Accuracy of Clinician Predictions of Future Self-Harm: A Systematic Review and Meta-Analysis of Predictive Studies

Suicide Crisis Syndrome: A systematic review

BACKGROUND

The objective of this systematic review is to describe the scientific evidence for the Suicide Crisis Syndrome (SCS), a presuicidal cognitive and affective state consisting of five symptomatic dimensions: entrapment, affective disturbance, loss of cognitive control, hyperarousal, and social withdrawal. The aim of this article is to summarize the emerging literature on the SCS and to assess the extent to which a uniform syndrome can be assumed.

METHODS

A systematic literature search was conducted in three different databases (PubMed, PsycInfo, and Google Scholar) using the search terms "Suicide Crisis Inventory," "Suicide Crisis Syndrome," "Narrative Crisis Model of Suicide," and "Suicide Trigger State."

RESULTS

In total, 37 articles from 2010 to 2022 were identified by search criteria. Twenty-one articles published between 2017 and 2022 were included in the systematic review. All but three studies were conducted in the United States and examined clinical samples of adult high-risk psychiatric in- and outpatients. Sample sizes ranged from N = 170 to 4846. The findings confirm the unidimensional structure of the proposed disorder and support the predictive validity for short-term suicidal behavior above and beyond suicidal ideation.

CONCLUSION

Despite the promising predictive validity of the SCS, a precise prediction of future suicidal behavior remains difficult.

Developing a risk prediction model for death at first suicide attempt-Identifying risk factors from Thailand's national suicide surveillance system data

More than 60% of suicides globally are estimated to take place in low- and middle-income nations. Prior research on suicide has indicated that over 50% of those who die by suicide do so on their first attempt. Nevertheless, there is a dearth of knowledge on the attributes of individuals who die on their first attempt and the factors that can predict mortality on the first attempt in these regions. The objective of this study was to create an individual-level risk-prediction model for mortality on the first suicide attempt. We analyzed records of individuals' first suicide attempts that occurred between May 1, 2017, and April 30, 2018, from the national suicide surveillance system, which includes all of the provinces of Thailand. Subsequently, a risk-prediction model for mortality on the first suicide attempt was constructed utilizing multivariable logistic regression and presented through a web-based application. The model's performance was assessed by calculating the area under the receiver operating curve (AUC), as well as measuring its sensitivity, specificity, and accuracy. Out of the 3,324 individuals who made their first suicide attempt, 50.5% of them died as a result of that effort. Nine out of the 21 potential predictors demonstrated the greatest predictive capability. These included male sex, age over 50 years old, unemployment, having a depressive disorder, having a psychotic illness, experiencing interpersonal problems such as being aggressively criticized or desiring plentiful attention, having suicidal intent, and displaying suicidal warning signals. The model demonstrated a good predictive capability, with an AUC of 0.902, a sensitivity of 84.65%, a specificity of 82.66%, and an accuracy of 83.63%. The implementation of this predictive model can assist physicians in conducting comprehensive evaluations of suicide risk in clinical settings and devising treatment plans for preventive intervention.

Developing a clinical decision support system software prototype that assists in the management of patients with self-harm in the emergency department: protocol of the PERMANENS project

BACKGROUND

Self-harm presents a significant public health challenge. Emergency departments (EDs) are crucial healthcare settings in managing self-harm, but clinician uncertainty in risk assessment may contribute to ineffective care. Clinical Decision Support Systems (CDSSs) show promise in enhancing care processes, but their effective implementation in self-harm management remains unexplored.

METHODS

PERMANENS comprises a combination of methodologies and study designs aimed at developing a CDSS prototype that assists clinicians in the personalized assessment and management of ED patients presenting with self-harm. Ensemble prediction models will be constructed by applying machine learning techniques on electronic registry data from four sites, i.e., Catalonia (Spain), Ireland, Norway, and Sweden. These models will predict key adverse outcomes including self-harm repetition, suicide, premature death, and lack of post-discharge care. Available registry data include routinely collected electronic health record data, mortality data, and administrative data, and will be harmonized using the OMOP Common Data Model, ensuring consistency in terminologies, vocabularies and coding schemes. A clinical knowledge base of effective suicide prevention interventions will be developed rooted in a systematic review of clinical practice guidelines, including quality assessment of guidelines using the AGREE II tool. The CDSS software prototype will include a backend that integrates the prediction models and the clinical knowledge base to enable accurate patient risk stratification and subsequent intervention allocation. The CDSS frontend will enable personalized risk assessment and will provide tailored treatment plans, following a tiered evidence-based approach. Implementation research will ensure the CDSS' practical functionality and feasibility, and will include periodic meetings with user-advisory groups, mixed-methods research to identify currently unmet needs in self-harm risk assessment, and small-scale usability testing of the CDSS prototype software.

DISCUSSION

Through the development of the proposed CDSS software prototype, PERMANENS aims to standardize care, enhance clinician confidence, improve patient satisfaction, and increase treatment compliance. The routine integration of CDSS for self-harm risk assessment within healthcare systems holds significant potential in effectively reducing suicide mortality rates by facilitating personalized and timely delivery of effective interventions on a large scale for individuals at risk of suicide.

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.

Using Electronic Health Records to Facilitate Precision Psychiatry

The use of clinical prediction models to produce individualized risk estimates can facilitate the implementation of precision psychiatry. As a source of data from large, clinically representative patient samples, electronic health records (EHRs) provide a platform to develop and validate clinical prediction models, as well as potentially implement them in routine clinical care. The current review describes promising use cases for the application of precision psychiatry to EHR data and considers their performance in terms of discrimination (ability to separate individuals with and without the outcome) and calibration (extent to which predicted risk estimates correspond to observed outcomes), as well as their potential clinical utility (weighing benefits and costs associated with the model compared to different approaches across different assumptions of the number needed to test). We review 4 externally validated clinical prediction models designed to predict psychosis onset, psychotic relapse, cardiometabolic morbidity, and suicide risk. We then discuss the prospects for clinically implementing these models and the potential added value of integrating data from evidence syntheses, standardized psychometric assessments, and biological data into EHRs. Clinical prediction models can utilize routinely collected EHR data in an innovative way, representing a unique opportunity to inform real-world clinical decision making. Combining data from other sources (e.g., meta-analyses) or enhancing EHR data with information from research studies (clinical and biomarker data) may enhance our abilities to improve the performance of clinical prediction models.

