Abstract
Purpose :
To evaluate the measurement characteristics of the Japanese Impact of Vision Impairment IVI (IVI-J) using Rasch analysis and investigate the relationship between the severity of visual impairment and vision-related quality of life (VRQoL) among outpatients with visual impairment in Japan.
Methods :
This was a clinic-based and cross-sectional study of 226 patients with Log MAR >0.3 visual acuity in the better eye, recruited from outpatient clinics at Toho University Omori Hospital. They underwent a clinical examination and completed the 28-item IVI-J. The IVI-J underwent Rasch analysis to assess the scale’s unidimensionality, item fit, measurement precision, response option functioning, differential item functioning (DIF; item bias) and targeting.
Results :
The initial analysis of IVI-J showed that it was multidimensional (1st contrast eigenvalue >2 and variance explained by measures <50%). The scale was split into its component domains of Reading, Mobility and Emotional, and a separate Rasch analysis was conducted for each. Response categories were disordered and improvements in precision were observed in Reading, Mobility, and Emotional upon collapsing categories (PSI-Person-Separation-Index: 1.60, 1.52 and 1.67, respectively). At the end of the modification process, each domain was unidimensional, had ordered thresholds, minimal item misfit and no DIF, although measurement precision remained lower than optimal (i.e., PSI <2.0). The Rasch-scaled IVI-J demonstrated acceptable criterion validity with domain scores decreasing as severity of better eye VI (normal, mild, moderate, and severe) increased (p<0.001). Test-retest reliability was also high overall (intraclass correlation coefficient range: 0.75 to 0.90).
Conclusions :
The re-engineered three-domain 28-item J-IVI is a psychometrically valid and reliable tool to assess the impact of VI and the effectiveness of ophthalmic interventions on VRQoL in a hospital-based sample.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.