Objective: To develop and validate a self-report screening tool for common mental disorders.
Design and setting: Sequential development and validation studies in three cohorts of patients in general practice and one cohort of patients in a specialist psychiatry clinic.
Participants: 1585 patients in general practice examined cross-sectionally and longitudinally; 46515 patients attending 386 general practitioners nationwide; 364 patients participating in a longitudinal study of psychiatric disorders in general practice; and 522 patients attending a specialist psychiatry clinic.
Main outcome measures: Performance of the 12 items from the 34-item SPHERE questionnaire against DSM-III-R and DSM-IV diagnoses of psychiatric disorder, self-reported Brief Disability Questionnaire findings, GPs' ratings of patients' needs for psychological care and degree of risk resulting from mental disorder, and patients' and GPs' reports of reasons for presentation.
Results: Six somatic and six psychological questions identify two levels (and three types) of mental disorder: patients reporting both characteristic psychological and somatic symptoms (Level 1, Type 1), and patients reporting either psychological symptoms (Level 2, Type 2) or somatic symptoms (Level 2, Type 3). This classification system predicts disability ratings (Level 1, 8.2 "days out of role in the last month" and Level 2, 4.1 and 5.4 "days out of role in the last month" for Types 2 and 3, respectively), rates of lifetime psychiatric diagnoses (Level 1, 63% and Level 2, 59% and 48%, respectively), both patients' and GPs' report of reasons for presentation, and doctors' ratings of risk as a result of mental disorder. There are important and differing sociodemographic correlates for the three types of mental disorders.
Conclusion: A classification system based on the 12 items from the 34-item SPHERE questionnaire can be used to identify common mental disorders. This system has acceptable validity and reliability, and is suited specifically for general practice settings.