The failed promise of interoperability
It's almost 2022 and health care still hasn't entered the information age
Throughout the summer, the health care system that employs my primary care physician, Advocate Aurora Health in Chicago, sent me reminders that I was eligible for the Covid-19 vaccine. I had received both shots in March from a rival health care system, which was quicker out of the gate.
In early October, when I visited Advocate’s nearest clinic for the booster shot, I showed the physician assistant my vaccine card. She finally recorded the earlier shots in my medical record. However, I am still getting reminders via Advocate’s patient portal that I am eligible for the shingles vaccine, another two-shot regimen that I received at a local Walgreens earlier this year.
My recent experiences are emblematic of the state of electronic medical record (EMR) interoperability in the U.S. Most patients’ medical histories still do not flow seamlessly between their different providers, which was one of the goals of the 2009 stimulus act that paid physicians and hospitals more than $30 billion to install EMRs.
We live surrounded by a vast medical-industrial complex: hospitals; doctors’ offices; walk-in clinics run by governments, pharmacies and worksites; stand-alone emergency rooms, ambulatory surgical centers, dialysis clinics, imaging centers and lab test sites. The profusion of organizations eager to dip into the 18% of gross domestic product devoted to curing the sick is testimony to both America’s entrepreneurial spirit and the wasteful competition it engenders.
The net result is that health care in America takes place in silos. And with a few isolated exceptions, those fragmented organizations still refuse to communicate with each other.
The cost of living in silos
That failure has many costs. It hamstrings clinicians’ ability to provide better care. Patients suffer when doctors, hospitals and clinics fail to communicate or coordinate the services they provide. Public health suffers when data isn’t aggregated, which prevents quick identification of disease outbreaks like the current COVID-19 pandemic. And taxpayers and employers suffer from the excessive cost (and the potential harm to patients) that comes from duplicative and unnecessary tests and procedures.
What prevented implementation of the law calling for health information interoperability? The short answer is the self-interest of every sub-industry within health care. A silo, after all, is a high circular wall designed to contain a commodity. In this case, you, the patient, are the commodity.
Providers want to hold onto their clientele. Interoperability allows record portability, which empowers patients to leave providers who offer substandard, expensive, or unfriendly care.
The EMR vendors that design the systems, a group dominated by a few large players, also have no interest in creating the common standards required for interoperability. Such standards would enable patient data to flow easily to rival EMR vendors, which in turn would ease vendor switching by their hospital and physician office customers.
Instead, the EMR vendor cartel (three firms have well over half the market) use the promise of interoperability to further their own dominance. Judith Faulkner, the CEO of the largest player in the industry, Verona, Wis.-based Epic Systems, offers interoperability to every hospital system through its Share Everywhere program. But, as a practical matter, it only works for those who buy EPIC software. “It takes gymnastics to get an Epic system to talk to a non-Epic system,” one hospital chief information officer told CNBC last year at the height of the pandemic’s first wave.
The unenforced law
The 21st Century Cures Act, signed into law by outgoing President Barack Obama in December 2016, called for the elimination of “information blocking” by health care providers, insurers and EMR vendors. Unfortunately, the Office of the National Coordinator (ONC) inside the Health and Human Services Department took until this past April to issue a rule requiring providers to share data and prohibiting vendors or others from engaging in information blocking.
Will it be enforced?
Even if the Biden administration takes a tougher stance than his predecessors (there’s no evidence yet that this will be the case), huge barriers remain. Interoperability requires common standards for entering data and information exchange; easier patient consent valid across multiple health care and social service settings; a common method for accurately identifying patients who may use slightly different names or addresses at different settings; and integration with the nation’s fragmented network of social service organizations, whose efforts are critical to improving health.
It also requires an easily accessible common platform for uploading and downloading data. The 2009 law called for creation of health information exchanges (HIEs) in every section of the country. But grant funding for HIEs expired years ago and the few that are still operating rely on user fees, which ensures limited participation.
