How Big Pharma misleads patients with relative numbers
In 2014, Drs. Adrienne Faerber and David Kreling carefully evaluated whether claims made in consumer-facing drug advertisements were accurate. Their finding? Only 33% of claims were objectively true. This isn’t new or groundbreaking information. The U.S. Food and Drug Administration (FDA) has had what it calls the “Bad Ad” program for years. According to the Bad Ad program:
“Prescription drug advertising must:
- Be accurate
- Balance the risk and benefit information
- Be consistent with the prescribing information approved by FDA
- Only include information that is supported by strong evidence”
The FDA’s Bad Ad program attempts to gather reports of false and misleading drug advertising. But it doesn’t seem to be doing very good at curtailing deceptive drug ads. The direct-to-consumer drug ad market is huge and growing, and media is filled with pitches for expensive, side-effect-laden drugs that use positive undertones to overemphasize benefits. In fact, Pharma ads can say basically whatever they want when they first air because the FDA doesn’t even review the content before it gets to the public. Pharma corporations just face a fine later for misleading consumers, which they’ll easily pay — if they get caught.
From the Bad Ad program’s list of requirements above, there’s one point that is woefully ignored by drug company advertisers, and the average sick patient has no understanding of the magnitude of the deception to which they’re falling victim. It’s the second bullet — “Balance risk and benefit.” It’s possible for drug companies to advertise benefits that are, statistically, “accurate,” but in practical terms are so misleading it ought to constitute fraud. They do this by using what statisticians call “relative risk.”
Now, before we move forward, we need to take a look at just what relative risk is. First, we’re going to learn to understand drug benefits in relative terms and absolute terms.
Relative benefits are represented by the probability of an event in a treated group relative to the probability of the event in an untreated group.
Absolute benefits are the real world benefits, outside the context of comparison to some other intervention. These are what we understand intuitively, and what patients assume all the numbers Pharma ads throw at them mean.
Let’s use some hypothetical numbers to examine absolute risk. For instance, if the average risk of having a heart attack is 1 in 1,000, the average absolute heart attack risk is 0.1%. That’s one-tenth of one percent absolute risk. If a new drug reduces the risk of having a heart attack by 0.5 per 1,000, the new absolute risk is 0.05%. So, the original absolute risk is 1/1,000, and the new absolute risk is 0.5/1,000. That’s a pretty meaningless difference in the real world at the individual level, and most of us intuit it as such. But now we’re going to look at this same scenario in relative terms, and we’ll start to understand why relative numbers get pharma marketing executives so excited.
In relative terms, the same drug produces a 50% reduction in heart attack risk. That’s right — the actual, real benefit is a 0.5% risk reduction, but using relative numbers allows an ad promoting this new drug to claim that it “reduces risk by 50%.” The psychological impact of reducing risk by “0.5%” or “50%” is quite different when the subject is a potentially fatal event.
The power of relative risks is astonishing, especially when the people on the receiving end of the statistics are distraught patients or their caregivers. In 2003, researchers at the University of Texas presented statistics about the effectiveness of chemotherapy for breast cancer to a group of 203 people. To keep the scenario authentic, they presented the statistics to actual adult children making actual treatment decisions for their mothers with breast cancer. When the researchers presented the same treatment effectiveness in terms of relative risk, the children were far more likely to endorse chemotherapy for their sick mothers than when the researchers presented less misleading, absolute statistics. For adult children given effectiveness stats framed in relative terms, over 70% of them favored chemotherapy. But when the children were given the same data framed in clearer statistical terms, only 45% of them chose to submit their moms to chemo.
Big Pharma leverages this misinformation in ads regularly. For instance, in 2016, Bristol-Myers Squibb launched a direct-to-consumer television ad they called “Live Longer.” The ad promoted their new lung cancer drug, Opdivo. In the commercial, apparent patients are shown enjoying life, walking in the park, and seeming generally healthy. The ad displays the following text: “In a clinical trial, Opdivo reduced the risk of dying by 41% compared to chemotherapy (docetaxel).”
However, the actual increase in survival time before the patients’ cancer started spreading was only 21 days. The overall survival benefit was only 3.2 months. But that small bit of additional time certainly wasn’t days spent as the actors in the ad appeared to be spending their time. The following quote appears directly in the fine print of the Opdivo patient brochure:
“OPDIVO was discontinued in 11% of patients, and was delayed in 28% of patients for an adverse reaction. Serious adverse reactions occurred in 46% of patients receiving OPDIVO.”