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.

Implicit Identification with Death, Clinician Evaluation and Suicide Ideation among Adolescent Psychiatric Outpatients-The Mediating Role of Depression

Implicit identification with death (i.e., subconsciously self-associating oneself with death), measured by the Death-Suicide Implicit Association Test (D/S-IAT), is associated with Suicide Ideation (SI). Our understanding of the mechanisms underlying this association is limited. The current study examined (1) the mediating role of depression between D/S-IAT and recent SI and (2) the association between SI, D/S-IAT, and clinician evaluation of SI among a clinical sample of adolescents. 148 adolescents aged 10-18 years (69.4% female) from two outpatient clinics were assessed at intake. Participants completed D/S-IAT and self-report measures for recent SI and depression during intake. Findings indicate that depression is a mediator between D/S-IAT and recent SI, controlling for gender, site differences, and past suicidal thoughts and behaviors. D/S-IAT and clinician evaluation were correlated with recent SI but not beyond depression. Our findings highlight the importance of examining the underlying psychological mechanisms regarding the association between D/S-IAT and suicide.

EMPOWERED trial: protocol for a randomised control trial of digitally supported, highly personalised and measurement-based care to improve functional outcomes in young people with mood disorders

OBJECTIVES

Many adolescents and young adults with emerging mood disorders do not achieve substantial improvements in education, employment, or social function after receiving standard youth mental health care. We have developed a new model of care referred to as 'highly personalised and measurement-based care' (HP&MBC). HP&MBC involves repeated assessment of multidimensional domains of morbidity to enable continuous and personalised clinical decision-making. Although measurement-based care is common in medical disease management, it is not a standard practice in mental health. This clinical effectiveness trial tests whether HP&MBC, supported by continuous digital feedback, delivers better functional improvements than standard care and digital support.

METHOD AND ANALYSIS

This controlled implementation trial is a PROBE study (Prospective, Randomised, Open, Blinded End-point) that comprises a multisite 24-month, assessor-blinded, follow-up study of 1500 individuals aged 15-25 years who present for mental health treatment. Eligible participants will be individually randomised (1:1) to 12 months of HP&MBC or standardised clinical care. The primary outcome measure is social and occupational functioning 12 months after trial entry, assessed by the Social and Occupational Functioning Assessment Scale. Clinical and social outcomes for all participants will be monitored for a further 12 months after cessation of active care.

ETHICS AND DISSEMINATION

This clinical trial has been reviewed and approved by the Human Research Ethics Committee of the Sydney Local Health District (HREC Approval Number: X22-0042 & 2022/ETH00725, Protocol ID: BMC-YMH-003-2018, protocol version: V.3, 03/08/2022). Research findings will be disseminated through peer-reviewed journals, presentations at scientific conferences, and to user and advocacy groups. Participant data will be deidentified.

TRIAL REGISTRATION NUMBER

ACTRN12622000882729.

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.

Assessment of machine learning algorithms in national data to classify the risk of self-harm among young adults in hospital: A retrospective study

BACKGROUND

Self-harm is one of the most common presentations at accident and emergency departments in the UK and is a strong predictor of suicide risk. The UK Government has prioritised identifying risk factors and developing preventative strategies for self-harm. Machine learning offers a potential method to identify complex patterns with predictive value for the risk of self-harm.

METHODS

National data in the UK Mental Health Services Data Set were isolated for patients aged 18-30 years who started a mental health hospital admission between Aug 1, 2020 and Aug 1, 2021, and had been discharged by Jan 1, 2022. Data were obtained on age group, gender, ethnicity, employment status, marital status, accommodation status and source of admission to hospital and used to construct seven machine learning models that were used individually and as an ensemble to predict hospital stays that would be associated with a risk of self-harm.

OUTCOMES

The training dataset included 23 808 items (including 1081 episodes of self-harm) and the testing dataset 5951 items (including 270 episodes of self-harm). The best performing algorithms were the random forest model (AUC-ROC 0.70, 95%CI:0.66-0.74) and the ensemble model (AUC-ROC 0.77 95%CI:0.75-0.79).

INTERPRETATION

Machine learning algorithms could predict hospital stays with a high risk of self-harm based on readily available data that are routinely collected by health providers and recorded in the Mental Health Services Data Set. The findings should be validated externally with other real-world, prospective data.

FUNDING

This study was supported by the Midlands and Lancashire Commissioning Support Unit.

Validation of the German Version of the P4 Suicidality Tool

For general practitioners (GPs), it may be challenging to assess suicidal ideation (SI) in patients. Although promising instruments exist for the use in primary care, only a few have been validated in German. The objectives of this study were to examine the validity of the brief P4 screener for assessing SI in a cross-sectional study including outpatients. Inclusion criteria were a PHQ-9 score ≥ 10 or an affirmative answer to its SI item. Construct validity of the P4 was examined by comparison with the four-item Suicide Behaviors Questionnaire-Revised (SBQ-R), the PHQ-9 (convergent), and the positive mental health (PMH) scale (divergent). The study sample included 223 patients (mean age 47.61 ± 15 years; 61.9% women) from 20 primary care practices (104 patients) and 10 psychiatric/psychotherapeutic clinics (119 patients). The first three items of the P4 correlate positively with most of the four items of the reference standard SBQ-R (convergent validity); the fourth item of the P4 (preventive factors) correlates significantly with the PMH scale. The most common preventive factor (67%) is family or friends. The German P4 screener can be used to assess SI in outpatient care. It explores preventive or protective factors of suicide, which may support the GP's decision on treatment. We recommend a further clinical interview for patients flagged by P4 assessment in order to more formally assess suicidal risk.

Implicit cognitions on self-injurious and suicidal behavior in borderline personality disorder

BACKGROUND AND OBJECTIVES

Performance on implicit measures of suicidality has been associated with suicidal and nonsuicidal self-injury. Despite the high prevalence of self-harm in patients with borderline personality disorder (BPD), no previous study has assessed implicit measures in this patient group.

