Pharma, FDA Wrestle with Social Media Tools

The fast-changing nature of the Internet continually opens new opportunities for communicating with consumers and physicians. It also brings new regulatory challenges, as the FDA struggles to keep up—and pharmaceutical companies try to take advantage of technology without falling foul of the rules.

Currently, the spotlight is on two technology areas: paid search; and the rapidly growing technologies collectively known as social media, including, most prominently, Facebook and Twitter.

Within the Internet realm, the two could hardly be more different: paid search is an extremely established advertising technique used throughout the pharmaceutical industry, while social media represent uncharted territory in which the primary concern is how to participate in sites that could include user-generated content containing adverse event reports (AERs), negative comments about brands, and discussions of off-label uses.

Both areas were extensively covered—with a huge amount of comment and discussion—in FDA hearings last November. The hearings were designed to gather input on five issues related to promotion via the Internet and social media: fair balance on the internet; industry responsibility for content posted on social media and other websites; posting corrective information; the use of links; and adverse event reporting.

Following the hearings, FDA’s DDMAC is currently working on updated guidance expected to cover at least some of these issues, and widely anticipated in draft form this year—though FDA won’t confirm the date. The guidance is expected to be technology-independent, to ensure it doesn’t quickly become worthless. “Given how fast the technology is evolving, DDMAC does not want to issue guidance that may become quickly outdated,” an FDA spokesperson says. “DDMAC continues to plan its focus for guidance in this area on concepts that have long-term applicability … rather than on a particular venue or tool.” It is possible that DDMAC may develop multiple draft guidance documents related to Internet and social media promotion of FDA-regulated medical products, she says.

The DDMAC guidance won’t cover AER reporting based on information generated on Internet and social media sites, an area of huge concern among pharmaceutical companies: The lead role in that area falls to CDER Office of Surveillance and Epidemiology. FDA says only that it is considering whether to issue future guidance in this area.

To put all this in perspective, pharmaceutical promotional spending on the Internet still accounts for only a tiny fraction of the industry’s total promotional expenditure, according to data from The Nielsen Company (New York)—though it is growing fast. For example, in 2009 spending on online direct-to-consumer display ads surged by almost a third in 2009 to about $117M, but still represented less than 4% of the amount that pharmaceutical companies spent on TV DTC ads.

Paid search

The story of paid search provides a curious illustration of FDA’s difficulty in regulating Internet-based promotion. In April 2009, the agency fired off warning letters to 14 pharmaceutical companies that were using the so-called—but non-existent—“one-click rule.” Using this approach, sponsored search results on the link took typically included both the drug’s brand name and its indication, along with a link. To view the risk information, users needed to click on the link, which led them to a brand site where the risk information was displayed. FDA‘s letters said that this approach violated the requirement for fair balance between risk and benefit information.

Two proposals presented at the November FDA hearings offered modified one-click approaches that attempted to address the need for fair balance. Google proposed including brief drug-specific risk information along with the drug’s indication on the search results page, with separate links to additional product and risk information; PhRMA proposed a generic “all drugs have risks” statement on the results page, together with an “FDA-approved” symbol to indicate that it refers to approved products and information (see Fig. 2).

The warning letters had a significant effect, but perhaps not the one that was intended. Initially, the industry’s use of paid search dropped sharply. However, companies quickly adopted two new approaches to staying compliant. In one, the drug’s name isn’t mentioned in the sponsored search link: The link appears to be an unbranded disease state site, but this is merely a “vanity URL” that redirects to the brand website. The other approach mentions the drug’s name but not the indication.

Experts say the new approaches have two disadvantages: they are less effective for industry, and more misleading for consumers. At the hearing, Google presented data showing that users click on these links less frequently. ComScore (Reston, VA) VP John Mangano agrees that the use of vanity URLS has drawbacks. “It allows you to ultimately follow the rules as they are understood today,” he says. “That said, I’m not sure it is the clearest for the consumer, because they don’t know where they are going until they get there.”

FDA appeared, at the hearing, to get the message, says John Kamp, executive director of the Coalition for Healthcare Communication. “I think that the agency understood some pretty compelling evidence that their enforcement wasn’t achieving what they wanted it to achieve. There were fewer patients learning about side effects after the warning letters than before—which is not what FDA intended.”

After the initial drop off, the industry’s use of paid search has rebounded, says comScore’s Mangano, though in general pharmaceutical companies are not buying search terms as aggressively as before. For some brands, paid search is generating almost as much traffic to the brand website as before the warning letters, according to comScore data.

Social media

As the use of social media has exploded—according to Nielsen data, Facebook is now among the top three most-visited websites—so has its value as a method for exchanging healthcare information. For example, at health-information sites such as WebMD and HealthCentral, users engage in lively discussions of their experiences with regulated drugs, frequently discussing off-label uses and side effects.

It’s the presence of this user-generated content that has caused pharmaceutical companies to be fearful of social media, and correspondingly slow to get involved. There are several concerns; the most frequently cited is the fear of being held responsible for unearthing and reporting AERs mentioned on social-media sites.

