Healthcare Analytics, Population Health Management, Healthcare Big Data


Can Application Programming Interfaces Inspire a Better EHR?

Application programming interfaces will help build a better electronic health record, but only if providers and vendors work together on data governance and interoperability.

Source: Thinkstock

Almost everybody can recall being told the stories of Archimedes in the bath and Isaac Newton under the apple tree experiencing spontaneous moments of ingenuity and inspiration that irrevocably changed how scientists – and everyone who depends on scientists – view the world. 

These picture-book moments have endured in popular culture due to their compelling storylines and attractive message: genius can strike at any time through the simplest of circumstances, prompting a sudden seismic shift in reality that immediately filters through the global consciousness.

But any programmer who has spent countless hours staring at code streaming across a computer monitor and every lab technician staying after midnight to watch humming equipment interminably process a test result knows very well that in reality, scientific discovery, clinical breakthroughs, and technical innovation are actually kind of boring.

They take time, trial and error, strict adherence to detailed processes, a great deal of caffeine, and occasionally a complete reinvention of a fundamental hypothesis that forces six months’ worth of hard work into the garbage bin.

Unfortunately, it’s a lot easier for observers to cling to grade school expectations about scientific advancement than accept the fact that developing a theorem, a new treatment, an algorithm, or a technical infrastructure can take years of extremely mundane and repetitive effort.

After years of slogging through productivity-draining workflows, restricted data access, and overwhelmingly irritating alarms and alerts, health IT users can perhaps be forgiven for longing for a lightning flash of insight to solve their EHR and data access woes.

If good data is supposed to be the bedrock of clinical care, then the industry is just starting to realize that it has built its costly infrastructure on shifting quicksand.

Even after years of commercialization, multiple certification criteria, massive federal investment, and endless industry discussions about best practices and optimization techniques, providers still intensely dislike their EHRs.

At the 2017 HIMSS Conference and Exhibition in Orlando this February, vendors were eager to throw out terms like application programming interfaces, third party developers, FHIR, and data standards, all of which encourage external organizations to come in and tinker with their systems.

There are several clear benefits to this strategy.  For one thing, it signals a recognition that no single EHR company has all the answers, and that collaboration is absolutely essential to make any progress from this point on. 

For another, it can be a cunning way for vendors to save some time and expense by letting eager young startups do the heavy lifting in exchange for the hope of access to a larger customer base than they could acquire on their own.

It’s a win-win situation for providers with the cash or knowhow required to integrate bolt-on applications after spending millions on a core system.  It may be less of a benefit for those organizations struggling to stay above water without sufficient funding and development expertise to take a step past their core systems.

But either way, it clearly indicates that the majority of big-name EHR vendors are still, after all these years, wrestling with the fundamental problem that has plagued the paper-to-digital conversion: users and developers both started the process by treating data governance, interoperability, and usability as little more than afterthoughts.

If good data is supposed to be the bedrock of clinical care, then the industry is just starting to realize that it has built its costly infrastructure on shifting quicksand – and it will take much more than a single eureka moment to reengineer the basics of how health IT users interact with their electronic health records.

Interoperability falls flat without strong data governance

It would certainly be nice if every major electronic health record vendor was simultaneously granted the ability to push out seamless data exchange to all their customers in 24 hours or less, but the truth of the matter is that interoperability is a difficult and messy business with multiple complex layers.

“There isn’t one simple solution for interoperability and integration,” said Meg Marshall, Senior Director of Health Policy at Cerner Corporation.  “There isn’t a single button that says, ‘if you implement this, everyone is going to have exactly what they need at all times.’”

Meg Marshall, Senior Director of Health Policy at Cerner

Meg Marshall, Senior Director of Health Policy at Cerner

Source: Xtelligent Media

“Exchanging data and optimizing workflows aren’t just difficult because of the zeroes and ones.  There are certainly basic technical issues that still need to be addressed, but there are also so many different vocabularies being used, and so many different workflows that need to be supported.”

The variety of EHR use cases within a single healthcare organization, the fiercely guarded individuality of how physicians practice medicine, and the chronic lack of time to compose a perfect magnum opus of documentation for each and every patient have presented some major challenges for developers.

Users will quickly rebel when vendors try to implement a one-size-fits-all approach to data creation.  But when EHRs leave documentation choices entirely up to the individual, data integrity can just as quickly degrade to a point of no return.

“We’re starting to get on the same phone line – that’s the first phase of interoperability,” said Marshall.  “Next, we need to make sure we’re all using the same language.  That’s where governance comes in.”

 “There isn’t a single button that says, ‘if you implement this, everyone is going to have exactly what they need at all times.’”

