Longitudinal Study of Suicide Attempts in Adolescents

Duke University, Durham NC

Investigators

Linked publications & trials

Abstract

[unreadable] DESCRIPTION (provided by investigator): In this competing continuation application, we propose to continue following for a final four years a largely intact sample (n=180) thought to be at high-risk for suicidal behavior (a fifth year will be devoted to final analyses and dissemination of results of the study). The individuals in this study have been followed for up to ten years and were ascertained initially as adolescents following psychiatric hospitalization. With this continued period of follow-up, we intend to address several issues not only of theoretical interest, but also with practical clinical implications. The first Specific Aim of this project is to examine an apparent increase in suicidal behavior in this sample in the early- to mid-20's, and the degree to which this increase in suicidal behavior is accounted for by repeat suicidal behavior. In this vein, we also are interested in the degree to which lifetime histories of suicidal behavior in this sample are consistent with a sensitization hypothesis (with increasing frequency and severity of repeat suicide attempts over time, and decreasing relationship between suicide attempts and discrete life events). The Second Aim of this project is to examine the predictive validity of the clinical characteristics of suicidal behavior. Specifically, given the fact that medical lethality in particular is weighted very highly among mental health professionals in clinical judgments of the risk for later suicidal behavior, we are interested in examining the degree to which the stated intent and medical lethality of suicide attempts are predictive both of the occurrence of later attempts, and of the intent and lethality of subsequent attempts. The final Specific Aim of this project is to examine the real-life (non-suicidal) outcomes (e.g., social dysfunction, school drop-out, job instability, incarceration) among suicidal as contrasted with non-suicidal but psychiatrically impaired individuals. This issue of prognosis is particularly important given several reports in the literature of poor non-suicidal outcomes among suicidal individuals. In this context, we recognize the heterogeneity among suicidal individuals and will use latent class trajectory models to characterize various groups of individuals based on the course of their suicidal ideation and behavior over time. The groups that differ in the courses of their suicidal behavior over time will be contrasted and characterized in terms of risk and protective factors for suicidal behaviors, as well as their risks for various non-suicidal outcomes. [unreadable] [unreadable]