For more than two decades, researchers have sought to identify "risk factors" for age-related macular degeneration (AMD), a major cause of irreversible vision loss in the Western world, particularly in the elderly. Two issues have complicated this search: failure to differentiate between different stages of AMD and misinterpretation of measures of association (odds ratios) and risk (risk ratios) derivable from different research designs. Fortunately, in more recent epidemiologic studies, more attention has been given to these issues. Three groups of potential "risk factors" that have been studied were reviewed: those known to be risk factors for cardiovascular disease, environmental factors, and racial and ethnic factors. Of these, only tobacco smoking, a known risk factor for cardiovascular disease, has been demonstrated to be associated with AMD consistently across many studies of different design, carried out within different populations. The available evidence supports at least a doubling of risk of late AMD associated with long-term smoking, a factor that is under the control of the individual. The preponderance of evidence has not supported other factors to the same degree. Presently, racial and ethnic factors are high priorities for further research.