Abstract
Purpose::
To describe the psychometric properties of the Impact of Vision Impairment for Children (IVI_C) questionnaire and appraise its validity and reliability.
Methods::
The 28-item IVI_C questionnaire, developed from focus groups without a priori domains, was designed to measure the impact of vision impairment on participation in school and daily activities for students aged 8 to 18 years. Each item is rated on a 5-point Likhert scale ranging from always (1) to never (5). The IVI_C was administered to 126 low vision students (<0.3LogMAR) and 36 age-matched, sighted students. Distance and near visual acuity (VA), contrast sensitivity, socio-demographic, service support and school activity data were also collected. 3 sub-sets of students completed the IVI_C twice, 3 to 6 weeks apart or by different forms of administration (face or telephone).
Results::
The mean age and distance VA values for the low vision group (63% male) were 12.8 years (±2.9) and 1.1LogMAR (SD=0.80). 24 items (86%) were considered relevant by 90% or more of the respondents and 25 items (89%) recorded responses across the full range. 2 items with >20% of irrelevant responses and a third with ambiguous scoring were removed. Only 2 pairs of items had Spearman correlation of 0.62 and 0.63 indicating minimal redundancy. The increase in IVI_C total score of the retained 25 items indicates greater difficulty in participation and correlated with worsening distance and near VA (r=0.35 and 0.34;p<0.01 respectively). Internal consistency of total mean score was high (α=0.84) and was not affected by sequential elimination of items. There was a significant difference in the total mean score between the sighted and low vision groups supporting discriminant validity (ANOVA;F=92;p<0.001). The Guttman split-half correlations for temporal, mode (telephone and face) and inter-observer were 0.95, 0.90 and 0.72, respectively. Principal components analysis demonstrated that the 25 items contributed to one underlying theme and that 3 factors (social & academic access, 21.8%; orientation & mobility, 10.2%; community acceptance, 6.8%) explained 39% of the variance in the data. The internal consistency for factors 1 and 2 was high and moderate for factor 3 (α=0.78, 0.79 and 0.61 respectively).
Conclusions::
The IVI_C possesses demonstrated reliability and validity. The findings represent an important step toward providing assessment of participation in daily activities and quality of life in children with low vision. Future validation of the IVI_C will be to determine the measurement characteristics, unidimensionality and item-person targeting using Rasch analysis.
Keywords: low vision • quality of life