Objective: Nonspecific somatic symptoms play an important role in the manifestation of psychiatric morbidity. This study examined the psychometric performance of an instrument developed to improve the rates of identification of psychiatric disorders, which incorporated somatic and psychological dimensions of distress (the Somatic and Psychological Health Report, or SPHERE).
Methods: Eight hundred twenty-one adults who were participating in an epidemiological longitudinal study of the psychological impact of childhood disaster exposure (mean age of 28.23 years, S.D. of 2.29, range of 22-34) were recruited out of the original cohort of 1531. The 34-item SPHERE was administered in a self-report booklet, and the subjects were interviewed using the Composite International Diagnostic Interview to ascertain current and lifetime psychiatric diagnoses.
Results: While the negative predictive power was high (96.1% for current disorder and 81.7% for lifetime disorder), the positive predictive power was low (56.8% for a lifetime disorder and 27.5% for current disorder). This was despite 61.6% of lifetime sufferers and 78.6% who met current criteria for any disorder screened positive using the SPHERE. Sensitivity was highest when the broad "PSYCH or SOMA" screen was used (78.6%). Specificity of 89.5% was obtained for the "PSYCH and SOMA" scale.
Conclusions: In this population of young adults, where age limited the prevalence of comorbid physical disease, the SPHERE was an acceptable screening measure. The psychometric performance was better for lifetime than current disorder. The psychometric characteristics of this instrument indicate that its particular use may be in defining individuals who need a more detailed assessment in a clinical setting.