Designing a Clinical Decision Support Tool That Leverages Machine Learning for Suicide Risk Prediction: Development Study in Partnership With Native American Care Providers

Clinical decision support systems in addiction and concurrent disorders: A systematic review and meta-analysis

INTRODUCTION

This review aims to synthesise the literature on the efficacy, evolution, and challenges of implementing Clincian Decision Support Systems (CDSS) in the realm of mental health, addiction, and concurrent disorders.

METHODS

Following PRISMA guidelines, a systematic review and meta-analysis were performed. Searches conducted in databases such as MEDLINE, Embase, CINAHL, PsycINFO, and Web of Science through 25 May 2023, yielded 27,344 records. After necessary exclusions, 69 records were allocated for detailed synthesis. In the examination of patient outcomes with a focus on metrics such as therapeutic efficacy, patient satisfaction, and treatment acceptance, meta-analytic techniques were employed to synthesise data from randomised controlled trials.

RESULTS

A total of 69 studies were included, revealing a shift from knowledge-based models pre-2017 to a rise in data-driven models post-2017. The majority of models were found to be in Stage 2 or 4 of maturity. The meta-analysis showed an effect size of -0.11 for addiction-related outcomes and a stronger effect size of -0.50 for patient satisfaction and acceptance of CDSS.

DISCUSSION

The results indicate a shift from knowledge-based to data-driven CDSS approaches, aligned with advances in machine learning and big data. Although the immediate impact on addiction outcomes is modest, higher patient satisfaction suggests promise for wider CDSS use. Identified challenges include alert fatigue and opaque AI models.

CONCLUSION

CDSS shows promise in mental health and addiction treatment but requires a nuanced approach for effective and ethical implementation. The results emphasise the need for continued research to ensure optimised and equitable use in healthcare settings.

Adolescent, Parent, and Provider Perceptions of a Predictive Algorithm to Identify Adolescent Suicide Risk in Primary Care

OBJECTIVE

To understand adolescent, parent, and provider perceptions of a machine learning algorithm for detecting adolescent suicide risk prior to its implementation primary care.

METHODS

We conducted semi-structured, qualitative interviews with adolescents (n = 9), parents (n = 12), and providers (n = 10; mixture of behavioral health and primary care providers) across two major health systems. Interviews were audio recorded and transcribed with analyses supported by use of NVivo. A codebook was developed combining codes derived inductively from interview transcripts and deductively from implementation science frameworks for content analysis.

RESULTS

Reactions to the algorithm were mixed. While many participants expressed privacy concerns, they believed the algorithm could be clinically useful for identifying adolescents at risk for suicide and facilitating follow-up. Parents' past experiences with their adolescents' suicidal thoughts and behaviors contributed to their openness to the algorithm. Results also aligned with several key Consolidated Framework for Implementation Research domains. For example, providers mentioned barriers inherent to the primary care setting such as time and resource constraints likely to impact algorithm implementation. Participants also cited a climate of mistrust of science and health care as potential barriers.

CONCLUSIONS

Findings shed light on factors that warrant consideration to promote successful implementation of suicide predictive algorithms in pediatric primary care. By attending to perspectives of potential end users prior to the development and testing of the algorithm, we can ensure that the risk prediction methods will be well-suited to the providers who would be interacting with them and the families who could benefit.

Trust and acceptability of data-driven clinical recommendations in everyday practice: A scoping review

BACKGROUND

Increasing attention is being given to the analysis of large health datasets to derive new clinical decision support systems (CDSS). However, few data-driven CDSS are being adopted into clinical practice. Trust in these tools is believed to be fundamental for acceptance and uptake but to date little attention has been given to defining or evaluating trust in clinical settings.

OBJECTIVES

A scoping review was conducted to explore how and where acceptability and trustworthiness of data-driven CDSS have been assessed from the health professional's perspective.

METHODS

Medline, Embase, PsycInfo, Web of Science, Scopus, ACM Digital, IEEE Xplore and Google Scholar were searched in March 2022 using terms expanded from: "data-driven" AND "clinical decision support" AND "acceptability". Included studies focused on healthcare practitioner-facing data-driven CDSS, relating directly to clinical care. They included trust or a proxy as an outcome, or in the discussion. The preferred reporting items for systematic reviews and meta-analyses extension for scoping reviews (PRISMA-ScR) is followed in the reporting of this review.

