Reliability of Postal Questionnaires for the Glasgow Outcome Scale

Abstract

The purpose of the study was to investigate if a questionnaire sent by mail can give a reliable assessment of outcome on the Glasgow Outcome Scale. A questionnaire was developed for the Glasgow Outcome Scale (GOS) and a second questionnaire for the Extended Glasgow Outcome Scale (GOSE). The questionnaires were self-contained and designed to be completed by either a head-injured person or a proxy. The questionnaires were studied in two ways: each questionnaire was administered twice (at an interval of approximately 2 weeks), and ratings from the postal questionnaires were compared to ratings from a structured interview conducted by telephone. The four studies were carried out in separate groups of head-injured participants consisting of 32-38 individuals. For the test-retest comparison, κ _w(quadratic weights) was 0.94 for the GOS questionnaire and 0.98 for the GOSE questionnaire. For the comparison with the telephone interview, k_w was 0.67 for the GOS and 0.92 for the GOSE. The values of κ _w indicate good agreement for all comparisons. We conclude that it is possible to obtain reliable outcome data after head injury using postal questionnaires.