A Comparison Of Economic, Administrative, Patient Preference, And Predictability Involving Three Methods Of Dry Eye Determination.

Our study involved a comparison of economic, administrative, patient preference, and the predictability of determining dry eye symptoms in 30 patients. The three tests compared were the Schirmer modified test, the Lactoplate Lactoferrin Immunoassay test, and the not as yet available in the U.S., Cotton Thread Tear test developed by Hamano, et.al. 100%
of the patients preferred the Cotton Thread test compared to the other two tests, while 80% preferred the Schirmer over the Lactoplate. A subjective survey of patient discomfort induced by the testing procedure (based on a scale of 0 - no discomfort to 10- extreme discomfort) revealed a Lactoplate mean of 7.43, Schirmer of 5.53, and a Cotton Thread mean discomfort level 0.93. Economic factors included test cost, test shelf life, and the total amount of practitioner time necessary for the test. There was a large disparity between the Schirmer and Cotton Thread Tests, which involved minimal practitioner time and cost (the exact cost of the Cotton Thread test was not ascertained at this time, but should be comparable to the Schirmer, based upon the simplicity of the material), and the Lactoplate which involves a significant increase in time and cost, and which has a limited shelf life. A subjective measurement of the ease of practitioner test administration and patient cooperation (based on a scale of o - very easy to 10 - very difficult) revealed a Lactoplate mean of 2.80, a Schirmer mean of 1.43, and a Cotton Thread mean of 0.48. Two methods of predictability were employed. The first compared the subjective symptoms with the objective findings
usinq the established cut-off for definitive dry eye. With this method the Schirmer Test demonstrated the highest level of predictability with a sensitivity of 50% and a specificity of
86.36% . . This compared with a Lactoplate sensitivity of 12.5% and a specificity of 100%, and a Cotton Thread sensitivity of 12.5% and specificity of 95.45%. The second method involved a comparison of the means and standard deviations of the test results of asymptomatic versus symptomatic patients. Utilizing this method, the mean of the asymptomatic group was greater than one standard deviation below the mean of the asymptomatic group for all three tests.