Monday, October 1, 2007

Sermo Confidential?

Filed under: Medgadget Exclusive , Net News

A recently surfaced document from Sermo, a physician networking website, casts serious doubts on the company's willingness or ability to protect identities of its physician members, MEDGADGET has learned. The document, designated as "Confidential," and titled Utility of a market-based online physicians' community to detect and clarify signals related to medical product safety, was likely presented to the FDA on March 7, 2007, in an effort to incorporate Sermo's knowledge base into FDA's Sentinel Network, a system, currently under construction, to monitor drug and device safety.

The document seems to show the workings of AlphaMD, the back entrance to Sermo that clients pay to get a peek into. Sermo's business model is based on an interesting idea of having a forum for physicians and, at the same time, presenting a platform for interested parties to monitor the conversation. Here's how Daniel Palestrant, MD, CEO of Sermo described their business model to Healthcare Vox in a November 14, 2006 interview:

In this model, the community generates "heat maps" around different subject areas or ideas. This is valuable information to our clients. We are able to sell this information, without compromising physician anonymity in way shape or form, while providing an environment that is free from outside influence and advertising. This business model is accepted by physicians and appears to be legitimizing Sermo in that information arbitrage is a real business model and physicians don't fear a "bait and switch" looming in their future.

According to the document shown above, and available on FDA's website (www.fda.gov/ohrms/dockets/dockets/07n0016/07n-0016-ts00028-frost.pdf), as well as an embedded copy at the end of this post, there is indication that Sermo might be providing users' real names to third parties that probably pay to read what's on the site. The above screen shot of the "Client View" from page 10 of the document (click on image to enlarge), shows the names of three doctors under the "Watched Physicians" section of the site. Yet, according to Sermo's Privacy Policy, Article 3 Section A, "Sermo may share aggregated demographic information with Sermo's partners. This is not linked to any personal information that can identify any individual person.". It is obvious, though, that individual names of doctors are clearly present in the screen shot.

On page 11 of the document, one can see a screen shot containing a section called "Most Active Authors Tagging with 'byetta'", and below it - a list of aliases of doctors. This clearly points to a violation of Sermo's Privacy Policy Article 3 Section E, which reads: "Your name or alias and IP address may be recorded for purposes of maintaining Your own account within the Forums and preventing abuses of the forum (see forum or online community rules for more details). This information is not used to monitor Your activity within a forum, nor is it used to identify You outside www.sermo.com in any way. In order to diffuse the information in the Site's Forums to a wider audience, Sermo may, from time to time, collect some of Your postings and group them together to use in a specific publication, print, electronic mailing or other public dissemination. At no point however will Your name, alias or IP address be revealed in any publication."

Moreover, there appears to be a "Physicians" tab on the top of the client page, leading us to wonder what information is provided under that tab. Sadly, the screen shot of that page is not presented in the document.

People familiar with the document speculated to MEDGADGET on possible reasons for Sermo to possibly reveal identities of its physicians, whether on purpose or not. Simple sloppiness on the part of developers, simplification of the accounts payable process, federal mandate to reveal identities of people reporting drug or device problems, and other reasons were cited. No one seems to have a satisfactory answer at this time. The big question still remains: Does Sermo protect the identities of its physicians?

MEDGADGET has not contacted Sermo regarding this, as we have had no response to our previous attempts to contact them. Once they address this important story, we'll point a link to the Sermo's position.

Testimony of Alex Frost, VP for Research Initiatives, Sermo Inc before FDA's Sentinel Network Public Meeting (March 7, 2007)...

Utility of a market-based online physicians' community to detect and clarify signals related to medical product safety document on FDA website...

Thanks to Edward , Andrew, and Judy.

UPDATE: Dr. Daniel Palestrant, CEO of Sermo, has thoughtfully addressed our concerns inside Sermo.com, and has posted screen shots of the company's backend, AlphaMD:

Well, in one of those meetings, a slide stack that included the first version of a mock-up of what our client interface could look like was used. Those slides are now a matter of public record, easily downloadable from the FDA site (I have previously referenced them elsewhere within Sermo). However, one of the slides includes a conceptual client interface that shows physician names. The physician names are fictitious and the feature was never built (indeed, the slide is clearly labeled "not actual data"). Software development is an iterative process. Mock-ups are part of a process where we get feedback from both our clients and our physicians. Sermo has gone through several different iterations of our product and ultimately we create something that best meets the interest of all parties, while adhering to the strict policies that we have set forth with regards to protecting physician confidentiality.

Sermo has never shared confidential information. Our model has always been and will continue to be based on physicians deciding who gets to see what information. I have attached a screen shot of the level of detail that the client interface includes. While we have never planned to include physician names, and this was a mock-up with fictitious name, if it has created confusion or acted as an opportunity for others to cast dispersions on this community, I apologize. I take full responsibility for it being used out of context.


