Depression and mortality in the visually-impaired, community-dwelling, elderly population of Quebec.

Purpose: To assess the effect of visual impairment (VI) on the risk of depression or death in the community-dwelling elderly population.

Methods: A population-based, retrospective fixed cohort study was conducted in the community-dwelling elderly (age > or = 65 years) outpatient population of Quebec. The cohort was assembled through the Quebec medical services database and consisted of the 5063 patients aged > or = 65 years who received a diagnosis of VI during the years 2000-2004. The reference cohort consisted of 16 932 elderly subjects who were randomly selected among members of the public drug programme. The outcome variables were depression and death. The main independent variable was VI and covariates included age, gender, chronic disease score, fracture and diabetes.

Results: Controlling for covariates, VI was associated with an increased risk of depression although the effect was not modified by severity (hazard ratio [HR] = 1.35, 95% confidence interval [CI] 1.10-1.66 for severe VI; HR = 1.35, 95% CI 1.09-1.69 for moderate VI). Visual impairment was associated with an increased risk of mortality; patients with moderate vision loss had a higher risk of death (HR = 1.70, 95% CI 1.55-1.87) than those with severe vision loss (HR = 1.34, 95% CI 1.21-1.48).

Conclusions: Given the ageing of the population, VI in elderly subjects is becoming a public health concern. These findings enhance the need to detect and treat VI in order to improve the quality of life and to prevent premature mortality in the elderly population.