Reliability and Validity of The Turkish Version of Cognitive Assessment Interview (CAI-TR)

Objective: The Cognitive Assessment Interview (CAI) is an interview-based scale developed to assess cognitive functioning of patients diagnosed with schizophrenia. It is scored by a clinician according to patient and informant interview. This study aimed to determine the reliability and validity of the CAI-Turkish Version (CAI-TR).

Methods: The Cognitive Assessment Interview (CAI) was translated to Turkish and back to English and the back-translated form of the scale was approved by Dr. Ventura, its original developer. The translated text was administered to five schizophrenia outpatients as a pilot study and was deemed appropriate following a joint review by the research team. Ninety clinically stable outpatients with schizophrenia or schizoaffective disorder diagnosis were evaluated by clinicians using SCID-I (CV), PANNS, and Social Functioning Scale. To assess the patients’ neurocognitive status, certified expert psychologists administered a neurocognitive test battery including Öktem Verbal Memory Process Test, Wechsler Memory Scale-Digit Span, Verbal Fluency Test, Continuous Performance Test, Trail Making Test-A, Tower of London Test, The ‘’Reading the Mind in the Eyes’’ test (Eyes Test), and Facial Emotion Identification and Discrimination Test.

Results: Internal consistency of CAI-TR was good, with Cronbach’s alpha value of 0.97. For patient scores, Cronbach’s alpha value was 0.91. Each item of CAI-TR was correlated with the related neurocognitive test (r=0.242-0.564; p<0.05). Moreover, overall scores of CAI-TR showed statistically significant correlations with Global Assessment of Functioning (GAF) (r=-0.538, p<0.001), social functioning (r=-0.520; p<0.01), and objective neurocognitive tests. As a measure of external validity of CAI-TR, statistically significant correlations were determined between patient, informant and interviewer evaluations independent of source of information (r=0.707, r=0.830, r=0.835, respectively; p <0.001 for all). Mean duration of patient interview was 18.7 minutes (8-30 min; SD=5.4), mean duration of informant interview was 18.0 minutes (10-25 min; SD=5.0) and total mean duration of CAI administration was 36. 6 minutes (18-55 min; SD=9.7).

Discussion: The analysis indicated that CAI-TR was a reliable and valid instrument to evaluate cognitive functioning. Ratings obtained from patient interview were also significant. CAI-TR is found to be a very practical and useful tool with some additional advantages such as being a clinically based interview with a brief administration time providing information about patients’ functioning.