To assess the association between smoking and long-term incident age-related macular degeneration (AMD).
Of 3654 Australians 49 years and older examined at baseline (January 14, 1992, through December 18, 1993), 2454 were examined 5 years later (January 11, 1997, through February 23, 2000), 10 years later (July 10, 2002, through November 4, 2005), or both. Retinal photographs were taken to assess AMD. Smoking status was recorded at each interview.
After controlling for age, sex, and other factors, current smokers had a 4-fold higher risk of late AMD than never smokers (relative risk, 3.9; 95% confidence interval, 1.7-8.8). Past smokers had a 3-fold higher risk of geographic atrophy (relative risk, 3.4; 95% confidence interval, 1.2-9.7). Joint exposure to current smoking and (1) the lowest level of high-density lipoprotein (HDL) cholesterol, (2) the highest total to HDL cholesterol ratio, or (3) low fish consumption was associated with a higher risk of late AMD than the effect of any risk factor alone. However, interactions between smoking and HDL cholesterol level, ratio of total to HDL cholesterol, and fish consumption were not statistically significant.
Smoking strongly increased the long-term risk of incident late, but not early, AMD, with a possibly greater effect in persons with a low HDL cholesterol level, a high ratio of total to HDL cholesterol, and low fish consumption.