Joan A Stelmack1, Robert W Massof. 1. Blind Rehabilitation Center, Edward Hines Veterans Affairs Hospital, Hines, Illinois 60141, USA. Joan.Stelmack@med.va.gov
Abstract
PURPOSE: This study was conducted to demonstrate use of a simple scoring algorithm for the 48-item Veterans Affairs Low Vision Visual Functioning Questionnaire (VA LV VFQ-48) that approximates the measure of persons' visual ability that would be calculated with Rasch analysis and to provide a short form version of the questionnaire for clinical practice and outcomes research. METHODS: Items were eliminated from the VA LV VFQ-48 to reduce redundancy and shorten the instrument. The approximation of persons' visual ability calculated with the scoring algorithm for vision function questionnaires developed by Massof was compared with the person measure estimated from Rasch analysis for a sample of 126 subjects entering a low vision rehabilitation program. RESULTS: The approximation captures 98% of the variability in the Rasch measure estimate of persons' visual ability and 97% of the variability in the change score estimate. The relationship does not hold in circumstances where patients with high visual ability find most items to be easy. A 20-item short form of the instrument was constructed for use in low vision service delivery. CONCLUSIONS: The scoring algorithm can be used with the VA LV VFQ-48 or short form versions of the questionnaire. Clinicians can use the algorithm to score the VA LV VFQ from examination of individual patients or as an outcome measure for their low vision rehabilitation programs. Research investigators can use the scoring algorithm with small samples when Rasch analysis is not reliable or in studies where Rasch analysis is not practical. Rasch analysis is still recommended for research studies that require more accurate assessments.
PURPOSE: This study was conducted to demonstrate use of a simple scoring algorithm for the 48-item Veterans Affairs Low Vision Visual Functioning Questionnaire (VA LV VFQ-48) that approximates the measure of persons' visual ability that would be calculated with Rasch analysis and to provide a short form version of the questionnaire for clinical practice and outcomes research. METHODS: Items were eliminated from the VA LV VFQ-48 to reduce redundancy and shorten the instrument. The approximation of persons' visual ability calculated with the scoring algorithm for vision function questionnaires developed by Massof was compared with the person measure estimated from Rasch analysis for a sample of 126 subjects entering a low vision rehabilitation program. RESULTS: The approximation captures 98% of the variability in the Rasch measure estimate of persons' visual ability and 97% of the variability in the change score estimate. The relationship does not hold in circumstances where patients with high visual ability find most items to be easy. A 20-item short form of the instrument was constructed for use in low vision service delivery. CONCLUSIONS: The scoring algorithm can be used with the VA LV VFQ-48 or short form versions of the questionnaire. Clinicians can use the algorithm to score the VA LV VFQ from examination of individual patients or as an outcome measure for their low vision rehabilitation programs. Research investigators can use the scoring algorithm with small samples when Rasch analysis is not reliable or in studies where Rasch analysis is not practical. Rasch analysis is still recommended for research studies that require more accurate assessments.
Authors: Lauren N Ayton; Joseph F Rizzo; Ian L Bailey; August Colenbrander; Gislin Dagnelie; Duane R Geruschat; Philip C Hessburg; Chris D McCarthy; Matthew A Petoe; Gary S Rubin; Philip R Troyk Journal: Transl Vis Sci Technol Date: 2020-07-16 Impact factor: 3.283
Authors: Lauren N Ayton; Nick Barnes; Gislin Dagnelie; Takashi Fujikado; Georges Goetz; Ralf Hornig; Bryan W Jones; Mahiul M K Muqit; Daniel L Rathbun; Katarina Stingl; James D Weiland; Matthew A Petoe Journal: Clin Neurophysiol Date: 2019-12-10 Impact factor: 3.708
Authors: Heather E Whitson; Sandra Woolson; Maren Olsen; Robert Massof; Stephanie M Ferguson; Kelly W Muir; John A Dziadul; Eleanor McConnell; Joan A Stelmack Journal: Optom Vis Sci Date: 2020-06 Impact factor: 2.106