(click to enlarge)

UPDATE: Sermo CEO Offers Answers...

email this article to a friend      print this!      add this article ... Add this article to: digg StumbleUpon Facebook      comments and peer reviews (29)






replies: 29 comments
Open comments are not moderated, although abusive and vulgar remarks may be deleted. Opinions expressed do not necessarily reflect the views of Medgadget.com. Please consult our disclaimer.

Hmmmm.....when I click on the FDA link you supplied, it seems that the slide that you reference has a clear disclaimer that states "-Not Actual Data", that you chose to edit out. As a longtime fan of this website (I have no idea how you guys do it, with REAL jobs on the side), can you let us know if you verified any of this? I have always held this site's journalistic integrity to the highest level. I assume you've confirmed with an independent source that this is something more than slideware, do we know that it is actually in production?


Posted by: Marc
on September 30, 2007 06:06 PM GMT

"Not Actual Data" as a discaimer does not change the fact that the application shown in the screen is clearly designed to provide real physician names and activity to Sermo's customers. This presentation was delivered to the FDA - do you believe that Sermo said during that presentation - oh, by the way, please ignore all of the individual physician names and the Physicans tab and the Watchlist(s) - there was some confusion when we printed these out this morning at Kinkos... ??? This application is designed to do something Sermo expressly claims they do not do.

Sermo highlights an article from the Wall Street Journal on their site, and consistently touts their agreement with the American Medical Association. I wonder what either of these organizations would do if they discover this? The AMA I have to imagine could not have known about this, or they would not have endorsed Sermo in any way.

Shame on you Sermo!


Posted by: Andrew
on September 30, 2007 06:25 PM GMT

So, I'll confess to being a medgagdet freak. I'm over here at Kaiser Oakland and I also also admit that meeting some of the contributors at the Health 2,0 cocktail thing was a bit of a rush. I check the site daily and I'm always thrilled by the constant and reliable information. During the week, I scan the site, and make a note of what I want to dive in and read on the weekend. Sunday's are my "catch-up", the first thing I thought when I saw this article, was, well, this is odd. Different tone from the normal story, no quotes from the company that you always seem to get, and then the last comment at the end, claiming that you chose not to contact the company....just seemed different. I started looking back at the other stories on this, they seem very accusatory, very expose. And then I realized what was so odd. I NEVER see new stories on this site on Sundays. It is my chance to catch up. I feel like I start the week "on top" of things. In fact looking back, medgadget has not posted a story on a Sunday for over 6 months (actually, I gave up looking after May, so it might be longer). So what gives? Is this a vendetta? Was this so burning that you needed to do something so unprecedented (the most pressing thing that you have reported on at least six months) and get something in a Sunday? I LOVE this site. Please, please do not turn it into some sort of an arena for open warfare with a competing company. I am here and I am loyal, just keep up what you've been doing. I don't get this obsession with Sermo.


Posted by: dr.imony.ca
on September 30, 2007 06:36 PM GMT

Journalistic integrity...are you kidding me? Click on the FDA link, there is a clear disclaimer on the Slide that Medgadget removed. So they doctored the information, rushed to get it onto the site, couldn't wait to get ANY quote to verify this, and they wants us to believe this. Hey Medgagdet advertisers are you reading this STUFF? I'm not kidding. This is journalism? What ever happened to verifying your sources, to not modifying the data? Any one coudhave accomplished what they did with just a google search, and they would get accurate data, not something doctored, by the doctors to make their point. Medgadget. So long and farewell. I'll see everyone over at CNET. Also, check out www.thehealthcareblog.com blows this crap out of the water.


Posted by: Feliciano Guimarães
on September 30, 2007 06:50 PM GMT

info@johnmuirhealth.com

Dear Sirs:

I can see that we are a major sponsor of the Medgagdet website. Our banner add is a fixture on the top of their landing page. As both an employee of JMH and a patient (delivered two babies here), I am disgusted by our sponsorship of this site. This site was once a bacon of IT innovation for Northern California. It has become a sad, pathetic match between this site and Sermo. Why are we sponsoring this? I am married to a rheumatologist who loves Sermo and we've discussed the buzz among doctors about Sermo several, several times. I wonder if Sermo accepts advertising.

Yours,


>


Posted by: mbono1
on September 30, 2007 08:02 PM GMT

Feliciano:

This post has an embedded full document in question (at the end of the post). Where is the problem? That they did not put all pages into a picture format? They cut out the important stuff to demonstrate possible violations.

I wonder if Sermo even going to reply.


Posted by: marina
on September 30, 2007 08:11 PM GMT

Feliciano:

The original document talks about the data, which is fictitious. But it says nothing about the interface, which seems to have lots of private data.