METHODS

Forty patients with BPD and 25 healthy controls completed three implicit association tests (IATs) (Death words - Me/Others words, Self-Harm pictures - Me/Others, and Self-Harm pictures - Good/Bad words) and a subliminal priming task (effect of the primes "dying"/"growing" on the categorization speed of positive/negative adjectives) as well as measures of psychopathology (suicidal ideation, previous nonsuicidal self-injury, borderline symptomatology, depression, and hopelessness).

RESULTS

Patients with BPD had higher scores on all three IATs than healthy controls. The subliminal priming procedure did not reveal group differences. Correlations between implicit measures and psychopathology among patients with BPD were mostly weak and nonsignificant with a few exceptions: Positive correlations were observed between IAT Self-Harm - Good/Bad and lifetime frequency of nonsuicidal self-injury, between IAT Self-Harm - Me/Others and depression, and between IAT Death - Me/Others and depression. Correlations between implicit measures were weak to moderate.

LIMITATIONS

The study was cross-sectional only, and the study had reduced power as the sample size was limited.

CONCLUSIONS

As expected, patients with BPD had higher scores than healthy controls on the IATs, which indicates higher implicit self-identification with self-harm and death as well as stronger implicit positive attitudes towards self-harm. The mostly weak correlations between implicit and explicit measures speak against the discriminative value of IATs in patients with BPD.

Suicide risk assessments: Why are we still relying on these a decade after the evidence showed they perform poorly?

Suicide deaths have a profound impact on whānau and community and are a tragic loss. However, from a statistical point of view, suicide is a relatively rare event. Predicting rare events is difficult, and the implications for suicide prevention were highlighted in an important editorial in this journal more than a decade ago, yet little seems to have changed. Risk assessment that focuses on accurate prediction of suicide in real-world contexts is given a great deal of attention in mental health services, despite the fact that current scientific knowledge and best practice guidelines in this area highlight that it is unlikely to be a good basis on which to provide access to treatment. It is our view that we have a good enough understanding of the common conditions people who struggle with suicidal distress experience and energy is better directed at acting to reduce exposure to these conditions and providing treatment for those who seek it. Blueprints for successful suicide prevention exist. If we lessen the focus on prediction, we will have greater resources to focus on treatment and prevention.

Suicide and Self-Harm Risk Assessment: A Systematic Review of Prospective Research

OBJECTIVE

Suicide and self-harm are widespread yet underreported. Risk assessment is key to effective self-harm and suicide prevention and management. There is contradicting evidence regarding the effectiveness of risk assessment tools in predicting self-harm and suicide risk. This systematic review examines the effect of risk assessment strategies on predicting suicide and self-harm outcomes among adult healthcare service users.

METHOD

Electronic and gray literature databases were searched for prospective research. Studies were screened and selected by independent reviewers. Quality and level of evidence assessments were conducted. Due to study heterogeneity, we present a narrative synthesis under three categories: (1) suicide- and self-harm-related outcomes; (2) clinician assessment of suicide and self-harm risk; and (3) healthcare utilization due to self-harm or suicide.

RESULTS

Twenty-one studies were included in this review. The SAD PERSONS Scale was the most used tool. It outperformed the Beck Scale for Suicide Ideation in predicting hospital admissions and stay following suicide and self-harm, yet it failed to predict repeat suicide and self-harm and was not recommended for routine use. There were mixed findings relating to clinician risk assessment, with some studies recommending clinician assessment over structured tools, whilst others found that clinician assessment failed to predict future attempts and deaths.

CONCLUSIONS

There is insufficient evidence to support the use of any one tool, inclusive of clinician assessment of risk, for self-harm and suicidality. The discourse around risk assessment needs to move toward a broader discussion on the safety of patients who are at risk for self-harm and/or suicide.HIGHLIGHTSThere is insufficient evidence to support using standalone risk assessment tools.There are mixed findings relating to clinician assessment of risk.Structured professional judgment is widely accepted for risk assessment.

Suicidal behavior-advances in clinical and neurobiological research and improvement of prevention strategies

Suicide is the 14^th leading cause of death worldwide. It is responsible for 1%-5% of all mortality. This article highlights the latest developments in universal, selective, and indicated prevention strategies. Concerning universal suicide prevention, current research has shown that strategies such as restricting access to lethal means (e.g., control of analgesics and hot-spots for suicide by jumping) and school-based awareness programs are most efficacious. Regarding selective prevention, substantial progress can be expected in psychological screening methods for suicidal behavior. The measurement of implicit cognition proved to be more valid in predicting future suicide attempts than classic clinical assessment. Latest developments are smartphone-based interventions and real-time monitoring of suicidal behavior. Great effort has been made to establish valid neurobiological screening methods (e.g., genetic and epigenetic risk factors for suicide, hypothalamic-pituitary-adrenal axis) without yielding a major bre-akthrough. Potentially, multiple biomarkers rather than a single one are necessary to identify individuals at risk. With regard to indicated prevention in form of psychopharmacological treatment, recent pharmacoepidemiological studies and meta-analyses have supported a protective role of antidepressants, lithium, and clozapine. However, the data concerning a specific anti-suicidal effect of these drugs are currently not consistent. Promising results exist for ketamine in reducing suicidal ideation, independently of its antidepressant effect. Concerning psychotherapy, recent findings suggest that psychotherapeutic interventions specifically designed to prevent suicide re-attempts are most efficacious. Specifically, cognitive behavioral therapy and psychodynamic therapy approaches proved to decrease the number of suicide re-attempts significantly.

Resting-state functional connectivity of the amygdala in major depressive disorder with suicidal ideation

Suicide is a common issue among major depressive disorder (MDD) patients and suicidal ideation (SI) is the first step toward it. There are no definitive objective biomarkers of SI relative to MDD. In this study, a seed-based correlation analysis was performed among 36 MDD patients with SI, 66 MDD patients without SI (NSI), and 57 healthy controls (HCs) using amygdala resting-state functional connectivity (RSFC). Furthermore, the correlation between amygdala RSFC and clinical features was examined in the SI group. When compared to the NSI group, SI group exhibited increased RSFC between the left amygdala seed and left medial superior frontal gyrus (SFGmed) as well as left middle frontal gyrus (MFG). In turn, a decreased RSFC was observed between the left amygdala seed and the following brain regions including the left inferior parietal lobule (IPL), right precentral gyrus (PrCG), and left superior parietal lobule (SPL) in SI group compared to NSI group. Moreover, the SI group exhibited increased RSFC of the right amygdala with left middle temporal gyrus (MTG); In addition, the RSFC of the left amygdala with left MFG was negatively associated with learning and memory (VSM), speed of processing (SOP). The RSFC of the amygdala is distinct between MDD patients with SI and without SI. Our findings reveal the neurobiological characteristics of MDD with respect to SI and provide new clues regarding vulnerability to mental illness. It is necessary to carry out repeated and more longitudinal researches using multimodal approaches on SI in the future.

