By now everyone has seen the announcement last Friday that Google Health is being formally retired. Thanks to the several years I worked on Google Health, my phone was ringing off the hook Friday afternoon and emails were pouring in all weekend long.
Let me first start by saying that I am not going to comment on any specific company details. I think the broader question to ask anybody that has worked in health IT and consumer tech for the past 15 years is what have you learned from this experience. Or how about, is there a market for PHRs in the future? Given that I started looking at PHRs back in 2005 while I was working for David Brailer at the Office of the National Coordinator (ONC), here is what I would say I have learned:
1. Healthcare is paternalistic – consumers are blind to costs and data.
Let’s face it. Our current healthcare system is set up to be extremely paternalistic. Health plans, hospitals, and physician practices steward patient data on the patient’s behalf. Patients don’t know the costs of a simple outpatient procedure or inpatient stay. Because health care is not a true market-based commodity in this country, patients end up being lousy healthcare consumers. Unlike the banking, airline, and retail industries, this makes it much harder to convince a broad array of consumers to engage in a service that helps them organize, manage, and share their medical records online. The value proposition becomes even harder when consumers are not rewarded with industry aligned incentives for taking the time to manage their personal health data (e.g., discount on health insurance premiums, lower co-pays at doctor office visits, or something beyond a measly $50.00 benefit credit for signing up for a PHR at their place of employment).
I liken PHRs to credit check services. You go to Equifax or Experian when you want to generate a credit report because you are making a large purchase. But, in general, you almost never pay attention to your credit score until you need to know it or have an issue. Health care is very similar, you only care about your medical data when you get sick, show up in the ER, or start taking care of someone else who is sick. Otherwise, out of sight, means out of mind.
2. PHRs are boring, too generic, and are not social enough.
In their current form, PHRs strive to do too much and as a result do not have enough functionality to appeal to a broad array of users. PHRs are generally designed to store structured medical data that comes from the established medical industry. But my mom does not understand what CCR or CCD means. PHRs are boring to the outside world and too insular within the established medical community. They generally do not take into account the broader definition of health such as: quality of sleep, food intake, stress, emotional wellness, frequency of exercise, intimacy and support, caretaking tasks, etc. If a PHR successfully imports data in from a medical provider, the data generally does not make sense to the consumer who is using the service, or may not even be current (e.g. prescription history from CVS over time). Even if the data is normalized or presented with sophisticated visualization tools, the data is still somewhat hard to break down for a layperson (e.g., what exactly is a “bad lab test result”).
PHRs that focus on one or two tasks or certain discrete types of data will find a market to serve in the future. An example of this type of specialized service is a “Pregnancy PHR” which would focus exclusively on storing and sharing fetal ultrasounds throughout the duration of a pregnancy. Imagine the social media you could build into this PHR. Friends and family members would love to be able share ultrasound images in a networked group and comment on who the baby most looks like. You could even post bets on baby names during the entire prenatal experience. Growth charts and immunization records for kids is also another area where PHRs could be really successful (check out the new startup Motherknows.com). Even OB/GYN care and fertility treatments have a ready-made market for specialized PHRs given the amount of women who are highly motivated to store and share their reproductive medical information. The market for specialized PHRs will have even a better chance for success due the explosion of “health apps.” Mobile platforms will make it much easier for niche PHRs to get distributed and adopted.
3. People Want Convenience Features Not Medical Archiving
At the end of the day, people want their lives to be made easier and to save time on everyday healthcare tasks. They are not as interested in archiving medical data at large. In my 15 years looking at PHR survey data and leading and hosting various consumer and provider focus groups, I kept hearing four main themes:
1. People want to securely email their doctor to ask a quick question or discuss a lab result;
2. People want to ask for and get Rx refills online;
3. People want to book medical appointments online; and
4. People want to get information that is personalized to their health needs.
While some PHRs have a few of these convenience features (e.g., Kaiser Permanente’s My Health Manager), most do not. Services like Zocdoc are tackling online medical appointments and scheduling across the nation. Teladoc is trying to offer secure messaging with licensed doctors via email, phone and video. Most retail pharmacy chains offer Rx refills online via their own patient portals. All of these services are trying to solve these problems outside of the PHR environment.
