A comparison of responsiveness indices in multiple sclerosis patients.

Responsiveness was measured in a number of health-related quality of life (HRQoL) instruments among which two generic (SF-36 and COOP/WONCA Charts) and one disease-specific instrument, the Disability & Impact Profile (DIP). Subjects were 162 multiple sclerosis patients. The following responsiveness indices were used: effect size, standardised response mean and smallest real difference (SRD). The latter measure gives an indication of the magnitude of real change, i.e. change not attributable to 'noise' or 'error' and can thus be used for the interpretation of change scores in clinical practice whereby change scores larger than the SRD value indicate real change. It is assumed that low SRD values indicate high responsiveness. The results confirmed our expectation that the effect size and standardised response mean are probably less suitable for use in slowly progressive diseases, since they use the average change as the numerator. Therefore, the article focused on the SRD. Compared to scales, items measured on a visual analogue scale show high SRD values. The DIP scales generally show lower SRD values compared to scales of other questionnaires. The SRD seems to be a promising new measure to study responsiveness. More research into the interpretation of this measure is necessary.