The reliability of interpretation of photoscreening results with the off PS-100 in Headstart preschool children.

R. Lewis,W. Marsh-Tootle
Abstract:BACKGROUND Photoscreening has emerged as one of the newest means of screening children to detect amblyogenic factors. In the preschool population, it is important to reliably screen children who are at risk for developing amblyopia. This research was undertaken to determine the reliability of the interpretation of photoscreening results in a Headstart preschool population. METHODS Fifty-four African-American children aged 3 to 5 years of age were examined using the MTI PS-100 photorefractor. Five health care professionals with no prior knowledge of photoscreening techniques were asked to perform independent, masked interpretations of a set of 54 Polaroid snapshots. Their interpretations followed a training session by a company consultant. RESULTS The kappa coefficient for reliability of interpretation between observers was 0.55. This kappa value indicates that there was moderate agreement between the observers when identifying amblyogenic conditions in a child. CONCLUSIONS Non-vision professionals using this device for the assessment of children will probably achieve only moderate levels of agreement on pass-fail decisions at first. Higher reliability reported previously by more experienced photoscreeners indicates that agreement could improve with advanced training, feedback regarding diagnostic findings, and experience. Methods to improve the technique, the device, or the training of the examiners are discussed.
What problem does this paper attempt to address?