Purchase this article with an account.
Marguerite Huff, Meghan Park, Michael Choi, Daniel Sarezky, Catherine Clark, Clara Choo, Jie Peng, Alfred Rademaker, Angelo Tanna; Inter-rater reproducibility in the assessment of video recordings of eye drop instillation by glaucoma patients. Invest. Ophthalmol. Vis. Sci. 2013;54(15):459.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To determine the level of agreement of eye drop instillation efficacy, safety and efficiency performed by glaucoma patients by three masked graders based on assessment of video recordings.
Subjects with open-angle glaucoma or ocular hypertension who had at least 6 months of experience with self-instillation of ocular hypotensive eye drops were recruited for a study on eye drop instillation. Subjects were video recorded while self-instilling artificial tears sequentially to both eyes. The video recordings of the first 39 subjects were independently graded by 3 masked observers. Efficacy was assessed based on whether the subject successfully instilled at least one eye drop on the ocular surface (binary). Safety was assessed based on whether the bottle tip made contact with the ocular surface or periocular skin (binary). Efficiency was graded as the number of drops expelled. Cases in which a continuous stream of drops was expressed were graded as 5 drops. Kappa statistics (weighted kappa for efficiency) were used to estimate inter-rater agreement. Analyses were done separately for left and right eye. Kappas and their standard errors were averaged across the three raters and two eyes.
Efficacy: The mean kappa level of agreement for efficacy was 0.69 (95% CI, 0.39-0.98). The range of grader scores of percentages of patients with successful eye drop instillation was 79.5-84.6% and 74.36-84.62% for the right and left eye, respectively. Safety: The mean kappa level of agreement for safety was 0.84 (95% CI, 0.66-1.00). The range of grader scores of percentages of patients making no contact between the bottle tip and ocular or skin surface was 61.5-66.7% and 64.1-74.4% for the right and left eye, respectively. Efficiency: The mean kappa level of agreement for efficiency was 0.57 (95% CI, 0.31-0.82). The range of grader scores of percentages of patients using 1 drop per instillation was 64.1-74.4% for both the right and left eye separately.
In reviewing digital video recordings of eye drop instillation by patients, there was excellent agreement among the three masked observers in grading safety. Agreement for efficacy was good. Agreement of eye drop instillation efficiency was only fair, indicating that the determination of the number of drops expressed from an eye drop bottle based on digital video recordings may be imprecise.
This PDF is available to Subscribers Only