JMIR Publications

ABSTRACT

Background:

Machine learning is a promising tool in the area of suicide prevention due to its ability to combine the effects of multiple risk factors and complex interactions. The power of machine learning has led to an influx of studies on suicide prediction, as well as a few recent reviews. However, the existing systematic reviews did not differentiate between data sources, and have largely failed to summarize the most important risk factors.

Objective:

The objective of our study was to assess the degree to which machine learning techniques capture incidents of suicidal thoughts and behaviors, and ascertain the most important predictors of suicidal thoughts and behaviors identified via machine learning.

Methods:

A systematic literature search of PubMed,‎ Medline, Embase, PsycINFO, Web of Science,‎ Cumulative Index to Nursing and Allied Health Literature‎ (CINAHL)‎, and Allied and Complementary Medicine Database (AMED) to identify all studies that have used machine learning to predict suicidal thoughts and behaviors using administrative or survey data‎ was performed. The search was conducted for articles published between January 1, 2019 and May 11, 2022. In addition, all articles identified in three recently published systematic reviews (the last of which included studies up until January 1, 2019) were retained, if they met our inclusion criteria. To evaluate the predictive power of machine learning methods in predicting suicidal thoughts and behaviors box plots were used to summarize the distribution of the area under the receiver operating characteristic curve (AUC) values by suicide outcome (i.e., suicidal thoughts, suicide attempt, and death by suicide, as well as all suicide outcomes combined) and machine learning method. Mean AUCs with 95% confidence intervals (CIs) were computed for each suicide outcome by study design, study data source, total sample size, and sample size of cases, and machine learning methods employed. The most important risk factors identified were summarized.

Results:

The search strategy identified a total of 2,200 unique records, of which 104 articles met the inclusion criteria. Machine learning algorithms achieved good prediction of suicidal thoughts and behaviors (i.e., an AUC between 0.80 and 0.89); however, their predictive power appears to differ across suicide outcomes. The AUCs for boosting algorithms achieved good prediction of suicidal thoughts, death by suicide, and all suicide outcomes combined, while neural network algorithm achieved good prediction of suicide attempt. The risk factors for suicidal thoughts and behaviors differed depending on the data source and the population under study.

Conclusions:

The predictive utility of machine learning for suicidal thoughts and behaviors is largely dependent on the approach used. The findings of the current review should prove helpful in preparing future machine learning models. Clinical Trial: N/A