Give It to Me Straight!

If you've ever flipped through a magazine in a checkout line, or even on your iPad, then you've probably seen a print DTC advertisement. If you bothered to look beyond the images and actually read the text, you may have noticed a marked absence of numbers. Instead, descriptive, non-numeric language such as "some," "frequently," "rare," or "common" is used to describe the drug's side effects, in combination with blunt statements like, "It may cause harm to your heart or brain," (from a recent Sanofi ad for Lantus), or "Serious allergic reactions may occur," (from a recent Eisai/Janssen ad for Aciphex).

As a defense against failure-to-warn claims, this qualitative, laundry-list articulation of a drug's risks and benefits may be smart, but as a way to communicate clearly with consumers and patients, it's obviously lacking. Enter the drug facts box, a concise, quantitative, easy-to-read format for sharing numeric information about a drug's efficacy and side effects. The goal, according to proponents of the facts box, is to make it easier for consumers and physicians (facts boxes would also appear in a drug's promotional labeling) to parse the underlying clinical data for a given product, which would in turn facilitate a more informed decision about treatment.

This theory has been tested directly by academic researchers, and a body of literature has emerged; the methodologies, participants, and objectives vary, but the resulting consensus in the academic community is that a standardized facts box or table is superior to current presentations of risk and benefit. Indeed, the current members of FDA's Risk Communication Advisory Committee (RCAC) released an extensive literary review in preparation for its most recent meeting, held Nov. 17 to 18 in Washington, DC. The report, titled "Quantitative Summary of the Benefits and Risks of Prescription Drugs," analyzes the findings of 52 journal articles, with a focus on answering two stated questions: 1) "What is the value of quantitative information or summaries about the risks and benefits of medical interventions for consumers, patients, and clinicians?" and 2) "How does presentation of the quantitative information influence consumers,' patients,' and clinicians' processing and understanding of the risks and benefits of medical interventions?"

These seem like pretty straightforward questions, but as anyone who has personally witnessed an RCAC meeting knows, very little is straightforward in the realm of risk communication theory. As the famed analytical psychologist Carl Jung once wrote, "The shoe that fits one person pinches another; there is no recipe ... that suits all cases." This point is very well taken, given the RCAC November session, but first, it's important to understand what happened the first time the drug facts box went to Washington.

In 2009, the FDA's RCAC—then chaired by Baruch Fischhoff, professor of social and decision sciences and of engineering and public policy at Carnegie Mellon University—took a vote on whether a standardized drug facts box should be adopted by FDA for use in pharmaceutical print advertising and promotional labeling. The committee said yes, unanimously. Then nothing happened for two years. When the Patient Protection and Affordable Care Act passed, it contained a mandate (Section 3507) ordering the US Department of Health and Human Services (HHS) to review the evidence for or against a quantitative summary of a drug's risk and benefit, "such as a table or drug facts box," and to make a judgment. HHS responded by filing a report to Congress last March, requesting at least three more years for additional research.

Lisa Schwartz and Steven Woloshin, a wife and husband team of professors at Dartmouth Medical School, and two of the primary contributors to the body of literature around facts boxes, don't get it. Asked about the justifications for another three years of research, Woloshin says he's "completely baffled," an opinion he made known in a New York Times editorial printed on July 4. "It's crazy that people—doctors, patients, tax payers—don't have ready access to summaries of [a drug's efficacy and risks] in a way that we've shown can help them make better decisions," says Woloshin. "No one has published any important criticism of the [facts box] studies."

Conducting research on any topic always leads to new questions. At some point action should be taken, especially when it comes to health outcomes. "One can always improve upon something, but you'd have to be able to claim that your research is going to produce powerfully different results, to justify delaying implementation of something which I think serves everybody," says Fischhoff. "Given the amount of work that's gone into the facts box, and the evidence that supports it, it's hard for me to believe that alternate design [studies conducted by HHS/FDA] would justify a delay of a couple of years." Fischhoff describes the facts box as having a "sound design, both analytically and empirically ... it follows the theories of risk communication, it has demonstrated efficacy in its field trials, and is clearly much better than what we have out now."

The November RCAC meeting set out to answer the two key questions by identifying perceived gaps in the literature. It quickly turned into a gap-a-thon, with participants identifying gaps everywhere they looked. "What about the consumer's literacy levels, and cultural diversity issues?" asked RCAC member Sokoya Finch Nealy. "Is that a gap?" Sure, "that's fair to say," answered another member. "What about graphs; are those really non-numeric? Some of them have numbers in them," said Nananda Col. "That's a gap," someone else declared. At one point, a member who only recently joined the committee asked, "What are we aiming to do with this exercise ... is it to review and critique the study?" Someone else answered, "To give advice and pointers to FDA on regulatory policy." Craig Andrews identified "two major gaps," including "the external validity of the various findings in realistic settings, like broadcast and print," and the fact that "people bring biases to [drug ads], different moods and emotional states ... the 'magic bullet' effect, etc. They may process these things differently." Valerie Reyna wondered if "it is even possible to get a script for a standardized presentation." Gap, gap, gap ...

Yolanda Fultz-Morris, a public affairs spokesperson at the FDA, said the November RCAC meeting reflects "what we need to know for the types of actual cases that are really quite common in drug regulation, where there is not a single applicable number, or where the meaning of numbers may vary based on the [clinical] studies producing them," such as when a product has different indications; was studied in different populations; contains differing endpoints, such as composite endpoints or surrogate markers; or when some products are notably more dangerous than others. "FDA does indeed wish to facilitate good communication of the efficacy and risks of prescription drugs, but we also wish to avoid unintended consequences that could cause a worse understanding. The only way we know to avoid unintended consequences—as much as possible at least—is to take the time to attend to the practical variations from the theoretical ideal," says Fultz-Morris. Short of implementing the facts box, though, it's hard to say how they would perform in a practical, real-world setting.

In an editorial piece commenting on Schwartz and Woloshin's 2009 facts box article in the Annals of Internal Medicine, titled "Communicating Drug Benefits and Risks Effectively: There Must Be a Better Way," authors Jerry Avorn and William Shrank call the current risk presentation in consumer drug ads "a virtual museum of poor communication: the print is tiny; the prose is usually dull, stiff, and hard to understand; and vital facts are buried in a sea of less-relevant data. The format of the information can mask important side effects, as well as—ironically—numb the reader with so many worries that a perfectly worthy treatment may seem too toxic to take." Even so, it seems FDA is at least three more years away from implementing a standardized, quantitative summary of a drug's risks and benefits.

Woloshin says he hasn't looked into broadcast media applications for the facts box, but suggests that certain warnings, such as approval on the basis of a surrogate outcome, or the main efficacy finding for a drug, could be easily inserted into broadcast spots. For broadcast, "we imagine the primary goal would be to point people to a drug [facts] box, either in print or online ... We have not tested these ideas, but hope to in the future," he says.

PhRMA, for its part, is taking a measured approach to the issue. "Careful thought must be given to how to convey risk-benefit information in a simplified format, in a way that remains relevant to the real-world practice of medicine and patients," says Matthew Bennett, PhRMA's senior vp of communications, in an email. "The desire to present simple 'facts' should not trump the overriding goal of ensuring that information about a drug's benefits and risks is accurate, and not misleading." Most people don't have a problem with facts, particularly when they're presented in a readable format. But perhaps more research is needed, to know for sure.