Reliability and validity of Turkish version of DSM-5 psychosis scale [DSM-5 psikoz ölçeği Türkçe formunun geçerliliği ve güvenilirliliği]

Objective: In this study we evaluated the reliability and validity of the Turkish version of DSM-5 Psychosis Scale developed in order to evaluate severity of psychosis in adult patients. Methods: This study was carried out with patients who met the diagnostic criteria for any psychotic disorder according to DSM-5 criteria and who are undergoing treatment in the inpatient and outpatient wards of Psychiatry Department of Celal Bayar University Medical School. Except for psychotic disorders patients with any mental or physical disease were excluded. A total of 49 patients (40 patients with schizophrenia, six patients with not otherwise specified psychotic disorder, three patients with schizoaffective disorder) were included in the study. One hundred and one volunteers without any mental or physical disease were included as the healthy control group. Reliability analysis was based on internal consistency and item-total score correlation. Validity analysis was based on exploratory factor analysis, comparison of mean scores of the patient and healthy control groups and ROC analysis. Results: The mean age of the study group was 26.0±10.3 years and 52.7% of the sample group (n=79) were female. Sixty-nine point three percent of the study group were college graduates, 8% primary school graduates, 19.3% is also high school graduates. Mean disease duration was 17.1±9.7 years in the psychotic disorder group. The internal consistency of the Psychosis Scale was 0.91 and, item-total correlation coefficients of the Psychosis Scale were between 0.25 and 0.82. In the factor analysis, for sample adequacy Kaiser- Meier-Olkin (KMO) coefficient was found as 0.81 and Bartlett coefficient as 1.03. Single-factor solution was obtained and the eigenvalue is 5.12, representing 64.0% of the total variance. Factor loadings of the items of the scale were between 0.33 and 0.88. The mean of score was 11.00±5.20 in patient group while it was calculated as 0.21±1.21 in control group. The area under the curve was calculated as 0.98 in the ROC analysis. Conclusion: These results indicate that the Turkish version of Psychosis Scale is reliable and valid. © 2017, Cukurova University, Faculty of Medicine. All rights reserved