BACKGROUND: The first aim of this study was to compare the subjective experiences of social, cognitive, and emotional problems of the patient and his/her next of kin, and explore if this related to cognitive testing. The second aim was to explore how these results reflect the patient's awareness.

METHODS: This is a subanalysis from a longitudinal study in the first year after discharge with comparisons of patient and next of kin scores on the European Brain Injury Questionnaire (EBIQ) and analyses of the relationship of their scores to an objective cognitive screening on 3 occasions. A paired t test was used to explore differences between 35 stroke patients and their next of kin on the EBIQ. Gamma analyses were made to explore the relationship between the EBIQ scores and the Barrow Neurological Institute Screening (BNIS) of higher cognitive functions, with the entire sample grouped into aware and unaware according to the BNIS item of awareness.

RESULTS: We found significant differences between the patient and next of kin assessments on all occasions, apart from at discharge. When grouped, only the group of aware patients differed significantly from their next of kin. Significant relations of the patient ratings on the EBIQ and BNIS were only found at the 1-year follow-up for the unaware group and the entire sample. Next of kin EBIQ scores did not correlate with the BNIS.

CONCLUSIONS: Neither next of kin ratings nor an objective measurement was feasible to use for evaluation of the patient's awareness of social, cognitive, and emotional problems.

OBJECTIVE: We sought to examine the reliability and validity of the European Brain Injury Questionnaire as an evaluation tool in a stroke population by applying the Rasch methodology.

METHODS: The European Brain Injury Questionnaire was distributed to 54 patients with stroke and 36 next of kin at discharge from the rehabilitation ward and at a 1-year follow-up. Rasch analysis was used for evaluating the psychometric characteristic and the quality of the measures produced with focus on reliability and validity. The reliability was determined by the separation. The construct validity was determined by examining the hierarchy of the tasks and by evaluating the fit of individual tasks to the latent construct. Analyses of differences across occasions, and across patients and next of kin, were made to examine the stability of the instrument.

RESULTS: The reliability was good as the instrument met the criteria for separation. Two items at discharge and 5 items at 1-year follow-up were misfits. If not more than 5% of the items fail to fit the Rasch model the instrument can be considered unidimensional, which in this case only concerned the 1-year follow-up. The instrument was stable over time and across patients and next of kin.

CONCLUSION: With some alterations the European Brain Injury Questionnaire seems to be a useful instrument in clinical practice and research, and a help to capture the social, emotional, and cognitive impacts of a stroke.