Performance of the VA low vision visual functioning questionnaire (VA LV VFQ)
 
Z. Ardickas, MA,1,2 J.P. Szlyk, Ph.D.,1,2 J. Stelmack, OD,2-4 R.W. Massof, Ph.D.,5 M. Kuzia, MS,1,2 T.R. Stelmack, OD,1,2,4 B.D. Wright, Ph.D.6 and the LVOS Group.
 
1Chicago VA Health Care System/West Side, 2University of Illinois at Chicago, 3Hines VA Blind Rehabilitation Center, 4Illinois College of Optometry, 5Johns Hopkins Wilmer Eye Institute, 6University of Chicago.
 
Objectives: Our goal was to develop an instrument that is sensitive to the effects of low vision rehabilitation.
 
Methods: Eighty-five patients (mean age=72.4 yrs), with a range of vision loss from mild impairment to total blindness, from three sites (Hines Blind Rehab Center, VA West Side VICTORS Program, Deicke Program) were recruited and interviewed by phone prior to their rehab program admission. Our interview guide contains 48-items on which the respondent is asked to rate (a) their difficulty to perform a task, (b) whether their difficulty was related to their vision, (c) whether they want training, and (d) how they currently perform the task. Results: Data were analyzed with a Rasch rating scale model. The difficulty question showed strong separation and high reliability (S=4.04, R=0.94). The item ordering for the difficulty question (items ranked from easiest to impossible) was highly correlated with the item ordering for the following questions: (1) whether difficulty was related to their vision (R=0.87), and (2) whether they wanted training (R=0.85). Both of these correlations increased to 0.98 with the removal of 9 ineffective items that included doing yard work and playing sports. The rankings for the vision and the training questions were highly correlated, R=0.98 with all 48 items.

Figure: Item difficulty scale generated from Rasch analysis showing the distribution of patient responses for those items rated not difficult to impossible. Red highlighted items at the top of the diagram indicate items rated as not difficult. Red highlighted items at the bottom of the diagram indicate items rated as extremely difficult. The vertical line drawn down the center of the diagram indicates the mean measure for the patient group for each question. The item ordering for the (subcomponent a) difficulty question was highly correlated with the item ordering for (subcomponent b) whether difficulty was related to vision (R=0.87). Also, the item ordering for the difficulty question was highly correlated with the item ordering for (subcomponent c) whether they wanted training (R=0.85).
 
Conclusions: The results show that the VA LV VFQ is an effective instrument to measure vision difficulties in low vision rehab patients in VA-based programs.
 
Funding Acknowledgment: This study was funded by the Department of Veterans Affairs, Rehabilitation Research and Development Service, project # O2086RA.