Machine Learning for Suicide Research–Can It Improve Risk Factor Identification?

Abstract

Machine learning is on the rise. According to Scopus (www2. scopus.com), the number of publications in medicine with machine learning in the title, abstract, or as a keyword during 2016 to 2018 increased from 1658 to 3904. In psychiatry, applications of machine learning are proposed to improve the accuracy of diagnosis and prognosis and determine treatment choice. At the same time, much of this research has given insufficient attention to high-quality methods, clinical applications, and ethical aspects. This is compounded by poor reporting of performative measures and misleading claims about the high accuracy of such approaches. In this issue of JAMA Psychiatry, the article by Gradus and colleagues 1 raises important questions about the place of machine learning in research and practice.Machine learning is useful when analyzing several predictors, particularly if there are nonlinear observations and interaction terms that are difficult to theoretically conceptualize and practically model. A strength of the study by Gradus et al 1 is that it uses a robust data set of patient-level predictors from Danish national health care registers without imposing or relying on theoretical models. The authors tested 1365 parameters based on 334 individual predictors, which were mostly associated with somatic and psychiatric diagnoses and by temporal proximity with the suicide event by being categorized into 4 points of 6, 12, 24, and 48 months. Another strength is the outcome of death by suicide. As the study illuminates risk factors rather than prediction, this is important, as risk factors for suicide are different from self-harm and suicidal ideation. For example, male sex is more highly associated with suicide, whereas female sex is a risk factor for suicide ideation and attempts. 2 The use of high-quality Danish register data means that suicide is ascertained with negligible missing data; however, the authors did not consider injuries of undetermined intent as an outcome, which could lead to possible misclassification. Based on 10 outcomes per predictor recommended by methodologists, 3 the authors also had sufficient statistical power to investigate 14 463 suicides. A study of such large suicide numbers combined with hundreds of potentially modifiable predictors is an important advance for the field.The most informative aspects of the study by Gradus et al 1 are in highlighting the importance of psychiatric disorders as risk factors for suicide. The analyses simultaneously accounted for somatic and psychiatric disorders and a few sociodemographic factors (eg, income, civil status, age bands, and being an immigrant). It reported that diagnoses, such as schizophrenia and adjustment disorders, and markers of psy- Related article page 25Opinion EDITORIAL jamapsychiatry.com (Reprinted) JAMA

Predicting suicide attempt or suicide death following a visit to psychiatric specialty care: A machine learning study using Swedish national registry data

“…There was a substantial overlap in top-ranked predictors between different models. While borderline personality disorder, substance use disorder, depression, dispensed benzodiazepines and/or antidepressants have already been identified as risk factors for suicide [ 38 , 39 ], the timing of these factors may play a vital role in predicting subsequent suicidal behavior [ 40 ]. Our study showed that temporally close predictors tended to be ranked higher than distal ones.…”

Multimodal neuroimaging and suicidality in a US population-based sample of school-aged children

“…The aim was to improve identification and prevention of suicidality; however, to date, the evidence is still weak for this purpose due to small sample sizes, heterogeneity and inconsistency across studies, and, as further shown in this study, small effects sizes with limited predictive value. There is therefore an urgent need to improve the study of neurobiological biomarkers, possibly in conjunction with psychosocial risk factors, using other methodologies such as machine learning 61,62 . That said, what our results show is that vulnerability to suicidality does not appear to have a “brain signature” with a strong enough effect in school-age children.…”

Which PHQ-9 Items Can Effectively Screen for Suicide? Machine Learning Approaches

“…ML algorithms are designed to maximize their clinical significance and generalization potential. In fact, ML models have been shown to be able to predict suicide tendencies [ 27 ]. Although ML algorithms have improved the accuracy of suicide risk detection; for a number of reasons, perfect predictions using ML are not possible and result in false positives and false positives [ 31 ].…”

“…Item 9 is widely used as a suicide screening tool; however, no studies have assessed the efficacy of PHQ-9 as a suicide screening tool using machine learning (ML) techniques. Machine learning algorithms have been proposed to improve diagnostic and prognostic accuracy and determine treatment options [ 27 ]. Various machine learning studies are focused on suicide prediction [ 28 , 29 , 30 ] and machine learning analysis has an advantage in its accuracy and scalability compared to conventional statistical approaches [ 29 ].…”