The industry’s activity now appears to be increasing rapidly. Jonathan Richman, director of strategic planning for ad agency Bridge Worldwide (Cincinnati), maintains a list of pharma and health-related social media activities at the Doseofdigital.com website. He says the number of social media sites listed has grown at least five-fold over the past year, with most of the growth taking place in the last six to nine months. The list spans everything from pharma-owned unbranded disease-management websites and blogs to industry-contributed content on Facebook, YouTube, and Twitter.

The growth is a little misleading, because pharmaceutical companies are in general still carefully restricting their efforts to efforts where they are less likely to encounter regulatory issues. Typical examples are unbranded disease-management information sites operated either as standalone websites or on Facebook, YouTube or other major social-media sites. “Because they are unbranded, they don’t need to worry as much about adverse event reporting and fair balance,” says Melissa Wells, who heads the healthcare group at Nielsen Online.

Often, the ability to provide user-generated comments is disabled, so the sites are “really not true social media,” she adds. Where comments are allowed, the sites almost invariably screen or “pre-moderate” them to filter out potential problems. An example of the possible problems—and pharmaceutical companies’ reaction—occurred on Sanofi-Aventis’ Voices Facebook page, which accepted comments until a reader repeatedly complained of side effects, after which the ability to comment was removed and a disclaimer added.

Meanwhile, pharmaceutical companies have generally avoided directly taking part in social media conversations that they don’t control. For example, as noted by FDA in its announcement of last year’s hearings, pharmaceutical companies have avoided posting corrective information, because of fears that they could be held responsible for all the content on the sites. As a result, while social-media discussions may include inaccurate information about drug uses and side-effects posted by just about anyone, the manufacturers are not able to respond with accurate approved content.

The effect of this was highlighted by WEGO Health (Boston), a patient community site that, for the FDA hearings, surveyed the “patient activists” that run its discussion groups. “They were saying that pharmaceutical companies are largely absent, and that it’s a glaring hole,” says WEGO CEO Jack Barrette, “As a result of their absence, due to antiquated regulations, we’re not getting good information. But by no means are we saying let’s lift the regulations—pharmaceutical companies need to watch for misinformation and correct it, and they need to follow regulations for adverse event reporting very closely.”

At some sites, pharmaceutical manufacturers also have held back from placing ads within discussions, where they could appear next to adverse events or off-label discussions, even though the companies advertise elsewhere on the site. “In actual discussions—where people are talking to each other or answering each other’s questions—by and large, there is no branded pharma advertising,” says Jeremy Shane, HealthCentral president and chief operating officer. “All publishers are looking for clarity from the FDA.”

However, Barrette says that a lack of advertising is largely due to other concerns. “There’s no question that pharmaceutical companies are very concerned with user-generated content—but that’s mostly because of the concern of negative adjacency—it’s mostly more of a brand issue than a regulatory issue. I see that going away very quickly—many large companies are beginning to change their position.”

Adverse events

Still, if the most-cited social media concern is adverse-event reporting, and one key aspect of this is how much of the Internet a pharma company is supposed to be monitoring to detect these events. “What is your responsibility to monitor all these conversations whether you are advertising or not? That is by no means clear, I believe, in the minds of pharmaceutical companies or regulators,” Barrette says.

Existing draft guidance states that companies should review Internet sites that they sponsor; they are not responsible for reviewing sites they don’t sponsor. One concern is what constitutes sponsorship; one specific proposal presented by HealthCentral’s CEO Chris Schroeder suggested that merely advertising on a page should not, in itself, constitute sponsorship; and that the definition of sponsorship should be limited to specific circumstances, such as those in which a pharmaceutical company owns or has editorial control or influence over the content of the web page.

If pharmaceutical companies are ultimately determined to responsible for monitoring vast areas of the Internet for adverse events, how much work would it take? One often-cited analysis, by Richman and former Nielsen analyst Melissa Davies, concluded that only 1 in 500 posts contain enough information to meet all four criteria required for a reportable event: an identifiable patient, an identifiable reporter, a specific drug or biologic involved in the event, and an adverse event. Even across nearly 1,400 health-related web sources tracked by Nielsen, this would result in only about 166 AEs per day for the entire pharmaceutical industry. However, notes Nielsen’s Wells, some companies use a more conservative approach when assessing whether an AE should be reported — they may require fewer than four criteria, which could greatly increase the number of reports.

Amid the current confusion over social media, some see at least of glimmer of hope that—if clear guidance emerges that facilitates two-way communication with consumers—social media could present a way for the pharmaceutical industry to rebuild the public’s trust in the industry. “So much of the industry’s reputation is based on the inability to communicate because of regulations,” says Barrette. “I believe transparently communicating through social media is the pharmaceutical industry’s last best hope to rebuild its reputation with consumers.”

Indeed, the pharmaceutical industry took a positive step in its approach to last year’s hearing, says Richard Minoff, managing partner of 1 Global Partners (King of Prussia, PA), an industry consulting firm. “It seems logical that the FDA, and pharma naysayers, would be pleased with the responsibility industry has shown by asking for a clear set of rules and regulations designed to ensure appropriateness,” he says. “This should not be an adversarial situation, but rather a unique collaborative opportunity. Is this really possible? Based on history, I’d say no.” PC