magnetoencephalography (MEG) in 75 adults with differing levels of suicide risk. During MEG scanning, participants completed a modified Life-Death Implicit Association Task. MEG data were source-localized in the gamma (30-58 Hz) frequency, a proxy measure of excitation-inhibition balance. Dynamic causal modeling was used to evaluate differences in connectivity estimates between risk groups. A proof-of-concept, open-label, pilot study of five high risk participants examined changes in gamma power after administration of ketamine (0.5 mg/kg), an NMDAR antagonist with rapid anti-suicide ideation effects.

RESULTS

Implicit self-associations with death were stronger in the highest suicide risk group relative to all other groups, which did not differ from each other. Higher gamma power for self-death compared to self-life associations was found in the orbitofrontal cortex for the highest risk group and the insula and posterior cingulate cortex for the lowest risk group. Connectivity estimates between these regions differentiated the highest risk group from the full sample. Implicit associations with death were not affected by ketamine, but enhanced gamma power was found for self-death associations in the left insula post-ketamine compared to baseline.

CONCLUSIONS

Differential implicit cognitive processing of life and death appears to be linked to suicide risk, highlighting the need for objective measures of suicidal states. Pharmacotherapies that modulate gamma activity, particularly in the insula, may help mitigate risk.Clinicaltrials.gov identifier: NCT02543983, NCT00397111.

Management of suicidal risk in the emergency department: A clinical pathway using the computerized adaptive screen for suicidal youth

Objective

Given the critical need for efficient and tailored suicide screening for youth presenting in the emergency department (ED), this study establishes validated screening score thresholds for the Computerized Adaptive Screen for Suicidal Youth (CASSY) and presents an example of a suicide risk classification pathway.

Methods

Participants were primarily from the Study One derivation cohort of the Emergency Department Screen for Teens at Risk for Suicide (ED-STARS) enrolled in collaboration with Pediatric Emergency Care Applied Research Networks (PECARN). CASSY scores corresponded to the predicted probabilities of a suicide attempt in the next 3 months and risk thresholds were classified as minimal (<1%), low (1%-5%), moderate (5%-10%), and high (>10%). CASSY scores were compared to risk thresholds derived from clinical consensus and ED complaints and dispositions. CASSY risk thresholds were also examined as predictors of future suicide attempts in the Study Two validation cohort of ED-STARS.

Results

A total of 1452 teens were enrolled with a median age of 15.2 years, 59.5% were female, 55.6% were White, 22% were Black, 22.3% were Latinx, and 42.8% received public assistance. The clinical consensus suicide risk groups were strongly associated with the CASSY-predicted risk thresholds. Suicide attempts in the Study Two cohort occurred at a frequency consistent with the CASSY-predicted thresholds.

Conclusions

The CASSY can be a valuable tool in providing patient-specific risk probabilities for a suicide attempt at 3 months and tailor the threshold cutoffs based on the availability of local mental health resources. We give an example of a clinical risk pathway, which should include segmentation of the ED population by medical versus psychiatric chief complaint.

Prevalence and related factors of first-time suicide attempts in the past 14 days in Chinese adult patients with first-episode drug-naïve major depressive disorder

Background

This study aimed to identify socio-demographic, physiologic, and psychologic related factors of the first-time suicide attempt (FSA) in the past 14 days in Chinese adult patients with first-episode drug-naïve (FEDN) major depressive disorder (MDD).

Methods

A total of 1718 adult patients with FEDN MDD were enrolled in this cross-sectional survey. Depression, anxiety symptoms, and suicide attempts were assessed. Additionally, biological samples were collected and measured, while Logistic regression analysis was employed to explore the risk factors for FSA in the past 14 days among FEDN MDD patients.

Results

Among suicide attempters, 12.11% (208 out of 1718) reported experiencing FSA in the past 14 days. Logistic regression analysis showed that the risk factors for FSA included more severe anxiety symptoms (OR=1.37, 95%CI: 1.28-1.48, p<0.001), higher levels of total cholesterol (TC) (OR=1.42, 95%CI: 1.13-1.77, p=0.003), and elevated thyroid-stimulating hormone (TSH) (OR=1.13, 95%CI: 1.03-1.25, p=0.01). The regression model exhibited good discriminatory power for FSA with an area under the curve (AUC) of 0.82.

Conclusion

FEDN MDD patients with more severe anxiety and higher levels of TSH and TC are more likely to develop FSA in the past 14 days. These factors are risk factors for short-term (in the past 14 days) FSA and may serve as indicators for early intervention.

Machine learning and the prediction of suicide in psychiatric populations: a systematic review

Machine learning (ML) has emerged as a promising tool to enhance suicidal prediction. However, as many large-sample studies mixed psychiatric and non-psychiatric populations, a formal psychiatric diagnosis emerged as a strong predictor of suicidal risk, overshadowing more subtle risk factors specific to distinct populations. To overcome this limitation, we conducted a systematic review of ML studies evaluating suicidal behaviors exclusively in psychiatric clinical populations. A systematic literature search was performed from inception through November 17, 2022 on PubMed, EMBASE, and Scopus following the PRISMA guidelines. Original research using ML techniques to assess the risk of suicide or predict suicide attempts in the psychiatric population were included. An assessment for bias risk was performed using the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) guidelines. About 1032 studies were retrieved, and 81 satisfied the inclusion criteria and were included for qualitative synthesis. Clinical and demographic features were the most frequently employed and random forest, support vector machine, and convolutional neural network performed better in terms of accuracy than other algorithms when directly compared. Despite heterogeneity in procedures, most studies reported an accuracy of 70% or greater based on features such as previous attempts, severity of the disorder, and pharmacological treatments. Although the evidence reported is promising, ML algorithms for suicidal prediction still present limitations, including the lack of neurobiological and imaging data and the lack of external validation samples. Overcoming these issues may lead to the development of models to adopt in clinical practice. Further research is warranted to boost a field that holds the potential to critically impact suicide mortality.

