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

Quantifying the interrelationships between physical, social, and cognitive-emotional components of mental fitness using digital technology

Mental fitness is a construct that goes beyond a simple focus on subjective emotional wellbeing to encompass more broadly our ability to think, feel, and act to achieve what we want in our daily lives. The measurement and monitoring of multiple (often interacting) domains is crucial to gain a holistic and complete insight into an individual's mental fitness. We aimed to demonstrate the capability of a new mobile app to characterise the mental fitness of a general population of Australians and to quantify the interrelationships among different domains of mental fitness. Cross-sectional data were collected from 4901 adults from the general population of Australians engaged in work or education who used a mobile app (Innowell) between September 2021 and November 2022. Individuals completed a baseline questionnaire comprised of 26 questions across seven domains of mental fitness (i.e., physical activity, sleep and circadian rhythms, nutrition, substance use, daily activities, social connection, psychological distress). Network analysis was applied at both a domain-level (e.g., 7 nodes representing each cluster of items) and an individual item-level (i.e., 26 nodes representing all questionnaire items). Only 612 people (12%) were functioning well across all domains. One quarter (n = 1204, 25%) had only one problem domain and most (n = 3085, 63%) had multiple problem domains. The two most problematic domains were physical activity (n = 2631, 54%) and social connection (n = 2151, 44%), followed closely by daily activity (n = 1914, 39%). At the domain-level, the strongest association emerged between psychological distress and daily activity (r = 0.301). Psychological distress was the most central node in the network (as measured by strength and expected influence), followed closely by daily activity, sleep and circadian rhythms and then social connection. The item-level network revealed that the nodes with the highest centrality in the network were: hopelessness, depression, functional impairment, effortfulness, subjective energy, worthlessness, and social connectedness. Social connection, sleep and circadian rhythms, and daily activities may be critical targets for intervention due to their widespread associations in the overall network. While psychological distress was not among the most common problems, its centrality may indicate its importance for indicated prevention and early intervention. We showcase the capability of a new mobile app to monitor mental fitness and identify the interrelationships among multiple domains, which may help people develop more personalised insights and approaches.

Dynamic learning of individual-level suicidal ideation trajectories to enhance mental health care

There has recently been an increase in ongoing patient-report routine outcome monitoring for individuals within clinical care, which has corresponded to increased longitudinal information about an individual. However, many models that are aimed at clinical practice have difficulty fully incorporating this information. This is in part due to the difficulty in dealing with the irregularly time-spaced observations that are common in clinical data. Consequently, we built individual-level continuous-time trajectory models of suicidal ideation for a clinical population (N = 585) with data collected via a digital platform. We demonstrate how such models predict an individual's level and variability of future suicide ideation, with implications for the frequency that individuals may need to be observed. These individual-level predictions provide a more personalised understanding than other predictive methods and have implications for enhanced measurement-based care.

Examining Predictors of Psychological Distress Among Youth Engaging with Jigsaw for a Brief Intervention

Risk factors for psychological distress among help-seeking youth are poorly understood. Addressing this gap is important for informing mental health service provision. This study aimed to identify risk factors among youth attending Jigsaw, a youth mental health service in Ireland. Routine data were collected from N = 9,673 youth who engaged with Jigsaw (Mean age = 16.9 years, SD = 3.14), including presenting issues, levels of psychological distress, age, and gender. Confirmatory Factor Analysis identified thirteen factors of clustering issues. Several factors, including Self-criticism and Negative Thoughts, were strongly associated with items clustering as psychological distress, however these factors were poorly predictive of distress as measured by the CORE (YP-CORE: R2 = 14.7%, CORE-10: R2 = 6.9%). The findings provide insight into associations between young people's identified presenting issues and self-identified distress. Implications include applying appropriate therapeutic modalities to focus on risk factors and informing routine outcome measurement in integrated youth mental health services.

Predicting Lifetime Suicide Attempts in a Community Sample of Adolescents Using Machine Learning Algorithms

Suicide is a major global health concern and a prominent cause of death in adolescents. Previous research on suicide prediction has mainly focused on clinical or adult samples. To prevent suicides at an early stage, however, it is important to screen for risk factors in a community sample of adolescents. We compared the accuracy of logistic regressions, elastic net regressions, and gradient boosting machines in predicting suicide attempts by 17-year-olds in the Millennium Cohort Study (N = 7,347), combining a large set of self- and other-reported variables from different categories. Both machine learning algorithms outperformed logistic regressions and achieved similar balanced accuracies (.76 when using data 3 years before the self-reported lifetime suicide attempts and .85 when using data from the same measurement wave). We identified essential variables that should be considered when screening for suicidal behavior. Finally, we discuss the usefulness of complex machine learning models in suicide prediction.

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.

Using electronic health record data to predict future self-harm or suicidal ideation in young people treated by child and youth mental health services

INTRODUCTION

Identifying young people who are at risk of self-harm or suicidal ideation (SHoSI) is a priority for mental health clinicians. We explore the utility of routinely collected data in developing a tool to aid early identification of those at risk.

