Smoking, urinary cotinine levels and incidence of visual impairment.

The longitudinal relationship between smoking status and risk of developing visual impairment (VI) remains unclear. We examined the relationship of smoking status and urinary cotinine level, an objective measure of smoking, with incidence of VI. This cohort study included 279,069 individuals free of VI who were followed for up to 8.8 years (median 4.8 years). VI was defined as when bilateral visual acuity was worse than 0.5 (cutoffs of 0.3 Logarithm of the Minimum Angle of Resolution). During 1,324,429.8 person-years of follow-up, 7852 participants developed new-onset bilateral VI. Self-reported current smoking status was associated with increased risk of developing VI in both men and women, with a stronger association in women (P for interaction = 0.01). Multivariable adjusted hazard ratios (95% confidence intervals) for incident VI comparing current smokers to never-smokers were 1.14 (1.04-1.25) in men and 1.52 (1.28-1.80) in women. Urinary cotinine levels of ≥ 100 ng/ml were significantly associated with increased risk of incident VI, and these associations remained when introducing changes in urinary cotinine and other confounders during follow-up as time-varying covariates. Cigarette smoking assessed based on self-report and urinary cotinine level was associated with increased incidence of VI. Our findings identify smoking as an independent risk factor for VI.