Cataract surgery is the most frequently performed surgical procedure on Medicare beneficiaries, with an annual cost to the Medicare program of more than $3.4 billion. In this study, the relationship between demographic, environmental, and provider-related factors, and the likelihood that cataract surgery will be performed on a Medicare beneficiary were assessed. The association between likelihood of cataract surgery and patient age, sex, race, income, and latitude of residence was examined, as was the association with the supply of ophthalmologists and optometrists in each region, and the allowed charge for cataract surgery and cost of practice in a region. This cross-sectional, population-based study used administrative data. Both regional models, using least-squares regression and person-based models, using logistic regression were employed. A random 5% sample of 1986 and 1987 Medicare beneficiaries, 65 years of age and older, were included in the study. Medicare beneficiaries who lacked continuous Part A and Part B coverage during 1986 and 1987, or who were enrolled in a health maintenance organization at any time during this 2-year period of observation were excluded from the study to make certain that complete utilization data were available for each individual. Rate of cataract surgery per 1,000 Medicare beneficiaries in each Bureau of Economic Analysis Economic Area (BEAEA) and the likelihood of an individual with particular characteristics undergoing cataract surgery were determined in separate regression models. The mean annual rate of cataract surgery during 1986 and 1987 in the 181 BEAEAs was 25.4 surgeries per 1,000 persons 65 years of age or older (standard deviation = 6.2, coefficient of variation = 0.24). Both the regional model and the person-based model detected an association between a higher rate of and personal likelihood of cataract surgery and female gender, more southerly latitude, higher concentration of optometrists per 1,000 Medicare beneficiaries, and higher allowed charge for cataract surgery, after adjusting for variation in practice expense. The person-based model additionally demonstrated that increased likelihood of undergoing cataract surgery was associated with increasing age from 65 to 94 years, white race, and living in a zip-code area with mean income greater than $15,000. Neither analysis detected a statistically significant association between the concentration of ophthalmologists per 1,000 Medicare beneficiaries and the regional rate of, or an individual's likelihood of, cataract surgery. Compared with the geographic variation in provision of other surgical procedures, the variation in cataract surgery across large geographic areas observed in this analysis was relatively low.(ABSTRACT TRUNCATED AT 400 WORDS)