I still remember the first time I logged onto Google Health. I dutifully inputted all my minor ailments and lab values and hit “save.” And then nothing happened. There was no medical advice, no additional reading, no pointers on what to do with all of the information. The service wasn’t anything more than a repository. I played with it for a day or two longer and never logged in again.
Still, it made more sense for me to tell Google about my health problems than to stick them elsewhere on the web. Forums — often chock full of wildly inconsistent information — weren’t for me. And I had little desire to bombard my Facebook or Twitter community with my medical record for personal reasons: my personal and professional relationships on those network don’t revolve around my medical history. (These issues were well articulated by a research team from the University of Michigan last month.) Given the inadequacy of existing tools, there was a need for a resource that was as serious about data as Google Health, but also one that was serious about community.
That need is now being filled. Two announcements in the last week have underscored the big way that online patient networks are coming of age.
Yesterday, researchers from Children’s Hospital Boston published a study in PLoS One that showed how information voluntarily provided by the TuDiabetes community could be used by researchers. The study’s goal was to track a measure of diabetes — hemoglobin A1c — and the researchers found that the value reported by users matched the level for people with diabetes that the government collects. But unlike the government data, researchers can interact with the TuDiabetes population, allowing a level of feedback that can’t be achieved through standard population-based research.
On Monday, the research team at PatientsLikeMe.com — perhaps the most robust of the online patient networks — published an article in Nature Biotechnology that found that lithium did not slow the progression of Lou Gehgrig’s disease. The study drew on the submitted data of 447 patients on the site and refuted earlier research that suggested a benefit. What’s more, PatientsLikeMe has now opened itself up to patients with any medical condition (previously, it had been restricted to a select set of diseases, including multiple sclerosis, mood disorders and Lou Gehrig’s disease), expanding the patient pool and creating the potential for more research collaborations down the road.
To be sure, there are limitations to using non-randomized, non-blinded, self-selected patients and self-reported data, and no one should suggest that data gathered from these communities can or should replace other studies. But they provide an interesting augmentation to more traditional tools.
The problem with Google Health was that I never got anything back for my use of the tool. There was no benefit to me — or anyone else — from keeping my data locked up. And there wasn’t any benefit to sharing on Facebook, either: my friends are very dear to me, but I don’t trust them with decisions about my health. TuDiabetes and PatientsLikeMe make sure that there is a benefit to participating: in addition to the support of a community, the data that’s gathered can be used to answer medical questions that are important to both users and society at large. It appeals to our best instincts.
That’s a sustainable model for the future, and kudos to those two communities for bringing us closer to that future.
[UPDATE: Some other smart takes on this week’s news:
- Are We the Source of Knowledge?, Pioneering Ideas (RWJF)
- Are PatientsLikeMe and Facebook Both ‘Social Media’?, DanielHooker.com]