Glaucoma and survival: the National Health Interview Survey 1986-1994

OBJECTIVE: Associations between glaucoma and survival have not been studied extensively, in part, because of the relatively low prevalence of this condition. This study examines associations between self-reported glaucoma and mortality in a nationally representative sample of U.S. adults. DESIGN: Annual cross-sectional multistage area probability survey of the U.S. civilian noninstitutionalized population living at addressed dwellings. PARTICIPANTS: Mortality linkage with >96% of participants from the 1986 to 1994 National Health Interview Survey was performed by the National Center for Health Statistics through 1997. Complete data were available on 116796 adults >or=018 years old. METHODS: Adults within randomly selected households were administered a chronic conditions list that included questions about glaucoma and visual impairment. Proxy information on these conditions was obtained when household members were unavailable for interview. Statistical methods included Cox regression models with adjustments for covariates, as well as for the complex sample survey design. MAIN OUTCOME MEASURE: All-cause mortality and cardiovascular and cancer mortality. RESULTS: A total of 1559 (1.3%) glaucoma cases were reported. Nearly 19% of participants with reported glaucoma also had reported visual impairment (n = 303). Mortality linkage identified 8949 deaths; the average follow-up was 7.0 years. After controlling for survey design, gender, age, race, marital status, education level, and self-rated health, participants with reported glaucoma but without reported visual impairment were at significantly increased risk of death relative to participants without reported glaucoma, irrespective of visual impairment status (hazard ratio HR], 1.35; 95% confidence interval CI], 1.19-1.53); similar associations were found for participants with reported glaucoma and visual impairment vs. participants with no reported glaucoma (HR, 1.39; 95% CI, 1.14-1.71). An increased risk of cardiovascular disease mortality was found for participants with reported glaucoma both without (HR, 1.31; 95% CI, 1.11-1.55) and with (HR, 1.53; 95% CI, 1.15-2.05) reported visual impairment. Risk of mortality due to cancer was increased only in participants with reported glaucoma but without reported visual impairment (HR, 1.57; 95% CI, 1.25-1.98); this association was stronger when the mortality analysis was restricted to cancers amenable to early screening, including breast, cervical, colon, and prostate cancer (HR, 1.99; 95% CI, 1.41-2.81). CONCLUSIONS: Among adults residing in the United States, reported glaucoma is associated with an increased risk of all-cause and cardiovascular disease mortality. Associations between glaucoma and cancer were inconsistent and may reflect, in part, a detection bias, in which glaucoma is more likely to be diagnosed in adults receiving health care because of other medical conditions.