In this cross-sectional study, 964 subjects (740 postmenopausal women and 224 men) who underwent ultrasound measurement of the calcaneus at the outpatient osteoporotic clinic in the years 1993-1996 were evaluated. The aim of the study was to compare the influence of age, years since menopause (YSM), and body size on ultrasound variables in normal and osteoporotic male and female populations as well as to assess the ability of quantitative ultrasound (qUS) to discriminate between healthy and osteoporotic individuals. The subjects were divided into four age-matched groups: normal women (n = 601, mean age 56.3 +/- 4.8 years); osteoporotic women (n = 139, mean age 56.5 +/- 4.8 years); normal men (n = 148, mean age 56.2 +/- 10.2 years); and osteoporotic men (n = 76, mean age 56.5 +/- 10.7 years). Persons with a history of a disease and/or medications known to affect bone metabolism were excluded. Broadband ultrasound attenuation (BUA, in decibels per megahertz) and speed of sound (SOS, in meters per second) were measured using Achilles device (Lunar, Madison, WI). Short- and long-term in vitro coefficients of variation (CVs) were: 1.23% and 0.54% for BUA and 0.12% and 0.14% for SOS, respectively. Short-term in vivo CVs were: in women, BUA 1.8% and SOS 0.22%; and, in men, 2.48% and 0.33%, respectively. SOS and BUA values were significantly higher in healthy men (1517.5 +/- 35.3 m/sec, 114.0 +/- 13.3 dB/MHz) than in healthy women (1511.1 +/- 25.6 m/sec, 108.7 +/- 9.5 dB/MHz) (p < 0.000001). The two ultrasound variables had higher values in osteoporotic men (SOS = 1492.6 +/- 24.6 m/sec, BUA = 106.1 +/- 11.6 dB/MHz) in comparison to osteoporotic women (SOS = 1490.4 +/- 19.5 m/sec, BUA = 103.2 +/- 8.6 dB/MHz), but the differences did not achieve significance. In both genders, ultrasound parameters were significantly lower in osteoporotic groups (p < 0.000001). The following age-adjusted odds ratios were obtained: in women, 1.7 (95% CI 1.42-2.03) for BUA, and 2.3 (95% CI 1.87-2.81) for SOS; in men, 1.05 (95% CI 0.03-2.07) for BUA, and 2.13 (95% CI 0.77-3.49) for SOS. ROC analyses performed in both genders showed the following area under the curve data: in women, 0.66 for BUA, and 0.74 for SOS; and, in men, 0.66 for BUA, and 0.71 for SOS. Multiple regression analysis showed age to be the main negative determinant of both ultrasound variables in healthy women, and YSM in osteoporotic women. In both genders, weight was found to have a positive influence on SOS and BUA values, whereas the effect of height was weaker and generally negative. It can be concluded that ultrasound measurement at the calcaneus is able to discriminate between normal male/female and osteoporotic male/female populations in a similar manner. Women had greater rates of decrease in BUA and SOS with age compared with rates in men, and the decrease was greater in normal individuals in both genders. Distinct gender-related differences were noted in regard to correlations of ultrasound parameters with body size.