Purchase this article with an account.
Mabel Crescioni, Terri L Warholak, John Daniel Twelker, Joseph M Miller, Amy L Davis, Erin M Harvey; Rasch Analysis of the Convergence Insufficiency Symptom Survey (CISS) Used to Assess Symptom Reduction with Spectacle Correction Among Students. Invest. Ophthalmol. Vis. Sci. 2014;55(13):4492.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The Convergence Insufficiency Symptom Survey (CISS) was designed to quantify the severity of symptoms associated with convergence insufficiency, such as visual fatigue, headaches and double vision. It has been validated with children aged 9 to 18 years. We assessed the validity and reliability of the CISS for measuring changes in visual symptoms with vs. without spectacle correction in 6th thru 11th grade students, most of whom had significant astigmatism.
The CISS is a 15-item questionnaire containing 5 point Likert scale items. It was administered to students enrolled in a longitudinal study on refractive error who were prescribed glasses. The CISS was administered as per instructions, with the exception that each question was asked twice: “when you are wearing your glasses” and “when you are NOT wearing your glasses”. It was administered approximately 1 year after spectacles were prescribed and provided. The fit of the 15 items to a single construct was assessed with Rasch analysis using Winsteps software (ver 3.74.0).
The response categories “infrequently (not very often)” and “fairly often” were underutilized and therefore were combined with the “sometimes” and “always” responses to improve scale function. Using fit statistics and PCA of the residuals, we determined that both versions of the questionnaire (without and with glasses) are unidimensional. As a result of fit statistics analysis, no items were removed from the “without glasses” version (person separation 1.76, infit range 0.87-1.13, outfit range 0.82-1.12) or the “with glasses” version (person separation 0.93, infit range 0.82-1.25, outfit range 0.75-1.26). Item separation and reliability for “without glasses” was 4.35 and 0.95 while for “with glasses” it was 2.31 and 0.84. The resulting 15-item without and with glasses symptoms survey appears to be reliable and target our cohort of 6th to 11th grade students appropriately.
For the purposes of our study, the CISS benefits from reducing the response options from 5 to 3 options. The refined CISS may prove to be a useful tool for measuring change in symptoms from using glasses correction among highly astigmatic children. Measuring changes in visual symptoms among school-aged children with correction deserves more attention to ensure children reap the most benefits from wearing their glasses.
This PDF is available to Subscribers Only