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.

Identifying long-term and imminent suicide predictors in a general population and a clinical sample with machine learning

BACKGROUND

Machine learning (ML) is increasingly used to predict suicide deaths but their value for suicide prevention has not been established. Our first objective was to identify risk and protective factors in a general population. Our second objective was to identify factors indicating imminent suicide risk.

METHODS

We used survival and ML models to identify lifetime predictors using the Cohort of Norway (n=173,275) and hospital diagnoses in a Saskatoon clinical sample (n=12,614). The mean follow-up times were 17 years and 3 years for the Cohort of Norway and Saskatoon respectively. People in the clinical sample had a longitudinal record of hospital visits grouped in six-month intervals. We developed models in a training set and these models predicted survival probabilities in held-out test data.

RESULTS

In the general population, we found that a higher proportion of low-income residents in a county, mood symptoms, and daily smoking increased the risk of dying from suicide in both genders. In the clinical sample, the only predictors identified were male gender and older age.

CONCLUSION

Suicide prevention probably requires individual actions with governmental incentives. The prediction of imminent suicide remains highly challenging, but machine learning can identify early prevention targets.

Psychotherapeutic interventions for the prevention of suicide re-attempts: a systematic review

A history of suicide attempt (SA) is a strong predictor of future suicide re-attempts or suicide. The aim of this systematic review is to evaluate the efficacy of psychotherapeutic interventions specifically designed for the prevention of suicide re-attempts. A systematic search from 1980 to June 2020 was performed via the databases PubMed and Google Scholar. Only randomized controlled trials were included which clearly differentiated suicidal self-harm from non-suicidal self-injury in terms of intent to die. Moreover, psychotherapeutic interventions had to be focused on suicidal behaviour and the numbers of suicide re-attempts had to be used as outcome variables. By this procedure, 18 studies were identified. Statistical comparison of all studies revealed that psychotherapeutic interventions in general were significantly more efficacious than control conditions in reducing the risk of future suicidal behaviour nearly by a third. Separate analyses revealed that cognitive-behavioural therapy as well as two different psychodynamic approaches were significantly more efficacious than control conditions. Dialectical behaviour therapy and elementary problem-solving therapy were not superior to control conditions in reducing the number of SAs. However, methodological reasons may explain to some extent these negative results. Considering the great significance of suicidal behaviour, there is unquestionably an urgent need for further development of psychotherapeutic techniques for the prevention of suicide re-attempts. Based on the encouraging results of this systematic review, it can be assumed that laying the focus on suicidal episodes might be the key intervention for preventing suicide re-attempts and suicides.

Do implicit measures improve suicide risk prediction? An 18-month prospective study using different tasks

BACKGROUND

There is accumulating evidence that implicit measures improve the prediction of suicidality within a 6-month follow-up period in psychiatric populations. Building upon these results, we set out to expand the follow-up period and to investigate various implicit methods.

METHODS

Seventy-nine inpatients completed the Beck Scale for Suicidal Ideation (BSS) and a range of implicit measures: three implicit association tests (IATs: Death; Self-harm-Me/Others; Self-Harm-Good/Bad) and a subliminal priming task (with separate scores for negative and positive adjectives, each indicating the association between the primes "dying" and "growing"). After 18 months, we reached n = 52 patients and reassessed suicidal ideation, plans, and attempts.

RESULTS

In a hierarchical regression, the five implicit task indices were entered after the patient's age, gender, and BSS score at baseline. The implicit scores improved prediction of BSS scores after 18 months compared to prediction based on age, gender, and BSS score at baseline alone. However, none of the implicit measures was associated with suicide plans or attempts during the follow-up period.

CONCLUSION

Results suggest that implicit measures can be a useful assessment tool for the prediction of suicidal ideation, even beyond the BSS. However, long-term prediction of suicide plans or attempts using implicit measures seems limited.

Prediction and network modelling of self-harm through daily self-report and history of self-injury

BACKGROUND

Self-harm is a significant public health issue, and both our understanding and ability to predict adverse outcomes are currently inadequate. The current study explores how preventative efforts could be aided through short-term prediction and modelling of risk factors for self-harm.

METHODS

Patients (72% female, M_age = 40.3 years) within an inpatient psychiatric facility self-reported their psychological distress, interpersonal circumstances, and wish to live and die on a daily basis during 3690 unique admissions. Hierarchical logistic regressions assessed whether daily changes in self-report and history of self-harm could predict self-harm, with machine learning used to train and test the model. To assess interrelationships between predictors, network and cross-lagged panel models were performed.

RESULTS

Increases in a wish to die (β = 1.34) and psychological distress (β = 1.07) on a daily basis were associated with increased rates of self-harm, while a wish to die on the day prior [odds ratio (OR) 3.02] and a history of self-harm (OR 3.02) was also associated with self-harm. The model detected 77.7% of self-harm incidents (positive predictive value = 26.6%, specificity = 79.1%). Psychological distress, wish to live and die, and interpersonal factors were reciprocally related over the prior day.

CONCLUSIONS

Short-term fluctuations in self-reported mental health may provide an indication of when an individual is at-risk of self-harm. Routine monitoring may provide useful feedback to clinical staff to reduce risk of self-harm. Modifiable risk factors identified in the current study may be targeted during interventions to minimise risk of self-harm.

Development and validation of the Durham Risk Score for estimating suicide attempt risk: A prospective cohort analysis

BACKGROUND

Worldwide, nearly 800,000 individuals die by suicide each year; however, longitudinal prediction of suicide attempts remains a major challenge within the field of psychiatry. The objective of the present research was to develop and evaluate an evidence-based suicide attempt risk checklist [i.e., the Durham Risk Score (DRS)] to aid clinicians in the identification of individuals at risk for attempting suicide in the future.

