Psychometric properties of an index of emotional distress in the U.S. National Comorbidity Survey.

Background: The National Comorbidity Survey (NCS; Kessler et al. 1994) was a nationwide household survey of the U.S. population designed to produce data on the prevalence and correlates of psychiatric disorders. The NCS dataset is now in public-use format and continues to be widely used for ongoing research efforts. The NCS dataset included a set of 14 items that have face validity as a measure of current emotional distress (depression and anxiety) and could serve as a potentially useful continuous measure of psychological distress. However, there have been no published studies on its psychometric properties and this measure has not yet been utilized by researchers using the NCS dataset. This paper provides an evaluation of the psychometric properties of the NCS Distress Index. Method: The NCS Part II public-use dataset (N = 5877) was used. Detailed diagnostic information was collected along with 14 items assessing current psychological distress and measures of Neuroticism and Openness to Experience. Results: The NCS Distress Index was found to be internally consistent (Alpha = 0.92) and a series of principal-components analyses demonstrated that the measure is most accurately conceptualized as a single-factor measure of general distress. The construct validity of the Distress Index was supported by its associations with the measures of Neuroticism and Openness to Experience. A series of comparisons between diagnostic groups also supported the construct validity of the measure. For example, those with disorders characterized by depressed mood and worry scored higher on the Distress Index than those with disorders characterized by fear and hyperarousal. Conclusions: The NCS Distress Index is a psychometrically sound measure of current emotional distress. Future studies utilizing the NCS public-use dataset could potentially benefit from the inclusion of this measure in addition to more commonly investigated categorical variables such as diagnosable disorders. [ABSTRACT FROM AUTHOR]