Abstract
Methods::
Medicare beneficiaries with AMD in 1995 were identified by International Classification of Diseases, Ninth Revision (ICD-9) diagnoses codes from a 5% random sample of Medicare databases (Centers for Medicare and Medicaid Services (CMS) 5% Physician/Supplier Part B Files), including neovascular AMD (ICD-9: 362.42, 362.43, 362.52), non-neovascular AMD (ICD-9: 362.51) and early AMD (ICD-9: 362.50, 362.57). Patients were excluded if they were under 65 years of age, did not reside in the 50 United States or the District of Columbia, did not have Part-B coverage, had HMO coverage that was not processed by CMS, or lacked follow-up information. Five-year mortality rate from 1995 to 1999 was then determined by the death code in the CMS 5% Denominator files. Multiple logistic regression models were applied to estimate the effects of AMD on the five-year mortality rate, adjusted for potential baseline risk factors, including demographic factors (age, gender, race/ethnicity, CMS region of residence, and length of follow-up) and clinical factors (the indicators for the presence of primary open-angle glaucoma, cataract, hypertension, cardiovascular disease, cerebrovascular disease, diabetes, and hyperlipidemia).
Results::
Among 1,223,796 eligible Medicare patients in the 5% sample, a total of 82,017 (6.7%) AMD patients were identified in 1995, including 9,224 (0.8%) neovascular AMD patients, 29,267 (2.4%) non-neovascular AMD patients, and 43,526 (3.6%) early AMD patients. The overall five-year mortality rate was 26%. There was no relationship between any AMD and five-year mortality rate in the adjusted model (OR=0.98; 95% CI: 0.96, 1.00). Compared with patients without any AMD, there was a significantly lower mortality rate in neovascular AMD patients (OR=0.87; 95% CI: 0.82, 0.92) and a slightly higher mortality rate in non-neovascular AMD patients (OR=1.03; 95% CI: 1.00, 1.07).
Conclusions::
Although there was no association between any AMD and the 5-year mortality rate in the fully-adjusted model, which is consistent with the results from the Beaver Dam Eye Study and the Blue Mountain Eye Study, neovascular AMD patients have significantly lower mortality rates and non-neovascular AMD patients have slightly higher mortality rates. These findings warrant further exploration.
Keywords: age-related macular degeneration • clinical (human) or epidemiologic studies: outcomes/complications