7. The Special Case of Validity and PROs
Questions: Did the results provide sufficient support for the validity of the questionnaire?
Answer: The results provided moderate support for the validity of the questionnaire, but more data are required before the instrument can be used with confidence. This example illustrates that validation is not an all-or-nothing process. We may have varying degrees of confidence that an instrument is really measuring what it is supposed to measure. The sort of a priori predictions that characterize the validation of the Inflammatory Bowel Disease Questionnaire strengthen the validation process. Without such predictions, it is too easy to rationalize the observed correlations. Validation does not end when the first study with data concerning validity is published but continues with repeated use of an instrument. The more frequently an instrument is used, and the more situations in which it performs as expected, the greater our confidence in its validity. Perhaps we should never conclude that a questionnaire has “been validated”, but rather we should suggest that strong evidence for validity have been obtained in a number of different settings and studies.