So, the advertisement used the relative phrase “…reduced the risk of dying by 41%…” to convey a survival improvement of 3.2 months accompanied by a host of side effects. That’s not what sick patients think when someone pitches them the phrase “…reduced the risk of dying by 41%...”
And the cost for those additional 3.2 months of survival (calling it “life” may be a misnomer)? A mere $150,000.
And, alas, here’s a quote from a Medscape article summarizing a newer clinical trial on Opdivo that was published after the ad originally aired:
“[Opdivo] did not yield longer progression-free survival (PFS) than platinum-based chemotherapy when used as a first-line therapy in patients with untreated stage IV or recurrent non–small cell lung cancer (NSCLC)… Furthermore, overall survival (OS) was similar between groups…”
It would be possible to summarize many additional examples of this kind of misrepresentation of benefit through relative numbers. But they’re all very similar stories. The narrative underlying all of them is the exploitation of sick people for profit at very little real benefit to the patient. Indeed, the FDA has surveyed physicians about this very practice due to its prevalence and detrimental impact:
“… about 75 percent of physicians surveyed believed that [direct-to-consumer] ads cause patients to think that the drug works better than it does, and many physicians felt some pressure to prescribe something when patients mentioned DTC ads.”
It should already be clear why relative risk statistics cause Big Pharma marketing executives to salivate. But it might not be immediately clear just who else benefits — the media, who write clickbait headlines based on relative risks; universities, who issue misleading relative-risk-based press releases about the impact of studies conducted at their facilities; medical journals, who leverage studies reporting relative risks to increase reader engagement; and the list goes on.
Patients may think they can rely on their doctors to shield them from Pharma’s misleading claims. Sadly, that’s not true. Not because the doctors are in cahoots with drugmakers, but because they don’t understand the statistics used to pitch drugs either. The FDA’s Office of Prescription Drug Promotion funded researched published in 2017 that showed average docs didn’t understand relative risk:
“ Physicians’ extant knowledge and skills were in the low to middle of the possible score ranges… Physicians with formal education in epidemiology, biostatistics, and research demonstrated higher levels of knowledge and skills. In hypothetical scenarios presenting equivalent effect sizes, the use of relative effect measures was associated with greater perceptions of medication effectiveness and intent to prescribe, compared with the use of absolute effect measures… Critical appraisal knowledge and skills are limited among physicians. The effect measure used can influence perceptions of treatment effectiveness and intent to prescribe.”
Despite not understanding it entirely, most physicians know about the use of relative risks and how misleading they can be. In fact, the topic is prevalent across the medical literature, and has been pointed out as problematic for decades. What’s fascinating is that basically every discussion of relative risk highlights how it can mislead patients. Yet, these deceptive statistics are still in widespread use today.
Before you take any new drug you hear about in any form of advertisement, ask your doctor to tell you the benefits in absolute terms.
Perhaps unsurprisingly, the United States is one of only two countries in the world where direct-to-consumer drug ads are allowed. New Zealand is the other. In Germany, for instance, health literacy organizations have rallied against misleading drug advertising, and the deceptive practice of reporting relative risks instead of absolute numbers has been eradicated there since about 2010.
Perhaps equally unsurprising — though maybe deeply insightful as to the current state of America — Kim Kardashian recently got in trouble for advertising a drug for a Pharma company on Instagram. (Editorial note: writing that sentence killed something inside me).
Despite all the advertisements, it may be that many patients understand only a more limited message. According to an FDA survey, 1 in five patients didn’t even know what disease an advertised drug treated. But the ad apparently looked good enough that they still asked their doctors about it. Doctors agree that drug ads often mislead.
Spending on Pharma ads is on the rise in the U.S., and no sight of tighter regulation is in sight. So where might this lead us? With the recent approval of the first digital pill, we may be headed toward a world of advertisements targeted far more intimately than Facebook can dream of currently. Pretty soon, you might start getting ads from Pharma companies — based on the hidden observations taken by the digital pill you swallowed yesterday — telling you how some new drug will reduce your risk of dying from a disease you didn’t yet know you had. Of course, that risk reduction statistic will be in relative terms.