4. Data Authentication is a Bitch.
In the world of healthcare, the burden of authenticating a patent’s identity, across multiple systems, falls on the manager of those systems via HIPPA. In lay terms, this means that in order to release any of your medical data (data that is about you), a HIPAA covered entity or business associate has to confirm that you are who you say you are. This is not so much a regulatory barrier as it is a user-design and user work-flow issue. In an untethered PHR model, linking your PHR to an integrated service requires that you get your identity authenticated by the service you are linking to. So you have to create an identity on the service side first, (e.g., Quest Diagnostics). You also have to remember your username and password for the service while you are going through the one time linking process. Most people do not remember their username and password or do not have one yet and get frustrated and abort linking all together. If you don’t already have an online account with the service, you are forced to create yet another online credential. Then you may have to wait for that provider to confirm your identity by sending you an email confirmation, activating a PIN, or possibly even mailing you something to your home address. Each provider may have a different form of identity authentication which can slow up the whole linking process. Until this piece gets worked out, PHRs (especially untethered ones) will have a hard time gaining speed.
5. To tether or not to tether – that is the question.
Let’s first pay homage to several early PHR trailblazers who were in the space over a decade ago and provided early data points on the tethered vs. untethered model.
- Ed Fotsch at Medem created a PHR within a provider portal called iHealth that was co-founded by the AMA and several other national medical speciality societies. Medem was sold to Medfusion in 2009 but was best known for its doctor-hosted PHR model.
- Wendy Angst at CapMed launched a commercial PHR in 1990. The company positioned itself as an untethered PHR in 2003 using a USB key then quickly learned that the hospital tethered model was more viable with regards to diffusion and a sustainable business model. CapMed was sold to BioImaging in 2003 and then sold again to Metavante Technologies in 2009.
- Cynthia Solomon started one of the first consumer-based PHRs called FollowMe back in 2000 and has since garnered moderate success on the untethered side by serving special populations. She spun off Mivia.com, a PHR for migrant and seasonal farm workers in 2004, and most recently HealthShack.info, a PHR for youth and those aging out of foster care.
All of these early PHR innovators worked very hard to explore where PHRs would have the most uptake and impact. Recent survey data (Markle Foundation, Jan 2011) suggests that PHR adoption is trending upward, from 3% in 2008 to 10% in 2011. We can assume this uptick is on the tethered side (62% of respondents in the Markle survey say their PHR was offered by their doctor or hospital). Those surveyed by the The California Healthcare Foundation, April 2010 reported doctors as the most trusted source of PHR offerings (58%). While tethered PHRs have historically shown higher adoption numbers, they still don’t solve the problem of consumer mobility and data liquidity. The heaviest adoption seems to come from the large provider tethered offerings through organizations like Kaiser Permanente (they claim they have 33% adoption nationwide), Palo Alto Medical Foundation (PAMF), The Cleveland Clinic, Beth Israel Deaconess Medical Center, and MyHealtheVet at The Department of Veterans Affairs (VHA). There have been dozens of other attempts at trying to deploy independent PHRs directly to consumers or through payors and self-insured employers, most notably are players like WebMD, Dossia and Revolution Health who closed down their PHR business in early 2010.
With the push towards Accountable Care Organizations (ACOs), medical homes, and the inclusion of EHR interfaces to PHRs as part of the meaningful use criteria, there is a place for PHR-type applications to evolve in the future for providers who are managing patient interactions between cycles of care.
While there is so much more I could say about what I have learned in the last 15 years in this space, I invite my colleagues in the industry to chime in as well. I applaud others who are continuing to do the hard work in this space, like my hardworking friends at Microsoft HealthVault and those in the government who are looking for ways to support data interoperability in name of the consumer.
Missy Krasner, is the former founding member, Google Health & Senior Advisor to David Brailer, National Coordinator for Health IT, Office of the National Coordinator.