METHODS AND FINDINGS

Three prospective cohort studies, including a population-based study from the United States [i.e., the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) study] as well as 2 smaller US veteran cohorts [i.e., the Assessing and Reducing Post-Deployment Violence Risk (REHAB) and the Veterans After-Discharge Longitudinal Registry (VALOR) studies], were used to develop and validate the DRS. From a total sample size of 35,654 participants, 17,630 participants were selected to develop the checklist, whereas the remaining participants (N = 18,024) were used to validate it. The main outcome measure was future suicide attempts (i.e., actual suicide attempts that occurred after the baseline assessment during the 1- to 3-year follow-up period). Measure development began with a review of the extant literature to identify potential variables that had substantial empirical support as longitudinal predictors of suicide attempts and deaths. Next, receiver operating characteristic (ROC) curve analysis was utilized to identify variables from the literature review that uniquely contributed to the longitudinal prediction of suicide attempts in the development cohorts. We observed that the DRS was a robust prospective predictor of future suicide attempts in both the combined development (area under the curve [AUC] = 0.91) and validation (AUC = 0.92) cohorts. A concentration of risk analysis found that across all 35,654 participants, 82% of prospective suicide attempts occurred among individuals in the top 15% of DRS scores, whereas 27% occurred in the top 1%. The DRS also performed well among important subgroups, including women (AUC = 0.91), men (AUC = 0.93), Black (AUC = 0.92), White (AUC = 0.93), Hispanic (AUC = 0.89), veterans (AUC = 0.91), lower-income individuals (AUC = 0.90), younger adults (AUC = 0.88), and lesbian, gay, bisexual, transgender, and queer or questioning (LGBTQ) individuals (AUC = 0.88). The primary limitation of the present study was its its reliance on secondary data analyses to develop and validate the risk score.

CONCLUSIONS

In this study, we observed that the DRS was a strong predictor of future suicide attempts in both the combined development (AUC = 0.91) and validation (AUC = 0.92) cohorts. It also demonstrated good utility in many important subgroups, including women, men, Black, White, Hispanic, veterans, lower-income individuals, younger adults, and LGBTQ individuals. We further observed that 82% of prospective suicide attempts occurred among individuals in the top 15% of DRS scores, whereas 27% occurred in the top 1%. Taken together, these findings suggest that the DRS represents a significant advancement in suicide risk prediction over traditional clinical assessment approaches. While more work is needed to independently validate the DRS in prospective studies and to identify the optimal methods to assess the constructs used to calculate the score, our findings suggest that the DRS is a promising new tool that has the potential to significantly enhance clinicians' ability to identify individuals at risk for attempting suicide in the future.

Investigating the relationship between childhood sexual abuse, self-harm repetition and suicidal intent: mixed-methods study

BACKGROUND

Research into the association between childhood sexual abuse (CSA) and self-harm repetition is limited.

AIMS

We aimed to examine the association between self-harm repetition, mental health conditions, suicidal intent and CSA experiences among people who frequently self-harm.

METHOD

A mixed-methods study was conducted including consecutive patients aged ≥18 years, with five or more self-harm presentations, in three Irish hospitals. Information was extracted from psychiatric records and patients were invited to participate in a semi-structured interview. Data was collected and analysed with a mixed-methods, convergent parallel design. In tandem, the association between CSA and self-harm repetition, suicidal intent and mental health conditions was examined with logistic regression models and independent sample t-test, with psychiatric records data. Thematic analysis was conducted with interview data, to explore CSA experiences and self-harm repetition.

RESULTS

Between March 2016 and July 2019, information was obtained on 188 consecutive participants, with 36 participants completing an interview. CSA was recorded in 42% of the total sample and 72.2% of those interviewed. CSA was positively associated with self-harm repetition (odds ratio 6.26, 95% CI 3.94-9.94, P = 0.00). Three themes emerged when exploring participants' CSA experiences: CSA as a precipitating factor for self-harm, secrecy of CSA accentuating shame, and loss experiences linked to CSA and self-harm.

CONCLUSIONS

CSA was frequently reported among people who frequently self-harm, and associated with self-harm repetition. Identification of patients at risk of repetition is key for suicide prevention. This is an at-risk group with particular characteristics that must be considered; comprehensive patient histories can help inform and tailor treatment pathways.

Skala Suizidales Erleben und Verhalten (SSEV)

Zusammenfassung. Suizidales Erleben und Verhalten ist in klinischen Kontexten sehr verbreitet. Während international diverse Messinstrumente zur Erfassung von Suizidalität entwickelt und validiert wurden, gibt es nur wenige deutsche Messinstrumente. In der vorliegenden Studie wurde die neu entwickelte Skala Suizidales Erleben und Verhalten (SSEV) in fünf Stichproben mit insgesamt N = 1 099 Proband_innen im Hinblick auf ihre psychometrischen Eigenschaften untersucht. Die faktorenanalytische Untersuchung (explorative und konfirmatorische Faktorenanalyse) ergab eine eindimensionale Struktur des Fragebogens. Die interne Konsistenz der SSEV ist hoch und es zeigten sich erwartungsgemäß positive Zusammenhänge zu diversen Maßen aktueller Theoriemodelle suizidalen Erlebens und Verhaltens, sowie zu Depressivität, Angst und Stress. Weitere Analysen zeigten erwartungskonform negative Zusammenhänge mit sozialer Unterstützung und positiver mentaler Gesundheit. Insgesamt verweisen die Ergebnisse darauf, dass es sich beim SSEV um ein reliables und valides Instrument zur Erfassung von akutem suizidalem Erleben und Verhalten handelt, welches in der Forschung und der klinischen Praxis angewendet werden kann.

Higher Suicide Intent in Patients Attempting Suicide With Violent Methods Versus Self-Poisoning

Background: Suicidal intent for patients attempting suicide using violent methods (VMs) is assumed to be higher than for those using self-poisoning (SP), which may explain the higher mortality observed in follow-up studies. However, this has not been studied prospectively. Aims: We aimed to compare patients attempting suicide using VMs with those using SP regarding suicidal intent, suicidal ideation, depression, and hopelessness during hospital stay and after 1 year. Methods: Patients hospitalized after suicide attempt by VMs (n = 80) or SP (n = 81) completed the Beck scales for Suicide Intent, Suicide Ideation, Depression Inventory, and Hopelessness on admission and at the 12-month follow-up. Results: On admission, those using VMs had higher suicidal intent than those using SP (M = 16.2 vs. 13.3, p < .001), but lower depression scores (M = 22.2 vs. 26.8, p < .05). No significant differences were found in suicidal ideation (M = 20.1 vs. 23.1) or hopelessness (M = 10.1 vs. 11.9). At 12-month follow-up, depression scores decreased significantly for both groups, while hopelessness decreased only for the SP group. Limitations: The statistical power achieved was lower than intended. Conclusion: The higher levels of suicidal intent, but lower levels of depression, may indicate more impulsivity among people attempting suicide using VMs. Suicidal ideation was relatively stable.