RESULTS

3291 papers were screened, with 85 primary research studies eligible for inclusion. Studies covered a diverse range of clinical specialisms and intended contexts, but hypothetical systems (24) outnumbered those in clinical use (18). Twenty-five studies measured trust, via a wide variety of quantitative, qualitative and mixed methods. A further 24 discussed themes of trust without it being explicitly evaluated, and from these, themes of transparency, explainability, and supporting evidence were identified as factors influencing healthcare practitioner trust in data-driven CDSS.

CONCLUSION

There is a growing body of research on data-driven CDSS, but few studies have explored stakeholder perceptions in depth, with limited focused research on trustworthiness. Further research on healthcare practitioner acceptance, including requirements for transparency and explainability, should inform clinical implementation.

Evaluation of the Risk Identification for Suicide and Enhanced Care Model in a Native American Community

Importance

There are many prognostic models of suicide risk, but few have been prospectively evaluated, and none has been developed specifically for Native American populations.

Objective

To prospectively validate a statistical risk model implemented in a community setting and evaluate whether use of this model was associated with improved reach of evidence-based care and reduced subsequent suicide-related behavior among high-risk individuals.

Design, Setting, and Participants

This prognostic study, done in partnership with the White Mountain Apache Tribe, used data collected by the Apache Celebrating Life program for adults aged 25 years or older identified as at risk for suicide and/or self-harm from January 1, 2017, through August 31, 2022. Data were divided into 2 cohorts: (1) individuals and suicide-related events from the period prior to suicide risk alerts being active (February 29, 2020) and (2) individuals and events from the time after alerts were activated.

Main Outcomes and Measures

Aim 1 focused on a prospective validation of the risk model in cohort 1. Aim 2 compared the odds of repeated suicide-related events and the reach of brief contact interventions among high-risk cases between cohort 2 and cohort 1.

Results

Across both cohorts, a total of 400 individuals identified as at risk for suicide and/or self-harm (mean [SD] age, 36.5 [10.3] years; 210 females [52.5%]) had 781 suicide-related events. Cohort 1 included 256 individuals with index events prior to active notifications. Most index events (134 [52.5%]) were for binge substance use, followed by 101 (39.6%) for suicidal ideation, 28 (11.0%) for a suicide attempt, and 10 (3.9%) for self-injury. Among these individuals, 102 (39.5%) had subsequent suicidal behaviors. In cohort 1, the majority (220 [86.3%]) were classified as low risk, and 35 individuals (13.3%) were classified as high risk for suicidal attempt or death in the 12 months after their index event. Cohort 2 included 144 individuals with index events after notifications were activated. For aim 1, those classified as high risk had a greater odds of subsequent suicide-related events compared with those classified as low risk (odds ratio [OR], 3.47; 95% CI, 1.53-7.86; P = .003; area under the receiver operating characteristic curve, 0.65). For aim 2, which included 57 individuals classified as high risk across both cohorts, during the time when alerts were inactive, high-risk individuals were more likely to have subsequent suicidal behaviors compared with when alerts were active (OR, 9.14; 95% CI, 1.85-45.29; P = .007). Before the active alerts, only 1 of 35 (2.9%) individuals classified as high risk received a wellness check; after the alerts were activated, 11 of 22 (50.0%) individuals classified as high risk received 1 or more wellness checks.

Conclusions and Relevance

This study showed that a statistical model and associated care system developed in partnership with the White Mountain Apache Tribe enhanced identification of individuals at high risk for suicide and was associated with a reduced risk for subsequent suicidal behaviors and increased reach of care.

Complex modeling with detailed temporal predictors does not improve health records-based suicide risk prediction

Suicide risk prediction models can identify individuals for targeted intervention. Discussions of transparency, explainability, and transportability in machine learning presume complex prediction models with many variables outperform simpler models. We compared random forest, artificial neural network, and ensemble models with 1500 temporally defined predictors to logistic regression models. Data from 25,800,888 mental health visits made by 3,081,420 individuals in 7 health systems were used to train and evaluate suicidal behavior prediction models. Model performance was compared across several measures. All models performed well (area under the receiver operating curve [AUC]: 0.794-0.858). Ensemble models performed best, but improvements over a regression model with 100 predictors were minimal (AUC improvements: 0.006-0.020). Results are consistent across performance metrics and subgroups defined by race, ethnicity, and sex. Our results suggest simpler parametric models, which are easier to implement as part of routine clinical practice, perform comparably to more complex machine learning methods.

