RESUMO
Previous suicide attempts, psychopathology symptoms, and pain significantly increase risk of suicide, a leading cause of death. It is possible that patients across these three groups exhibit key differences that could provide insights into unique interventions for suicide-related outcomes. Data were collected using a standardized form at 432 emergency departments (EDs; 14,018 participants [females, n = 8,042; 57.4%; males, n = 5,976; 42.6%]). We conducted a series of ANOVAs to investigate if patients presenting for (1) suicide attempts (n = 33; 0.2%), (2) psychopathology symptoms (n = 1,104; 7.9%), or (3) pain (n = 12,881; 91.9%) varied across a variety of healthcare-relevant variables. Findings indicated that patients presenting with suicide attempts were seen with more urgency (F[2,12054] = 66.41, p < .001) and were more likely to be admitted to hospitalization (F[2,14015] = 187.296, p < .001), observation unit overall (F[2,14015] = 78.572, p < .001), or transferred to another hospital (F[2,14015] = 406.568, p < .001); they also required longer visits (F [2, 12054] = 66.41, p < .001) as compared to patients with psychopathology symptoms or pain. Notably, potentially important similarities between groups emerged: groups did not differ across leaving without medical screening, leaving against medical advice, or contact with healthcare providers in the long-term (i.e., twelve months) or short-term (i.e., 72 hours) preceding ED admission. These findings in particular indicate that there could be ample time (1) prior to admission to intervene and (2) during care in EDs to connect patients to goal-oriented, time-limited evidence based psychotherapies at a time when they may be particularly willing to engage in care.
RESUMO
Theoretically-driven models of suicide have long guided suicidology; however, an approach employing machine learning models has recently emerged in the field. Some have suggested that machine learning models yield improved prediction as compared to theoretical approaches, but to date, this has not been investigated in a systematic manner. The present work directly compares widely researched theories of suicide (i.e., BioSocial, Biological, Ideation-to-Action, and Hopelessness Theories) to machine learning models, comparing the accuracy between the two differing approaches. We conducted literature searches using PubMed, PsycINFO, and Google Scholar, gathering effect sizes from theoretically-relevant constructs and machine learning models. Eligible studies were longitudinal research articles that predicted suicide ideation, attempts, or death published prior to May 1, 2020. 124 studies met inclusion criteria, corresponding to 330 effect sizes. Theoretically-driven models demonstrated suboptimal prediction of ideation (wOR = 2.87; 95% CI, 2.65-3.09; k = 87), attempts (wOR = 1.43; 95% CI, 1.34-1.51; k = 98), and death (wOR = 1.08; 95% CI, 1.01-1.15; k = 78). Generally, Ideation-to-Action (wOR = 2.41, 95% CI = 2.21-2.64, k = 60) outperformed Hopelessness (wOR = 1.83, 95% CI 1.71-1.96, k = 98), Biological (wOR = 1.04; 95% CI .97-1.11, k = 100), and BioSocial (wOR = 1.32, 95% CI 1.11-1.58, k = 6) theories. Machine learning provided superior prediction of ideation (wOR = 13.84; 95% CI, 11.95-16.03; k = 33), attempts (wOR = 99.01; 95% CI, 68.10-142.54; k = 27), and death (wOR = 17.29; 95% CI, 12.85-23.27; k = 7). Findings from our study indicated that across all theoretically-driven models, prediction of suicide-related outcomes was suboptimal. Notably, among theories of suicide, theories within the Ideation-to-Action framework provided the most accurate prediction of suicide-related outcomes. When compared to theoretically-driven models, machine learning models provided superior prediction of suicide ideation, attempts, and death.