The identification of a client's suicidal ideation is one of the more important tasks faced by social work practitioners in mental health settings. There is also a growing recognition of the importance of social workers in health settings identifying persons who may be at risk of suicide. This potential role has been underscored by findings suggesting, for example, that nearly 50% of depressed persons seek help from physicians. Standardized short-form measures of suicidal thinking are one of the more important tools for detecting suicidal ideation. The Multi-Problem Screening Inventory, or MPSI, is a multi-dimensional measurement tool with 27 short-form subscales, each designed to measure the magnitude or severity of a problem in an area of personal, interpersonal, or social functioning. One of the MPSI subscales is the Suicidal Thoughts (ST) subscale. Scores on this eleven item subscale can range from 0 to 100, with higher scores indicative of greater magnitude problems with suicidal thinking, and vice versa. The psychometric properties of scores from this subscale have only recently been studied (Nugent, in press). The results of the research by Nugent (in press) suggested the plausibility of using a very small number of items from this subscale to create an ultra-short-form measure of suicidal thinking that could be used in a wide range of service settings to help identify persons contemplating suicide. The research reported in this paper focuses on the development of an ultra-short-form measure of suicidal ideation, the Ultra-Short-Form Suicidal Thinking Measure (USSTM), which contains only four items. The brevity of this measure makes it potentially very useful in a wide range of mental health (and other) service settings, particularly those in which social workers face severe time constraints. The results of psychometric analyses of scores from this measure, based upon classical, item response, and generalizability theories, are also reported. Five-hundred-forty-two responses to the MPSI were obtained from persons seeking help, in one of several non-profit family service agencies, for personal and/or family problems. The MPSI ST subscale responses were analyzed using SPSS version 11 and MULTILOG version 7.03. The results of IRT analyses were used to identify MPSI ST subscale items with the most desirable item information curves for detecting and measuring suicidal thinking. These information curves, together with item content considerations, led to the identification of four MPSI ST subscale items to be placed onto the USSTM. Results of reliability analyses suggest that scores from this measure will have reliability coefficients in excess of .90, and dependability indices for clinical classification decisions as high as .92. The results of IRT analyses further indicate excellent information function characteristics for both identification of persons at risk of suicide and for monitoring suicidal ideation across time. Preliminary evidence also supports the validity of scores from this measure. These results provide preliminary forecasts that suggest this measure will produce scores with excellent psychometric properties for detecting and measuring suicidal thinking.

Nugent, W. (In press). A Psychometric Study of the MPSI Suicidal Thoughts Subscale. Forthcoming in Crisis Intervention and Time Limited Treatment.