Functional network alterations differently associated with suicidal ideas and acts in depressed patients: an indirect support to the transition model

The transition from suicidal ideas to a suicide act is an important topic of research for the identification of those patients at risk of acting out. We investigated here whether specific brain activity and connectivity measures at rest may be differently associated with suicidal thoughts and behaviors. A large sample of acutely depressed patients with major depressive disorder was recruited in three different centers (Montreal/Canada, Stanford/USA, and Jena/Germany), covering four different phenotypes: patients with a past history of suicide attempt (n = 53), patients with current suicidal ideas but no past history of suicide attempt (n = 40), patients without current suicidal ideation nor past suicide attempts (n = 42), and healthy comparison subjects (n = 107). 3-T resting-state functional magnetic resonance imaging (fMRI) measures of the amplitude of low-frequency fluctuation (ALFF) and degree centrality (DC) were obtained and examined in a whole-brain data-driven analysis. Past suicide attempt was associated with a double cortico-subcortical dissociation in ALFF values. Decreased ALFF and DC values mainly in a frontoparietal network and increased ALFF values in some subcortical regions (hippocampus and thalamus) distinguished suicide attempters from suicide ideators, patient controls, and healthy controls. No clear neural differences were identified in relation to suicidal ideas. Suicide attempters appear to be a distinct subgroup of patients with widespread brain alterations in functional activity and connectivity that could represent factors of vulnerability. Our results also indirectly support at the neurobiological level the relevance of the transition model described at the psychological and clinical levels. The brain bases of suicidal ideas occurrence in depressed individuals needs further investigations.

Identifying Distinguishable Clinical Profiles Between Single Suicide Attempters and Re-Attempters

More than 800,000 individuals die from suicide each year in the world, which has a devastating impact on families and society. Ten to twenty times more attempt suicide. Previous studies showed that suicide attempters represent a heterogeneous group regarding demographic characteristics, individual characteristics of a suicidal attempt, and the assumed clinical factors, e.g., hopelessness or impulsivity, thus differently contributing to the likelihood of suicidal behavior. Therefore, in the present study, we aim to give a comprehensive clinical description of patients with repeated suicide attempts compared to single attempters. We explored putative differences between groups in clinical variables and personality traits, sociodemographic information, and specific suicide attempt-related information. A sample of patients with a recent suicide attempt (n = 252), defined according to DSM-5 criteria for a suicidal behavior disorder (SBD), was recruited in four psychiatric hospitals in Thuringia, Germany. We used a structured clinical interview to assess the psychiatric diagnosis, sociodemographic data, and to collect information regarding the characteristics of the suicide attempt. Several clinical questionnaires were used to measure the suicide intent and suicidal ideations, depression severity, hopelessness, impulsivity, aggression, anger expression, and the presence of childhood trauma. Univariate and multivariate statistical methods were applied to evaluate the postulated risk factors and, to distinguish groups based on these measures. The performed statistical analyses indicated that suicide attempters represent a relatively heterogeneous group, nevertheless associated with specific clinical profiles. We demonstrated that the re-attempters had more severe psychopathology with significantly higher levels of self-reported depression, suicidal ideation as well as hopelessness. Furthermore, re-attempters had more often first-degree relatives with suicidal behavior and emotional abuse during childhood. They also exhibited a higher degree of specific personality traits, i.e., more "urgency" as a reaction to negative emotions, higher excitability, higher self-aggressiveness, and trait anger. The multivariate discriminant analysis significantly discriminated the re-attempters from single attempters by higher levels of self-aggressiveness and suicidal ideation. The findings might contribute to a better understanding of the complex mechanisms leading to suicidal behavior, which might improve the early identification and specific treatment of subjects at risk for repeated suicidal behavior.

Clinicians' Perspectives on Self-Harm in Pakistan: A Qualitative Study

Background: Suicide is a serious public health problem, ranked amongst the leading causes of death worldwide. There are no official data on self-harm and suicide in Pakistan; both are illegal acts, and are socially and religiously condemned. This study explored the views of clinicians, including general practitioners (GPs) and hospital physicians (HPs) on self-harm, about their management of people who self-harm and what interventions might be appropriate in Pakistan. Methods: This qualitative study, generating data using semi-structured interviews, was nested within a Randomized Controlled Trial (RCT) of a psychosocial intervention for people following self-harm. Clinicians (n = 18) with experience of treating people who self-harm were recruited from public hospitals and general practices. Results: Face-to-face interviews were conducted in Urdu and digitally recorded with consent, transcribed and translated into English. Transcripts were checked for cultural and interpretive interpretations by the research team, then analyzed thematically using the principles of constant comparison. The following themes will be presented: encountering people with self-harming behaviors; challenges encountered in managing people who self-harm; barriers to accessing care, and what ideal care might look like. Participants identified their lack of training and expertise in the management of people with self-harm behavior. Conclusions: This is the first study to explore clinicians' perspectives on self-harm in Pakistan. The study highlighted the need for training for doctors in the identification and management of mental health problems, including the management of people who self-harm.

Predicting self-harm within six months after initial presentation to youth mental health services: A machine learning study

BACKGROUND

A priority for health services is to reduce self-harm in young people. Predicting self-harm is challenging due to their rarity and complexity, however this does not preclude the utility of prediction models to improve decision-making regarding a service response in terms of more detailed assessments and/or intervention. The aim of this study was to predict self-harm within six-months after initial presentation.

METHOD

The study included 1962 young people (12-30 years) presenting to youth mental health services in Australia. Six machine learning algorithms were trained and tested with ten repeats of ten-fold cross-validation. The net benefit of these models were evaluated using decision curve analysis.

