Psychiatric morbidity among medical in-patients: a standardized assessment (GHQ-12 and CIS-R) using 'lay' interviewers in a Brazilian hospital.

The 12-item General Health Questionnaire (GHQ-12) and the revised Clinical Interview Schedule (CIS-R) were used to estimate the prevalence of psychiatric morbidity among 78 consecutive admissions to a general medical ward in a Brazilian university hospital (43 males and 35 females; mean age = 43.2 years). The CIS-R was administered by three 5th-year medical students after a brief training. A prevalence rate of 36% was found for psychiatric disorders. The most frequent symptoms were sleep disorders (48.7%), worry (35.9%), depression (28.2%) and anxiety (26.9%). The sensitivity and specificity of the GHQ-12 were 71% and 76%, respectively. The CIS-R was simple to administer and acceptable both to patients and interviewers. Misunderstanding was most likely to occur with the poorly educated (20% were illiterate) in questions involving time calculation. Alternative options might be used to specify the length of time in future studies. The findings support the feasibility of the CIS-R and the use of 'lay' interviewers to produce epidemiological information on psychiatric disorders in developing countries at lower costs.