Suicide Risk Screening Tools for Pediatric Patients: A Systematic Review of Test Accuracy

CONTEXT

Health care settings have increasingly adopted universal suicide risk screening tools into nonpsychiatric pediatric care; however, a systematic review examining the accuracy of these tools does not yet exist.

OBJECTIVE

Identify and review research on the test accuracy of suicide risk screening tools for pediatric patients in nonpsychiatric medical settings.

DATA SOURCES

PubMed and PsycINFO were searched to identify peer-reviewed articles published before March 23, 2023.

STUDY SELECTION

Articles that quantified the accuracy of a suicide risk screening tool (eg, sensitivity, specificity) in a nonpsychiatric medical setting (eg, primary care, specialty care, inpatient or surgical units, or the emergency department) were included.

DATA EXTRACTION

A total of 13 studies were included in this review. Screening tool psychometric properties and study risk of bias were evaluated.

RESULTS

Sensitivity among individual studies ranged from 50% to 100%, and specificity ranged from 58.8% to 96%. Methodological quality was relatively varied, and applicability concerns were low. When stratifying results by screening tool, the Ask Suicide-Screening Questions and Computerized Adaptive Screen for Suicidal Youth had the most robust evidence base.

LIMITATIONS

Because of considerable study heterogeneity, a meta-analytic approach was deemed inappropriate. This prevented us from statistically testing for differences between identified screening tools.

CONCLUSIONS

The Ask Suicide-Screening Questions and Computerized Adaptive Screen for Suicidal Youth exhibit satisfactory test accuracy and appear promising for integration into clinical practice. Although initial findings are promising, additional research targeted at examining the accuracy of screening tools among diverse populations is needed to ensure the equity of screening efforts.

Predicting suicide death after emergency department visits with mental health or self-harm diagnoses

OBJECTIVE

Use health records data to predict suicide death following emergency department visits.

METHODS

Electronic health records and insurance claims from seven health systems were used to: identify emergency department visits with mental health or self-harm diagnoses by members aged 11 or older; extract approximately 2500 potential predictors including demographic, historical, and baseline clinical characteristics; and ascertain subsequent deaths by self-harm. Logistic regression with lasso and random forest models predicted self-harm death over 90 days after each visit.

RESULTS

Records identified 2,069,170 eligible visits, 899 followed by suicide death within 90 days. The best-fitting logistic regression with lasso model yielded an area under the receiver operating curve of 0.823 (95% CI 0.810-0.836). Visits above the 95th percentile of predicted risk included 34.8% (95% CI 31.1-38.7) of subsequent suicide deaths and had a 0.303% (95% CI 0.261-0.346) suicide death rate over the following 90 days. Model performance was similar across subgroups defined by age, sex, race, and ethnicity.

CONCLUSIONS

Machine learning models using coded data from health records have moderate performance in predicting suicide death following emergency department visits for mental health or self-harm diagnosis and could be used to identify patients needing more systematic follow-up.

Multi-method assessment of suicidal thoughts and behaviors among patients in treatment for OCD and related disorders

Obsessive-compulsive and related disorders (OCRDs) are associated with increased risk of suicidal thoughts and behaviors (STBs), yet research characterizing suicidality in OCRDs remains limited. A major challenge in assessing STBs is the reliance on explicit self-report. This study utilized multi-method assessment to examine changes in both implicit and explicit STBs in 31 adults receiving partial/residential treatment for OCRDs. Assessments were administered at admission and weekly during treatment. Approximately three-quarters of participants reported lifetime suicidal thoughts, with 16 % reporting a prior suicide attempt. OCD severity was significantly correlated with lifetime suicidal thoughts, and was significantly higher for those with lifetime suicidal thoughts and prior attempts compared to those without. Implicit biases towards death were not associated with OCD severity, and did not predict explicitly endorsed STBs. This is the first study to measure both explicit and implicit STBs in adults with OCRDs. Limitations included small sample size and lack of racial/ethnic diversity. Given the majority had recent suicidal thoughts and one in six had a prior attempt, we emphasize the importance of STB assessment in OCD treatment settings.

Suicide Risk Screening for Head and Neck Cancer Patients: An Implementation Study

OBJECTIVES

 There is limited research on suicide risk screening (SRS) among head and neck cancer (HNC) patients, a population at increased risk for suicide. To address this gap, this single-site mixed methods study assessed oncology professionals' perspectives about the feasibility, acceptability, and appropriateness of an electronic SRS program that was implemented as a part of routine care for HNC patients.

METHODS

Staff who assisted with SRS implementation completed (e.g., nurses, medical assistants, advanced practice providers, physicians, social workers) a one-time survey (N = 29) and interview (N = 25). Quantitative outcomes were assessed using previously validated feasibility, acceptability, and appropriateness measures. Additional qualitative data were collected to provide context for interpreting the scores.

RESULTS

 Nurses and medical assistants, who were directly responsible for implementing SRS, reported low feasibility, acceptability, and appropriateness, compared with other team members (e.g., physicians, social workers, advanced practice providers). Team members identified potential improvements needed to optimize SRS, such as hiring additional staff, improving staff training, providing different modalities for screening completion among individuals with disabilities, and revising the patient-reported outcomes to improve suicide risk prediction.

