Abstract
Abstract
Background:
The purpose of this study is to identify which patient factors might be useful in developing a simplified testosterone deficiency (TD) questionnaire.
Methods:
A total of 5071 men who participated in a community-based screening program had testosterone (T) determinations and completed a questionnaire including lifestyle, coexisting diseases, and body mass index. Associations were determined by ANOVA and chi-square analyses for TD (T <300 ng/dL). The effect of multiple variables on T levels was determined by linear regression. Significant variables (p<0.001) were used to model the male androgen deficiency syndrome (MADS) questionnaire.
Results:
The mean T level was 378.6 ng/dL, and 1950 (38.5%) had TD. African Americans (AA) had higher T levels (p<0.001). Nonsmokers had lower T (377.4 vs. 418, p<0.001). Men who exercised >5 times per week had T levels of 391.9 ng/dL compared with 360.4 ng/dL for those without exercise (p<0.001). Men overweight by 30 lbs had T of 293.3 ng/dL compared with 421.8 ng/dL for normal weight (p<0.001). On regression analysis overweight status (p<0.001), adult onset diabetes mellitus (AODM) (p<0.001), race (p<0.001), heart disease (p=0.05), and exercise frequency (p=0.001) remained significant. The final model for the MADS questionnaire included AODM and overweight >20 lbs, and for non-AODM, overweight status, exercise frequency, and erectile dysfunction (ED). For men with AODM and greater overweight status, the risk of TD increased from 49.1% to 62.6% (p=0.010), and for non-AA men without AODM, the increase in TD was from 31.1% to 39.3% for one, to 48.2% for two, and to 61.7% for three of the significant conditions (p<0.001).
Conclusions:
Health factors were identified that can be used with a high degree of predictability to screen for TD. The MADS questionnaire, which includes queries on diabetes, race, obesity, exercise frequency, and ED, can be easily introduced into practice to screen men for low T.
Get full access to this article
View all access options for this article.
