Lessons from Google Health

On Friday (5 days ago), Google announced that Google Health is being retired.  What does that mean?  Does that mean that all PHRs are necessarily on their deathbed, that Microsoft will soon follow with the demise of HealthVault, and that 15 (or so) of the nation’s top employers will write off their investment in Dossia?  Or does it just mean that Google’s approach to Personal Health Records was off the mark?

I’ve waited a few days to see what other commentators would say before weighing in.  Our trusted sources, including John Moore at Chilmark Research, have done an excellent job of summarizing most of the issues.  Still, there is a bit to add.

Our perspective.

We formed White Pine Systems in January, 2006 to address the clear market need for technology-based patient engagement.  The term “Personal Health Record” had yet to achieve its still limited recognition and patient portals were almost non-existent.  In 2006, we built prototypes of our system, which we call SPINN, the Secure Personal Information & Notification Network, but quickly learned the market was just wasn’t ready and we were therefore too early.  Although consumers and even providers loved it, we could not find a way to quickly achieve enough traction in any particular community to ensure sustainability—still a problem, although the environment is much more encouraging.   So we put it in the drawer and waited.

By the fall of 2007, Microsoft HealthVault, Google Health and Dossia had entered the market and shown some early traction, each with tremendous promise.  We knew the market was still too early, but we also feared that not entering then would mean missing the opportunity.  We approached all three about a partnership.  Our added value was clear—at least to us:  We would build personal health applications on these PHR platforms that would bridge the gap between EMR systems and the PHR platforms while giving new functions and features to all stakeholders.  And we would provide layers of live customer support—something not offered by either Google Health or Microsoft HealthVault.

Microsoft said “Great.  How can we help?”

Dossia said “Well, if one of our client members asks, we’ll support you.  (One has since asked and we are in the process of working with Dossia now.)

And Google said, effectively, “Whatever.”  A less than embracing response.

At a Robert Wood Johnson conference in the fall of 2008, a panel of executives from all three companies expressed their vision that one day companies like ours would build applications that could inter-operate with all three platforms.  There it was:  Our vision acknowledged by the three giants.  We continued to build our application suite focused on HealthVault but kept it open to add Google Health or Dossia when the time came.

But the Google approach never quite made sense to us.  Google’s almost singular focus on the consumer was not supported by adequate attention to engaging physicians and care coordinators.  We competed against and lost to Google Health on two occasions.  This week we learned that neither of these projects made it off the ground.

Like others, we’ve been expecting the end of Google Health for the better part of a year.  At the same time, there is a lingering feeling that they could come roaring back at some point when the market direction is clear—which we’d embrace.

So, what lessons do we take from the Google Health experiment?

Lesson One:  Personal Health Records aren’t for everyone—and that’s just fine.

In Google’s post announcing the end of Google Health it says:

“Now, with a few years of experience, we’ve observed that Google Health is not having the broad impact that we hoped it would. There has been adoption among certain groups of users like tech-savvy patients and their caregivers, and more recently fitness and wellness enthusiasts. But we haven’t found a way to translate that limited usage into widespread adoption in the daily health routines of millions of people.”

That idea was echoed by other commentators who cited the study from earlier this year that only about 7 percent of Americans have a PHR today.  The Google expectation appears to be that PHR adoption should become as ubiquitous as online banking.  To use a health-related metaphor, that’s like saying that Lipitor must be a complete failure solely because the number of patients who take it are only a tiny fraction of those who take aspirin, ibuprofen or acetaminophen—right evidence; wrong conclusion.

The conclusion should be that patient engagement, particularly PHRs, can give huge value to families dealing with chronic medical conditions but are just not as important to folks who do not have chronic health issues.

Very early in our research, back in 2006 and 2007, we tried to understand who would be most likely to adopt and regularly use a PHR.  We broke the population into four categories:

  1. Healthy.  Individuals who were not focused on specific medical conditions either personally or as a care giver.
  2. Recent onset chronic illness.  Individuals for whom their recent diagnosis, or that of someone close, is the most important thing in the world, and who are searching for tools and solutions to manage this new life state.
  3. Long term chronic illness.  Individuals who have had a chronic condition so long that their behaviors are established, and for whom new solutions have relatively little perceived value.
  4. People who are “disenfranchised.”  Individuals who, for a wide range of reasons not the least of which is a lack of insurance, operate mostly outside of the healthcare system.

Although I’m sure it exists, we couldn’t find statistical evidence that shows levels of interest based on both a person’s own personal health and that of others for whom he or she is a guardian or care giver.  After digesting the information we could find, we guesstimated this pie chart, which we think is still sufficient for this purpose.

We decided our initial product design and marketing would address the needs of people with the recent onset of a chronic illness.  These are the folks we think are most inclined to adopt a new health process, to sustain their use over time, and to benefit from it.  Other profiles would develop as the market evolved.  Noting also that people with chronic medical conditions typically suffer from multiple co-morbidities, see several different physicians and specialists, with lab results and interactions with a range of health providers, and, finally, account for a vastly disproportionate percentage of healthcare cost, it is clear that even modest levels of adoption by the right people will have a huge impact and create a thriving business.

