The effectiveness of questionnaires in the determination of Contact Lens Induced Dry Eye.

PURPOSE

This study investigates how well the recently introduced Ocular Comfort Index (OCI) and the more established McMonnies Dry Eye Index are able to detect contact lens induced dry eye (CLIDE).

METHODS

Seventy volunteer contact lens wearers (mean age = 28.7 +/- 8.0; 52 female, 18 male) were grouped as CLIDE positive or CLIDE negative depending on three sets of criteria: Method I: objective (tear meniscus height, lid parallel conjunctival folds, phenol red thread test and tear break-up time); Method II: objective and/or subjective [contact lens dry eye questionnaire (CLDEQ)] (i.e. being positive by Method I and/or Method III); Method III: subjective (CLDEQ). Symptoms were measured using the McMonnies Dry Eye Index and the OCI. Discrimination was analysed by the receiver operating characteristic (ROC) curves; positive predictive values (PPV) and cut-off values were calculated.

RESULTS

ROC results for Methods I and II were statistically significant only for the McMonnies (p = 0.008; p < 0.001; respectively). In Method III ROC results were statistically significant for both the McMonnies (p = 0.001) and OCI (p = 0.005). The McMonnies showed the best prediction of Method I [0.646/10.5 (PPV/cut-off value)], while the OCI gave the best predictability in Method III only (0.518/30.6).

CONCLUSION

A McMonnies scoring of > or = 10.5 or an OCI scoring of > or = 30.6 indicates CLIDE. The McMonnies performs better in prediction of CLIDE than the OCI.