METHOD

We used electronic health records of 4610 young people aged 5-19 years who were treated by Child and Youth Mental Health Services (CYMHS) in greater Brisbane, Australia. Two Lasso models were trained to predict the risk of future SHoSI in young people currently rated SHoSI; and those who were not.

RESULTS

For currently non-SHoSI children, an Area Under the Receiver Operating Characteristics (AUC) of 0.78 was achieved. Those with the highest risk were 4.97 (CI 4.35-5.66) times more likely to be categorized as SHoSI in the future. For current SHoSI children, the AUC was 0.62.

CONCLUSION

A prediction model with fair overall predictive power for currently non-SHoSI children was generated. Predicting persistence for SHoSI was more difficult. The electronic health records alone were not sufficient to discriminate at acceptable levels and may require adding unstructured data such as clinical notes. To optimally predict SHoSI models need to be tested and validated separately for those young people with varying degrees of risk.

The temporal dependencies between social, emotional and physical health factors in young people receiving mental healthcare: a dynamic Bayesian network analysis

AIMS

The needs of young people attending mental healthcare can be complex and often span multiple domains (e.g., social, emotional and physical health factors). These factors often complicate treatment approaches and contribute to poorer outcomes in youth mental health. We aimed to identify how these factors interact over time by modelling the temporal dependencies between these transdiagnostic social, emotional and physical health factors among young people presenting for youth mental healthcare.

METHODS

Dynamic Bayesian networks were used to examine the relationship between mental health factors across multiple domains (social and occupational function, self-harm and suicidality, alcohol and substance use, physical health and psychiatric syndromes) in a longitudinal cohort of 2663 young people accessing youth mental health services. Two networks were developed: (1) 'initial network', that shows the conditional dependencies between factors at first presentation, and a (2) 'transition network', how factors are dependent longitudinally.

RESULTS

The 'initial network' identified that childhood disorders tend to precede adolescent depression which itself was associated with three distinct pathways or illness trajectories; (1) anxiety disorder; (2) bipolar disorder, manic-like experiences, circadian disturbances and psychosis-like experiences; (3) self-harm and suicidality to alcohol and substance use or functioning. The 'transition network' identified that over time social and occupational function had the largest effect on self-harm and suicidality, with direct effects on ideation (relative risk [RR], 1.79; CI, 1.59-1.99) and self-harm (RR, 1.32; CI, 1.22-1.41), and an indirect effect on attempts (RR, 2.10; CI, 1.69-2.50). Suicide ideation had a direct effect on future suicide attempts (RR, 4.37; CI, 3.28-5.43) and self-harm (RR, 2.78; CI, 2.55-3.01). Alcohol and substance use, physical health and psychiatric syndromes (e.g., depression and anxiety, at-risk mental states) were independent domains whereby all direct effects remained within each domain over time.

CONCLUSIONS

This study identified probable temporal dependencies between domains, which has causal interpretations, and therefore can provide insight into their differential role over the course of illness. This work identified social, emotional and physical health factors that may be important early intervention and prevention targets. Improving social and occupational function may be a critical target due to its impacts longitudinally on self-harm and suicidality. The conditional independence of alcohol and substance use supports the need for specific interventions to target these comorbidities.

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.

Predictive modelling of deliberate self-harm and suicide attempts in young people accessing primary care: a machine learning analysis of a longitudinal study

PURPOSE

Machine learning (ML) has shown promise in modelling future self-harm but is yet to be applied to key questions facing clinical services. In a cohort of young people accessing primary mental health care, this study aimed to establish (1) the performance of models predicting deliberate self-harm (DSH) compared to suicide attempt (SA), (2) the performance of models predicting new-onset or repeat behaviour, and (3) the relative importance of factors predicting these outcomes.

METHODS

802 young people aged 12-25 years attending primary mental health services had detailed social and clinical assessments at baseline and 509 completed 12-month follow-up. Four ML algorithms, as well as logistic regression, were applied to build four distinct models.

RESULTS

The mean performance of models predicting SA (AUC: 0.82) performed better than the models predicting DSH (AUC: 0.72), with mean positive predictive values (PPV) approximately twice that of the prevalence (SA prevalence 14%, PPV: 0.32, DSH prevalence 22%, PPV: 0.40). All ML models outperformed standard logistic regression. The most frequently selected variable in both models was a history of DSH via cutting.

CONCLUSION

History of DSH and clinical symptoms of common mental disorders, rather than social and demographic factors, were the most important variables in modelling future behaviour. The performance of models predicting outcomes in key sub-cohorts, those with new-onset or repetition of DSH or SA during follow-up, was poor. These findings may indicate that the performance of models of future DSH or SA may depend on knowledge of the individual's recent history of either behaviour.

