Selective Disclosure of the Truth: When Is the Line Crossed?

In this month's edition of The New England Journal of Medicine, a review of data submitted to the U.S. Food and Drug Administration (FDA) for a dozen popular antidepressants suggests that their effectiveness was exaggerated by the selective publication of favorable results. In fact, nearly a third of those antidepressant drug studies were never published in the medical literature and nearly all happen to show that the drug being tested did not work, researchers reported. This is a very common tactic used in many different places, including in industry, advertising among salesmen (and women) and even job candidates.

It is what like to call "selective disclosure of the truth". It's not exactly lying (admittedly, it's a very fine line) rather its just omitting those things that happen work against whatever is trying to be accomplished. There are many examples of using "selective disclosure of the truth" -- even in healthcare. For example, New York City Health Officials did this when they pushed (some would say steamrolled) to implement mandatory surveillance of glycosated hemoglobin (better know as A1c) test results, which was the first for a non-infectious disease. Diabetes was addressed mainly because it was an easy target, a simple blood test to measure A1c is done and that is indicative of glycemic control. The same cannot be done for other health problems such as asthma or hypertension.

In an interview with CNN and the media, New York City Health Commissioner, Dr. Thomas R. Frieden stated, "There will be some people who will say, 'What business of the government is it to know that my diabetes is not in control?'" The answer, he said, is that diabetes costs an estimated $5 billion a year to treat in New York and was the fourth leading cause of death in the city in 2003, killing 1,891. Sounds like a pretty convincing argument, only it wasn't exactly truthful.

Frieden's statement was very carefully worded to suggest that diabetes costs New York City far more than the city actually spends on the disease. Of the estimated $5 billion spent to treat diabetes in New York, private health insurance and the Federal government via Medicare and Medicaid pay the overwhelming majority of this figure, not the City of New York.

In fact, as of February 2006, the NYC Health Department had an annual budget of just $950,000 allocated to diabetes and a staff of just 3 people, again according to The NEJM. I wrote more about that in my posting about that topic, but I don't want to stray too far away from the central point: how prevalent is "selective disclosure of the truth" in drug marketing?

I suspect the latest example is only the tip of the iceberg, and we are starting to see a growing body of evidence suggesting that in drug marketing in particular, its far more common for the negative studies never to see the light of day than for the positive studies. This has ramifications beyond getting FDA approvals. Doctors never see the negative studies in their medical journals, and therefore are persuaded (or some would say misled) to believe they are safe and to continue prescribing them.

Clearly, not all cases of "selective disclosure of the truth" are the same. Let's be honest, in your last job interview, you probably didn't mention those things which made you look bad, and why would you? Why raise such an issue unless someone asks? Consider when someone has had a long absence from the workplace. Although someone who was in jail for a few years might have different reasons for not highlighting this than someone who took a leave from the workplace to raise their children, but the company interviewing a candidate is free to question it. Few (if any) people are hurt by leaving it out. And while no one would question that your local, polyester leisure suit wearing used car salesman probably pushes those limits of selective truth disclosure too far, even so, it usually only impacts the individual buying the car. But when drug companies do this, millions of people risk dying as a result.

If this posting seems a bit like deja vu, its because it is. Last year, I wrote about the type 2 diabetes drug Avandia being as being another, all-too-painful example of what is wrong with the process. Unfortunately, now we have yet another example of this in action, this time with antidepressants. It is the job of our regulators to mandate where that line happens to be, but having the pharmaceutical industry pay for a majority of the FDA's budget with user fees (see here and here for more background) clearly blurs where that line should be. In 2008, let's get Congress to kick the FDA user-fee habit and start funding it from taxpayer dollars again!