Thursday, August 05, 2010

Implications of FDA’s Warning Letter to Novartis Regarding Facebook Share Widget

You know you are getting old when the FDA knows more about Facebook than you do! In a July 29, 2010 letter to Novartis, FDA cited violations with regard to a "Facebook Share" social media widget that generates Novartis-created information for Tasigna that can be shared with Facebook users (i.e., “shared content”). FDA said: "The shared content is misleading because it makes representations about the efficacy of Tasigna but fails to communicate any risk information associated with the use of this drug."

To help the recipient of the letter -- Lisa Drucker, Director, Regulatory Affairs -- understand what a Facebook Share social media widget is, the letter includes several footnotes that supplied technical details and explanations:
"Facebook Share is a way for users of Facebook to share articles, pages, video, or flash content of a site with other Facebook users. Over two billion pieces of content are shared each week through Facebook. With two clicks, visitors to a website can share any page of that website through Facebook by generating a link to the page, along with a thumbnail image and a brief description (i.e., 'shared content') that will appear on the users’ profiles and, depending on privacy settings, in the home page stream of all of the users’ friends. Each time a link is shared by one user, potentially hundreds of new people may see and/or click through on the link."
and
"As described below, the shared content for Tasigna generated by the Facebook Share social media widget was developed by Novartis and, although Facebook users can add additional comments that are displayed separately from the Tasigna information, the shared content cannot be modified by Facebook users who use this Facebook Share social media widget."
and
"We also note multiple Tasigna web pages contain widgets that allow users to share content via other social media applications offered via the “Share This” tool (http://sharethis.com). Some of the content available to share through these other social media applications raise similar issues to those discussed in this letter."
In a separate document, FDA included a screen shot showing what one of the widgets looked like before Novartis edited its Facebook page:

This looks very much like one of those infamous Google bAdwords (see Lunesta, Google, and "bAdWords")!

Luckily, Digitas Health published a "Facebook Regulatory Alert" that clarified a few issues and sheds some light on how this widget was created.
"The descriptive content displayed by Facebook on the user's page via the Share feature pulled from the [Tasigna] website metadata [my emphasis], which is invisible to most users but relied on by search engines, such as Google, Yahoo!®,and Bing™, to provide a description of the page's content," notes the Digital Health document.
I Warned This May Happen!
In a previous post, I warned that the FDA may cite certain content generated from metadata (see "Are Organic Search Results Next on FDA's Chopping Block?"). While the Novartis example is not exactly what I was talking about, it's pretty close because the FDA is now citing content that is usually hidden but made visible by widgets. If it can do that, why not cite natural search content that is also picked up from metadata tags?

Possibly thinking along the same lines as I am, Digitas Health ended its Alert with this recommendation:
"Because website metadata is used both by search engines in generating organic search listings and by social media channels, such as the Facebook Share functionality, Digitas Health advocates that all site metadata should be included in internal medical/legal/regulatory review and as part of mandatory FDA submissions."

P.S. Novartis has updated its meta tag on the Tasigna Web site. It now reads: "Tasigna (nilotinib) 150-mg and 200-mg capsules from Novartis is a treatment for Ph+ Chronic Myeloid Leukemia in newly diagnosed adult patients in chronic phase or patients in chronic or accelerated phase who are resistant to Gleevec."

FDA had complained about the previous wording: "These statements of Tasigna’s indication are incomplete, and misleadingly imply that the drug is approved to treat all individuals with Ph+ CML, when this is not the case. At the time this shared content was originally disseminated, Tasigna was only approved as a second-line option after failure or intolerance to prior therapy that included imatinib. Furthermore, Tasigna is only approved for use in patients with Ph+ CML in the chronic or accelerated phases."

8 comments:

  1. Thanks for this John. Just another example of when it comes to social media and the Internet - really nothing has changed - the SAME RULES APPLY!

    ReplyDelete
  2. Hi John,

    It's the message NOT the medium! This content did not follow existing FDA guidelines, it just so happened it was to be shared on FAcebook.

    ReplyDelete
  3. Wendy,

    Yes, but... did you see this tweet posted by @ddwebster, cancer survivor and student of social media:

    "Ah, chicken or egg debate? RT @pharmaguy: FDA lttr 2 Novartis: Is it msg or medium? w/o medium-specific metadata, FDA wuld not hv seen msg."

    ReplyDelete
  4. The point is no matter where pharma puts information it has to follow DDMAC guidelines for fair balance.

    ReplyDelete
  5. Hi John

    Just a point regarding meta *tags* and Google Search:

    According to Matt Cutts, head of Web spam at Google, Google does not really use the keyword meta tag in its search.

    More here:

    http://www.mattcutts.com/blog/keywords-meta-tag-in-web-search/

    Other search engines might, but Google doesn't. Meta Description might be used, but keyword meta tags aren't really used in their keyword ranking algorithm.

    If you think about it, the Web and search engines have dramatically evolved over the last 15 years, and so it makes sense that the early need for HTML meta data isn't nearly as important all other off-site meta data (eg links, backtweets, etc.)

    Which raises another question: Is this as an important issue as the FDA's letter implies? Is the FDA truly up to snuff, or is this a reflection of late 1990s understanding of the Web.

    That is: aren't there other priorities at this point?
    (Eileen's point about fair balance is duly noted - my question goes to the true impact here: Does this *really* protect the public?)

    So here's the question:

    Which is

    @PhilBaumann

    ReplyDelete
  6. Sorry, Phil. What was the question?

    ReplyDelete
  7. Ooops - (Disregard those 2 last lines - forgot to delete them) The questions was raised. Here it is again:

    Which raises another question: Is this as an important issue as the FDA's letter implies? Is the FDA truly up to snuff, or is this a reflection of late 1990s understanding of the Web.

    ReplyDelete
  8. just a point regarding the point about meta data -

    Google still does use meta data from the tags.

    Keyword tags, for instance, while no longer a direct organic rank factor, can impact the time it takes a site to begin to appear in the results for specific terms.

    Description tags and title tags are still very important to Google. Description tags often appear as the blurb under a result in the SERP. It's important enough that if you don't provide a description tag - Google will create their own!

    In my experience, Google, Yahoo and Bing reward sites with higher rank based on how well the site meets the needs of the user. The search algorithms are designed to measure the factors Google ( or another search engine) believe are most important.

    Given a choice between:

    1) a well optimized site, with good internal linking and great home page content with no back links or other off page optimization

    or

    2) a site with a ton of manufactured back links


    I'd take option 1 every time. So would Google.

    In terms of the FDA - I agree with John - it's all about fair balance. I think the FDA recognizes that the channel(s) will continue to modify so trying to regulate channels is a non-starter.

    Rather then it being evidence of 1990's understanding - I think it reflects the realities of ongoing technology changes.

    ReplyDelete