Clinical staging and the differential risks for clinical and functional outcomes in young people presenting for youth mental health care

BACKGROUND

Clinical staging proposes that youth-onset mental disorders develop progressively, and that active treatment of earlier stages should prevent progression to more severe disorders. This retrospective cohort study examined the longitudinal relationships between clinical stages and multiple clinical and functional outcomes within the first 12 months of care.

METHODS

Demographic and clinical information of 2901 young people who accessed mental health care at age 12-25 years was collected at predetermined timepoints (baseline, 3 months, 6 months, 12 months). Initial clinical stage was used to define three fixed groups for analyses (stage 1a: 'non-specific anxious or depressive symptoms', 1b: 'attenuated mood or psychotic syndromes', 2+: 'full-threshold mood or psychotic syndromes'). Logistic regression models, which controlled for age and follow-up time, were used to compare clinical and functional outcomes (role and social function, suicidal ideation, alcohol and substance misuse, physical health comorbidity, circadian disturbances) between staging groups within the initial 12 months of care.

RESULTS

Of the entire cohort, 2093 young people aged 12-25 years were followed up at least once over the first 12 months of care, with 60.4% female and a baseline mean age of 18.16 years. Longitudinally, young people at stage 2+ were more likely to develop circadian disturbances (odds ratio [OR]=2.58; CI 1.60-4.17), compared with individuals at stage 1b. Additionally, stage 1b individuals were more likely to become disengaged from education/employment (OR=2.11, CI 1.36-3.28), develop suicidal ideations (OR=1.92; CI 1.30-2.84) and circadian disturbances (OR=1.94, CI 1.31-2.86), compared to stage 1a. By contrast, we found no relationship between clinical stage and the emergence of alcohol or substance misuse and physical comorbidity.

CONCLUSIONS

The differential rates of emergence of poor clinical and functional outcomes between early versus late clinical stages support the clinical staging model's assumptions about illness trajectories for mood and psychotic syndromes. The greater risk of progression to poor outcomes in those who present with more severe syndromes may be used to guide specific intervention packages.

Introduction to the PLOS ONE collection on 'Understanding and preventing suicide: Towards novel and inclusive approaches'

More than 700,000 people lose their lives to suicide each year and evidence suggests that the current COVID-19 pandemic is leading to increases in risk factors for suicide and suicide-related behaviour, in particular among young people. It is widely documented that some sectors of the population are over-represented in the suicide statistics. It is also well established that the pathways that lead someone to a suicidal crisis are complex and differ across regions and sectors of the population; as such a multi-faceted approach to prevention is required. Many of us would also argue that novel approaches, that combine broad population-based strategies with individual interventions, and approaches that capitalise on new technologies and methodologies are also required. For these reasons, when bringing together this collection, we deliberately sought studies that focused upon those groups who are over-represented in the suicide statistics yet under-represented in research. We also called for studies that reported on novel approaches to suicide prevention and for studies that reflected the voices of people with lived experience of suicide, also often unheard in research efforts.

Editorial: Disrupting Pathways to Self-Harm in Adolescence: Machine Learning as an Opportunity

Self-harm, hurting oneself with or without suicidal intent, is associated with poor mental health. Domains of risk known to be associated with self-harm include sociodemographic factors such as female gender, negative life events, family adversity, and psychiatric diagnoses.¹ However, the heterogeneous nature of self-harm makes predicting risk and prevention challenging. The behaviors can be occasional or repetitive, suicidal in nature or not. Only about half of youths with deliberate self-harm present significant suicide risk.¹ We are left with these remaining questions: What are the early signs of risk for self-harm? Who are the children and adolescents most at risk? Machine learning is the scientific discipline that focuses on how computers learn from data with efficient computing algorithms and prediction models.² If we can use this analytic tool wisely, it could help us to predict risk of self-injury and offer prevention and treatment with precision. However, we need to be careful not to replicate the human biases that already permeate our health care system by failing to include data from diverse populations or considering the ways they are marginalized in building prediction models.

PSCNN: PatchShuffle Convolutional Neural Network for COVID-19 Explainable Diagnosis

Objective: COVID-19 is a sort of infectious disease caused by a new strain of coronavirus. This study aims to develop a more accurate COVID-19 diagnosis system. Methods: First, the n-conv module (nCM) is introduced. Then we built a 12-layer convolutional neural network (12l-CNN) as the backbone network. Afterwards, PatchShuffle was introduced to integrate with 12l-CNN as a regularization term of the loss function. Our model was named PSCNN. Moreover, multiple-way data augmentation and Grad-CAM are employed to avoid overfitting and locating lung lesions. Results: The mean and standard variation values of the seven measures of our model were 95.28 ± 1.03 (sensitivity), 95.78 ± 0.87 (specificity), 95.76 ± 0.86 (precision), 95.53 ± 0.83 (accuracy), 95.52 ± 0.83 (F1 score), 91.7 ± 1.65 (MCC), and 95.52 ± 0.83 (FMI). Conclusion: Our PSCNN is better than 10 state-of-the-art models. Further, we validate the optimal hyperparameters in our model and demonstrate the effectiveness of PatchShuffle.