Abstract
Purpose :
Cerebral visual impairment (CVI) is the leading cause of visual impairment in children. Given the diversity of clinical presentations of CVI, we are interested in whether questionnaires appropriately target the spectrum CVI cases, specifically the TeachCVI Screening Tool. Rasch analysis is a common psychometric technique for assessing the targeting of questionnaire items, however to date, this analysis technique has not yet been applied to this questionnaire.
Methods :
We performed a retrospective review of clinical CVI cases from the NECO Center for Eye Care and Perkins School for the Blind from January 1, 2016 to December 31, 2021. Electronic medical records were reviewed to identify patients with an ICD-9 or ICD-10 code of CVI or other neurological visual impairment. Age, gender, diagnoses, visual acuity, contrast sensitivity, visual fields, ocular alignment, and TeachCVI survey responses were collected.
We applied the method of successive dichotomizations, a polytomous Rasch model, to estimate item measures and person measures from the three levels of the TeachCVI survey. Targeting of questionnaire items to the sample population was explored by comparing estimated item measures to person measures. Multiple linear regression was used to determine which factors influence patient functional ability (i.e., person measure).
Results :
102 patients fit our inclusion criteria; 51% were female. The mean (SD) age was 9.49 (5.7) years, with a range of 0 to 33. 8.8% of patients had no impairment, 5.9% only had CVI, 38% had CVI and an ocular impairment, 8.8% had CVI and a physical disability, and 38% were multiply impaired (CVI, ocular, and physical). Mean (SD) visual acuity was 0.40 (0.40) logMAR, mean (SD) contrast sensitivity was 1.58 (0.47) logCS, and 33% had a visual field deficit.
Estimated item measures were well-targeted to the persons: the mean item measure was 0 by convention with an SD = 0.92 logits, and the mean (SD) person measure was 0.27 (1.19) logits. The only statistically significant factor in influencing patient functional ability was the level of impairment, i.e., the presence of multiple comorbidities. Visual acuity, age, gender, contrast sensitivity and visual field loss were not statistically significant.
Conclusions :
This study shows that TeachCVI surveys are well targeted and are appropriate screening tools for CVI.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.