PURPOSE: To evaluate the validity of a dry eye (DE) screening method in visual display terminal (VDT) workers using functional visual acuity (VA) measurement combined with a symptoms questionnaire, and to make recommendations regarding practical use of the screening. METHODS: This prospective cross-sectional study included 369 young and middle-age office workers, who use VDTs. Subjects completed two types of DE questionnaires and underwent DE testing and functional VA measurement. Subjects were assigned to DE and non-DE groups. Sensitivity and specificity of the DE prediction value were assessed. According to the probability distribution of DE and non-DE calculated by the discriminant analysis formula, we classified subjects into three categories based on the most appropriate cut-off levels. Positive predictive value (PPV) and negative predictive value (NPV) of the DE prediction value were also assessed. RESULTS: Of 369 subjects, 218 and 151 subjects were assigned to the DE and non-DE groups, respectively. Sensitivity and specificity were 93.1% and 43.7%, respectively. Based on the probability prediction, the cut-off levels were defined as probability prediction ≥ 85%, 45% ≤ probability prediction < 85% and probability prediction < 45%. Subjects with a probability prediction ≥ 85% were defined positive (N = 130), while those with values < 45% were defined as negative (N = 73). The PPV and NPV were 83.8% and 80.8%, respectively. CONCLUSIONS: Acceptable PPV and NPV were obtained by setting three discriminate categories. This improved version of the DE screening method may be acceptable for practical use in VDT workers. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/
PURPOSE: To evaluate the validity of a dry eye (DE) screening method in visual display terminal (VDT) workers using functional visual acuity (VA) measurement combined with a symptoms questionnaire, and to make recommendations regarding practical use of the screening. METHODS: This prospective cross-sectional study included 369 young and middle-age office workers, who use VDTs. Subjects completed two types of DE questionnaires and underwent DE testing and functional VA measurement. Subjects were assigned to DE and non-DE groups. Sensitivity and specificity of the DE prediction value were assessed. According to the probability distribution of DE and non-DE calculated by the discriminant analysis formula, we classified subjects into three categories based on the most appropriate cut-off levels. Positive predictive value (PPV) and negative predictive value (NPV) of the DE prediction value were also assessed. RESULTS: Of 369 subjects, 218 and 151 subjects were assigned to the DE and non-DE groups, respectively. Sensitivity and specificity were 93.1% and 43.7%, respectively. Based on the probability prediction, the cut-off levels were defined as probability prediction ≥ 85%, 45% ≤ probability prediction < 85% and probability prediction < 45%. Subjects with a probability prediction ≥ 85% were defined positive (N = 130), while those with values < 45% were defined as negative (N = 73). The PPV and NPV were 83.8% and 80.8%, respectively. CONCLUSIONS: Acceptable PPV and NPV were obtained by setting three discriminate categories. This improved version of the DE screening method may be acceptable for practical use in VDT workers. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/