Vision Impairment, Poor Diet, and Food Insecurity in the United States

Abstract

Abstract Objectives To test the hypotheses that poor diet and food insecurity are associated with both measured and self-reported vision impairment (VI). Methods We analyzed data from adults age ≥50 years in the National Health and Nutrition Examination Survey (NHANES) between 1999–2008 in the United States (US). To assess diet quality, Healthy Eating Index 2015 scores (HEI-2015) were applied to participants’ 24-hour recall dietary intake. Participants were categorized as having full, marginal, low, or very low food security using the US Food Security Survey Module. Presenting VI (PVI) was defined as presenting distance visual acuity worse than 20/40 in the better-seeing eye. Self-reported VI (SRVI) was defined as vision reported as fair, poor, or very poor. Accounting for NHANES’ complex survey design, separate logistic regression models were constructed to assess the associations between HEI-2015 quintile and PVI, HEI-2015 quintile and SRVI, food insecurity category and PVI, and food insecurity category and SRVI. Models were adjusted for age, gender, race, education, income, cigarette use, alcohol use, body mass index, and physical activity. Results Of the 10078 adults in this analysis, mean age was 63.4 years, 54% were women, and 80% were white. Mean HEI-2015 score (from 0–100) was 52.5. The prevalence of full food security was 89.9%, that of PVI was 7.2%, and that of SRVI was 20.1%. In adjusted logistic regression models, those with HEI-2015 scores in the lowest quintile (poor diet) had significantly higher odds of SRVI (OR: 1.29; 95% confidence interval [CI]: 1.04–1.61), but not PVI (OR: 1.13; CI: 0.86–1.49) compared to those in the fifth quintile. Compared to full food security, marginal (OR: 1.31; CI: 0.97–1.76), low (OR: 1.61; CI: 1.17–2.23), and very low (OR: 2.71; CI: 1.75, 4.20) food security was associated with increased odds of PVI. Similarly, marginal (OR: 1.58; CI: 1.23–2.02), low (OR: 1.46; CI: 1.11–2.92), and very low (OR: 1.85; CI: 1.41, 2.41) food security were associated with increased odds of SRVI compared to full security. Conclusions In this nationally representative sample of US adults age 50 years and older, poorer overall diet quality was associated with greater prevalence of SRVI. In a dose-response pattern, increasingly severe food insecurity was associated with greater prevalence of both PVI and SRVI. Funding Sources None