CONCLUSION

 Staff perspectives about implementing SRS as a part of routine cancer care for HNC patients varied widely. Before screening can be implemented on a larger scale for HNC and other cancer patients, additional implementation strategies may be needed that optimize workflow and reduce staff burden, such as staff training, multiple modalities for completion, and refined tools for identifying which patients are at greatest risk for suicide.

Estimated Average Treatment Effect of Psychiatric Hospitalization in Patients With Suicidal Behaviors: A Precision Treatment Analysis

Importance

Psychiatric hospitalization is the standard of care for patients presenting to an emergency department (ED) or urgent care (UC) with high suicide risk. However, the effect of hospitalization in reducing subsequent suicidal behaviors is poorly understood and likely heterogeneous.

Objectives

To estimate the association of psychiatric hospitalization with subsequent suicidal behaviors using observational data and develop a preliminary predictive analytics individualized treatment rule accounting for heterogeneity in this association across patients.

Design, Setting, and Participants

A machine learning analysis of retrospective data was conducted. All veterans presenting with suicidal ideation (SI) or suicide attempt (SA) from January 1, 2010, to December 31, 2015, were included. Data were analyzed from September 1, 2022, to March 10, 2023. Subgroups were defined by primary psychiatric diagnosis (nonaffective psychosis, bipolar disorder, major depressive disorder, and other) and suicidality (SI only, SA in past 2-7 days, and SA in past day). Models were trained in 70.0% of the training samples and tested in the remaining 30.0%.

Exposures

Psychiatric hospitalization vs nonhospitalization.

Main Outcomes and Measures

Fatal and nonfatal SAs within 12 months of ED/UC visits were identified in administrative records and the National Death Index. Baseline covariates were drawn from electronic health records and geospatial databases.

Results

Of 196 610 visits (90.3% men; median [IQR] age, 53 [41-59] years), 71.5% resulted in hospitalization. The 12-month SA risk was 11.9% with hospitalization and 12.0% with nonhospitalization (difference, -0.1%; 95% CI, -0.4% to 0.2%). In patients with SI only or SA in the past 2 to 7 days, most hospitalization was not associated with subsequent SAs. For patients with SA in the past day, hospitalization was associated with risk reductions ranging from -6.9% to -9.6% across diagnoses. Accounting for heterogeneity, hospitalization was associated with reduced risk of subsequent SAs in 28.1% of the patients and increased risk in 24.0%. An individualized treatment rule based on these associations may reduce SAs by 16.0% and hospitalizations by 13.0% compared with current rates.

Conclusions and Relevance

The findings of this study suggest that psychiatric hospitalization is associated with reduced average SA risk in the immediate aftermath of an SA but not after other recent SAs or SI only. Substantial heterogeneity exists in these associations across patients. An individualized treatment rule accounting for this heterogeneity could both reduce SAs and avert hospitalizations.

Determining Distinct Suicide Attempts From Recurrent Electronic Health Record Codes: Classification Study

BACKGROUND

Prior suicide attempts are a relatively strong risk factor for future suicide attempts. There is growing interest in using longitudinal electronic health record (EHR) data to derive statistical risk prediction models for future suicide attempts and other suicidal behavior outcomes. However, model performance may be inflated by a largely unrecognized form of "data leakage" during model training: diagnostic codes for suicide attempt outcomes may refer to prior attempts that are also included in the model as predictors.

OBJECTIVE

We aimed to develop an automated rule for determining when documented suicide attempt diagnostic codes identify distinct suicide attempt events.

METHODS

From a large health care system's EHR, we randomly sampled suicide attempt codes for 300 patients with at least one pair of suicide attempt codes documented at least one but no more than 90 days apart. Supervised chart reviewers assigned the clinical settings (ie, emergency department [ED] versus non-ED), methods of suicide attempt, and intercode interval (number of days). The probability (or positive predictive value) that the second suicide attempt code in a given pair of codes referred to a distinct suicide attempt event from its preceding suicide attempt code was calculated by clinical setting, method, and intercode interval.

RESULTS

Of 1015 code pairs reviewed, 835 (82.3%) were nonindependent (ie, the 2 codes referred to the same suicide attempt event). When the second code in a pair was documented in a clinical setting other than the ED, it represented a distinct suicide attempt 3.3% of the time. The more time elapsed between codes, the more likely the second code in a pair referred to a distinct suicide attempt event from its preceding code. Code pairs in which the second suicide attempt code was assigned in an ED at least 5 days after its preceding suicide attempt code had a positive predictive value of 0.90.

CONCLUSIONS

EHR-based suicide risk prediction models that include International Classification of Diseases codes for prior suicide attempts as a predictor may be highly susceptible to bias due to data leakage in model training. We derived a simple rule to distinguish codes that reflect new, independent suicide attempts: suicide attempt codes documented in an ED setting at least 5 days after a preceding suicide attempt code can be confidently treated as new events in EHR-based suicide risk prediction models. This rule has the potential to minimize upward bias in model performance when prior suicide attempts are included as predictors in EHR-based suicide risk prediction models.

Digital therapeutics in the hospital for suicide crisis - content and design recommendations from young people and hospital staff

Objective

Hospital emergency departments lack the resources to adequately support young people who present for suicidal crisis. Digital therapeutics could fill this service gap by providing psychological support without creating additional burden on hospital staff. However, existing research on what is needed for successful integration of digital therapeutics in hospital settings is scant. Thus, this study sought to identify key considerations for implementing digital therapeutics to manage acute suicidal distress in hospitals.

