The 1st half of the study made an appearance in Investigative Ophthalmology earlier this year. Related StoriesAAO recognizes outstanding eye doctors and surgeons for technology in patient careGood information for patients experiencing dry age-related macular degenerationNightstar announces $35 million Series B funding roundAbout 23 % got some sort of visible impairment, and 4.2 % were blind. Cataracts had been the most common cause of blindness, with glaucoma second.For women, position with respect to postmenopausal hormone therapy was included in multivariate models also. Significantly less than 5 percent of the cohort lacked any solitary covariate; for each covariate, we included an indicator for lacking data in the regression versions, if required. In a sensitivity evaluation, we altered for propensity scores34 that reflected associations of coffee intake with the various other variables in the multivariate-adjusted models. Results obtained by using propensity-score adjustment were nearly the same as those from multivariate-adjusted versions . Assessments of linear craze across categories of coffee usage had been performed by assigning participants the midpoint of their coffee-intake category and getting into this new variable right into a separate Cox proportional-hazards regression model.