The lesson should be clear:   The goal for PHR adoption should not be measured as a percentage of the entire population, but should be considered as a percentage of those people who directly or indirectly are affected by chronic medical conditions.  In that framework, an adoption rate by as few as 15 or 20 percent of the total population could still be a resounding success by all other measures.

Lesson Two:  When it comes to patient engagement, one size does not fit all.

In an age when market segmentation means that Coke comes in 18 different flavors, it didn’t make sense for Google to offer a its generic interface even when that UI gave users considerable range and flexibility in how to configure and use it.  That would be like offering a Coke flavor kit and allowing consumers to mix and match to create their own 18 (or more) flavors. (I know I’ll be challenged on this, so let me be clear: we like the range and flexibility of Google Health, we just don’t think it was easy for end users to derive value for their specific health interests.)

We believe that pre-defined dashboards designed around particular health interests that are closely connected to care coordinators through EMR and other health information systems are key to achieving regular, sustained use.  For example, a patient with diabetes will want to have the shortest possible path to information related to living with diabetes, while a person whose loved one is struggling with cancer will want a very different experience, as will someone who is pregnant, etc.  And they don’t want to have to figure it out on their own.  Configurability around specific health interests is key to successful adoption and sustained use.  Equally, close linkage to care coordinators reinforce behaviors and help identify problems at the earliest moment.  We call these personalized dashboards “Care Units.”

Studies of PHRs and Patient Portal usage and outcomes are still quite limited.  As described in the Program Announcement for a present AHRQ grant program titled “Understanding User Needs and Context to Inform Consumer Health Information Technology (IT) Design”, “there is still a lack of basic research around these end users’ PHIM [Personal Health Information Management] practices and needs and how these methods are influenced by a multitude of other contextual factors.” Still, success with similar telemedicine programs and Patient Portal/PHR applications at The Cleveland Clinic, the VA and a few others suggest that this kind of personalized experience is successful.

The lesson is simple:  Unlike an intuitive single purpose application like on-line banking, personal health applications need to be pre-configured for the specific health interests of divergent groups of users and closely linked to care coordinators.

Lesson Three:  Patient engagement must give value to all stakeholders.

As noted above, Google’s clear, almost exclusionary focus was on the consumer.  By contrast, Dossia focuses on employers and insurers as vested stakeholders along with the consumer, and Microsoft pursues a broad range of stakeholders with a heavy focus on provider engagement.  We believe that patient engagement must have the consumer at the center and in control, but must give significant value to all stakeholders.

So who are the stakeholders?

  • Patients
  • Families (and un-official caregivers)
  • Providers (including physicians, hospitals, labs, pharmacies, home healthcare providers and nursing homes, to name the most obvious)
  • Employers/insurers/third-party administrators (including Medicaid and Medicare)
  • Patient advocacy groups

Some commentators reduce this to a battle between tethered patient portals and un-tethered PHR systems, concluding that the battle will be won by large integrated providers that offer an integrated patient portal, with the provider at the center of ownership, control and design.  Certainly patient portals will be the preferred choice for many users, particularly consumers not dealing with chronic medical conditions.  We expect that will be the case in some markets but it will not prevail in all markets.  Particularly in markets where there are multiple providers with no integrated patient portal, the PHR becomes the platform for a multi-provider patient portal, providing tremendous value to all stakeholders.  A likely example of this will be the proposed Accountable Care Organization rule, where wide-ranging providers, including physicians, hospitals, labs, home healthcare services and nursing homes are expected to collaborate closely but cannot be expected to operate off a single EMR system.  Another setting is in communities supported by Health Information Exchanges.  Relying on tethered patient portals at the provider level effectively precludes the patient/consumer from the benefit of health information exchange.  In these settings, a multi-patient provider portal at the HIE level that incorporates a PHR platform is an ideal solution.

At the same time, we see a host of new personal health applications coming to market including health and fitness, games, and personal disease management tools.  In the larger scheme, we don’t see providers or EMR vendors as the moderator of all these applications for all of these stakeholders.  We see the consumer as ultimately dictating the combination of functions and features, while recognizing that all stakeholders will expect value.

The lesson is simple:  While the patient must be at the center and in control, other stakeholders must be embraced.


Google Health’s exit from the stage is a reflection of its own strategy and expectations.  It is not representative of the market for PHR systems or PHR platforms as a whole.  With the huge changes in healthcare delivery and payment models (not just the Affordable Care Act but also changes being driven by employers and insurers) now is the time to promote care coordination and patient engagement, not take a step away from it.

The inherent value of the remaining PHR platforms, Dossia and Microsoft HealthVault (along with a few other smaller players and possible new entrants) and patient engagement solutions like SPINN is the ability to create a community of services and participants that can be engaged by the patient/consumer that simply cannot be offered through a provider-controlled, tethered patient portal.  Ultimately there will be a convergence between tethered patient portals and PHR solutions that empowers consumers while giving value to all stakeholders.  PHRs will certainly evolve through this process, but, to borrow from Mark Twain, the reports of their death are premature.

Posted in Uncategorized

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s