Method

Participants were 17 young people who recently presented at the hospital for suicide-related crisis, and 12 hospital staff who regularly interacted with young people experiencing mental ill-health in their day-to-day work. Interviews were conducted via videoconference. Framework analysis and reflexive thematic analysis were used to interpret the data obtained.

Results

Qualitative insights were centred around three major themes: hospital-specific content, therapeutic content, and usability. Digital therapeutics were seen as a useful means for facilitating hospital-based assessment and treatment planning, and for conducting post-discharge check-ins. Therapeutic content should be focused on helping young people self-manage suicide-related distress while they wait for in-person services. Features to promote usability, such as the availability of customisable features and the use of inclusive design or language, should be considered in the design of digital therapeutics.

Conclusions

Digital therapeutics in hospital settings need to benefit both patients and staff. Given the unique context of the hospital setting and acute nature of suicidal distress, creating specialty digital therapeutics may be more viable than integrating existing ones.

Risk Factors for Suicide in Men

Suicide and the risk factors associated with it have been researched with increasing interest over the last 5 decades with respect to socioeconomic status, age, geographic location, and ethnic background. There has been less focus related to the risk factors specific to gender and how to incorporate clinical screening and interventions to reduce the mortality of suicide in males. With men accounting for a disproportionate number of deaths from suicide in the United States and worldwide, how gender could impact suicidal behavior and ideations remains a topic understudied and with great potential for significant improvement in clinical recognition and treatment.

A practical risk calculator for suicidal behavior among transitioning U.S. Army soldiers: results from the Study to Assess Risk and Resilience in Servicemembers-Longitudinal Study (STARRS-LS)

BACKGROUND

Risk of suicide-related behaviors is elevated among military personnel transitioning to civilian life. An earlier report showed that high-risk U.S. Army soldiers could be identified shortly before this transition with a machine learning model that included predictors from administrative systems, self-report surveys, and geospatial data. Based on this result, a Veterans Affairs and Army initiative was launched to evaluate a suicide-prevention intervention for high-risk transitioning soldiers. To make targeting practical, though, a streamlined model and risk calculator were needed that used only a short series of self-report survey questions.

METHODS

We revised the original model in a sample of n = 8335 observations from the Study to Assess Risk and Resilience in Servicemembers-Longitudinal Study (STARRS-LS) who participated in one of three Army STARRS 2011-2014 baseline surveys while in service and in one or more subsequent panel surveys (LS1: 2016-2018, LS2: 2018-2019) after leaving service. We trained ensemble machine learning models with constrained numbers of item-level survey predictors in a 70% training sample. The outcome was self-reported post-transition suicide attempts (SA). The models were validated in the 30% test sample.

RESULTS

Twelve-month post-transition SA prevalence was 1.0% (s.e. = 0.1). The best constrained model, with only 17 predictors, had a test sample ROC-AUC of 0.85 (s.e. = 0.03). The 10-30% of respondents with the highest predicted risk included 44.9-92.5% of 12-month SAs.

CONCLUSIONS

An accurate SA risk calculator based on a short self-report survey can target transitioning soldiers shortly before leaving service for intervention to prevent post-transition SA.

P4 suicidality screener: Literature synthesis and results from two randomized trials

OBJECTIVE

To synthesize the literature on use of the P4 suicidality screener since its introduction in 2010 and to summarize results from 2 randomized clinical trials.

METHOD

A PubMed search was conducted from 2010 to 2023 to retrieve studies reporting on use of the P4. Also, data was extracted from the CAMMPS and SCOPE trials in which the P4 was periodically administered over 12 months when the 9th item of the PHQ-9 was endorsed.

RESULTS

A total of 21 research studies using the P4 were found, of which 12 provided some data on P4 findings. Additionally, another 7 protocol papers reported intended use of the P4 as a study measure. In our 2 trials, the 9th item was endorsed 259 (12.5%) times in 2068 administrations of the PHQ-9. Higher risk suicidal ideation was identified in 4.1% (12/294) of CAMMPS participants and 2.8% (7/250) of SCOPE participants. No suicide attempts occurred over the 12 months in either trial.

CONCLUSIONS

The P4 has had moderate use as a brief suicidality screener and is an efficient way to identify the small proportion of depressed patients with higher risk suicidality. Studies comparing the P4 with other common suicidality screeners would further inform use.

Toward objective characterizations of suicide risk: A narrative review of laboratory-based cognitive and behavioral tasks

Although suicide is a leading cause of preventable death worldwide, current prevention efforts have failed to substantively mitigate suicide risk. Suicide research has traditionally relied on subjective reports that may not accurately differentiate those at high versus minimal risk. This narrative review supports the inclusion of objective task-based measures in suicide research to complement existing subjective batteries. The article: 1) outlines risk factors proposed by contemporary theories of suicide and highlights recent empirical findings supporting these theories; 2) discusses ongoing challenges associated with current risk assessment tools and their ability to accurately evaluate risk factors; and 3) analyzes objective laboratory measures that can be implemented alongside traditional measures to enhance the precision of risk assessment. To illustrate the potential of these methods to improve our understanding of suicide risk, the article reviews how acute stress responses in a laboratory setting can be modeled, given that stress is a major precipitant for suicidal behavior. More precise risk assessment strategies can emerge if objective measures are implemented in conjunction with traditional subjective measures.