RESULTS

Out of 1962 young people, 320 (16%) engaged in self-harm in the six months after first assessment and 1642 (84%) did not. The top 25% of young people as ranked by mean predicted probability accounted for 51.6% - 56.2% of all who engaged in self-harm. By the top 50%, this increased to 82.1%-84.4%. Models demonstrated fair overall prediction (AUROCs; 0.744-0.755) and calibration which indicates that predicted probabilities were close to the true probabilities (brier scores; 0.185-0.196). The net benefit of these models were positive and superior to the 'treat everyone' strategy. The strongest predictors were (in ranked order); a history of self-harm, age, social and occupational functioning, sex, bipolar disorder, psychosis-like experiences, treatment with antipsychotics, and a history of suicide ideation.

CONCLUSION

Prediction models for self-harm may have utility to identify a large sub population who would benefit from further assessment and targeted (low intensity) interventions. Such models could enhance health service approaches to identify and reduce self-harm, a considerable source of distress, morbidity, ongoing health care utilisation and mortality.

Incidence of suicide death in patients with cancer: A systematic review and meta-analysis

BACKGROUND

Growing evidence indicated the higher risk of suicide in cancer survivors compared with general population. Our aim is to systematically quantify the extent of suicide death and identify risk factors associated with the incidence of suicide in cancer patients.

METHODS

We conducted a meta-analysis of relevant studies published in English or Chinese before May 20, 2020. Suicide rate and the number of suicide death were extracted. Our main outcome was suicide rate per 100,000 person-years with 95% CIs using random-effects model.

RESULTS

The pooled incidence of suicide death was 39.72 per 100,000 person-years (95%CI, 33.91-46.52, I ²= 99.6%, P <0 .001). The suicide rate for cancer patients was higher in men (57.78, 95%CI, 47.31-70.56) than in women (14.47, 95%CI, 11.27-18.57). For both sexes combined, esophagus cancer had the highest rate of suicide (87.71, 95%CI, 27.42-280.54). By sex, suicide rates ranked first in males and females were pancreas cancer (195.70, 95%CI, 129.55-295.61) and esophagus cancer (18.34, 95%CI, 5.92-56.84), respectively. The highest suicide rate was 61.02(95%CI, 53.66-69.40) in Asia, and Oceania (24.07, 95%CI, 20.78-27.88) had lowest suicide rate. Suicide rate had a downward trend by years after diagnosis, with the first six months after cancer diagnosis clearly standing out (89.33, 95%CI, 50.64-157.58).

LIMITATIONS

Included studies came from high-income countries and our results might not represent the suicide rate among cancer patients in low- and middle-income countries.

CONCLUSIONS

The incidence of suicide among cancer patients was high despite the declined trend recent years, which emphasized psychological health aspects of interventions and perfecting suicide prevention programs.

A multi-method psychological autopsy study on youth suicides in the Netherlands in 2017: Feasibility, main outcomes, and recommendations

OBJECTIVES

In the Netherlands, there was a sharp increase in the number of suicides among 10- to 19-year-olds in 2017. A multi-method psychological autopsy study (PA) was conducted to assess feasibility, identify related factors, and study the interplay of these factors to inform suicide prevention strategies.

METHODS

Coroners identified youth suicides in 2017 in their records and then general practitioners (GPs) contacted the parents of these youths. Over a period of 7 months, 66 qualitative interviews were held with the parents, peers, and teachers, providing information on precipitating factors and five topics involving 35 cases (17 boys and 18 girls, mean age 17 years). Furthermore, 43 parents and care professionals filled in questionnaires to examine risk and care-related factors. Qualitative and quantitative analyses were performed.

RESULTS

Although registration problems faced by coroners and resistance to contacting bereaved families by GPs hampered the recruitment, most parents highly appreciated being interviewed. Several adverse childhood experiences played a role at an individual level, such as (cyber) bullying, parental divorce, sexual abuse, as well as complex mental disorders, and previous suicide attempts. Two specific patterns stood out: (1) girls characterized by insecurity and a perfectionist attitude, who developed psychopathology and dropped out of school, and (2) boys with a developmental disorder, such as autism, who were transferred to special needs education and therefore felt rejected. In addition, adolescents with complex problems had difficulty finding appropriate formal care. Regarding potential new trends, contagion effects of social media use in a clinical setting and internet use for searching lethal methods were found.

CONCLUSION

This first national PA study showed that, as expected, a variety of mostly complex clusters of problems played a role in youth suicides. An infrastructure is needed to continuously monitor, evaluate, and support families after each youth suicide and thereby improve prevention strategies.

Suicide prediction models: a critical review of recent research with recommendations for the way forward

Suicide is a leading cause of death. A substantial proportion of the people who die by suicide come into contact with the health care system in the year before their death. This observation has resulted in the development of numerous suicide prediction tools to help target patients for preventive interventions. However, low sensitivity and low positive predictive value have led critics to argue that these tools have no clinical value. We review these tools and critiques here. We conclude that existing tools are suboptimal and that improvements, if they can be made, will require developers to work with more comprehensive predictor sets, staged screening designs, and advanced statistical analysis methods. We also conclude that although existing suicide prediction tools currently have little clinical value, and in some cases might do more harm than good, an even-handed assessment of the potential value of refined tools of this sort cannot currently be made because such an assessment would depend on evidence that currently does not exist about the effectiveness of preventive interventions. We argue that the only way to resolve this uncertainty is to link future efforts to develop or evaluate suicide prediction tools with concrete questions about specific clinical decisions aimed at reducing suicides and to evaluate the clinical value of these tools in terms of net benefit rather than sensitivity or positive predictive value. We also argue for a focus on the development of individualized treatment rules to help select the right suicide-focused treatments for the right patients at the right times. Challenges will exist in doing this because of the rarity of suicide even among patients considered high-risk, but we offer practical suggestions for how these challenges can be addressed.

