Uveitis in the elderly: epidemiological data from the National Long-term Care Survey Medicare Cohort

There is a paucity of population-based data on the epidemiology of uveitis in the elderly. In the past 40 years, only 2 U.S. population-based studies have examined the epidemiology of uveitis. The conclusions of these studies on the burden of uveitis in the elderly differ greatly. In this analysis, we use Medicare claims data to define the population-based incidence and prevalence of uveitis in the United States elderly population. Cohort study. A cohort of 21644 Medicare beneficiaries drawn for the National Long-term Care Survey, a random sample of U.S. adults 65 years and older, was followed up from 1991 through 1999. The International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis codes specific to uveitis were queried. To decrease the inclusion of uveitis caused by surgery, diagnoses occurring within 3 months of intraocular surgery were excluded. The incidence and prevalence of uveitis by anatomic location was calculated for each year in the study. The onset of uveitis during a given year and its presence in subsequent years. The cumulative yearly incidence of uveitis ranged from 302/100000 to 424/100000 persons per year and averaged 340.9/100000 persons per year. Anterior uveitis was the most common form of uveitis in this population, with a mean incidence of 243.6 cases per 100000 persons per year. The incidence of posterior uveitis averaged 76.6/100000 and the incidence of panuveitis/endophthalmitis averaged 41.7/100000. Only 2 cases of intermediate uveitis occurred during the study period. The cumulative prevalence of uveitis doubled from 511/100000 in 1991 to 1231/100000 in 1999, with anterior uveitis accounting for most prevalent cases in every year. The burden of uveitis in the elderly is substantial and is higher than previously thought. Longitudinal analysis of Medicare claims data may provide a useful tool for monitoring uncommon diseases, such as uveitis, in the elderly population.