Short Screening Instrument For Psychiatric Morbidity J Formos Med Assoc 2003 • Vol 102 • No 10 687 1 Departments of Psychiatry and Social Medicine, National Taiwan University College of Medicine, Taipei; 2 Department of Psychiatry, Taipei City Psychiatric Center, Taipei; Departments of 3 Psychiatry and 4 Rehabilitation, National Taiwan University Hospital, Taipei, Taiwan. Received: 15 July 2003 Revised: 11 August 2003 Accepted: 2 September 2003 Reprint requests and correspondence to: Dr. Ming-Been Lee, Department of Psychiatry, National Taiwan University College of Medicine, No. 7, Chung Shan South Road, Taipei, Taiwan. DEVELOPMENT AND VERIFICATION OF VALIDITY AND RELIABILITY OF A SHORT SCREENING INSTRUMENT TO IDENTIFY PSYCHIATRIC MORBIDITY Ming-Been Lee, 1,2,3 Shih-Cheng Liao, 3 Yue-Joe Lee, 3 Chia-Hsuan Wu, 3 Mei-Chih Tseng, 3 Shur-Fen Gau, 1,3 and Chi-lun Rau 4 Background and Purpose: The accurate identification of minor mental disorders associated with depression and anxiety in non-psychiatric medical settings is an important component of mental health care. The present study aimed to develop a reliable and valid short screening tool to improve the identification of psychiatric morbidity. Methods: Data from the 50-item Brief Symptom Rating Scale (BSRS-50) obtained from 721 medical inpatients were used to develop a short screening tool (BSRS-5) to identify psychiatric morbidity. The BSRS-5 comprises 5 symptom items, selected from the BSRS-50, each of which has the highest correlation with the corresponding subscale score of Anxiety, Depression, Hostility, Interpersonal Sensitivity and Additional Symptoms in the BSRS-50. Various types of reliability and validity of the BSRS-5 were assessed in different populations, including 253 human immunodeficiency virus-1 infected outpatients, 257 psychiatric outpatients, 56 psychiatric inpatients, 100 rehabilitation outpatients with chronic low back pain, 2915 university freshmen, and 1090 community members. Results: Internal consistency (Cronbach alpha) coefficients of the BSRS-5 ranged from 0.77 to 0.90. The test-retest reliability coefficient was 0.82. Concurrent validity coefficients between the sum score of BSRS-5 and the General Severity Index of BSRS-50 ranged from 0.87 to 0.95. Choosing 6+ as the cut-off score for psychiatric cases, the rate of accurate classification of BSRS-5 was 76.3% (78.9% sensitivity, 74.3% specificity, 69.9% positive predictive value, 82.3% negative predictive value). The BSRS-5 could differentiate the severity of illness in psychiatric outpatients based on psychiatrist’s ratings using the Clinical Global Impression scale, severity of psychopathology of psychiatric inpatients between admission and discharge, levels of pain indicated by 4 dimensions of the Dallas Pain Questionnaire for outpatients with chronic low back pain, and the severity of psychopathology between university students and community members with and without suicidal ideation. Conclusion: The BSRS-5 can be used to identify psychiatric morbidity in both medical practice and the community. Key words: Mental disorders; Psychiatric status rating scales; Psychometrics; Psychopathology J Formos Med Assoc 2003;102:687-94 Psychiatric disorders presenting with anxiety, depres- sion, sleep disturbances, or other related symptoms are very common in the community, as well as in psychiatric and non-psychiatric medical settings including primary care outpatient clinics and inpatient units. 1–6 The disability and burden resulting from these disorders not only decreases an individual’s productivity, but also increases the level of health care utilization. 7 There- fore, mental health has been selected as the main topic to contend with in this century by the World Health Organization. Very low rates of correct diagnosis of psy- chiatric disorders made by non-psychiatric physicians in clinical practice have been reported. 6–8 Multiple data sets suggest that primary care physicians miss about 33% of major depressive disorder and at first pass, 90% of panic disorders. 8 The coexistence of psychiatric morbidity in physically ill patients can aggravate physical conditions, lengthen hospital stay, and increase medical health care costs. 9 Early identification and treatment of psychiatric comorbidity in physically ill inpatients could shorten hospital stay. 10 Thus, it is important to develop a reliable and valid screening tool to help non-psychiatric physicians improve their rate of early detection of psychiatric disorders.