Vitamin T for Sex (cont.)

Interestingly, the answer is no.

Extra testosterone can definitely improve the lives of men with extremely low levels of testosterone due to disease -- say, 5% of normal, due to conditions such as the removal of a pituitary tumor, removal of the testes (in the case of testicular cancer, for example), or in Sullivan's case, HIV. In these cases, the lowered level would likely affect libido. If you were to boost such a man's testosterone levels by giving him supplemental doses of the hormone, you would expect his behavior to return to normal. That was Sullivan's experience, with some healthy placebo effects tossed in on top -- a possibility he barely raises in his piece.

But most men simply don't need AndroGel. Here's why.

Testosterone and the Aging Male

Testosterone levels tend to decline gradually in men, starting in early middle age. The popular perception, and one trumpeted throughout Sullivan's piece, is that this decline is a cause of the typical decrease in energy, edge, and sex drive that often accompanies aging.

But there is simply no scientific data to support a cause-and-effect relationship between the (slow) decline in testosterone experienced during normal aging and a negative impact on libido, sexual performance, or level of energy. Furthermore, even the temporary declines in testosterone experienced by many men during periods of stress (traffic jams, a poor evaluation by a superior) typically do not appear to make any long-term difference in their libido and performance, either.

Behavior Drives Hormones, Not Vice Versa

For argument's sake, let's suppose that for every smidgen of a decline in testosterone, there was a proportional decline in sexual drive, energy, and libido, while for every smidgen of a rise, there was an increase. If that were the case, researchers should be able to select a group of normal, healthy guys for a study, measure their testosterone levels, and show that the men with the higher testosterone levels are more libidinous, sexually active, or more aggressive.

Indeed, some studies have suggested such a correlation. But there turns out to be a confounding factor: Acting aggressively and having sex both raise testosterone levels, so the correlation seems due to behavior driving the hormones, rather than the other way around. This phenomenon has been discussed in many leading texts on the subject.

When experiments control for those factors -- acting aggressively and having sex -- the vast majority of studies, whether of humans or animals, show that individual differences in testosterone levels across the normal range don't particularly predict individual differences in aggressiveness or sexual activity. My own study, "Testicular function, social rank, and personality among wild baboons," published in the journal Psychoneuroendocrinology (Vol. 16, No. 4, 1991), is among the many to have found this.