The Endocrine Society and various endorsing organizations recently published recommendations for improving and standardizing testosterone testing. “Toward Excellence in Testosterone Testing: A Consensus Statement” asserts that urologists and other health-care providers must understand the importance of using accurately calibrated testosterone tests—a process that the groups hope to have fully in place in 2012 (J Clin Endocrinol Metab. 2010;95:4542-4548; available at, accessed November 12, 2010 ). Endocrinologist William Rosner, MD—professor of medicine at Columbia University in New York, and a lead author of the consensus statement—explains to Renal & Urology News what impending testosterone-testing standardization—and the current lack of it—means to urology practice and research.


The American Urological Association has endorsed your consensus statement. Has the lack of standardization in testosterone testing been a long-recognized problem in the urology community as well as in the endocrinology community?


Dr. Rosner: There is a subset of urologists who are familiar with this issue, but it has been a long-recognized problem in the endocrine community.

But will you need some kind of buy-in from the wider urology community?


Dr. Rosner: Yes. We need some kind of buy-in from everybody: from physicians, from insurance payers, from the NIH [National Institutes of Health], from medical journals.. Everybody has to insist that this be done properly. That’s why they were invited to the [February 2010] conference [convened to help formulate the consensus statement].

How would that work?


Dr. Rosner: Journals would have to insist that any papers they accept have numbers generated from acceptable methods. And for NIH-funded studies, the NIH would have to agree that it’s more expensive to pay for improper determinations in the long run and allow budgets that may cost a bit more but provide reliable results. In the short run, standardization looks more expensive. But in the long run, it’s less expensive because you don’t have to repeat tests a million times; you don’t give patients the wrong medicine and run into complications because you’re dealing with wrong numbers. So, insurers should also insist on standardization because it’s expensive in the long run to go with the wrong numbers. It’s also the only way to get useful evidence-based guidelines.

So this issue has been brewing for a long time?


Dr. Rosner: Yes. It’s been brewing for about eight to nine years, I would say. There’s been an enormous amount of inertia while trying to get this done. A lot of it is financial, but it’s also a matter of simplicity. A platform assay takes the blood sample and quickly yields 40 answers—that’s a quick turnaround time and it’s relatively inexpensive.  It happens to give you the wrong answer in the case of testosterone testing, but we have opted for simplicity and lower costs over accuracy. Money and inertia are always the biggest pitfalls.

You’re saying that testosterone testing as part of that platform assay does not yield accurate results?


Dr. Rosner: That is correct. Immunoassay-based platform assays are demonstrably inaccurate, especially at the low end, where the patient population that needs treatment resides—e.g., hypogonadal men and normal women or, more accurately, women with androgen-based disorders.


How will standardization change the way urologists practice?


Dr. Rosner: There’s a big unknown area about when to replace testosterone, especially in the aging man. It’s easy if men have really low testosterone, but in the borderline area—say between 200-300 ng/dl—you need an accurate assay to know who needs replacement. Of course, we don’t replace testosterone on the numbers alone; we replace it based on the combination of symptoms and blood tests.  Because values for testosterone in this borderline region are very often measured inaccurately by most current methods, we are faced the very real possibility of under- or overtreating.

How does the plan look cost-wise?


Dr. Rosner: We don’t know yet; we’re not married to a method but rather to ensuring that all tests can be traced back to single agreed upon standard. The standardization part should not be too costly.  The cost of using methods that match the standard is in evolution.  We should have a better idea of costs in a year or so.

The trouble is that there has been no universally agreed upon testosterone standard. However, a standard now has been established by the CDC [Centers for Disease Control and Prevention], and individual laboratories can subscribe to the [CDC’s] program, a quality control program that will ensure accuracy.  Now that we have a standard, we have to insist that everybody adhere to it. We’re looking for compliance with the standard, not with any particular method of testing.


Has an effort like this ever been launched?


Dr. Rosner: For testosterone it’s a first, but the same thing was done with cholesterol. Initially, there was no standardization. The CDC led the way on that. Think about it – how could one promulgate guidelines for cholesterol in blood if different labs and methods gave widely differing results on the same sample.  The same situation existed with  hemoglobin A1c; we had guidelines, but everybody was getting different answers from their tests— so the guidelines weren’t meaningful. Now the guidelines have meaning because the tests are traceable to a single standard.



Do you anticipate a lot of controversy over which numbers should represent treatment thresholds?


Dr. Rosner: Since we didn’t have good methods and standards [until now], we have to correct all our normal ranges. Therefore, one of the things you’ll see [in the consensus statement] is that part of the goals is to establish normal ranges using proper methods.

Who would be opposed to standardization of testosterone testing, and on what grounds?


Dr. Rosner: Nobody can really oppose it in public, because it’s like motherhood, America, and apple pie—who wants to promote numbers that are meaningless? But the big argument against it is that it looks like it’s going to be more expensive to get it right in the short term.  But who can argue for an inexpensive way to do things when the wrong answers are generated?


Are there any other grounds for objection?


Dr. Rosner: It’s tradition. It’s status quo. It’s what’s been acceptable in the past. For your quality control, if you get the same answer as everybody else who uses the same method you pass. But all the methods yield divergent answers so there’s no right answer in the truest sense of the word. That has to go away. Quality control has to be accuracy-based and not based upon a method—so called peer-based quality control

Urologists have no reason to oppose standardization?


Dr. Rosner: No. They will be very supportive.

Can the standardized tests be analyzed in about the same amount of time as the conventional blood test?


Dr. Rosner: It depends. First of all, we don’t know what the ultimate testing method will be.  The manufacturers who are making inexpensive but inaccurate tests will find it makes sense for them to make better tests for less money. That’s up to them. Right now the gold standard kind of a test is more expensive.

What is the gold-standard test?


Dr. Rosner: At the moment, the best and most sensitive method seems to be the mass spectroscopy-based measurements, but we are not married to this test. If someone comes up with an  immunoassay-based test that’s just as accurate, that’s fine. But it’s got to be done not only quickly and cheaply, but accurately. Accuracy is the emphasis not the method. 



At present, how widely used is mass spectroscopy?


Dr. Rosner: It’s generally available in large reference laboratories. Very few hospital-based laboratories have it. They will be faced with continuing with inaccurate testing, sending appropriate samples to reference laboratories, investing in new technologies, and hoping for the development of accurate methods based on the technologies already in their armamentarium.

It doesn’t sound like urologists will be affected by the testing method used.


Dr. Rosner: It should be transparent to the urologists. They take the same tube of blood and send it out. It just depends what happens at the other end. Of course, the urologists have a choice of where to send tests and what kind of test to ask for.

What sort of timeline are you looking at?


Dr. Rosner: The move toward accuracy-based testing has begun already; we hope to have everybody on board within the next couple of years: journals, third-party payers, the NIH, physicians, and so on. It’s been a struggle, but the end is just over the horizon.



Organizations endorsing consensus statement:

  • American Association for Clinical Chemistry
  • American Association of Clinical Endocrinologists
  • Androgen Excess/PCOS Society
  • American Society for Bone and Mineral Research
  • American Society for Reproductive Medicine
  • American Urological Association
  • Association of Public Health Laboratories
  • The Endocrine Society
  • Laboratory Corporation of America
  • North American Menopause Society
  • Pediatric Endocrine Society (formerly known as Lawson Wilkins Pediatric Endocrine Society)

Source: The Endocrine Society