Tuesday, July 05, 2011

Is there a Piazza in medicine's future?

One of the holy grails of clinical information systems is to build decision support programs that would provide doctors with real time advice during the care of patients. The industry is moving along in this realm. The idea is to mine the clinical data in a hospital's records to look for patterns that would make the practice of medicine more evidence-based.

That is all to the good and will likely bear fruit. But let's think differently and explore the potential value of social networking in helping out clinicians.

I'm drawn to this idea by an article in yesterday's New York Times, entitled "Homework help site has a social networking twist." The story is about a start-up company called Piazza, created by a young woman who found herself alone while other students collaborated on engineering problem sets.

When Pooja Nath was an undergraduate at the Indian Institute of Technology Kanpur, an elite engineering school in India, she felt isolated. She was one of the few women on campus. While her male classmates collaborated on problem sets, Ms. Nath toiled in the computer lab alone.

The experience as a young woman in that culture formed the foundation of her start-up....

Students post questions to their course page, which peers and educators can then respond to. Instructors moderate the discussion, endorse the best responses and track the popularity of questions in real time. Responses are also color-coded, so students can easily identify the instructor’s comments.

Although there are rival services, like Blackboard, an education software company, Piazza’s platform is specifically designed to speed response times. The site is supported by a system of notification alerts, and the average question on Piazza will receive an answer in 14 minutes.

OK, I know clinical decision-making is different from getting the right answer on a homework assignment. Wait, is it all that different?

Students post questions to their course page, which peers and educators can then respond to. Instructors moderate the discussion, endorse the best responses and track the popularity of questions in real time. Responses are also color-coded, so students can easily identify the instructor’s comments.

What if a clinician had access to a cadre of like-minded clinicians and posed questions and got answers in real time? We can imagine this at different levels. Think of medical students on rotation, performing patient histories. They are not allowed to actually treat the patients, but what if they wanted to test their diagnostic and prescriptive skills, by quickly tossing questions to the medical world version of Piazza and seeing how well they have analyzed the situation? Participants could include other medical students and the course instructors.

Ditto for residents, except they actually deliver care as well as analyzing patients. And, of course, ditto for attending physicians as well.

In short, decision support systems do not need to be based on mining computerized databases. They could, in addition, be based on mining the experiential database of thousands of clinicians.


Anonymous said...

Not my cup of tea, but it exists already. visit SERMO.


Andrew said...

Ditto cup of tea.

The closest thing I've seen so far to what you're describing is a video of Doximity (same guy that made Epocrates). It lets you connect with any physician or group of docs based on their research expertise, specialty, location.

I know a doc that uses Sermo to chat about random stuff, but it's all anonymous, questionable what you're getting back. It's a different need, I think both have their place in the world.

Anonymous said...

I don't find Sermo that useful. For one thing they have some weird system that pays you a bonus for either posting a question or answering it, or something, which turns me off - in my specialty, pathology, the site is dominated by one doc who clearly sees this as a supplemental source of income!
For another, as Andrew mentions, you are certainly not sure what you're getting back.
Third, they make $$ by allowing pharma and other companies to 'spy' on what the docs are saying for marketing purposes. This smells bad.
And last, it is also used as a miscellaneous bitching site about how doctors are persecuted by various bad guys.

I think what Paul has in mind is a bit more highbrow...


Dr.Theresa Willett said...

A similar social-networking and data-gathering site aimed at patient questions is HealthTap. Once they get comfortably established in that domain, I wouldn't be surprised if they turned their energies toward the providers.

Jim said...

Agree with the idea of Piazza for medical decision making.

However, we will need to navigate the real and perceived exposure to malpractice claims.

This is analogous to a “curbside consult” where the consultant doesn’t get the full picture – just the tidbits provided by the requesting MD. If this was the best way to provide an opinion, we would use it for all consults.
I assume you are familiar with “Wisdom of Crowds” (http://www.amazon.com/Wisdom-Crowds-James-Surowiecki/dp/0385721706/ref=sr_1_1?ie=UTF8&qid=1309956869&sr=8-10) by Surowiecki.

Paul Levy said...

I don't see the malpractice issue, as this is just sharing information, with the ultimate responsibility (as always) remaining with the attending physician. So maybe it is just a perception problem.

Jim said...


But if you don’t think an industrious lawyer would bring me into a suit if I offer a medical opinion (for which there is a clear electronic paper trail) for a patient who I have never interviewed or examined….
It’s not the same as publishing an article if it relates to a single patient. You are offering a medical opinion related to a single case.

I do, however, think that perception is more of a problem than reality. We will all be sued. Being sued and even the outcome of the case does not change how you acted. Best bet is to do what you believe to be the right thing every time. This includes apologizing, acknowledging error, etc. However, not all of us buy into this.

Anonymous said...

Paul: I am afraid that it is a threat. Twitters, tweets, blogs, facebook pages etc. can all be easily searched and plaintiff lawyers have an incentive to bring as many providers into the picture as they reasonably can.

Fortunately we have the Tribunal in Massachusetts and many providers (up to 30%) are dropped out at this point. If a provider only rendered a passing opinion, it is also difficult to prove negligence and causation.

I would never want to discourage informal interaction among providers but advice in writing, in any form or forum, should be qualified as based on “limited” information or coupled with a recommendation to consult with someone who does see the patient and has access to testing results. There is no absolute protection but the paper trail rules in these cases.

Anonymous said...

The above comments are tribute to the extent to which malpractice risks are perceived by our profession, accurately or not, as affecting everything we do. Many new and innovative ideas are viewed through this lens and then rejected.
I am not passing judgment on this phenomenon as perception often becomes reality.
But to those who claim malpractice concerns are not a significant contributing factor to medical decision-making (and therefore not worthy of efforts at malpractice reform) - here is more evidence that you are wrong.

nonlocal MD

Bruce said...

From Facebook:

I believe that compassionate, team-based patient care and care communities built around problems from the patient's perspective are critical to a healthy future for our health care system. This concept fits right in with that.