Is PatientsLikeMe an effective platform

Medicine and New Media

Some time ago there was an interesting takeover in the Health 2.0 / Quantified Self environment: the online genetic test service bought the health community platform

What are the two startups doing and why do they fit together ?!

Mirror ball and tarot cards were yesterday. Health Prediction promises 23andMe for $ 299 through DNA analysis.

Actually does 23andMe nothing new. DNA analysis. In every major city there are laboratories that can do DNA analysis. It is not unlikely that they would even use the same OmniExpress Plus Genotyping BeadChip from the Illumina from the USA that are also used at 23andMe. Similar products are also manufactured in Germany, for example by QIAGEN from Düsseldorf.
There is a genetics consultation hour in every university clinic where you can get advice on your DNA test results.

Of course nobody does. Nobody just googles the next genetic consultation hour, then goes over there and says: “Was just curious about my DNA. Can you check it out? ”. You generally only end up there by referral if, for example, a doctor has suggested a genetic disease. If you show up there on your own without a referral slip, you can also pay for the laboratory and consultation hours out of your own pocket.

This is where the 23andMe business model comes in: the whole process is greatly simplified and comes from a single source. There is no need to go to the clinic and you don't have to justify yourself to anyone. "Direct-to-customer genetic tests" is the technical term. A trendy, contemporary design in Web 2.0 style creates a product that is sexy and suitable for the masses:

  • Register online and order a test kit with test strips
  • spat once on the strip
  • send back to 23andMe and
  • you get your results via email. Finished.

With high test numbers, 23andMe can lower the purchase prices for test strips and laboratory services and thus offer a price that is below the inhibition threshold for many curious people. The acquisition of CureTogether suggests that substantial profits will also be made. Rather, they want to go on a “shopping spree”: the official press release mentions the deal as 1. Acquisition spoken.

That itself as Personal Genetics Conpany descriptive startup 23andMe comes from sunny California, who wouldn't have thought it. It is no coincidence that the Bay Area is also the nucleus of the Quantified Self movement.

This loose association is made up of people who permanently collect, document and (with sometimes more, sometimes less success) different body data (e.g. pulse, temperature, weight) or activities (exercise, coffee consumption or what ever). The main goal is self-optimization.

The last decades have been shaped by practices of self-exploration of the mind and the own soul (yoga, New Age, meditation etc.) followers of quantified self, so-called “self trackers”, interestingly, the truth and self-knowledge are again more in the physical. Or just what data can be read out from the human machine using interfaces / sensors. When we were teenagers we did our hairdos, later our processors were turned ontuned, the iPhone was yesterdayjailbreaked. Knowing how, you can always get more out of it. Now it's your own body's turn. Body hacking is the buzzword and the goal. Thanks to 23andMe, you no longer have to stay on the (body) surface, the front end, so to speak. No, we can look into the code: the DNA.

The Self Tackers usually make no secret of their measurement results. This data is often shared with others, either at one of the countless meetings (more about Quantified Self Meetups in German-speaking countries here) or online on a now unmanageable flood of portals, networks and communities. One of the largest of these is CureTogether. From Mountain View, right, even in that sunny California.

What CureTogether does not reveal itself at first glance. The slogan offers an explanation "Open Source Health Research". You can register there and provide your personal health experience using various checklists. For example, there are checklists for more than 500 special diseases or symptoms. It is also asked how helpful the corresponding medication or forms of therapy were. The data of all users are then accumulated and presented visually. Considerable number of cases are achieved for common diseases, as the example shown here for acid reflux (heartburn) shows (number of cases n> 2,800).

Click on the picture to read the article on the CureTogether blog.

Ultimately, this “crowdsourcing” should give the user suggestions for optimizing their own therapy. By means of swarm intelligence (or alternatively also swarm lay knowledge) one wants to get well together, so "cure together". What helped Lieschen Müller will also help me.

