Thursday, February 21, 2008

Google Health starts pilot test at Cleveland Clinic - and my reflections on Personal Health Records 2.0 (PHR 2.0)

Please cite as: Eysenbach, Gunther. Google Health starts pilot test at Cleveland Clinic - and my reflections on Personal Health Records 2.0 (PHR 2.0). Gunther Eysenbach Random Rants Blog. 2008-02-21. URL: Accessed: 2008-03-08. (Archived by WebCite® at

Through the wires I hear that Google Health is finally kicking off:

Google Inc. will begin storing the medical records of a few thousand people as it tests a long-awaited health service that's likely to raise more concerns about the volume of sensitive information entrusted to the Internet search leader.

Google will test a new service that will allow the Internet search leader to store patients' medical records.

The pilot project to be announced Thursday will involve 1,500 to 10,000 patients at the Cleveland Clinic who volunteered to an electronic transfer of their personal health records so they can be retrieved through Google's new service, which won't be open to the general public.

Each health profile, including information about prescriptions, allergies and medical histories, will be protected by a password that's also required to use other Google services such as e-mail and personalized search tools.

(Source: CNN, Archived at

The announcement is timely for me as I am just preparing my keynote for the Australian 10th Annual Health Care Congress (WebCite) in Sydney next week, and this keynote will focus a fair bit on Personal Health Records and Personal Health Applications. This is partly because the new Australian government under its new prime minister Kevin Rudd has set a couple of priorities for reforming health care, among them is "focussing on preventative health care and health promotion to help keep Australians healthy and out of hospital", which is a goal that can - in my opinion - be attained or at least greatly supported with Personal Health Records, or more specifically with what I call second generation PHRs or PHR 2.0.

I created a series of slides to illustrate what I mean with PHR 2.0.

The first slide illustrates the current state-of-the-art of PHRs: "tethered" PHRs and "stand-alone" PHRs. Tethered PHRs could also be called Patient-Accessible Electronic Health Records (PAEHR), and they are basically a patient-view of an Electronic Medical Record which is tethered to a certain health care organization, for example a hospital making a part of its EMR accessible through the web. Some may also offer the possibility for patients to not only view but also to edit/annotate certain information.

Stand-alone PHRs are just that - stand-alone personal health applications, for example a software program installed on a local computer, a web-server, or on a USB flashdrive, for users to enter important health information, but with no connection to provider EMRs.

The second slide illustrates the "ideal" solution (at least this seems to be the current mainstream thinking) - the "interconnected" EMR/PHR, where people can keep and manage their own personal health information, which is connected to multiple EMRs (clinical information systems, e.g. in hospitals), which may feed part of their information into the PHR. As Tang et al. mention in their seminal JAMIA article on PHRs, interconnected systems are highly complex, especially due to issues of vocabularies and standards. The approach is very similar to what consumers do when they do online banking - account information is downloaded from banks into personal financial software packages and vice versa, but of course health care information is much more complex than financial information. The beauty of this approach is that it provides a solution to the fragmented health system, where patient information may be scattered across many different providers. Who else than the patient is better suited to be an aggregator of this information?

Now my idea on PHR 2.0: The previous approaches all assume that personal health information is a highly private matter, and for the most part that is true. All the literature and definitions of PHR stress the "private" nature of this information. However, what I feel should be stressed more in the current definitions and discussions is the fact that people freely share and want to share certain health information with others. For example, they share their story in discussion groups, if they ask for help or information, they proudly share good news such as a pregnancy or success in quitting smoking with others, and share successes and failures e.g. to stick to a diet at the hairdresser and elsewhere. Sharing information is a prerequisite for receiving information and emotional support from others, a prerequisite to find peers and "health buddies".

In the PHR 2.0 approach, illustrated above, certain sections in the PHR can be opened up by the patient for others to see. This is obviously a quite disruptive proposal, and certainly not risk free, giving a huge responsibility to consumers to decide on what can be shared, but ultimately this is what web 2.0 and patient empowerment is all about: Trust your users. I am convinced that PHR 2.0 is how we should think about personal health applications.
The other component of the PHR 2.0 approach would be an approach to model the relationships between all healthcare actors - including health care professionals and peers - and to develop a sophisticated infrastructure that enables patients to define exactly what kind of actor has what level of access to what kinds of data.
PHR 2.0 is all about building a system that not only contains personal health information, but one that builds a community and a social network around that information.

At my research group at the Centre for Global eHealth Innovation in Toronto we are currently developing an application dubbed Healthbook(TM) which is based on this PHR 2.0(TM) approach (more on that will be presented at the Medicine 2.0 Conference).

To what degree Google Health or Healthvault will venture into what I call the PHR 2.0 approach remains to be seen. It appears to me that implementing such a system in the US - where people have to be much more paranoid - ahem, I mean concerned - about their private health insurers or employers misusing this kind of data - poses many more privacy problems than in most other countries with universal health insurance such as Canada.

Note: PHR 2.0, Healthbook, and Medicine 2.0 are trademarks registered by Gunther Eysenbach


Dr. Bonis said...

I am a medical doctor. I see patients every day. Including the Emergency Room where I work.

During thousands years physician have follow this hippocrates oath sencente: What I may see or hear in the course of the treatment or even outside of the treatment in regard to the life of men, which on no account one must spread abroad, I will keep to myself, holding such things shameful to be spoken about.

So at the moment I designed the keyose ( service, I have a very clear idea: privacy must be the priority number one!

Read more here.

phillip said...

I agree "PHR 2.0" is certainly the way to go, but I think we may be building the Empire State Building without a foundation.

PHR 2.0 hinges on 2 factors

1. Users that want to be Social

2. Hospitals and Physicians wanting to freeup information

Neither of these factors are completely in place. Let's face it, consumers don't want to manually enter health information into a PHR. First off most patients don't have the slightest idea about their medical history in the first place. There is a need for a mechanism releases this info for them.

I believe patient portals maybe the key. Physician feel it frees resources and saves money. This gives them incentive to begin to releasing info to the web.

I also believe once patients have this information, it will be easy to get them to search and share it.

I propose creating a large open patient portal business. Once this is done you can then entice users to become social. IMHO this is the only way for PHR 2.0 to be 100% successful.