Digital Health Commons

Are larger health institutions and hospitals capable of supporting a digital health commons?

Last Thursday, I attended a Stanford Medicine X meetup with Teresa "On Peer-to-Peer Healthcare and Human Centered Design." The speaker, Sean Ahrens, a down-to-earth guy that spoke to the audience on a beanbag chair, beer in hand, spoke on how public good can be created with the advent of an online community that allows truly shared health data (a digital health commons).

Sean talked about his website, Crohnology, which serves patients diagnosed with Crohn's disease, a chronic condition with no cure. Inflammation can occur anywhere within the digestive system, taking a huge toll on a patient's quality of life. Sean developed a website in many ways reminiscent of sites like Livestrong and Weight Watchers because it builds a community where people can talk about their health progress, share information, and motivate each other. However, the site also serves a broader public good: it accumulates health data that contributes to research for, at the very least, quality of life management. If you want to think with a little more grandeur, it may one day help researchers find a cure for Crohn's. I was most impressed with the intuitiveness of patient journals - they report on how they are feeling in their journal on a scale of 0 to 100. This perceived health would then be charted against the diet, medications, supplements, and lifestyle changes they report to manage their disease. Patients can look at their own data over time and see how that compares with other people who have Crohn's.

Ultimately, the goal is to build a "Patient-Powered Research Network." Sean shared his struggles with creating Crohnology - he journeyed from trying to make a business out of his patient-powered research network to realizing that monetizing the site didn't align with his personal values. But if there is no return on investment, funding and support for the site is extremely difficult to find. Currently, Crohnology is kept alive with the help of the Patient-Centered Outcomes Research Institute (PCORI).

Key issues for this type of platform include:
1. Funding: How do you keep true to your philanthropic goals without selling yourself out? The website needs to be supported monetarily.
2. Open Access: How do you maintain open access and protect privacy?
3. Integration: Can data be shared between other information sources? For example, health data from Fitbit?

I thought that the challenges presented here were particularly interesting because the idea of "patient powered" research seems erroneous. Scientists would be skeptical in the validity of this research because of the sheer amount of subjectivity and myriad confounding variables (McKenna 2011). However, more and more, patient-reported outcomes (PROs) are becoming a valid research model - because a patient's happiness, outlook, and support group correlates with survival (Kyte 2013). I would argue that the more data they can obtain over a specific population, the more we could try to deduce and intrepret. Traditionally, we look at PROs with questionnaires that calculate an outcome measure on a scale or a score so it is less subjective and mores scientific. 

The key to creating a digital health commons, however, is giving patients the tools and the guidelines for them to look at their health objectively. In Crohnology's case, the system--a perceived health rating scale alongside other tools for recording objective data and the potential to add in data from other sources like wearable devices--is simple enough for patients to understand and comprehensive enough for self-awareness development.

In thinking about some of the problems Sean was dealing with, I thought about the potential of the VisionTree platform my team is working on implementing with Dr. Damato and Dr. Dunn. The cloud-based platform stores clinical information from existing EHR and creates a portal for patients to log in and answer questionnaires--thus creating a central space for data analytics to take place between PROs and clinical outcomes. Our main goal, ergo, is to enable research in PROs, based on validated psychometric and psychosocial measures. But a platform for patients to simply log in and fill in questionnaires seems too limited in scope.

What if we built in a forum or an analogous health data commons within VisionTree? UCSF patients could have a single portal where they learn about their disease and look at their medical records and images, fill in questionnaires for the hospital's own research, and provide a personal account of their own outcomes, thoughts, and feelings. Provided a patient was willing to volunteer their anonymous personal data via informed consent, they could freely participate in the data commons and learn about how other anonymous patients are dealing with their issues. Patients could then build a community within VisionTree that doctors could also see - enabling unprecedented feedback and understanding of their patients.

However, I foresee plenty of things wrong with this idea.

1. Institutional protection of data and lack of open access to other people outside of the community may conflict with the values of a truly "shared" community. This could create an "Uber" of the health industry.
2. Privacy concerns are always present. Although a patient signs informed consent, it does not mean they will always know how to manage their anonymity on the internet. This is an inherent "internet of things" problem.
3. A radically different relationship between doctors and patients could develop as patients will naturally discuss their misgivings about the care they receive from the healthcare institutions. This could have a negative impact on the economic incentive of the institution enabling a Data Commons within the platform. Antagonism may be inherent in this kind of community. On the other hand, this would encourage hospitals and providers to become more patient-centric.
4. A lot of developer work and maintenance is required. A large commission would be needed to develop a platform and maintain it. As mentioned, funding is very difficult to obtain because, in general, the return on investment for obtaining patient-reported outcomes data is uncertain. The trade-off, of course, is the sheer amount of subjective data we could obtain in conjunction with the objective data collected in the clinics, including medical images, tumor sizes, genetic factors, visual acuity, etc. But thinking of the kinds of questions we could answer from a research funding standpoint is a whole new ballpark.

So is this a feasible idea? Is it better to leave digital health commons to more separate entities, subject to altruistic funding? Or would institutions be willing to take on the potential challenges associated with hosting a platform?

I don't really know much about the landscape from a clinical or logistics standpoint - I am mostly, out of curiosity, laying out my thoughts and wanted to bounce it off of someone.

References:
1. Stanford Medicine X <https://medicinex.stanford.edu>
2. Crohnology <https://crohnology.com/>
3. Patient-Centered Outcomes Research Institute <http://www.pcori.org/about-us>
4. Measuring patient-reported outcomes: moving beyond misplaced common sense to hard science. Opinion. McKenna, Stephen. doi:10.1186/1741-7015-9-86 <http://www.biomedcentral.com/1741-7015/9/86>
5. Patient-reported outcomes helped predict survival in multiple myeloma using partial least squares analysis. Kyte, Derek. doi:10.0001/jama.2013.277222 <http://jama.jamanetwork.com/article.aspx?articleid=1741830>
6. VisionTree Software, Inc. <http://www.visiontree.com/>