Bringing Health Data to a Doctor’s Heart: An Interview with Open mHealth
In our brave new world of health data, two things are keeping Watson from becoming our ipso-facto doctor: The human touch and gut instincts. Would it not be a Singular Utopia to arm doctors with enough data to
make them as smart as Watson?
Truth of the matter is, medical schools don’t teach personalized medicine, and it’s tough to get doctors on the bandwagon of data when it is presented in a way that is alien to them. They don’t have the luxury of time to make sense of it all.
As software gurus and designers in the digital health space try to fix all the problems in health care with code and slick design, they often forget what kind of needs a doctor actually requires.
As a result, many companies end up building beautiful products that are far from clinically practical solutions. Open mHealth is on a mission to solve that exact problem.
I had a very enjoyable conversation with David Haddad, Executive Director of Open mHealth, which is an open API platform that allows toolmakers and data consumers to access digital health data with the right clinical context.
It was founded in 2011 as a non-profit (501c3) and is led by an international team of health care veterans, researchers, designers and a rapidly growing developer community.
Omar Shaker: Can you tell us more about the work you guys are doing?
David Haddad: The problem we are solving is the “siloed” approach to digital health data. We developed, in collaboration with both industry and academia, an open API standard that would provide developers the ability to read, write, and authorize against disparate digital health systems all with using a common clinical schema.
OS: How do you create this ubiquitous interoperability?
DH: You need two things: (1) a common logic and (2) a common language. What do we mean by that? We first work with clinical experts and clinical data scientists to understand how they would want to see data given a specific measure.
So for physical activity we might coalesce a cardiologist, an internal medicine doc and a clinical informaticist to think about how to represent physical activity. We then define a JSON Schema and reference existing clinical standards (like LOINC, Snowmed, etc) to give that schema clinical validity.
We then have a data point API that allows a user to read, write and authorize data from disparate sources that use those clinical schemas. Together these two components allow for greater interoperability with the right clinical context around what that data exactly means.
You can see some of our case studies on our website, where we helped clinicians analyze and visualize different datasets from patients with PTSD patients, diabetes and other diseases. The data has to be familiar to the doctor according to the clinical codes he already knows, and our team’s goal is to provide them with exactly that.
OS: Interesting, so how is Open mHealth different from similar platforms in the market?
DH: First off, our API framework is free and open. If you take a look at our offerings you can get started today and contribute back to the community.
Second, we are community driven. We are constantly brokering conversations between the clinical and toolmaker communities to be able to make the access of digital health data easier with the right clinical context around that data. Our schemas have been vetted by select toolmakers and clinical experts to ensure that there is a common language for digital health data.
Third, the Open mHealth API infrastructure is modular. Which means that we if you want to plug-in a new module for processing or visualization you can do so without having to refactor your entire system.
OS: What are some of the main challenges you face?
DH: Well, some clinical measures are easier than others. For example, blood glucose and weight are pretty straightforward to represent. On the other hand, things like nutrition and medication adherence are far more challenging to present in a useful way, because behaviors around those measures vary so much. We are open to bringing in the right clinical experts and clinical data scientists to help us think through this.
OS: How do you see solutions from giant tech companies coming into play in mobile health?
DH: When we first started out, telecos were providing their own plug and play APIs solutions but I’m not sure how much traction they’ve gotten. I’m excited to see how Apple’s Healthkit or Google’s Fit will improve interoperability. One thing’s for sure; we need a neutral party with a community that trusts it to act as a translator between the clinical and technology worlds.
OS: Could you tell us more about your product Linq, which is currently being piloted in Stanford Hospital’s Preventive Cardiology Clinic?
DH: We ate our own dog food. We wanted to demonstrate the power of what you can do with Open mHealth from 3rd party data APIs, data storage, processing and visualization. We want to show how this data can be used to strengthen the partnership between a clinician and patient in achieving a positive health outcome.
OS: So what’s the verdict so far?
DH: Both clinicians and patients we’ve showed it to love how simple and easy it is to use. One of our guiding design principles is that it has to work like magic, without obstacles for the patient and clinicians to waste time on.
With platforms like Open mHealth, we start to realize how health data can be brokered between its source and its user. As these schemas start being weaved into the ecosystem, we might start seeing doctors with computer-like capabilities rather than the other way around.
Each success attained by such platforms is a stride closer to realistic personalized medicine.
More information is available on www.openmhealth.org including their schemas, APIs and access to their developer community.
Omar Shaker completed medical school in Egypt, followed by internships in the US. He soon left primary care for the world of digital health, moving to San Francisco to work on his own projects. These posts represent his reflections on a series of interviews he conducted with some of the more exciting entrepreneurs working in digital health today. Omar can be reached at firstname.lastname@example.org.