Prediction of Suicide Attempts and Suicide-Related Events Among Adolescents Seen in Emergency Departments

IMPORTANCE

Screening adolescents in emergency departments (EDs) for suicidal risk is a recommended strategy for suicide prevention. Comparing screening measures on predictive validity could guide ED clinicians in choosing a screening tool.

OBJECTIVE

To compare the Ask Suicide-Screening Questions (ASQ) instrument with the Computerized Adaptive Screen for Suicidal Youth (CASSY) instrument for the prediction of suicidal behavior among adolescents seen in EDs, across demographic and clinical strata.

DESIGN, SETTING, AND PARTICIPANTS

The Emergency Department Study for Teens at Risk for Suicide is a prospective, random-series, multicenter cohort study that recruited adolescents, oversampled for those with psychiatric symptoms, who presented to the ED from July 24, 2017, through October 29, 2018, with a 3-month follow-up to assess the occurrence of suicidal behavior. The study included 14 pediatric ED members of the Pediatric Emergency Care Applied Research Network and 1 Indian Health Service ED. Statistical analysis was performed from May 2021 through January 2023.

MAIN OUTCOMES AND MEASURES

This study used a prediction model to assess outcomes. The primary outcome was suicide attempt (SA), and the secondary outcome was suicide-related visits to the ED or hospital within 3 months of baseline; both were assessed by an interviewer blinded to baseline information. The ASQ is a 4-item questionnaire that surveys suicidal ideation and lifetime SAs. A positive response or nonresponse on any item indicates suicidal risk. The CASSY is a computerized adaptive screening tool that always includes 3 ASQ items and a mean of 8 additional items. The CASSY's continuous outcome is the predicted probability of an SA.

RESULTS

Of 6513 adolescents available, 4050 were enrolled, 3965 completed baseline assessments, and 2740 (1705 girls [62.2%]; mean [SD] age at enrollment, 15.0 [1.7] years; 469 Black participants [17.1%], 678 Hispanic participants [24.7%], and 1618 White participants [59.1%]) completed both screenings and follow-ups. The ASQ and the CASSY showed a similar sensitivity (0.951 [95% CI, 0.918-0.984] vs 0.945 [95% CI, 0.910-0.980]), specificity (0.588 [95% CI, 0.569-0.607] vs 0.643 [95% CI, 0.625-0.662]), positive predictive value (0.127 [95% CI, 0.109-0.146] vs 0.144 [95% CI, 0.123-0.165]), and negative predictive value (both 0.995 [95% CI, 0.991-0.998], respectively). Area under the receiver operating characteristic curve findings were similar among patients with physical symptoms (ASQ, 0.88 [95% CI, 0.81-0.95] vs CASSY, 0.94 [95% CI, 0.91-0.96]). Among patients with psychiatric symptoms, the CASSY performed better than the ASQ (0.72 [95% CI, 0.68-0.77] vs 0.57 [95% CI, 0.55-0.59], respectively).

CONCLUSIONS AND RELEVANCE

This study suggests that both the ASQ and the CASSY are appropriate for universal screening of patients in pediatric EDs. For the small subset of patients with psychiatric symptoms, the CASSY shows greater predictive validity.

The use of machine learning and artificial intelligence within pediatric critical care

The field of pediatric critical care has been hampered in the era of precision medicine by our inability to accurately define and subclassify disease phenotypes. This has been caused by heterogeneity across age groups that further challenges the ability to perform randomized controlled trials in pediatrics. One approach to overcome these inherent challenges include the use of machine learning algorithms that can assist in generating more meaningful interpretations from clinical data. This review summarizes machine learning and artificial intelligence techniques that are currently in use for clinical data modeling with relevance to pediatric critical care. Focus has been placed on the differences between techniques and the role of each in the clinical arena. The various forms of clinical decision support that utilize machine learning are also described. We review the applications and limitations of machine learning techniques to empower clinicians to make informed decisions at the bedside. IMPACT: Critical care units generate large amounts of under-utilized data that can be processed through artificial intelligence. This review summarizes the machine learning and artificial intelligence techniques currently being used to process clinical data. The review highlights the applications and limitations of these techniques within a clinical context to aid providers in making more informed decisions at the bedside.