Issue: BCMJ, Vol. 54,
January, February 2012,
page(s) 13-14 Letters
Gerald Tevaarwerk, MD
“I keep six honest serving men (They taught me all I knew)”
The article by Dr Bebb is based on a survey that, left unchallenged, may lead to overservicing and cost increases for our already overburdened medical care system, while its translation in the Vancouver Sun as “Nearly half of men over 45 affected by testosterone deficiency” may cause unnecessary anxieties. The article is based on the Hypogonadism in Men (HIM) study and gives a misleading impression of the prevalence of hypogonadism in the general population.
To minimize bias and ensure clinical relevance of any publication, we must ask six questions:
1. Who published it?
2. In what group of subjects was the study done?
3. Where was the study done?
4. How was it done?
5. Is it statistically significant?
6. Is it clinically important?
The answer to “who?” is Thomas Mulligan and four co-authors. Mulligan acknowledged having received funding from and having acted as a consultant for Solvay Pharmaceuticals, Inc., the manufacturer of AndroGel, a transdermal testosterone preparation. One of his co-authors (CM) was an employee of Solvay and the remaining three co-authors (MF, QZ, and AS) were employees of Covance Periapproval Services, Inc., which “conducted the study on behalf of Solvay Pharmaceuticals, Inc.”
At this point we may be excused if we stop reading as there is significant potential for bias and we are unlikely to gain evidence from the article that we might usefully apply in practice. If we, nevertheless, do press on and ask in what group (sample) and where the study was done we find “2162 men… aged = 45 years visiting primary care practices in the United States… regardless of the reason for their visit,” a group of subjects not representative of the population targeted by Bebb in his article.
Reading on we find that the answer to how hypogonadism was diagnosed is defined as a total testosterone (TT) value of 300 ng/dl (10 nmol/L) in “a blood sample between… 8 a.m. and noon,” a time interval that misses the peak testosterone values reported to occur in many men before 8 a.m.[3,4]
The only question of statistical significance that can reasonably be asked of this survey is how does it compare to the results of an observational cohort study in a population-based random sample of men in the Massachusetts Male Aging Study (MMAS) conducted in 2004. It reported a prevalence rate of 6% using a cutoff for TT of 400 ng/dl (13.3 nmol/L). Had it used the same cut-off value as in the Mulligan survey the prevalence would have been lower yet, estimated at approximately 1 in 20, not “Nearly half of men age 45…”
As to clinical importance, not only does the article give a misleading impression of the prevalence of hypogonadism even when low testosterone levels are found, one has to ask whether replacement is indicated, as gradually declining testosterone levels may be part of the normal aging process. As a point of interest the cost of AndroGel is between $5.30 and $12 per day in Canada, depending on the dose used.
—Gerald Tevaarwerk, MD