With just a few days until the long awaited FDA hearings on pharma’s use of social media, the buzz in our little community has certainly picked up. Heck, there’s even a site to track it all courtesy of Fabio Gratton. If you want the latest about these hearings, this site is a pretty good place to start. You can get a bunch of logistics information, see the Twitter buzz, and even download the presentations of those people who have made them available (you can get mine here).
It’s all great. We’re finally going to have some guidelines around social media, which we all have been clamoring about for a long time.
Enter the cynic part of me. I’m not so sure that most people are going to be happy with the results of this meeting. I was talking with a client yesterday who asked if I expected the guidelines to be vague/ambiguous or specific. Presumably, the latter being preferable to marketers. Of course, the guidelines can’t really be specific. For example, they can’t say, “Do this, this and this on Twitter, but not this and that.” The guidelines HAVE to be vague otherwise they’ll be outdated next year as the technology changes. I wouldn’t have it any other way. Ultra-specific and highly directive guidelines is not what you want…trust me. Don’t you already have that with everything else? Have you seen a pharma print ad recently?
That’s 40% ad and 60% fair balance. Do you need more rules like this?
My point is here is that slightly ambiguous rules, will allow many marketers to continue to justify their lack of participation in social media. Sure, nevermind that their peers are already doing quite a lot (see it all on the Pharma Social Media Wiki). Doesn’t matter. They’ve always had one objection and they’ll continue to have it.
That’s right, you guessed it: “adverse events.”
I’ll say this, if I never hear the words “adverse events” from another brand marketer, I’ll live my life out a much happier and less stressed person. When I was a brand manager and social media really wasn’t all the rage as it is today, I don’t think I ever said “adverse event” one time in three years in the context of reporting. It was something that happened and we had a whole team just to handle them, but I didn’t concern myself too much with them as a marketer (thankfully, on a product with very, very few reported adverse events).I relied on more classic “rationale” when I didn’t want to do something…you know, things like “no budget,” “regulatory,” and “ROI.”
But now, that there’s social media, it’s a whole new set of “rationale.” Of course, the one that rises to the top is adverse events. To figure out a bit more about this, I informally asked some pharma marketing people about what they thought “adverse events,” specifically “reportable adverse events,” really were and was pretty surprised by the results. None of the five I asked knew the completely accurate answer. I realized that this might be part of the problem.
It became clear to me and I crafted this analogy: Being afraid of “adverse events” in social media is like being afraid of swimming right after you eat.
Someone always told you that if you went swimming right after you ate that you’d cramp up and sink like a stone and drown a horrible, if sated, death. Except you won’t. No one knows where this “old wives tale” came from or certainly the scientific proof for it, but yet it persists. There’s probably a hint of science in there…maybe you’re body is spending resources digesting the food that would ordinarily be used to help you swim, so therefore…well, you can see how these things happen. Same thing with adverse event reporting…someone heard that one time someone posted an adverse event on a website and a pharma employee saw it and didn’t report it. This employee was subsequently tarred and feathered, fired, spent 12 months in real prison (no white collar stuff) and the company was fined $1 billion.
Ah, urban legends.
So, knowing that there’s a bit of a misunderstanding about adverse event reporting, I’m going to clear it up with some good old-fashioned numbers that I know your average marketer (myself included) can relate to. When the guidelines come back from the FDA and you don’t like them, you won’t have the whole adverse event issue to hide behind anymore. If you still want that security blanket, then stop reading now.
[Important safety tip: I am not an attorney, much less your attorney, so this should not be considered legal advice.]
By now, everyone has seen Nielsen’s report on the incidence of reportable adverse events on health-related sites. If you haven’t, get the paper now. In summary, Nielsen pulled out 500 random posts from the massive amount that they monitor (more on that in a minute). They then analyzed these to see if any of them had a “reportable adverse event.” What do you need to have a “reportable adverse event”? Simple. You need four things: an identifiable patient, an identifiable reporter, a specific drug or biologic involved in the event, and an adverse event or fatal outcome. If you don’t have all four, don’t bother submitting it because the FDA won’t accept it. In their words, “[Without these four pieces] a report on the incident should not be submitted to the FDA because reports without such information make interpretation of their significance difficult, at best, and impossible, in most instances.” [ital. added]
Of the 500 posts Nielsen reviewed, only one (yes, 1) had all four criteria. That’s 0.2% for those scoring at home. So, 0.2% of all posts, should contain a reportable adverse event. Impossible you say? Here’s why it’s so low: the rate-limiting factor here isn’t what many people think. It’s not that the, say, identifiable patient piece is difficult because of the anonymity of the Internet, for example. It’s much simpler than that. Most of what you might think is an “event” is not required to be reported.