One-Week Suicide Risk Prediction Using Real-Time Smartphone Monitoring: Prospective Cohort Study

BACKGROUND

Suicide is a major global public health issue that is becoming increasingly common despite preventive efforts. Though current methods for predicting suicide risk are not sufficiently accurate, technological advances provide invaluable tools with which we may evolve toward a personalized, predictive approach.

OBJECTIVE

We aim to predict the short-term (1-week) risk of suicide by identifying changes in behavioral patterns characterized through real-time smartphone monitoring in a cohort of patients with suicidal ideation.

METHODS

We recruited 225 patients between February 2018 and March 2020 with a history of suicidal thoughts and behavior as part of the multicenter SmartCrisis study. Throughout 6 months of follow-up, we collected information on the risk of suicide or mental health crises. All participants underwent voluntary passive monitoring using data generated by their own smartphones, including distance walked and steps taken, time spent at home, and app usage. The algorithm constructs daily activity profiles for each patient according to these data and detects changes in the distribution of these profiles over time. Such changes are considered critical periods, and their relationship with suicide-risk events was tested.

RESULTS

During follow-up, 18 (8%) participants attempted suicide, and 14 (6.2%) presented to the emergency department for psychiatric care. The behavioral changes identified by the algorithm predicted suicide risk in a time frame of 1 week with an area under the curve of 0.78, indicating good accuracy.

CONCLUSIONS

We describe an innovative method to identify mental health crises based on passively collected information from patients' smartphones. This technology could be applied to homogeneous groups of patients to identify different types of crises.

Association rule mining of real-world data: Uncovering links between race, glycemic control, lipid profiles, and suicide attempts in individuals with diabetes

Aims

The increased risk of suicide among individuals with diabetes is a significant public health concern. However, few studies have focused on understanding the relationship between suicide attempts and diabetes. Association rule mining (ARM) is a data mining technique to discover a set of high-risk factors of a given disease. Therefore, this study aimed to utilize ARM to identify a high-risk group of suicide attempts among patients with diabetes using Cerner Real-World Data™ (CRWD).

Methods

The study analyzed a large multicenter electronic health records data of 3,265,041 patients with diabetes from 2010 to 2020. The Least Absolute Shrinkage and Selection Operator regression with ten-fold cross-validation and the Apriori algorithm with ARM were used to uncover groups of high-risk suicide attempts.

Results

Of the 52,217,517 unique patients in the CRWD, 3,266,856 were diagnosed with diabetes. There were 7764 (0.2%) patients with diabetes who had a history of suicide attempts. The study revealed that patients with diabetes who were never married and had average blood glucose levels below 150 mg/dl were more likely to attempt suicide. In contrast, patients with diabetes aged 60 and older who had diabetes for less than five years and A1C levels between 6.5 and 8.9% were less likely to attempt suicide. Risk factors were strongly associated with suicide attempts, including never married, White, blood glucose levels below 150 mg/dl, and LDL levels below 100 mg/dl.

Conclusions

This is the first study utilizing ARM to discover the risk patterns for suicide attempts in individuals with diabetes. ARM showed the potential for knowledge discovery in large multi-center electronic health records data. The results are explainable and could be practically used by providers during outpatient clinic visits. Further studies are needed to validate the results and investigate the cause-and-effect relationship of suicide attempts among individuals with diabetes.

Suicide risk detection using artificial intelligence: the promise of creating a benchmark dataset for research on the detection of suicide risk

Suicide is a leading cause of death that demands cross-disciplinary research efforts to develop and deploy suicide risk screening tools. Such tools, partly informed by influential suicide theories, can help identify individuals at the greatest risk of suicide and should be able to predict the transition from suicidal thoughts to suicide attempts. Advances in artificial intelligence have revolutionized the development of suicide screening tools and suicide risk detection systems. Thus, various types of AI systems, including text-based systems, have been proposed to identify individuals at risk of suicide. Although these systems have shown acceptable performance, most of them have not incorporated suicide theories in their design. Furthermore, directly applying suicide theories may be difficult because of the diversity and complexity of these theories. To address these challenges, we propose an approach to develop speech- and language-based suicide risk detection systems. We highlight the promise of establishing a benchmark textual and vocal dataset using a standardized speech and language assessment procedure, and research designs that distinguish between the risk factors for suicide attempt above and beyond those for suicidal ideation alone. The benchmark dataset could be used to develop trustworthy machine learning or deep learning-based suicide risk detection systems, ultimately constructing a foundation for vocal and textual-based suicide risk detection systems.

Dynamic suicide topic modelling: Deriving population-specific, psychosocial and time-sensitive suicide risk variables from Electronic Health Record psychotherapy notes

In the machine learning subfield of natural language processing, a topic model is a type of unsupervised method that is used to uncover abstract topics within a corpus of text. Dynamic topic modelling (DTM) is used for capturing change in these topics over time. The study deploys DTM on corpus of electronic health record psychotherapy notes. This retrospective study examines whether DTM helps distinguish closely matched patients that did and did not die by suicide. Cohort consists of United States Department of Veterans Affairs (VA) patients diagnosed with Posttraumatic Stress Disorder (PTSD) between 2004 and 2013. Each case (those who died by suicide during the year following diagnosis) was matched with five controls (those who remained alive) that shared psychotherapists and had similar suicide risk based on VA's suicide prediction algorithm. Cohort was restricted to patients who received psychotherapy for 9+ months after initial PTSD diagnoses (cases = 77; controls = 362). For cases, psychotherapy notes from diagnosis until death were examined. For controls, psychotherapy notes from diagnosis until matched case's death date were examined. A Python-based DTM algorithm was utilized. Derived topics identified population-specific themes, including PTSD, psychotherapy, medication, communication and relationships. Control topics changed significantly more over time than case topics. Topic differences highlighted engagement, expressivity and therapeutic alliance. This study strengthens groundwork for deriving population-specific, psychosocial and time-sensitive suicide risk variables.

