Laboratory studies have suggested a role for cholesterol in the pathogenesis of both osteoporosis and atherosclerosis. The purpose of this prospective study was to assess whether cholesterol levels, repeatedly measured over three decades in young and middle-aged adult women and men, predicted bone mineral density (BMD) at advanced age. Study participants included 712 women and 450 men enrolled in the Framingham Osteoporosis Study, aged 32-61 years at baseline (1953-55) who underwent bone densitometry 34 years later (1988-1989). BMD was measured at the proximal femur (neck, trochanter, and Ward's triangle) and lumbar spine using dual-photon absorptiometry and at the one-third radial shaft and ultradistal radius using single-photon absorptiometry. Sex-specific multivariable linear regression was used to model each BMD site as a function of total cholesterol level, adjusted for age, cigarette smoking, alcohol consumption, body mass index, systolic blood pressure, diabetes, and estrogen use (women). No significant association between total cholesterol and BMD was found in women for any of the bone sites considered. For example, adjusted mean BMD at the lumbar spine was similar in women from the lowest to highest quartile of total cholesterol, respectively, 1.07, 1.08, 1.06, 1.07 g/cm2; P for trend=0.98. Similarly, the findings in men largely showed no association between cholesterol and BMD, although there was an isolated finding of a statistically significant trend in decreasing mean radial shaft BMD with increasing total cholesterol, 0.73, 0.72, 0.72, 0.70 g/cm2, lowest to highest quartile, P for trend=0.02. Cholesterol levels in women and men from young adulthood to middle age years do not appear to have long-term clinical implications for osteoporosis later in life.