When Jonathan Hirsch cut his hand on a rusty piece of metal last year, he cleaned up the wound and promptly made his way to an emergency room.
Hirsch, a San Francisco-based technology entrepreneur, had to rush to the hospital, because he couldn’t recall whether he had received a tetanus shot in the past five years.
Two hours later, Hirsch emerged with a bill in the high hundreds of dollars, despite having received very little medical care. What he did get, after all that time and money, was a copy of his immunization record -- but only after five hospital staff members finally hunted down a scanned attachment of the record from another hospital in its network. It turned out he had been vaccinated just four years ago and didn't need another tetanus shot.
For hospitals and clinics all over the country, it’s still a technical and logistical nightmare to pull up a complete, accurate and up-to-date copy of patients’ medical records, whether it’s a family history, or a list of allergies and immunizations. Hirsch was lucky to receive this fragment of his medical record at all. It would have been far more cumbersome if he had selected a hospital in a different system or network, or if he had a complex set of medical problems.
The idea behind electronic medical records is to make health care more efficient and less costly by digitizing all of our health information so medical histories are more accessible. Hoping to finally make this goal a reality, in 2009, the Obama Administration allocated billions of dollars to incentivize hospitals, clinics and other care providers to adopt electronic medical record systems.
Although many hospitals and other medical facilities have digitized their records, the systems they are using are often proprietary and it’s difficult to easily share health information from one hospital or clinic to another. These challenges can be costly and slow down access to health care.
“The health care delivery system today treats patients as they engage it: on a one-off basis,” said Aneesh Chopra, who served as the first chief technology officer of the White House from 2009 to 2012. Chopra has dedicated much of his time in and out of office to solving this problem.
Modernizing medical records has been a major challenge, and requires cooperation from the government and the private sector. While the big medical records corporations, hospitals and federal officials attempt to solve these problems, some startup companies are hard at work trying to find solutions.
Chopra, who is now CEO of Hunch Analytics, a Virginia company that is working to use public and private databases to help health care companies become more efficient, is part of a new initiative called the Argonaut Project that aims to make exchanging health data easier.
Last fall, Chopra and a diverse group of policymakers, health executives and entrepreneurs sat down at a popular Turkish restaurant in Washington D.C. and discussed a solution that would “put patients at the center,” using the same types of technologies that have been widely adopted by Internet companies like LinkedIn and Facebook.
The Argonaut approach is to use web-based, open standards instead of much more complex, health care-specific standards. The project relies on public APIs, or application programming interfaces, which are commands the tech industry commonly uses to make it easier for software programs to share data. It’s basically developer speak for what Chopra refers to as a “handshake” between pieces of software, such as websites.
The goal for the project is to promote more flexible, easier access to key groups of health information. If it succeeds, mobile app developers will have a much easier time working with electronic medical records, which could make it possible for patients and doctors in the future to call up important patient information on smart phones and tablets.
According to Chopra, another key objective is to build tools to protect patient privacy, so patients will be aware and can consent when their data is shared with a mobile app or any other service.
Two of the nation’s largest electronic medical records companies, rivals Cerner, in Kansas City, and Epic Systems, in Wisconsin, are both founding members of Argonaut. Experts say that’s a good sign, and signals an increasing willingness from the industry to tackle the problem. And the project has a tight deadline: to come up with a solution by June.
“We are heading into health care's perfect storm - a shift to value-based, rather than volume-based, health care that will reshape the care delivery system and spur greater demand for open data apps,” said Chopra.
Silicon Valley Looks For Solutions
Seeing a business opportunity, a growing number of entrepreneurs in Silicon Valley and other tech hubs are also trying to come up with easier ways of sharing health data.
Some health startups are doing the tough job of collecting data from a patient’s medical record and storing it in one, easily accessible place. This is easier said than done. Patient data is often stored in formats that a computer can’t ingest and share, like PDF files. Hirsch, the patient who had to rush to the emergency room after wounding his hand, started a company in 2008 to make sense of this unstructured data. His startup, Syapse, based in Palo Alto, is working with large and regional health systems, like Intermountain Healthcare and the University of California San Francisco, to aid in the development of a database that combines medical and genetic information.
That type of project is one of the biggest potential promises of electronic medical records: To take huge amounts of data and use them to broaden and speed up medical research studies. But with many of the largest record vendors failing in this regard, dozens of other startups are stepping in.
And some hospitals, like Kaiser Permanente, are turning to entrepreneurs to help solve their own health data-sharing challenges.
Kaiser was ahead of the curve in adopting an electronic medical record system. Southern California Kaiser’s chief medical information officer, John Mattison, was one of the key figures behind that landmark decision in 2003, which helped put Epic Systems on the map when Kaiser announced it would be using that company’s software at its hospitals.