In the Nielsen study, they looked for events that would need to be reported within 15 days of receipt per FDA regulations. These are the most serious events and have the strictest reporting standards. Events that must be reported within 15 days are those that are BOTH serious/life-threatening AND “unexpected.” The former is pretty simple and includes outcomes such as “death, a life-threatening…experience, inpatient hospitalization or prolongation of existing hospitalization, a persistent or significant disability/incapacity, or a congenital anomaly/birth defect.” The latter, “unexpected,” is a bit trickier, but also is pretty straight forward. “Unexpected” is “any adverse drug experience that is not listed in the current labeling for the drug product.” [Read the full code on FDA's site]
This means that if someone reports they got a headache while taking your drug, for example, and headache is listed in your labeling as a known side effect, then you do not need to report this in 15 days or any other day. If, on the other hand, they report that their hair fell out and that’s not in your label, then it might be reportable since it’s “unexpected.” However, it’s not serious or life threatenting, so it would not be required to be reported within 15 days. Events that are EITHER serious/life-threatening OR “unexpected” need to be periodically reported to FDA (hint: you’re already doing this). Drugs launched in the three previous years needs need to do quarterly updates on these types of events. Those drugs that have been around longer than three years need to do annual reports. For the purposes of this discussion, I’m really only considering the events that would fall under the 15-day rule (as did Nielsen), as these are the ones that are most labor intensive and require immediate attention. Those that fall outside this rule can use your normal channels for reporting back to FDA, which you’re doing already.
Most of the confusion I’ve seen is around this issue is around defining exactly what qualifies as an adverse event that requires action. I mentioned already that only 1 posting in the 500 Nielsen reviewed met all four criteria including the adverse event part. However, I asked Nielsen for a bit more information on this and got some great help from my colleague at Nielsen, Melissa Davies. She informed me that only 4 of the 500 posts had events that were reportable under the 15-day rule (i.e., an event that was BOTH serious/life-threatening AND “unexpected.”) . Three of these were missing one of the other criteria required for reporting, so you’re left with one in 500. I’ve heard a bunch of debate about this study because it doesn’t consider some companies’ required due diligence around investigating reports that are missing one of the four components. For example, there might a company policy that says you must investigate to find a reporter’s name including sending them a direct message or email from the site where the event was posted. Nielsen didn’t (and shouldn’t) account for every company’s policy on this.
However, regardless of your due diligence policy, you can’t report something that didn’t happen. If there’s no adverse event, it doesn’t matter if you have the other three pieces or not. Show’s over. No need to go further. So, the way I see it, the absolute worst case scenario is that only 4 out of 500 posts are potentially reportable, which is 0.8%. That’s not a big number, of course, but how many posts are there every day? That would tell you how many potentially reportable adverse events there are in a month or year.
Nielsen shared with me a bunch of data for my FDA testimony. They shared that their BuzzMetrics product regularly scans more than 100 million sites. Around 1,350 of these are healthcare-specific and have some discussion component to them (blogs, forums, Q&A, etc.). These 1,350 generate almost 83,000 new posts each DAY.
Here comes the math:
Math problem #1: Multiply the total number of posts per day by the number of posts that contain adverse events that are both serious/life-threatening AND unexpected and have the other three criteria required for an adverse event = adverse events generated per day that need to be reported in 15 days to the FDA.
Math problem #1 answer: 83,000 X 0.2% = 166
If you want to use the more conservative number…
Math problem #2: Multiply the total number of posts per day by the number of posts that contain adverse events that are both serious/life-threatening AND unexpected, but do not necessarily have one or more of other three criteria to have a reportable event = the maximum number of adverse events generated per day that need to be reported in 15 days to the FDA.
Math problem #2 answer: 83,000 X 0.8% = 644
Let me put that to you another way…the ENTIRE PHARMA INDUSTRY, assuming they were responsible for EVERY SINGLE discussion online, would have to manage 166 reportable adverse events per day. Divide that out across the number of companies out and there’s not a lot of work for people to do.
Of course, you are not required to monitor everything out there and you only have to report events that are reported to you (which would include posts or comments on sites you manage or control) or events that you come across elsewhere online. Unless you’re a really fast searcher, then you’re probably not going to read through 83,000 posts a day (hint: that’s about one per second). If you only concentrate on your little neck of the woods, then you might never come across a reportable adverse event in months. How many posts or comments does your social media program get? Looking at what’s out there, not many. So, if it takes 500 posts to see one, then you might not see one in your lifetime.
The point of all this is simple. Are we really holding back everything there is to gain from properly engaging in social media because of 166 reportable events a day? Does that add up for anyone else?