Conducting Trustworthy Medical Trials in a High-Stakes Industry

From the Lecture Series: The Skeptic's Guide to Health, Medicine, and the Media

By Professor Roy Benaroch, M.D.

Unfortunately, ethics often clash with profits in this trillion dollar industry, which can result in either big profits or huge losses. Discover the ingredients of trustworthy medical trials.

Team of Research Scientists Working With Personal Computer, Analysing Test Trial New Generation Drug Data. They Work in a Modern Laboratory/ Medical Center
(Image: Gorodenkoff/Shutterstock)

Is Medical Treatment Always the Best Option?

Here’s a classic quote about taking medication to help with a common cold. A doctor might say, “If you take this medicine, you’ll be better in seven days. If not, you’ll get better in a week.”

Think about the last time you had a cold. You got better, probably in a week or so. But if you saw a doctor and were prescribed an antibiotic, you might think you got better because of the medicine.

In fact, though, the medicine did not help. We know antibiotics cannot help with a viral illness like a common cold, but they might seem to because you will get better if you take them.

Of course, you will also get better if you don’t take them. But in your mind, you’re likely to give credit to the medicine, the doctor, or the therapy, and not just to Mother Nature, though she’s the one who really did the work.

This is a transcript from the video series The Skeptic’s Guide to Health, Medicine, and the Media. Watch it now, on The Great Courses Plus.

Accounting for the Placebo Effect

So, scientific studies of medications or therapies can’t just look at people who take the drug to see if they get better. We must compare what happens with and without the therapy to really know if it was the therapy that made the difference.

Placebo drugs concept
A legitimate clinical study must include at two groups of people: the study group, takes the new medicine or tries the new therapy; the control group takes the placebos (sugar pills, sometimes, or a fake surgery that makes you think you’ve really had something done) (Image: AllaBond/Shutterstock)

To do this, a legitimate clinical study must include at two groups of people: one group, called the study group, takes the new medicine or tries the new therapy; the other group does not—the control group.

Comparing two groups, the study and the control group, lets you separate out the effect of the natural history of the disease. But there’s something else we must consider: the effects of wishful thinking.

We know that people want to get better, and we know that people want to believe that whatever they’re doing is going to help, especially if what they’re doing is expensive or time-consuming.

Placebos, as they’re called, are sham treatments—sugar pills, sometimes, or a fake surgery that makes you think you’ve really had something done. So the control group in a study is there not just to cancel out the natural history of the disease, but also to measure this placebo effect.

For that reason, the control group should be given what they think is the drug or the therapy. It should look, taste, and seem like the same thing as the real therapy, so the study participants in both groups don’t know if they’re getting real medicine or a placebo.

A double-blind study means that neither the patients nor their doctors and nurses know who’s taking the real drug. The best studies randomize the study participants into these two groups, in a pattern that cannot be guessed.

The ideal clinical studies, then, are randomized and placebo-controlled and double-blinded. If studies like these are done well, they’re the best way to know that a specific treatment really works for a given condition.

But as we’ll see, even these kinds of studies aren’t always done or reported in a way that helps us easily understand that simple question of whether a drug is effective.

Learn more about how to better understand and evaluate medical data

Medical Drug Development: A Trillion Dollar Business

Drug development and sales is a huge business—the market worldwide is about a trillion dollars a year, and that’s dominated by huge U.S. companies. The Tufts Center for the Study of Drug Development estimated in 2014 that it took, on average, $2.6 billion to develop a new drug and bring it to market.

Now, that number has been hotly disputed, especially since the Tufts Center is heavily funded by the pharmaceutical industry. However, there’s no doubt that a big time successful drug, while fiendishly expensive to develop, can potentially generate huge revenue.

There’s a huge risk, too. If the drug doesn’t pan out, the development costs are pretty much money down the drain.

But if it does work out, there’s serious money to be made. Gambling in Las Vegas? That’s small potatoes. You want a risky gamble that can pay off in spades, look at the pharmaceutical business. 

