The participants ranged in age from 67 to 97, with a mean age of 75.3. Men comprised 37.5% of the sample. The participants were relatively well-educated, reporting an average of 13.1 years of education. The average family had an annual income between $16,000 and $24,999. Based on their reported physical activity, African-American participants in CHS expended about 1,500 kcals per day above basal metabolic rate. Only 12% of African-American women were taking estrogen. Few of the participants screened positive for depression. Five-to 6% of the participants reported difficulty with more than one ADL or IADL, and 63.5% reported their health as good or better. Seventy-five percent of the cohort was overweight, defined as a BMI >25.0 kg/m².

Among whites, gender is known to play an important role in bone density. In regression models, including men and women, sex was highly associated with BMD (p<0.001), explaining 34.7%, 25.0% and 10.2% of the variability for BMD of the total body, total hip, and lumbar spine, respectively.
erectalis tablets

Table 1. Means and Standard Deviations (Where Appropriate), P Values for Investigated Variables

Table 1 reports the means and standard deviations (where applicable) of variables that were tested. Due to the exploratory nature of this analysis, only the p value for most significant BMD site is reported. The correlations are reported for the entire sample, women and men. Many of these variables may be collinear with gender and, thus, appear in this univariate analysis to be significant for the entire sample and not in the gender-specific analysis. There was no significant correlation noted (p>0.1) for calcium supplementation, thiazide diuretic use or kcals of physical activity contrary to what has been reported in the literature. Steroid use was marginally significant in women (p=0.087) but not in the combined group or the men alone. (There were no data from CHS available on dietary intake or family history.) Table 2 provides a comparison of BMD between African-American and white participants in the CHS. Among African-American women, current estrogen use failed to show a significant association with BMD at any of the three tested sites (total body: p=0.60, total hip: p=0.25, lumbar spine: p=0.56). However, only 12% of African-American women were taking estrogen at the time of the BMD measurements. online pharmacy no prescription

Table 2. Mean and Standard Deviation for BMDs in African Americans and Whites

Table 3 shows the results of the final regression models for BMD. The percent R² for any variable that entered the model is included. (Those with an R² of less than 1% are not shown.) The addition of weight to the models eliminated BMI; weight but not height² was significantly correlated with BMD (Table 4).

Table 3. Percent Variability Explained by Individual Variables

Age has a strong, negative correlation with BMD in whites. Age similarly showed a negative correlation with BMD at all three sites in this cohort of African-American women and men over 64 years old. However, in African-American women, age contributed just 2% of the variability in total body BMD, and only 4% of the variability in the total hip. It did not enter the model for lumbar spine in women or for any of the BMD sites in men. cialis canadian pharmacy

Table 4. Partial Correlations of Weight and Height Squared with BMD

Figure 1. Total Hip BMD as a Function of Tertiles of BMI and Weight

Weight was strongly and positively correlated with BMD at all three sites in both men and women. The use of BMI, weight/(height), was less powerful in predicting BMI than weight alone. Figure 1 shows the relationship of BMI and weight to BMD for the total hip in African-American women. For a given tertile of weight, the contribution of BMI remains essentially constant from low to high, demonstrating that BMI’s relationship to BMD is driven almost entirely by weight alone. Similar relationships were seen at the other two sites in women and all three sites in men. kamagra tablets

Figure 2. Association of Weight in Pounds and BMD at the Total Hip

Figure 2 plots the correlation of BMD and weight and shows the relationship of weight to BMD for total hip. The mg/cm² change in BMD per kg of weight ranged from 5.3 to 6.8.