Discovering the Unclassified Suicide Cases Among Undetermined Drug Overdose Deaths Using Machine Learning Techniques

Abstract

Objective The Centers for Disease Control and Prevention (CDC) monitor accidental and intentional deaths to answer questions that are critical for the development of effective prevention and resource allocation. CDC's National Violent Death Reporting System (NVDRS) is a major innovation in surveillance linking individual‐level data from multiple sources. However, suicide underreporting is common, particularly from drug overdose deaths. This study sought to assess machine learning (ML) techniques in quantifying drug overdose suicide underreporting rates. Methods Clinical, sociodemographic, toxicological, and proximal stressor data on overdose decedents (n = 2,665) were extracted from Utah's NVDRS from 2012 to 2015. The existing well‐determined cases were used to train and test our ML models. We assessed and compared multiple machine learning methods including Logistic Regression, Random Forest Classifier, Support Vector Machines, and Artificial Neural Networks. We applied a majority voting methodology to classify undetermined drug overdose deaths. Results Overdose suicide rates were estimated to be underreported by 33% across all years, increasing yearly from 29% in 2012 to 37% in 2015. The overall test accuracies for all models ranged from 92.3% to 94.6%. Conclusions This research identifies a cost‐effective, replicable, and expandable ML‐based methodology to estimate the true rates of suicide which may be partially masked during the opioid epidemic.

Association of Suicide Risk With Transition to Civilian Life Among US Military Service Members

“…Second, the potential for variability in the manner in which coroners and medical examiners ascertained death is a recognized limitation of death certificate records in general. 23 However, based on additional analyses we conducted, including deaths of undetermined intent and suicides combined as the outcome, we found negligible changes to HRs and statistical significance.…”

“…Second, the potential for variability in the manner in which coroners and medical examiners ascertained death is a recognized limitation of death certificate records in general. 23…”

Utility of support vector machine and decision tree to identify the prognosis of metformin poisoning in the United States: analysis of National Poisoning Data System

“…The application of DT and SVM have been widely investigated among patients with drug overdose in recent times [ 13 – 15 ]. However, despite the serious outcomes, there has been no explicit application of machine learning techniques to study metformin overdose and provide a realistic strategy for determining the prognosis based on clinical and laboratory findings.…”

Using natural language processing to improve suicide classification requires consideration of race

“…However, limitations in the data used provide directions for future work. NVDRS does not contain records of deaths that were ruled as accidental, though some participating states have initiated programs to record unintentional/accidental drug overdose deaths within NVDRS (Liu et al, 2020). Given the issues raised in this and other studiesy (Bohnert et al, 2013; Rockett et al, 2020; Stone et al, 2017), and the challenges of identifying suicide in general, it is possible that some or many deaths due to suicide were incorrectly classified as accidental and thus not included in this analysis.…”