Using Administrative Data to Predict Suicide After Psychiatric Hospitalization in the Veterans Health Administration System

There is a very high suicide rate in the year after psychiatric hospital discharge. Intensive postdischarge case management programs can address this problem but are not cost-effective for all patients. This issue can be addressed by developing a risk model to predict which inpatients might need such a program. We developed such a model for the 391,018 short-term psychiatric hospital admissions of US veterans in Veterans Health Administration (VHA) hospitals 2010-2013. Records were linked with the National Death Index to determine suicide within 12 months of hospital discharge (n=771). The Super Learner ensemble machine learning method was used to predict these suicides for time horizon between 1 week and 12 months after discharge in a 70% training sample. Accuracy was validated in the remaining 30% holdout sample. Predictors included VHA administrative variables and small area geocode data linked to patient home addresses. The models had AUC=.79-.82 for time horizons between 1 week and 6 months and AUC=.74 for 12 months. An analysis of operating characteristics showed that 22.4%-32.2% of patients who died by suicide would have been reached if intensive case management was provided to the 5% of patients with highest predicted suicide risk. Positive predictive value (PPV) at this higher threshold ranged from 1.2% over 12 months to 3.8% per case manager year over 1 week. Focusing on the low end of the risk spectrum, the 40% of patients classified as having lowest risk account for 0%-9.7% of suicides across time horizons. Variable importance analysis shows that 51.1% of model performance is due to psychopathological risk factors accounted, 26.2% to social determinants of health, 14.8% to prior history of suicidal behaviors, and 6.6% to physical disorders. The paper closes with a discussion of next steps in refining the model and prospects for developing a parallel precision treatment model.

Forecasting a Fatal Decision: Direct Replication of the Predictive Validity of the Suicide–Implicit Association Test

A previous study by Nock et al. (2010) suggested that people’s implicit identification with “death” or “suicide” can accurately predict whether they will attempt suicide several months in advance. We report the first direct and independent replication of this promising finding. Participants were 165 patients seeking treatment at a psychiatric unit in France. At baseline, patients completed the Suicide–Implicit Association Test (S–IAT), a semistructured interview, and a self-report measure of suicide ideation. Six months later, we contacted participants by phone and examined their hospital medical records to determine whether they had made a new suicide attempt. Results showed that the S–IAT did not distinguish between patients who were admitted to the hospital following suicide attempts and those who were admitted for other reasons. As in the original study, however, the S–IAT predicted suicide attempts within the 6-month follow-up period beyond well-known predictors. The test correctly classified 85% of patients (95% confidence interval = [76.91, 91.53]), supporting its diagnostic value for identifying who will make a suicide attempt.

Temperament and character as risk factor for suicide ideation and attempts in adults with autism spectrum disorders

Persons with autism spectrum disorders (ASDs) are suggested to have an increased risk for suicide ideation and suicide attempts, but this topic is largely understudied. Research indicates that temperament and character traits are associated with suicidal behavior in persons without ASD, with higher scores for novelty seeking (NS), harm avoidance (HA), and self-transcedence (ST), and lower scores for self-directedness (SD) and cooperativeness (CO). Usually persons with ASD have temperament and character profiles with high HA, and low NS, reward dependence (RD), SD, and CO. The aim is to investigate whether there is a relationship between temperament and character traits and suicide ideation and attempts in adults with ASD. Seventy-four adults with ASD participated by completing self-report measures on suicide thoughts and behavior, depression, and temperament. Independent sample t-tests were conducted to compare scores between attempters versus nonattempters and between ideators versus nonideators. Regression analysis was performed to explore the predictive value of temperament and character. T-tests showed lower NS and SD, and higher HA for ideators versus nonideators, but not for attempters versus nonattempters. Regression models showed no significant relation between suicide ideation and NS, SD, HA after the latter were controlled for the significant influence of depression. Temperament and character can probably not be used for predicting suicide ideation and attempts, based on results from the current sample. Clinicians must take note of the high prevalence and risk of depression among persons with ASD, which may be under-reported. Autism Res 2020, 13: 104-111. © 2019 International Society for Autism Research, Wiley Periodicals, Inc. LAY SUMMARY: The aim of the current study was to uncover risk factors for suicide ideation and attempts in adults with autism, since this urgent topic is largely understudied. We initially investigated whether temperament and character could be risk factors, but found no association. However, we did find that depression might be a high predictor for suicide ideation, which could remain under-reported in adults with autism, due to impaired communication and problems with expressing emotions and thoughts.

Machine learning methods for developing precision treatment rules with observational data

Clinical trials have identified a variety of predictor variables for use in precision treatment protocols, ranging from clinical biomarkers and symptom profiles to self-report measures of various sorts. Although such variables are informative collectively, none has proven sufficiently powerful to guide optimal treatment selection individually. This has prompted growing interest in the development of composite precision treatment rules (PTRs) that are constructed by combining information across a range of predictors. But this work has been hampered by the generally small samples in randomized clinical trials and the use of suboptimal analysis methods to analyze the resulting data. In this paper, we propose to address the sample size problem by: working with large observational electronic medical record databases rather than controlled clinical trials to develop preliminary PTRs; validating these preliminary PTRs in subsequent pragmatic trials; and using ensemble machine learning methods rather than individual algorithms to carry out statistical analyses to develop the PTRs. The major challenges in this proposed approach are that treatment are not randomly assigned in observational databases and that these databases often lack measures of key prescriptive predictors and mental disorder treatment outcomes. We proposed a tiered case-cohort design approach that uses innovative methods for measuring and balancing baseline covariates and estimating PTRs to address these challenges.

Distinguishing suicides of people reported missing from those not reported missing: retrospective Scottish cohort study

BACKGROUND

Understanding what distinguishes the suicide of individuals reported missing (missing-suicides) from those of individuals not reported missing (other-suicides) may have preventative and/or operational utility and inform our knowledge of suicide.AimsTo assess whether specific epidemiological, sociodemographic or circumstantial characteristics differ between individuals reported missing and those not reported missing who take their own life.

METHOD

Content analysis of Scottish Police Death Reports, detailing 160 suicides/undetermined deaths over a 3-year period in the North-East of Scotland.

RESULTS

Those in the missing-suicide group were more likely to be older but did not differ from the other-suicide group on any other epidemiological or sociodemographic characteristics. Individuals in the other-suicide group were more likely to be found inadvertently by people known to them. The missing-suicide group took longer to find and were more likely to be located in natural outdoor locations by police/searchers or members of the public.

CONCLUSIONS

Individuals who die by suicide and who are reported as a missing person differ from those not reported as missing in terms of factors relating to location and how they are found but not epidemiological or sociodemographic characteristics.Declaration of interestNone.