Personally, I'm a bit at war with CureTogether. On the one hand, the website, which is bursting with self-love and empty words (selection of quotations: "We work hard to help you ...", "We are here to be good", Feel better faster "), in my opinion only moderately relevant data, which is also incorrectly evaluated and processed. Inadequate knowledge or ignorance of statistical procedures repeatedly results in incorrect or at least consumed images. (I explained this in my lecture “Electric body music - When the blood pressure cuff becomes pop culture” at re: publica2012).

Lecture at re: publica2012 as part of the re: health track.
The section on CureTogether starts at 13Min: 21Sec and ends at 23Min: 44Sec.

It is difficult to answer whether the idealism of the makers of CureTogther is real or played. In any case, the fact is that you can also earn money with “patient-generated content”. The veteran of all Health 2.0 applications, PatientsLikeMe, makes no secret of the fact that the data collected by users on their platform is specifically evaluated and, for example, sold to the pharmaceutical industry. However, PatientsLikeMe has significantly more users with higher quality content at the same time. The users sometimes track their individual disease progression down to the smallest detail. Basically a digital version of the good old curve sheet known from every hospital, where all data and every action by nurses and doctors is permanently entered. With CureTogether, on the other hand, only checklists (questionnaires) are queried according to a rigid scheme, whereby even the questions in the checklists are often inadequate. At PatientsLikeMe we have a point in time for each event and can thus observe processes, trends or interactions. The event can also be defined much more finely, whereas with CureToghether you can use the selection options Yes or Noand a 1 to 5 star rating comes from this and, for example, leaves out questions about "when", "how often" or "in which dose". It is not difficult to guess where the pharmaceutical manufacturer previously preferred to tap and evaluate data.

Yes or no - and a “zero to five star” rating.
Screenshot of an online questionnaire on

So why didn't 23andMe grab the cream cake PatientsLikeMe right away ?! Isn't there enough money for that ?! Or is that a no-go because it's Boston and thus East Coast ?!

Or does the data structure of CureTogether ultimately fit better with 23andMe ?! And is the overlap between the users of 23andMe and CureTogether greater than that of PatientsLikeMe?

The fact is that there are already efforts on the Internet to collect the results of 23andMe (or the competitor deCODEme) on web portals and then to evaluate them scientifically. I would like to mention openSNP as an interesting example.

Screenshot of the web portal

The users start to play with the data, researchers start to analyze them. The new trend in personalized medicine is largely based on genetic observation. This is where 23andMe wants to participate: they not only want to supply the raw data, but also want to participate in the refinement process. To do this, they need a community platform and without further ado they bought the already established CureTogether “from next door”. If you get the users of 23andMe to fill out the “questionnaires” on CureTogether, a very interesting data overlap could arise.

"The acquisition will improve our own ability to gather data for research and give customers more tools to explore and participate in online communities", you can read about it on the 23andMe blog. In the press release mentioned above, reasons are given as follows: "CureTogether brings to 23andMe additional tools and systems for gathering data from health-based communities that are complementary to the existing 23andMe platforms, allowing customers to share quantitative information on more than 500 medical conditions, talk about sensitive symptoms and compare which treatments work best for them as they track their health. "And the “incumbents” don't make an unhappy impression either, as the founders of CureTogether were immediately taken over into the 23andME management team. Alexandra Carmichel said in an interview immediately after the takeover as follows:“We are joining the team as Senior Product Managers. Our backgrounds as genetics researchers, serial entrepreneurs, developers, and community builders should bring a variety of options to the table for how we can help out. "

It is obviously a win-win situation and could have potential. At least if you believe that the data mixed together in this way is really useful for science (and thus its commercial value increases).
The PatientsLikeMe, which has now fallen behind, is eagerly trying to row in the same direction: their blog is increasingly about genome and personalized medicine ...

Posted in General, Health 2.0, Personalized Medicine, Quantified Self, re: publica, Start up | Tagged23andme, curetogether, deCODEme, DNA, Genotyping, Genotypisierung, openSNP, PatientsLikeMe, Personalized medicine, Quantified Self Labs | Write an answer