Skip to main content
. Author manuscript; available in PMC: 2009 Dec 2.
Published in final edited form as: J Am Geriatr Soc. 2008 Nov;56(11):2000–2008. doi: 10.1111/j.1532-5415.2008.01965.x

TABLE 1.

Study sample characteristics (N=810)

Variable Mean ± Standard Deviation
or N (Percent)
Age, yr 46.3 ± 12.4
Race / Ethnicity
 Black 220 (27.2%)
 Hispanic 267 (33.0%)
 White 323 (39.9%)
Education, yr 15.3 ± 4.0
Household income
 < $10k 190 (15.4%)
 $10k–29,9k 219 (20.1%)
 $30k–69,9k 218 (32.8%)
 ≥ $70k 149 (31.8%)
Weight, kg 87.8 ± 15.0
Height, cm 176.2 ± 7.4
Self-rated health
 Excellent 139 (21.7%)
 Very Good 227 (37.1%)
 Good 291 (29.8%)
 Fair/Poor 153 (11.5%)
Smoking
 Never 320 (45.4%)
 Former 251 (28.4%)
 Current 236 (26.3%)
Alcohol, drinks/day
 0 247 (23.1%)
 < 1 275 (39.2%)
 1–3 188 (28.4%)
 > 3 99 (9.4%)
Total testosterone, ng/dL 438.9 ± 180.1
Free testosterone, ng/dL 9.0 ± 3.5***
Total estradiol, pg/mL§ 23.7 ± 9.2
Free estradiol, pg/mL§ 0.7 ± 0.3**
Luteinizing hormone, IU/L 4.6 ± 1.6***
Fat mass, kg 21.9 ± 8.3
Appendicular lean mass, kg 26.7 ± 4.1***
Physical activity (PASE) 193 ± 112***
Grip strength, kg 38.3 ± 11.1***
Walking speed, meters/s 1.4 ± 0.2***
Chair stand speed, stands/s 0.4 ± 0.1***
Physical function score (0–7) 4.2 ± 2.0***
*

Age trend, p < .05

**

Age trend, p < .01

***

Age trend, p < .001

With the exception of race/ethnicity, estimates are weighted according to sampling design (see Methods ).

To convert nmol/L, multiply values by 0.0347.

§

To convert pmol/L, multiply values by 3.671

Geometric means have been transformed from the natural log scale to original units for presentation. Sample sizes (N) vary according to missing data.