Dan O'Neill

I am interested in the policy, economics, and technology of complex systems, particularly in American healthcare.

San Francisco, California

Durable interoperability? Maybe a market isn’t such a bad idea

This article was originally published in Becker’s Healthcare.

Senate hearings on St. Patrick’s Day revived debate about electronic health record “interoperability” – the ease with which health information follows a patient across settings of care.

After the ONC’s subsequent report on “information blocking,” some major vendors rushed to drop fees for data transmission services, and Dr. DeSalvo outlined her perspective in a thoughtful presentation and post on the Health Affairs blog. After years of billion-dollar subsidies, scrutiny is warranted, but the rush to thump the big IT vendors may prove short-sighted. Perhaps the simplest approach to incentivize interoperability “in a durable and sustainable way,” as Dr. DeSalvo suggests, would in fact be a vibrant market for health information exchange services and infrastructure. Such a market would ensure continued investment and innovation to support interoperability, and would give providers, patients and healthcare leaders more options to handle, transfer and consume information. The furor currently directed at Epic could well deter exactly the sort of data services the system needs.

The idea is hardly novel. This morning, I exchanged location data with an Uber driver to arrange a pickup, sent payment information (securely) to a coffee shop for my morning Joe, and then joined a conference call, where I had no trouble sharing all of the information on my computer screen with a colleague in Chicago, in real time and through a secure connection. Within 90 minutes, I used a half dozen different network services to make these interactions possible – banks, credit card networks, cellular providers and Uber itself all collected fees to support payment flows, absorb non-payment risk, and transmit mobile data to and from my iPhone. Other technology and telecoms firms collected fees from my employer to handle the conference call traffic and enable screen sharing. Yet, even as we champion the value of seamless, secure health information sharing, many politicians and pundits seem oddly squeamish about the emergence of similar network business models in the healthcare arena. To be clear, this is not a market for protected health information itself, but a market for the infrastructure and transmission services that let it move smoothly across settings of care. This seems illogical – when a leading health IT vendor describes a $2.35 fee to support patient data sharing, as occurred before the Senate committee, policymakers should celebrate a burgeoning market for a sought-after service, not respond with misplaced Jeffersonian outrage at a business with the chutzpah to charge for its products. Banks and phone companies respond to consumer demand for easy payments and mobile data; as hospitals and health plans write checks for connectivity, the same is starting to occur in healthcare. No tax dollars required.

Strangely, those who rail against health data transmission fees also seem unaware that this exact model underpins the most successful existing example of digital health information exchange. Fifteen years ago, virtually all prescriptions were scrawled on slips of paper and carried from the doctor’s office to the pharmacy; today, roughly two-thirds of all prescriptions arrive at the pharmacy electronically, and most physicians can readily access a patient’s medication history at the point of care. Secure, real-time health information exchange, for millions of patients every day. Why? Simply put, pharmacies and health plans believed they would benefit commercially by digitizing the data flow, through processing efficiencies and healthier patients, and so began paying for the service. Today, every time a healthcare IT vendor transmits an electronic prescription, pharmacies and health plans pay a small transaction fee to that vendor. Unsurprisingly, some 600 different EHR applications have connected to the largest prescribing network using a single standard, half a million physicians write prescriptions electronically, and illegible paper scripts are a (dangerous) anachronism.

With so much at stake, it’s fair to ask why a broader market for clinical data exchange has not already emerged. In part, of course, it has, though only in fits and starts, and often with an emphasis on claims and other administrative data, instead of the critical clinical information which will power better care. Firms like Emdeon, RelayHealth, Orion Health and Medicity have long charged various transactional and subscription fees to build and maintain clearinghouses and local interfaces. A newer generation of population health tools, such as Explorys and Health Catalyst, occasionally charge interface fees to capture and crunch data from source systems. And then, of course, EMR vendors have dipped a few toes in the water. All of this, however, pales next to the scale of the opportunity to coordinate care and eliminate duplicative and unnecessary tests and procedures. In 2012, the Institute of Medicine pegged the overall annual cost of avoidable and wasteful care at $750 billion. If robust, fluid data-sharing can eliminate even 1% of that waste – which seems highly conservative – we have a $7.5 billion breakeven budget for health information exchange infrastructure and services. More realistically, the value would likely reach the tens or hundreds of billions. From another perspective, if the American healthcare bill runs to nearly $10,000 for every man, woman and child, surely a few dollars for data transmission is worthwhile, if it ensures that the $10,000 is wisely spent?

The friction, as ever, stems from fee-for-service reimbursement, which often incents doctors and hospitals to skimp on care coordination. Companies embrace technology which improves productivity, by reducing costs or powering new revenue opportunities. Banks, for example, invest massive sums in their web and mobile services because deposits are far cheaper to process online than by staffing a branch, and because convenience attracts customers. Healthcare is no different. In areas where technology creates cost savings – such as claims processing – digital banished paper long ago. On the clinical side, hospitals and physicians are still primarily paid for each test and procedure they perform, not for the speed with which a patient returns to full health. Investments chase returns – hospitals purchase new CT scanners instead of technology to coordinate care and eliminate duplicative tests. Since health information requires the same robust infrastructure as banking or mobile telephony, it carries a cost. When providers have a greater incentive to deliver more care than to ensure better care, they balk at this cost, and allocate budgets elsewhere.

Meanwhile, although the Meaningful Use program certainly nudged many providers to purchase and use software, payments to doctors and hospitals were generally not contingent on actual, ongoing health information exchange, nor on the achievement of any particular improvement in care outcomes using the new tools. Arguably, the program is akin to a subsidy for the cost of Microsoft Outlook, with one test message sent as proof of purchase, even if the user then returns to the warm embrace of the postal service, or never again sends mail in any form.

Finally, tangled regulations add complexity and compliance cost to clinical data transmission. For example, various rules intended to guard against anti-competitive and genuinely fraudulent behavior – such as the federal Anti-Kickback Statute – are often misinterpreted as limitations on the ability of some institutions (particularly hospitals) to pay for the very data exchange those same rulemakers demand. The legislative objectives are worthy, but unintended consequences abound, not least because poorly understood rules and unbounded potential penalties ensure risk aversion and interminable legal wrangling before the engineers can get to work.

Happily, green shoots are visible. Medicare and many private health insurers are rapidly shifting incentives from care quantity to quality. In response, hospitals and physician groups have started to invest in the networks and interfaces they need to coordinate care and earn quality bonuses. Many laboratories have adopted the pharmacy model and now fund electronic data transmission; some forward-thinking radiology chains are starting to follow suit. Strikingly, leading health insurers, such as Humana, United Healthcare, Aetna, Blue Shield of California and Blue Cross Blue Shield of Michigan, have recognized their own bottom line interest in coordinated, lower cost care, and are investing directly to support data exchange. In response, EHR vendors are building business models around data sharing and quality management instead of the software itself, and new startups like Moxe Health and par8o offer tools to simplify transmission. Health information exchange is tremendously valuable, and like most valuable things, it does cost money. Congress should encourage an emerging market for clinical data transmission, rather than strangling a few promising seedlings with misguided regulation, or drenching the field in taxpayers’ cash.