Learn more about health, medicine, and the media

Tamiflu: An Effective Flu-Fighting Drug?

One drug that hit it big in the United States and abroad is Hoffmann-La Roche’s tamiflu. It was synthesized, originally, from an extract that came from the Chinese star anise, and licensed for sale to help fight influenza infections in the United States in 1999.

Tamiflu antiviral
 Tamiflu is an antiviral medicine used to treat flu and it was originally synthesized from an star anise extract. (Image: Nelson Hale/Shutterstock)

Since its launch, cumulative sales have totaled at least $18 billion in the United States, about half of which was for governments and companies around the world to stock up for a feared global pandemic. Most of that, of course, has never been used.

The drug tamiflu is a clever one. It deactivates a protein on the outside of the influenza virus so the virus can’t travel through cell membranes. It doesn’t destroy the virus, but it can prevent it from replicating and spreading and infecting more cells.

Hypothetically, especially if given early in the illness, tamiflu ought to prevent people from getting sick with influenza, perhaps shortening the illness, making it milder, or making it less likely that serious complications will occur.

That’s all hypothetical, of course. What really matters are the studies, especially the double-blind, placebo-controlled clinical trials, that compare tamiflu with placebo among people who’ve caught the flu.

Influenza, of course, is a significant cause of misery, illness, and death, though the exact numbers here aren’t so straightforward. Influenza alone doesn’t commonly kill people, but it often contributes to death in people with heart or lung disease.

Learn more about Influenza: the Past and Future Threat

NPR reported this well in their piece, “How Many People Die from Flu Each Year? Depends How You Slice the Data.” A number that had been often quoted was that flu caused 36,000 deaths a year in the United States. That was the figure from the CDC until 2010, when reporters were asked to give a more accurate range of deaths, from 3,300 to 49,000 per year.

That’s a lot of variability from year to year, which makes studying drug and vaccine effectiveness more difficult. A lot of the variability from year to year is because of the circulating virus itself, not the interventions that we use to try to stop it.

Though difficult and expensive to perform, Roche did do several clinical trials of tamiflu. One study, published in the Journal of the American Medical Association in 2000, showed that starting tamiflu early in the course of a flu infection reduced the duration of illness by 30 percent, and significantly reduced the complication rate.

A similarly positive study appeared in the The Lancet that same year. In 2003, a meta-analysis of all of the tamiflu studies to date was published. A meta-analysis is sort of a study of studies, combining data from clinical trials to make one huge study.

The 2003 meta-analysis concluded that tamiflu was effective in reducing complications and hospitalization, and helped support the idea of stockpiling tamiflu for future emergencies. Roche claimed overall that tamiflu reduced hospital admissions by over 60 percent, and reduced serious complications of flu by two-thirds.

Tamiflu became part of the standard, recommended care for flu in many countries, and the World Health Organization added it to the list of “essential medications” that should be available to everyone.

However, this would change soon.

Q: Who conducts clinical trials?

Clinical trials are conducted by professional researchers, both in private industries and public institutions.

Q: What is the role of clinical trials in the drug approval process?

Clinical trials play an important role in four phases of the drug approval process. Each phase requires a clinical trial on humans to study different effects of the drug in question.

Q: What makes a good clinical trial?

Good clinical trials share a number of common factors, including blind controls, double-blind randomization, and clearly documented data. These processes help ensure the trial provides a useful result.

Q: How do clinical trials work?

Clinical trials work by studying human in a scientific experiment such as evaluating a new drug. Clinical trials follow a plan that includes a study’s objectives, design, relevant background, and testing on participants.

This article was updated on 7/31/2019

Keep Reading
Can Mindfulness and the Power of Placebos Benefit Our Mental and Physical Health?
A History of Placebos and Their Regulations
The Dangers of Alternative Medicine—Why Scientific Standards Matter