Suicidal Ideation Among Patients with Gastrointestinal Cancer

BACKGROUND

Mental illness (MI) and suicidal ideation (SI) often are associated with a diagnosis of cancer. We sought to define the incidence of MI and SI among patients with gastrointestinal cancers, as well as ascertain the predictive factors associated with SI.

METHODS

Patients diagnosed between 2004 and 2016 with stomach, liver, pancreatic, and colorectal cancer were identified from the SEER-Medicare database. County-level social vulnerability index (SVI) was extracted from the Centers for Disease Control database. Multivariable logistic regression was used to identify factors associated with SI.

RESULTS

Among 382,266 patients, 83,514 (21.9%) individuals had a diagnosis of MI. Only 1410 (0.4%) individuals experienced SI, and 359 (0.1%) committed suicide. Interestingly, SI was least likely among patients with pancreatic cancer (ref: hepatic cancer; odds ratio [OR] 0.67, 95% confidence interval [CI] 0.52-0.86; p = 0.002), as well as individuals with stage III/IV disease (OR 0.59, 95% CI 0.52-067; p < 0.001). In contrast, male (OR 1.34, 95% CI 1.19-1.50), White (OR 1.34, CI 1.13-1.59), and single (OR 2.03, 95% CI 1.81-2.28) patients were at higher odds of SI risk (all p < 0.001). Furthermore, individuals living in relative privilege (low SVI) had markedly higher risk of SI (OR 1.33, 95% CI 1.14-1.54; p < 0.001). Moreover, living in a county with a shortage of mental health professionals was associated with increased odds of developing SI (OR 1.21, 95% CI 1.04-1.40; p = 0.012).

CONCLUSIONS

Oncology care teams should incorporate routine mental health and SI screening in the treatment of patients with gastrointestinal cancers, as well as target suicide prevention towards patients at highest risk.

Predictors of suicide attempt within 30 days of first medically documented major depression diagnosis in U.S. army soldiers with no prior suicidal ideation

BACKGROUND

Understanding mental health predictors of imminent suicide attempt (SA; within 30 days) among soldiers with depression and no prior suicide ideation (SI) can inform prevention and treatment. The current study aimed to identify sociodemographic and service-related characteristics and mental disorder predictors associated with imminent SA among U.S. Army soldiers following first documented major depression diagnosis (MDD) with no history of SI.

METHODS

In this case-control study using Army Study to Assess Risk and Resilience in Servicemembers (STARRS) administrative data, we identified 101,046 active-duty Regular Army enlisted soldiers (2010-2016) with medically-documented MDD and no prior SI (MDD/No-SI). We examined risk factors for SA within 30 days of first MDD/No-SI using logistic regression analyses, including socio-demographic/service-related characteristics and psychiatric diagnoses.

RESULTS

The 101,046 soldiers with documented MDD/No-SI were primarily male (78.0%), < 29 years old (63.9%), White (58.1%), high school-educated (74.5%), currently married (62.0%) and < 21 when first entering the Army (56.9%). Among soldiers with MDD/No-SI, 2,600 (2.6%) subsequently attempted suicide, 16.2% (n = 421) within 30 days (rate: 416.6/100,000). Our final multivariable model identified: Soldiers with less than high school education (χ²₃ = 11.21, OR = 1.5[95%CI = 1.2-1.9]); combat medics (χ²₂ = 8.95, OR = 1.5[95%CI = 1.1-2.2]); bipolar disorder (OR = 3.1[95%CI = 1.5-6.3]), traumatic stress (i.e., acute reaction to stress/not PTSD; OR = 2.6[95%CI = 1.4-4.8]), and "other" diagnosis (e.g., unspecified mental disorder: OR = 5.5[95%CI = 3.8-8.0]) diagnosed same day as MDD; and those with alcohol use disorder (OR = 1.4[95%CI = 1.0-1.8]) and somatoform/dissociative disorders (OR = 1.7[95%CI = 1.0-2.8]) diagnosed before MDD were more likely to attempt suicide within 30 days. Currently married soldiers (χ²₂ = 6.68, OR = 0.7[95%CI = 0.6-0.9]), those in service 10 + years (χ²₃ = 10.06, OR = 0.4[95%CI = 0.2-0.7]), and a sleep disorder diagnosed same day as MDD (OR = 0.3[95%CI = 0.1-0.9]) were less likely.

CONCLUSIONS

SA risk within 30 days following first MDD is more likely among soldiers with less education, combat medics, and bipolar disorder, traumatic stress, and "other" disorder the same day as MDD, and alcohol use disorder and somatoform/dissociative disorders before MDD. These factors identify imminent SA risk and can be indicators for early intervention.

The use of advanced technology and statistical methods to predict and prevent suicide

In the past decade, two themes have emerged across suicide research. First, according to meta-analyses, the ability to predict and prevent suicidal thoughts and behaviours is weaker than would be expected for the size of the field. Second, review and commentary papers propose that technological and statistical methods (such as smartphones, wearables, digital phenotyping and machine learning) might become solutions to this problem. In this Review, we aim to strike a balance between the pessimistic picture presented by these meta-analyses and the optimistic picture presented by review and commentary papers about the promise of advanced technological and statistical methods to improve the ability to understand, predict and prevent suicide. We divide our discussion into two broad categories. First, we discuss the research aimed at assessment, with the goal of better understanding or more accurately predicting suicidal thoughts and behaviours. Second, we discuss the literature that focuses on prevention of suicidal thoughts and behaviours. Ecological momentary assessment, wearables and other technological and statistical advances hold great promise for predicting and preventing suicide, but there is much yet to do.

