Reliability and validity of the Hebrew version of the International Spinal Cord Injury Activities and Participation Basic Data Set.

Forty-seven people were analysed, and the paper says it is unclear whether this represented the spinal cord injury population of Israel, partly because 20-25% of the Israeli population does not speak Hebrew fluently. Three of the fifty recruited participants were removed after consent, and answers of unknown or not relevant nullified whole domains, which shrank the number of usable responses, especially for subjective occupation. The Basic Data Set was administered in different ways: 44 face to face, 3 by phone, and repeat testing was mainly self-administered, with some face-to-face if more than 10 days had elapsed. The same interviewer may have paraphrased, and phone participants had no form or comments. Only 31 of 43 participants completed the repeat questionnaire within the allotted time with unchanged clinical status. The original English Basic Data Set does not appear to have been validated, so it is hard to know whether poor results come from translation or the source instrument. The authors created total scores by standardizing items into ordinal categories and adding them, even though the original Basic Data Set does not have a summation score. Only one interviewer administered the questionnaire, so different raters were not tested.

Do not read this as proof that the Hebrew Basic Data Set measures all spinal cord injury activity and participation domains well. The self-care and total objective scores were the most dependable, but validity was low to moderate overall, and subjective occupation did not reach significance on repeat testing.