Using a nationally-representative, population-based sample, this cross-sectional study investigated the relationship between GBE use over the past 12 months and self-reported glaucoma. Although past animal and human studies have shown promising results with regard to the effects of GBE use on visual function [4
], the results of this study suggest that there is not a significant association with glaucoma.
However, several limitations need to be addressed to draw firmer conclusions. One possible explanation for the lack of a significant association relates to the cross-sectional nature of the study design. The NHIS does not collect information regarding the duration of GBE use nor glaucoma, and thus it was not possible to address the issue of temporality. Even with such data, while it would be possible to limit GBE use to that which occurred prior to glaucoma diagnosis, the etiologically relevant duration and amount of use remains unknown and could only be investigated in an exploratory manner. Another possible explanation relates to the self-reported nature of the NHIS data. If misclassification exists with respect to GBE use and/or glaucoma then a truly protective association may be obscured. However, while non-differential misclassification may bias associations towards the null it should have little effect on the relative strength of the associated risk factors to the disease. Thus, given that other known risk factors such as age, race, smoking status, hypertension, CHD, angina, stroke, myocardial infarction, diabetes, cataracts, diabetic retinopathy, and macular degeneration did show significant associations with glaucoma as defined in this study, the finding that GBE use was not associated indicates that it is a relatively small association compared with other known risk factors.
A third possible explanation is that information regarding suspected risk factors for glaucoma (e.g., IOP, family history) was not available thus introducing the possibility that the observed results are subject to residual confounding. Finally, it is possible that the observed effects of GBE use in human crossover trials and animal studies, if true, are not sufficient to reduce the risk of glaucoma.
In summary, despite research suggesting its use is associated with improvements in certain glaucoma-related physiologic parameters, the results of the current study provide no evidence for a strong association between GBE use and glaucoma compared to the effects of known risk factors (e.g., age, race, smoking status, hypertension, CHD, angina, stroke, myocardial infarction, diabetes, cataracts, diabetic retinopathy, and macular degeneration). However, the cross-sectional design and the possibility of misclassification limits the ability to accurately measure the true association between GBE use and glaucoma other than its effect relative to other known risk factors. Thus, additional research is warranted that is able to address these methodological limitations.