Predictive value of suicidal risk assessment using data from China's largest suicide prevention hotline

BACKGROUND

Suicide hotlines are widely used, with potential for identification of callers at especially high risk.

METHODS

This prospective study was conducted at the largest psychological support hotline in China. From 2015 to 2017, all distressed callers were consecutively included and assessed, using a standardized scale consisting of 12 elements, yielding scores of high risk (8-16), moderate risk (4-7), and low risk (0-3) for suicidal act. All high-risk and half of moderate- and low-risk callers were scheduled for a 12-month follow-up. Main outcomes were suicidal acts (nonlethal attempt, death) over follow-up.

RESULTS

Of 21,346 fully assessed callers, 5822, 11,791, and 3733 were classified as high-, moderate-, or low-risk for suicidal acts, with 8869 callers (4076 high-, 3258 moderate-, and 1535 low-risk) followed up over 12 months. Over follow-up, 802 (9.0 %) callers attempted suicide or died by suicide. The high-risk callers (15.1 %) had 3-fold higher risk for subsequent suicidal acts than moderate- (5.1 %) and 12-fold higher risk than low-risk callers (1.3 %). The weighted sensitivity, specificity, and positive predictive value of high risk scores were 56.4 %, 74.9 %, and 14.4 %.

LIMITATIONS

Assessed callers with different risk levels were followed disproportionally.

CONCLUSIONS

Suicidal risk assessment during a hotline call is both feasible and predictive of risk, guiding resource allocation to higher risk callers.

An efficient landmark model for prediction of suicide attempts in multiple clinical settings

Growing evidence has shown that applying machine learning models to large clinical data sources may exceed clinician performance in suicide risk stratification. However, many existing prediction models either suffer from "temporal bias" (a bias that stems from using case-control sampling) or require training on all available patient visit data. Here, we adopt a "landmark model" framework that aligns with clinical practice for prediction of suicide-related behaviors (SRBs) using a large electronic health record database. Using the landmark approach, we developed models for SRB prediction (regularized Cox regression and random survival forest) that establish a time-point (e.g., clinical visit) from which predictions are made over user-specified prediction windows using historical information up to that point. We applied this approach to cohorts from three clinical settings: general outpatient, psychiatric emergency department, and psychiatric inpatients, for varying prediction windows and lengths of historical data. Models achieved high discriminative performance (area under the Receiver Operating Characteristic curve 0.74-0.93 for the Cox model) across different prediction windows and settings, even with relatively short periods of historical data. In short, we developed accurate, dynamic SRB risk prediction models with the landmark approach that reduce bias and enhance the reliability and portability of suicide risk prediction models.

Clinical Considerations for Digital Resources in Care for Patients With Suicidal Ideation

Smartphone apps offer accessible new tools that may help prevent suicide and that offer support for individuals with active suicidal ideation. Numerous smartphone apps for mental health conditions exist; however, their functionality is limited, and evidence is nascent. A new generation of apps using smartphone sensors and integrating real-time data on evolving risk offers the potential of more personalized support, but these apps present ethical risks and currently remain more in the research domain than in the clinical domain. Nevertheless, clinicians can use apps to benefit patients. This article outlines practical strategies to select safe and effective apps for the creation of a digital toolkit that can augment suicide prevention and safety plans. By creating a unique digital toolkit for each patient, clinicians can help ensure that the apps selected will be most relevant, engaging, and effective.

Future Directions in Understanding and Interpreting Discrepant Reports of Suicidal Thoughts and Behaviors Among Youth

Both the quality and utility of youth suicide research depend on how we assess our outcomes of interest: suicidal thoughts and behaviors (STBs). We now have access to more STB assessments than ever before, with measures for youth that vary in what exact experiences are asked about, how such measures elicit information, when and how frequently measures are administered, and who the informants are. This growing armamentarium of assessments has the potential to improve the study and treatment of STBs among youth, but it hinges on meaningful interpretation of assessment responses. Interpretation can be especially challenging when different STB assessments yield conflicting information. Determining how to manage discrepant reports of STBs is a pivotal step toward achieving meaningfully comprehensive STB assessment batteries. Here, we outline several discrepant reporting patterns that have been detected, discuss the potential significance of these observed discrepancies, and present initial steps to formally investigate discrepant reports of STBs among youth. Developing coherent, interpretable, and comprehensive batteries assessing STBs among youth would address a fundamental step to uncovering etiology, improving clinical decision-making and case management, informing intervention development, and tracking prognosis.

Polyphenic risk score shows robust predictive ability for long-term future suicidality

Suicides are preventable tragedies, if risk factors are tracked and mitigated. We had previously developed a new quantitative suicidality risk assessment instrument (Convergent Functional Information for Suicidality, CFI-S), which is in essence a simple polyphenic risk score, and deployed it in a busy urban hospital Emergency Department, in a naturalistic cohort of consecutive patients. We report a four years follow-up of that population (n = 482). Overall, the single administration of the CFI-S was significantly predictive of suicidality over the ensuing 4 years (occurrence- ROC AUC 80%, severity- Pearson correlation 0.44, imminence-Cox regression Hazard Ratio 1.33). The best predictive single phenes (phenotypic items) were feeling useless (not needed), a past history of suicidality, and social isolation. We next used machine learning approaches to enhance the predictive ability of CFI-S. We divided the population into a discovery cohort (n = 255) and testing cohort (n = 227), and developed a deep neural network algorithm that showed increased accuracy for predicting risk of future suicidality (increasing the ROC AUC from 80 to 90%), as well as a similarity network classifier for visualizing patient’s risk. We propose that the widespread use of CFI-S for screening purposes, with or without machine learning enhancements, can boost suicidality prevention efforts. This study also identified as top risk factors for suicidality addressable social determinants.