Posted by: marina
on September 30, 2007 08:14 PM GMT

dr.imony.ca:

Many thanks! The post is actually marked as Monday story. We did publish Monday stories late on Sunday before, but you are right, not that early. We thought it was an important story, so we wanted to run it asap.

To answer your other questions, this particular story is indeed about Sermo. All other stories, however, are not about Sermo per se. They are about security vulnerabilities of all these MDs networks. It is our belief that the lax security can sway doctors (via impostors, pharma reps-impersonators, medtech trolls, and others) to use drugs and devices inappropriately. Pharma/medtech companies put much money and effort to bring their stuff to the market, and the temptation is strong to get into these networks. We think that we raised important issues in all these posts.

So, no worry: no Sermo obsession.


Posted by: DrO
on September 30, 2007 08:43 PM GMT

OK, let me get this straight. You are talking about conflict of interest, protecting your "members" privacy and you have this line on YOUR OWN WEB SITE:
"
Our website traffic is in the hundreds of thousands unique visits a month. If you are interested in advertising, we will gladly provide you with the latest readership statistics, as well as our affordable advertising rates."

So what aspect of MY PRIVATE information are you sharing? Are you sharing the email address I am supplying on this comment? Either you have no stats on your readers, so your advertisers should go elsewhere, or you really have information ABOUT ME? I'm sorry, is this the pot calling the kettle black, or what? At least Sermo does NOT accept advertising. If the readers of your blog could see what the 30,000 doctors on Sermo are saying about you and this web site, I think you'd be LUCKY to have "Ads by Google", the ABSOLUTE lowest denominator in Internet advertising. Good Luck and Good Bye!


Posted by: Gene.O
on September 30, 2007 08:47 PM GMT

vulgar comment deleted by administrator


Posted by:
on September 30, 2007 09:01 PM GMT

Gene.O:

The only information that we have about you is your IP address. Anytime you visit any website, your IP is recorded. That's how the internet works. We don't even require you to submit your email when you post a comment. However, to reassure you, we have never shared anyone's email with anybody.

In terms of readership statistics, any time you enter the site, our stat counter records your visit. So we know how many people visit us.

To summarize: your private information stays private on our site.


Posted by: DrO
on September 30, 2007 09:18 PM GMT

WTF, the slide you posted has clearly been doctored. Are you kidding? Does someone co-sign your anesthesia records? This is nothing short of falsifying data. DO NOT submit any IRBs, articles, or grants to me or my institution (check the email address, I think you know where I am from). For all the debate here, YOU HAVE NO JOURNALISTIC INTEGRITY, YOU MODIFIED THE MEDIA TO MEET YOUR NEEDS, YOU NEVER GOT ANY SUPPORTING DATA, YOU RUSHED THE END PRODUCT. YOU MAKE THE NEJM LOOK LIKE A BUNCH OF CHOIR BOYS. WHY WOULD ANYONE BELIEVE ANYTHING YOU POST ON THIS SITE?


Posted by: bul1
on September 30, 2007 09:19 PM GMT

bul1:

We have a link to the original document, and a full original document is embedded in this post. We stand by this story: the interface presented to the FDA has data fields that feature private info.

NO CAPS!


Posted by: DrO
on September 30, 2007 09:28 PM GMT

Are the people who are criticizing Medgadget for modifying the document not capable of clicking the FDA link we provide? Are you under the impression that we hacked the FDA? Go to Google, search for "Sermo" using the FDA as the source, and see what you come up with.

As far as your use of Medgadget, we do not provide any IP addresses, email addresses, or any identifying information to any outside parties. Period.


Posted by: Bruder
on September 30, 2007 09:45 PM GMT

You'd think I was in Salem, Massachusetts with an angry, torch-carrying mob crying "Witch! Witch!" You people need to count to 10 (100?) before posting and embarrassing yourselves. The photo has not been "doctored" at all! If you look at the original, which was provided in good faith and in the interest of full disclosure, you'll see that it's a PDF of a PowerPoint presentation, with a title ("Client View: Dashboard Overview") and a footer ("�Copyright 2006, Sermo Inc. - CONFIDENTIAL", logo, etc.) which contained the text, "Sample Information -Not Actual Data-" as part of that footer. Both the header AND the footer were not included above, and the image in the slide is the same as the one above with the exception of red arrows added for emphasis. This is an otherwise unedited image. This is what's called cropping, and is the graphic equivalent of taking a paragraph and quoting a sentence. They didn't even hide the fact that it's a simple crop...Do you see the pea green speech balloon on the right that's chopped off? Had they included the entire page from the PDF and manually removed the "Sample Information" text, then I'd be crying foul with the rest of you.

Besides, as has been mentioned already, the real point here is that this is the "Client View" (ie, paying customers such as Big Pharma) that is intended to be seen. Having any actual names listed (much less under a header of Big-Brother-like "Watched Physicians") is in blatant violation of their privacy agreement which assures data is shared to clients only in aggregate, non-personally-identifiable fashion.

