Validation of the Behavioural Inattention Test (BIT) in patients with acquired brain injury in Turkey.

The aim of this descriptive study was to evaluate the construct validity and reliability of the Behavioural Inattention Test (BIT) in patients with acquired brain injury in Turkey. One hundred and eighteen acquired brain injury patients undergoing rehabilitation were assessed by the BIT. Internal construct validity was tested by Rasch analysis; reliability by internal consistency and the Person Separation Index; and external construct validity by associations with physical and cognitive disability. Analysis of the data revealed that some subtests deviated from Rasch model expectation and the conventional subscale of the BIT had an unsatisfactory reliability for individual use. Consequently, a common 10-item scale (BIT-10) was derived from both the behavioural and conventional subscales of the BIT. Reliability of .87 met expectation for individual use. The BIT-10 correlated at .52 with cognitive disability upon admission. As a conclusion the original BIT adapted for use in Turkey was shown to lack reliability and internal construct validity. A revised 10-item new version, BIT-10, gave a valid unidimensional summed score, with high sensitivity and specificity to the original cut points. Reliability of the BIT-10 was high and external construct validity was as expected.