Leveraging unstructured electronic medical record notes to derive population-specific suicide risk models

Electronic medical record (EMR)-based suicide risk prediction methods typically rely on analysis of structured variables such as demographics, visit history, and prescription data. Leveraging unstructured EMR notes may improve predictive accuracy by allowing access to nuanced clinical information. We utilized natural language processing (NLP) to analyze a large EMR note corpus to develop a data-driven suicide risk prediction model. We developed a matched case-control sample of U.S. Department of Veterans Affairs (VA) patients in 2015 and 2016. We randomly matched each case (all patients that died by suicide in that interval, n = 5029) with five controls (patients that remained alive). We processed note corpus using NLP methods and applied machine-learning classification algorithms to output. We calculated area under the curve (AUC) and risk tiers to determine predictive accuracy. NLP-derived models demonstrated strong predictive accuracy. Patients that scored within top 10% of risk model accounted for up to 29% of suicide decedents. NLP-derived model compares positively to other leading prediction methods. Our approach is highly implementable, only requiring access to text data and open-source software. Additional studies should evaluate ensemble models incorporating NLP-derived information alongside more typical structured variables.

Effects of stress on endophenotypes of suicide across species: A role for ketamine in risk mitigation

Suicide is a leading cause of death and morbidity worldwide, yet few interventions are available to mitigate its risk. Barriers to effective treatments involve a limited understanding of factors that predict the onset of suicidal thoughts and behaviors. In the context of suicide risk, stress is a precipitating factor that is largely overlooked in the literature. Indeed, the pathophysiology of stress and suicide are heavily interconnected, underscoring the need to target the stress system in suicide prevention. In this review, we integrate findings from the preclinical and clinical literature that links stress and suicide. We focus specifically on the effects of stress on underlying biological functions and processes associated with suicide, allowing for the review of research using animal models. Owing to the rapid anti-suicidal effects of (R,S)-ketamine, we discuss its ability to modulate various stress-related endophenotypes of suicide, as well as its potential role in preventing suicide in those with a history of chronic life stress (e.g., early life adversity). We highlight future research directions that could advance our understanding of stress-related effects on suicide risk, advocating a dimensional, endophenotype approach to suicide research.

•

Suicide and chronic stress pathophysiology are interconnected.

•

Chronic stress has profound impacts on several endophenotypes of suicide.

•

Animal and human research points to stress as a precipitating factor in suicide.

•

Ketamine modulates specific biological processes associated with stress and suicide.

•

Suicide research into endophenotypes can help inform risk-mitigation strategies.

Implementing Machine Learning Models for Suicide Risk Prediction in Clinical Practice: Focus Group Study With Hospital Providers

Background

Interest in developing machine learning models that use electronic health record data to predict patients’ risk of suicidal behavior has recently proliferated. However, whether and how such models might be implemented and useful in clinical practice remain unknown. To ultimately make automated suicide risk–prediction models useful in practice, and thus better prevent patient suicides, it is critical to partner with key stakeholders, including the frontline providers who will be using such tools, at each stage of the implementation process.

Objective

The aim of this focus group study is to inform ongoing and future efforts to deploy suicide risk–prediction models in clinical practice. The specific goals are to better understand hospital providers’ current practices for assessing and managing suicide risk; determine providers’ perspectives on using automated suicide risk–prediction models in practice; and identify barriers, facilitators, recommendations, and factors to consider.

Methods

We conducted 10 two-hour focus groups with a total of 40 providers from psychiatry, internal medicine and primary care, emergency medicine, and obstetrics and gynecology departments within an urban academic medical center. Audio recordings of open-ended group discussions were transcribed and coded for relevant and recurrent themes by 2 independent study staff members. All coded text was reviewed and discrepancies were resolved in consensus meetings with doctoral-level staff.

Results

Although most providers reported using standardized suicide risk assessment tools in their clinical practices, existing tools were commonly described as unhelpful and providers indicated dissatisfaction with current suicide risk assessment methods. Overall, providers’ general attitudes toward the practical use of automated suicide risk–prediction models and corresponding clinical decision support tools were positive. Providers were especially interested in the potential to identify high-risk patients who might be missed by traditional screening methods. Some expressed skepticism about the potential usefulness of these models in routine care; specific barriers included concerns about liability, alert fatigue, and increased demand on the health care system. Key facilitators included presenting specific patient-level features contributing to risk scores, emphasizing changes in risk over time, and developing systematic clinical workflows and provider training. Participants also recommended considering risk-prediction windows, timing of alerts, who will have access to model predictions, and variability across treatment settings.

Conclusions

Providers were dissatisfied with current suicide risk assessment methods and were open to the use of a machine learning–based risk-prediction system to inform clinical decision-making. They also raised multiple concerns about potential barriers to the usefulness of this approach and suggested several possible facilitators. Future efforts in this area will benefit from incorporating systematic qualitative feedback from providers, patients, administrators, and payers on the use of these new approaches in routine care, especially given the complex, sensitive, and unfortunately still stigmatized nature of suicide risk.