advertisement
Editors Selection IGR 13-4
Quality of Life: Visual disability assessment
Comment by Rohit Varma on:
23945 An evaluation of the reliability and validity of the visual functioning questionnaire (VF-11) using Rasch analysis in an Asian population, Lamoureux EL; Pesudovs K; Thumboo J et al., Investigative Ophthalmology and Visual Science, 2009; 50: 2607-2613
In this cross-sectional assessment Lamoureux et al. (506) used Rasch models to determine whether an instrument used to assess visual disability ‐ the VF 11 and a shortened version the VF 9, provide a valid measure in 618 participants in the Singapore Malay Eye Study. When assessing patient reported outcomes, two approaches were used: (i) the classic test theory and (ii) the item response theory. The classic test theory reates a mean score for a response to a question and assumes that the distance between the various responses (e.g.: never, infrequently, sometimes, a lot of the time or all the time) is similar and thus should be weighted similarly. The IRT does not assume such a similar weight but rather models responses assuming that the responses can be varied depending on the person and the question being asked. In this study, the authors utilize a specific type of IRT ‐ the Rasch model to determine that the VF 11 instrument could be collapsed into a VF 9. Further, the authors determined that while the VF 9 was a valid questionnaire, it was unable to discriminate between the impact of visual impairment between persons who had mild, moderate or severe visual impairment (based on visual acuity assessment). The authors are to be commended for utilizing this technique in assessing questionnaires. However, this newer IRT approach should be utilized with appropriate caution as there are some important caveats that should be considered: (i) Rasch models are concerned primarily with the responses of individuals not the distribution in a population ‐ thus, use of these models in measuring differences in the population may not be appropriate; (ii) Rasch models determine whether the a priori requirements for measurements have been met ‐ in this case the relationship between responses to items and visual impairment based on central visual acuity. The response of the person is based not solely on one aspect of vision such as central visual acuity, but also on other independent measures related to vision-specific tasks, such as scotopic contrast sensitivity, stereo-acuity, peripheral visual field abnormalities. Thus, while this new approach provides us with some insight into the validity of a questionnaire and its relationship to visual acuity, it would be unwise to stop assessing mean differences in sub-scale and summary scores for various patient-reported outcomes. Further work assessing the overall differences of the two approaches utilizing other instruments need to be conducted.
Comments
The comment section on the IGR website is restricted to WGA#One members only. Please log-in through your WGA#One account to continue.Log-in through WGA#One
