Inter-rater reliability of the Dysexecutive Questionnaire (DEX): comparative data from non-clinician respondents-all raters are not equal.

PRIMARY OBJECTIVE The Dysexecutive Questionnaire (DEX) is used to obtain information about executive and emotional problems after neuropathology. The DEX is self-completed by the patient (DEX-S) and an independent rater such as a family member (DEX-I). This study examined the level of inter-rater agreement between either two or three non-clinician raters on the DEX-I in order to establish the reliability of DEX-I ratings. METHODS AND PROCEDURES Family members and/or carers of 60 people with mixed neuropathology completed the DEX-I. For each patient, DEX-I ratings were obtained from either two or three raters who knew the person well prior to brain injury. MAIN OUTCOMES AND RESULTS This study obtained two independent-ratings for 60 patients and three independent-ratings for 36 patients. Intra-class correlations revealed that there was only a modest level of agreement for items, sub-scale and total DEX scores between raters for their particular family member. Several individual DEX items had low reliability and ratings for the emotion sub-scale had the lowest level of agreement. CONCLUSIONS Independent DEX ratings completed by two or more non-clinician raters show only moderate correlation. Suggestions are made for improving the reliability of DEX-I ratings.