Derivation of the SPAN, a brief diagnostic screening test for post-traumatic stress disorder

The Davidson Trauma Scale (DTS) is a validated 17-item self-rating scale us ed in the diagnosis of post-traumatic stress disorder (PTSD), which is sens itive to the effects of treatment. It was felt that a shorter version of th e scale might provide a better diagnostic screening tool. Subjects were dra wn from a sample of 243 patients obtained from multiple cohorts that includ ed a group of survivors of various forms of trauma, including natural disas ter, rape and combat. All subjects had diagnostic assessments for PTSD with a clinical interview and completed the DTS. The data were randomly divided between two subsamples. and frequency and severity scores were calculated for the DTS. A four-item scale, the SPAN (named for its top four items: Sta rtle, Physiological arousal, Anger, and Numbness), was developed. It demons trated an efficiency of 0.88, sensitivity of 0.84, specificity of 0.91 and positive likelihood ratio of 9.1. In a replication sample, values were slig htly lower but still acceptable (efficiency = 0.80). A subgroup of PTSD pat ients received either fluoxetine or placebo in a clinical trial, and a sign ificant SPAN score improvement was observed on fluoxetine. The SPAN, which correlated significantly with the Impact of Events Scale, the Sheehan Disab ility Scale, and the Structured Interview of PTSD, was found to have a diag nostic accuracy of 88%. (C) 1999 Elsevier Science Ireland Ltd. All rights r eserved.