G E Quinn1, J A Berlin, M James. 1. Division of Pediatric Ophthalmology, Children's Hospital of Philadelphia, PA 19104.
Abstract
PURPOSE: The Teller acuity card procedure provides a quantitative measure of grating acuity in infants and young children and has been used in clinical and laboratory settings. The purpose of this study is to address the influence of the individual testers in a clinical setting. The authors report a comparison of acuity results obtained by three well-trained, experienced testers in a busy hospital-based clinical practice. METHODS: Three different testers (A, B, C) performed 971 consecutive acuity assessments using the Teller acuity card procedure on 570 patients. The testers used similar techniques and adaptations, depending on clinical need. Diagnostic categories include nystagmus, aphakia, strabismus, glaucoma, retinopathy of prematurity (ROP), ptosis, optic nerve abnormalities, etc. Patients ranged in age from 2 weeks to 17 years (mean +/- standard deviation, 23 +/- 20 months). RESULTS: When each patient was represented only once, A tested 144 patients; B, 251; and C, 175. The geometric mean right eye acuity scores of A were 3.00 cycles/degree (cy/deg) (0.34 octaves, standard error [SE]; n = 134); B, 4.05 cy/deg (0.20 octave, SE; n = 225); and C, 3.71 cy/deg (0.30 octave, SE; n = 163). An analysis of covariance showed a significant difference in the mean log scores across testers (P = 0.02 adjusted for both age and diagnosis). Age was a strong predictor of acuity (P = 0.0001), and the interactions between tester and age and tester and diagnosis were not significant. The geometric mean left eye acuity scores for A were 2.86 cy/deg (0.33 octave, SE; n = 132); B, 3.46 cy/deg (0.23 octave, SE; n = 218); and C, 3.32 cy/deg (0.30 octave, SE; n = 160). The differences for left eyes among testers were not significant (P = 0.37 adjusted for age and diagnosis). CONCLUSIONS: These results suggest that the tester may influence the acuity results measured by the Teller acuity card procedure in a clinical setting. Evaluation of interobserver reliability in clinical subjects should be a high priority of further work using the Teller acuity card procedure.
PURPOSE: The Teller acuity card procedure provides a quantitative measure of grating acuity in infants and young children and has been used in clinical and laboratory settings. The purpose of this study is to address the influence of the individual testers in a clinical setting. The authors report a comparison of acuity results obtained by three well-trained, experienced testers in a busy hospital-based clinical practice. METHODS: Three different testers (A, B, C) performed 971 consecutive acuity assessments using the Teller acuity card procedure on 570 patients. The testers used similar techniques and adaptations, depending on clinical need. Diagnostic categories include nystagmus, aphakia, strabismus, glaucoma, retinopathy of prematurity (ROP), ptosis, optic nerve abnormalities, etc. Patients ranged in age from 2 weeks to 17 years (mean +/- standard deviation, 23 +/- 20 months). RESULTS: When each patient was represented only once, A tested 144 patients; B, 251; and C, 175. The geometric mean right eye acuity scores of A were 3.00 cycles/degree (cy/deg) (0.34 octaves, standard error [SE]; n = 134); B, 4.05 cy/deg (0.20 octave, SE; n = 225); and C, 3.71 cy/deg (0.30 octave, SE; n = 163). An analysis of covariance showed a significant difference in the mean log scores across testers (P = 0.02 adjusted for both age and diagnosis). Age was a strong predictor of acuity (P = 0.0001), and the interactions between tester and age and tester and diagnosis were not significant. The geometric mean left eye acuity scores for A were 2.86 cy/deg (0.33 octave, SE; n = 132); B, 3.46 cy/deg (0.23 octave, SE; n = 218); and C, 3.32 cy/deg (0.30 octave, SE; n = 160). The differences for left eyes among testers were not significant (P = 0.37 adjusted for age and diagnosis). CONCLUSIONS: These results suggest that the tester may influence the acuity results measured by the Teller acuity card procedure in a clinical setting. Evaluation of interobserver reliability in clinical subjects should be a high priority of further work using the Teller acuity card procedure.