Self-reported visual impairment and mortality: a French nationwide perspective.

Purpose: To estimate the association between self-reported visual impairment and mortality.
Methods: Two national surveys in community and institutionalized populations were combined. First, 2,075 institutions for children with impairments, adults with impairments aged persons, and psychiatric patients were selected randomly. The sample comprised 15,403 subjects of whom 14,603 (94.9%) were interviewed. Second, a random, stratified sample of 21,760 persons living in the community was selected, and 16,945 (77.9%) were interviewed. Types of impairment were identified by face-to-face interviews. Two years later, 14,497 subjects in institutions and 15,648 in the community were revisited. Data on death were obtained from either the National Register or households. Death rates were related to age, gender, and impairment. A logistic regression was performed including impairments, activities of daily living, age, gender, type of residence, and geographical area.
Results: Strong, independent associations were found between particular impairments, institutional residence, activities of daily living, age, gender, and risk of death. Associations between mortality and type of impairment could be ranked as follows: motor (OR = 1.235), brain (OR = 1.552), low vision (OR = 1.681), speech (OR = 2.090), visceral (OR = 2.233) and blindness (OR = 2.262).
Conclusions: Self-reported visual impairment is an independent factor associated with mortality.