Predicting Suicides After Psychiatric Hospitalization in US Army Soldiers

Kessler, Ronald C.; Warner, Christopher H.; Ivany, Christopher; Petukhova, Maria; Rose, Sherri; Bromet, Evelyn J.; Brown, Millard; Cai, Tianxi; Colpe, Lisa J.; Cox, Kenneth L.; Fullerton, Carol S.; Gilman, Stephen E.; Gruber, Michael J.; Heeringa, Steven G.; Lewandowski‐Romps, Lisa; Li, Junlong; Millikan-Bell, Amy M.; Naifeh, James A.; Nock, Matthew K.; Rosellini, Anthony J.; Sampson, Nancy A.; Schoenbaum, Michael; Stein, Murray B.; Wessely, Simon; Zaslavsky, Alan M.; Ursano, Robert J.

doi:10.1001/jamapsychiatry.2014.1754

Cited by 388 publications

(164 citation statements)

References 40 publications

Supporting

Mentioning

159

Contrasting

Order By: Relevance

“…A longitudinal investigation of a larger cohort of individuals with suicidal ideation could repeatedly assess the altered neural representations to determine whether there is a neural signature of an imminent attempt. Such information would be invaluable in the case of the small percentage (e.g., 5%) of patients in psychiatric inpatient care who make up as much as half of suicides subsequent to discharge from a hospital 35 . In future prospective studies, it would be of great interest to learn if our neurosemantic assessments are useful in monitoring for current suicidal risk and in predicting future suicide attempts.…”

Section: Discussionmentioning

confidence: 99%

RETRACTED ARTICLE: Machine learning of neural representations of suicide and emotion concepts identifies suicidal youth

et al. 2017

Self Cite

View full text Add to dashboard Cite

The clinical assessment of suicidal risk would be significantly complemented by a biologically-based measure that assesses alterations in the neural representations of concepts related to death and life in people who engage in suicidal ideation. This study used machine-learning algorithms (Gaussian Naïve Bayes) to identify such individuals (17 suicidal ideators vs 17 controls) with high (91%) accuracy, based on their altered fMRI neural signatures of death and life-related concepts. The most discriminating concepts were death, cruelty, trouble, carefree, good, and praise. A similar classification accurately (94%) discriminated 9 suicidal ideators who had made a suicide attempt from 8 who had not. Moreover, a major facet of the concept alterations was the evoked emotion, whose neural signature served as an alternative basis for accurate (85%) group classification. The study establishes a biological, neurocognitive basis for altered concept representations in participants with suicidal ideation, which enables highly accurate group membership classification.

show abstract

Section: Discussionmentioning

confidence: 99%

RETRACTED ARTICLE: Machine learning of neural representations of suicide and emotion concepts identifies suicidal youth

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…Studies that assess this issue more directly and extend beyond existing methodological confines are needed. For instance, approaches that involve machine learning algorithms to combine large number of risk factors over a short follow up period may be particularly promising (Kessler et al 2015). …”

Section: Discussionmentioning

confidence: 99%

Self-injurious thoughts and behaviors as risk factors for future suicide ideation, attempts, and death: a meta-analysis of longitudinal studies

et al. 2015

View full text Add to dashboard Cite

Background A history of self-injurious thoughts and behaviors (SITBs) is consistently cited as one of the strongest predictors of future suicidal behavior. However, stark discrepancies in the literature raise questions about the true magnitude of these associations. The objective of this study is to examine the magnitude and clinical utility of the associations between SITBs and subsequent suicide ideation, attempts, and death. Method We searched PubMed, PsycInfo, and Google Scholar for papers published through December 2014. Inclusion required that studies include at least one longitudinal analysis predicting suicide ideation, attempts, or death using any SITB variable. We identified 2179 longitudinal studies; 172 met inclusion criteria. Results The most common outcome was suicide attempt (47.80%), followed by death (40.50%) and ideation (11.60%). Median follow-up was 52 months (mean = 82.52, s.d. = 102.29). Overall prediction was weak, with weighted mean odds ratios (ORs) of 2.07 [95% confidence interval (CI) 1.76–2.43] for ideation, 2.14 (95% CI 2.00–2.30) for attempts, and 1.54 (95% CI 1.39–1.71) for death. Adjusting for publication bias further reduced estimates. Diagnostic accuracy analyses indicated acceptable specificity (86–87%) and poor sensitivity (10–26%), with areas under the curve marginally above chance (0.60–0.62). Most risk factors generated OR estimates of <2.0 and no risk factor exceeded 4.5. Effects were consistent regardless of sample severity, sample age groups, or follow-up length. Conclusions Prior SITBs confer risk for later suicidal thoughts and behaviors. However, they only provide a marginal improvement in diagnostic accuracy above chance. Addressing gaps in study design, assessment, and underlying mechanisms may prove useful in improving prediction and prevention of suicidal thoughts and behaviors.

show abstract

“…24 But to be useful to a clinician, new methods of suicide risk stratification would need to be several orders of magnitude more powerful than the existing methods. Future research might also concentrate on suicide among primary care patients or the general population.…”

Section: Is Ongoing Research Likely To Provide Relevant Evidence?mentioning

confidence: 99%