Mattison admits the electronic medical record systems are “a little delayed” in modernizing their systems, particularly their ability to meet legitimate requests for information. He said he has spent years pushing Epic executives to adopt more open software standards, so medical records can more easily share information with other vendors and apps.
Mattison said doctors at Kaiser and at other hospitals are frustrated by the amount of clicks of a mouse it takes to do any simple task, such as order medication. In the future, he suggests, the medical record may become a platform, or thin clinical layer of sorts, with newer, user-friendly applications for patients and doctors developed on top.
“I think we can get to a happy medium between fully opening everything, and the ‘walled gardens’ of the past,” he said.
According to Mattison, another area of investment is in identity management tools, which are desperately needed in the current system. As he puts it:
“I have a Maria Gonzales in my emergency room. She said she had a car accident three years ago in Chicago and had some stuff taken out, but she doesn’t remember what. The hospital in Chicago has hundreds of Maria Gonzales’s in their system. I do not want to act on the information of the wrong Maria Gonzales.”
Mattison said Kaiser is starting to evaluate several startups working on improving identity management, but declined to name them, saying it was still too early in the process. Some have called for a more drastic solution to this problem, such as a national patient identifier number, similar to a Social Security number. But that has been controversial and has raised privacy and security concerns. If information is indexed in this way, some fear that it could make a hacker's or snoop’s job easier.
But others say the risks are worth it, given the frequency that patients are mistakenly associated with someone else, and the serious medical problems that can develop if that occurs.
Another major issue with electronic medical records is typographical errors.
Boston-based patient advocate Dave deBronkart, known to his TED fans as “e-Patient Dave,” said his mother was recently discharged from a hospital and sent to a rehabilitation center. When her medical information was typed back into the IT system used by the rehab clinic, her thyroid condition was entered incorrectly, he said. DeBronkart said the “best doc in the world” could have given her the wrong medicine, if no one had spotted the error.
A handful of Bay Area startups have emerged to try to solve this problem. One company called Augmedix, based in San Francisco, has developed a doctor-focused app for Google Glass, the Internet-connected headgear that looks like a pair of high-tech glasses.
Another Silicon Valley-based startup, called Drchrono, is developing an application for Google Glass that it hopes will become the first “wearable medical record.” Doctors who use it can record a consultation with a patient’s consent, and notes are automatically stored in the electronic medical record.
When a physician sports one of these geeky-looking gadgets, they can use verbal cues to access the patient’s medical record. Notes are created and automatically entered into the record, which may reduce the rates of human error. Augmedix executives say they’ve experienced growing demand for Glass in health care.
Seemingly simple problems like data entry are the bane of many doctor’s days, as it means less time spent with patients. Many physicians are turning to solutions from the startup world to ease the transition to electronic medical records and open up the exchange of data.
Experts like Kaiser’s Mattison say that these new clinical applications will prove to be so valuable to doctors and hospitals over time that the medical record vendors will be forced to open up their systems.
“That will play out in the marketplace, as the forces are in place,” he said.
Another fertile area of growth is mobile health.
In Apple’s operating system alone, users can download more than 900 apps that track health and fitness. These apps can track everything from a users’ sleep to their moods. A growing segment of these apps helps doctors monitor their patients with one or more chronic diseases, like diabetes.
But sharing potentially valuable patient data has proven to be a challenge. This data, some of which is gathered on popular devices like activity trackers that people wear on their wrists, isn’t particularly useful to doctors for diagnostic purposes unless it is viewed alongside other important information, like family history and a list of medications.
Apple recently set out to solve this problem by developing a service called HealthKit, which aims to integrate data from the apps and devices that patients use. Apple partnered up with many of the largest electronic medical record systems, including Epic and Cerner, for this initiative.
But the process for patients to send data back to their electronic health record is wonky, and requires a number of steps. The patient has to consent to send his or her data from each app or device to HealthKit, where it will remain on the iPhone or iPad. That data can be shared with Epic’s patient-facing mobile app, for instance, and then eventually sent to the treating physician and stored in the medical record.
Apple has stressed that it will not share data with any third party, like an advertiser or marketer. It claims it will take on the responsibility of vetting out the apps that broker patient data, by rejecting the apps on that basis. Still, some people fear having their private information stored in one place, where it may be more susceptible to hackers.
Another key problem with Apple’s approach, some experts say, is that it requires patients to have an iPhone. Health experts are already calling for a set of common standards to ensure that data can be gathered from both Apple’s system and rivals like Android.
The bottom line: Patient-generated data increasingly is finding a place in health care. And systems will need to open up and modernize to keep up.
“I would not want to be treated as a cancer patient anywhere that didn’t have digital data. It’s a non starter,” said Mattison. “Then the question becomes, how do we make the experience more seamless for patients?”
Get the best of KQED’s science coverage in your inbox weekly.