There are public-spirited groups working on each of these issues, including some inside government. The National Information Exchange Model (NIEM), launched in 2005 to coordinate data flow between the new Department of Homeland Security and the Justice Department, is now being used by dozens of social service, police, education, and health-related agencies, including the Centers for Disease Control and Prevention.
“Exchanging data on foster children as they move from state to state; alarm companies and police dispatch centers; it’s really taking off,” said Paul Wormeli, an information technology consultant who works closely with agencies using NIEM. It’s also going international with Canada and France working on their own versions of the system.
However, the decision to use the model, which sets common standards for information exchange, is left to local officials and individual departments. HHS, which signed on in 2010, never implemented the program.
The ONC, under both Obama and Trump, chose instead to cater to the whims of EMR vendors and their hospital system and physician practice clients. The agency supported efforts over the past decade to create a common interface for exchanging data, but only among health care providers. That interface, dubbed FHIR for Fast Healthcare Interoperability Resources, does not communicate with the broader social service, education and criminal justice worlds that impact health and health care outcomes.
The ONC also prioritized creating an interface for third-party vendors wishing to sell providers their applications for scheduling, billing, wayfinding (yes, there’s an app for finding your way around the hospital) and other administrative functions that have little to do with care coordination. “The big hole is at HHS,” said Wormeli. “The new administration isn’t in place yet. We’ve talked to some bureaucrats who’ve been there, but there’s no leadership.”
Those trying to bridge the gap are taking a one-step-at-a-time approach. A handful of local groups working on integrating social service provision with health care – crucial to better outcomes for poor and disadvantaged populations – met for two years before deciding to work on creating a common patient consent form. It would eliminate the need for people to fill out similar forms every time they visit a different agency or provider.
“We focused on consent because that’s where we could have the most immediate structural impact,” said Dan Stein, CEO of Stewards of Change Institute, which is involved with the effort. “While there’s a tremendous distrust of bureaucracy by poor people, they know they have a complicated life and are eligible for multiple programs.”
Another point of light
A new consortium of hospital systems recently established a non-profit organization called Graphite Health to vet new software applications, including making sure they are safe from hackers. It’s a major concern. There were at least 80 ransomware attacks against health care organizations in 2020 with a median payout of nearly $50,000 to restore access to their data, according to a recent HHS report.
But Graphite’s larger objective is to expedite adoption of interoperable digital technologies that will improve the care process and lead to better patient outcomes. It chose the non-profit path to eliminate concerns the new venture may be promoting technologies that benefit its owners. “If it’s a for-profit entity, its charter is to generate profits for Wall Street. I don’t want to create another large predatory animal,” said Ries Robinson, the CEO of Graphite Health and chief innovation advisor for Presbyterian Healthcare Services, the largest health care provider in New Mexico.
“The mechanism by which you improve health care is the integration of the electronic medical record with payment information with agencies dealing with the social determinants of health (housing, food, transportation and other social services),” he said. “We want to do this in a way that gets health care systems aligning around the common good.” It modeled its effort on CivicaRx, the non-profit drug manufacturer created by a consortium of hospital systems to produce generic drugs that are either in short supply or are subject to price gouging by unscrupulous generic drug makers.
I wish them well. But let’s be frank. There are health care reporters working today who have spent their entire careers chronicling the thousand points of light that never move beyond pilot project status or are limited in their geographic reach.
Promoting a national system for data exchange and interoperability is not a major concern inside the Beltway. The recently passed $1 trillion infrastructure package contained no money for building out health information exchanges – or, for that matter, any money for health care beyond the benefits that expanded broadband coverage will bring to hospitals in rural areas. The pending $1.7 trillion Build Back Better package also ignores how better health information exchange can facilitate the provision of the many new social services contained in the bill.
That’s why, more than a decade after I began writing about health information technology, I remain skeptical about the U.S. health care system embracing the patient-centric benefits of interoperability anytime soon. The self-interest of too many powerful players stands in the way, and the government remains in their thrall.