Believe me, if this "Dashboard view" is accurate, all it'll take is one disgruntled client who feel they didn't get their money's worth with some screenshots to share and it's all over.


Posted by: enrico
on September 30, 2007 10:31 PM GMT

comment by an impostor deleted by admin

impostor has been banned


Posted by: deleted
on October 1, 2007 05:57 AM GMT

curious that so many making insinuations about medgadget here find it imperative to name specific destinations whence they'll take their blogging/ranting business in future.

none of the named options remotely tackle the particular health care turf medgadget has staked out, and aced.

One has become so enamored of a recent confab it put on that a dismaying assortment of self-reflection is about all the new content that's been posted there in 3+ weeks. C'mon, Matt! Put down the mirror and get back to business!

Nah, the other mentions are good in their way, but it's medgadget for developments in health tech.


Posted by: gjudd
on October 1, 2007 06:56 AM GMT

Gruntdoc - on another add a comment page, I quoted a Bruder from this site and at first they went back in and added the full statement rather than just the part I quoted and when objected they added in brackets that they added the rest of the quote so that I could not censor Bruder. I asked and am still waiting as to how I, not part of this site, could possibly censor someone on this site, when I cannot even change what I write after I send it in. When it comes to journalism, perhaps this site in not quite ready for prime time.


Posted by: Michael
on October 1, 2007 07:55 AM GMT

Enrico - Noticed your input.

How are your studies going as a third year medical school in Guadalajara. I am very impressed that you have that much time, with the typical medical school curriculum, to spend on your web site and other site's comments sections about Sermo. Will be interesting, if when you go into practice, if you move to the US to practice and if you might just then understand about credibility in the medical community.


Posted by: Michael
on October 1, 2007 09:46 AM GMT

Michael: you are just incredible non-credible troll. You are not adding anything to this debate. Your only goal is to besmirch our reputation.

You've been banned. Goodbye.

PS You can continue to dwell for the rest of your life on what double boarded means inside Sermo.


Posted by: DrO
on October 1, 2007 10:00 AM GMT

deleted


Posted by:
on October 1, 2007 10:16 AM GMT

Comments by someone pretending to be GruntDoc have been deleted. We verified with GruntDoc, and he was not the poster of the comments.


Posted by: Bruder
on October 1, 2007 11:04 AM GMT

Michael,

You're damn right you should be impressed.

Now that you're banned here, if you bring your trolling comments to my site, you'll be met with the same fate. You've been warned.


Posted by: enrico
on October 1, 2007 02:25 PM GMT

you tell him man Yo is the man


Posted by: enrica
on October 1, 2007 07:04 PM GMT

Very interesting. Last night I posted a reply, and it showed up on the list of replies under enrica's comment. Now, today, the reply is gone.

My reply said:

DrO says:

>Michael: you are just incredible non-credible troll. You are not adding anything to
> this debate. Your only goal is to besmirch our reputation.

>You've been banned. Goodbye.


DrO. I find it interesting that you would ban a person from this site just because you personally disagree with their comments (or don't feel that they are adding anything to the discussion). It makes me wonder whether you are really interested in having a discussion about this, or are just looking for people to pat you on the back. I hate to say it, but in my mind this reinforces the questions about your credibility.


Posted by: UnbrknCh8n
on October 2, 2007 03:47 PM GMT

UnbrknCh8n:

He's been banned for failure to debate. It has nothing to do with my (or our) credibility. Our readers can decide for themselves.


Posted by: DrO
on October 2, 2007 09:40 PM GMT

Sorry to disagree with you Dr O, but in some ways you viewed him as perhaps you are viewed on Sermo and yet you are free to post without being deleted or your words being changed.

If what you have said about him is true it should be obvious to all readers. It is like the charge that an a poster can in any way censor someone on your site is so patently absurd that you have not answered it and it stoppped. The way to deal with trolls is to ignore them and they go away

There were two issues and one was about journalism and trusting what a journalist says and counting on an attention to detail.


Posted by: jaydoc
on October 3, 2007 03:06 AM GMT

Good follow-story at:
http://www.xconomy.com/2007/10/02/sermo-ceo-offers-answers-to-xconomy-readers-questions-about-privacy/


Posted by: Paul
on October 4, 2007 07:41 AM GMT

Nice summary: http://www.mexicomedstudent.com/2007/09/679 this especially diggs into the part where money is made in sermo.com


Posted by: Martin
on October 16, 2007 12:43 AM GMT

add a comment
html tags: <b>, <i>, and <a>
examples: <b>Bold</b> <i>Italic</i>









Remember personal info?
(anonymous comments allowed)



click to make your selection boldclick to make your selection italicclick to add a link


Hello Human!

Enter the above anti-spambot
Turing code:





Click the "Post" button only once!