Food Insecurity and Vision Impairment Among Adults Age 50 and Older in the United States.

PURPOSE: To test the hypothesis that food insecurity is associated with greater prevalence of vision impairment (VI). DESIGN: Cross-sectional, nationally representative survey of US adults. METHODS: Adults 50 years and older were included. Food security category (ie, full, marginal, low, or very low) was assessed by the US Household Food Security Survey Module. Objective presenting VI (PVI) and self-reported VI (SRVI) were assessed by examination and survey, respectively. Separate logistic regression models were constructed with food insecurity category as a predictor of PVI or SRVI. Models were adjusted for age, gender, race, education, income, cigarette use, alcohol use, body mass index, and physical activity. Age-stratified analyses (age 50 through 64 years vs 65 years and older) were also conducted. RESULTS: Mean age of participants (n = 10,078) was 63.4 years; 8,518 (89.9%) were fully food secure, 1,033 (7.2%) had PVI, and 2,633 (20.1%) had SRVI. Compared with full food security, adjusting for sociodemographic confounders, those with marginal (adjusted odds ratio [aOR], 1.31; 95% CI, 0.97-1.76), low (aOR, 1.61; 95% CI, 1.17-2.23), and very low (aOR, 2.71; 95% CI, 1.75-4.20) food security had higher odds of PVI. Compared with full food security, those with marginal (aOR, 1.58; 95% CI, 1.23-2.02), low (aOR, 1.46; 95% CI, 1.11-1.92), and very low (aOR, 1.85; 95% CI, 1.41-2.41) food security had higher odds of SRVI. The associations between food insecurity and PVI were greater in magnitude in those 65 years and older compared with those age 50 through 64 years. CONCLUSIONS: In this nationally representative sample of US adults 50 years and older, severe food insecurity was increasingly associated with greater prevalence of VI in a dose-response manner.