Osteoporosis is a silent disease that is often undiagnosed until there is a bone fracture. This study was done to determine the prevalence of physician-diagnosed osteoporosis in U.S. men and women and relate this to osteoporosis prevalence determined by bone mineral density (BMD) measurement. Data were from the National Health and Nutrition Examination Surveys (NHANES) III (1988--1994) and IV (1999--2002) which were designed to obtain a representative sample of the non-institutionalized, U.S. civilian population. Demographic data and self-reported physician diagnosis of osteoporosis were collected by questionnaire. Femoral BMD was measured using dual-energy X-ray absorptiometry. Classification of osteoporosis was based on total femur BMD (t-score<-2.5) using data for non-Hispanic white women as the referent. Prevalence of osteoporosis at hip+spine was projected from the NHANES III hip data. Age- and gender-specific hip osteoporosis prevalence estimates from NHANES were multiplied by gender-specific ratios of osteoporosis prevalence at hip+spine/hip from published reports. Weighting variables provided by NHANES were used to project to the U.S. population and prevalence estimates were calculated using “surveymeans” procedure in SAS (Cary, NC). The prevalence of osteoporosis and diagnosed osteoporosis increased with age and was higher in women than men. Prevalence of osteoporosis and diagnosed osteoporosis differed by racial group in women, with highest prevalence among non-Hispanic whites, lowest in non-Hispanic blacks, and intermediate in Hispanics (data not shown). The biggest discrepancy between prevalence of osteoporosis determined by BMD and self-reported, physician-diagnosed osteoporosis was in older men and women. Prevalence of osteoporosis projected to hip and spine may be an underestimate of the true osteoporosis prevalence; therefore, the discrepancy between physician-diagnosed and BMD-determined osteoporosis might be even greater. The recent Surgeon General report cited that about 10 million U.S. individuals over age 50 have osteoporosis at the hip. When projected to hip and spine, to reflect physician practice, osteoporosis prevalence is likely double that. Osteoporosis appears to be under-diagnosed by physicians, suggesting that a sizeable portion of the population may have unrecognized and untreated disease and is therefore at significant fracture risk.