That is where organizations consistently struggle, and while they are likely to blame their vendors for designing unusable systems that force them into shortcuts and workarounds, developers would not be entirely wrong to say that a great deal of the onus also lies upon the person at the keyboard – and the organization overseeing how that user is trained to create and input data.   

“A lot of providers look at technology as a deterrent and a challenge, and I think in some cases it is,” said Jason Burke, Vice President of Data Informatics at 3M HIS. “There needs to be a lot more emphasis on improving workflows so they can spend more time treating patients and less time documenting what they’ve already done.”

READ: The Role of Healthcare Data Governance in Big Data Analytics

Most EHR vendors are working on that, Burke readily acknowledged, but provider organizations have to match those efforts with a strong focus on creating clean, complete, and accurate data that can flow easily through those hard-won interoperability pipelines.

“If you want to get real value out of your systems, you need to take a holistic approach to managing, securing, and operationalizing data,” said Burke. “That’s what data governance is.  It’s an end-to-end initiative that must take a number of things into account, such as security, data rights, and interoperability.” 

Every organization should develop – and adhere to – a comprehensive framework that governs data from creation to deletion, says AHIMA.  Doing so will prevent some of the most common EHR pain points, such as duplicate records, copy-and-paste errors, upcoding, wrong-patient events, and data breaches that can tank an organization’s public reputation.

Strong governance creates a sense of trust in the accuracy and usability of the data, which is particularly important as information starts moving back and forth between systems.

“There’s no room anymore for inconsistent quality and inconsistent data,” said Ann Chenoweth, MBA, RHIA, FAHIMA, the 2017 President and Chair of the AHIMA Board of Directors. “Trusted data must be reliable, accurate, accessible, where and when it’s needed.  It’s not the data that comes out of here verses the other system.  It has to be an enterprise-wide framework that you can rely on.” 

Governance is also essential for quality measurement and physician benchmarking, whether or not the organization is currently engaged in risk-based reimbursements or accountable care contracting, said Burke.

READ: Health Information Governance Strategies for Unstructured Data

“Over the past ten years or so, the measurement has gotten very aggressive,” he said.  “Everything physicians do is being reviewed by multiple entities.  That makes data integrity even more important, because they have to ensure that what they’re documenting is correct, that it accurately reflects their care, because that’s what payers have to go on for value-driven reimbursement.”

Even if pay-for-performance dollars aren’t on the line, no clinician wants to look like he or she is underperforming just because of lackluster documentation, Burke added.

“I’m not sure everyone is going to enjoy the conversations we need to have about governance and documentation improvement, but we have to address it.  Using the EHR properly is just what it means to be a part of healthcare these days, and clinicians will have to realize that.”

EHR vendors need to help physicians heal themselves

While data governance and integrity have to take root at the organizational level, vendors do in fact acknowledge that providers can only progress as far as their systems allow them to. 

“As vendors, it’s our obligation to find better ways to make documentation less intrusive and find better ways to surface data to providers without overwhelming or frustrating them,” Burke stated.

“I’m not sure everyone is going to enjoy the conversations we need to have about governance and documentation improvement, but we have to address it.”

“We need to figure out how to reduce the amount of digging required to get an answer.  If a patient had an MRI a week ago, it might not be necessary to do another one.  But if the provider doesn’t know about that existing test because it’s buried somewhere inaccessible, guess who’s going to be spending time and money on a new MRI?”

Inaccessible data can have serious ramifications for patient care, but an unmitigated “data dump” could be even worse, pointed out Dr. Michael Blackman, Chief Medical Officer at McKesson Enterprise Information Solutions.

“The initial complaint about interoperability was that providers weren’t getting any information at all,” he said.  “But now we’re moving on to the second complaint, which is that they have information, but don’t know what to do with it – or they have information, but it’s just not useful.”

Dr. David Blackman, CMO of McKesson Enterprise Information Solutions

Dr. David Blackman, CMO of McKesson Enterprise Information Solutions

Source: Xtelligent Media

“There is a very strong temptation just to ignore all of it, especially given the time constraints most clinicians face.  But then if something goes wrong, the question becomes, ‘Well, you had access to 5000 pages of information.  How come you didn’t read them in the ten minutes that you had with the patient?’”

For years, providers have been demanding that vendors find the magic balance point and roll out tools that can intelligently highlight the right data for the right person at the right time.  After all, many argue, that was the promise of the electronic health record in the first place.  And that promise has largely gone unfulfilled.

The simmering discontent puts vendors in the difficult position of having to take responsibility for shortcomings that are also deeply rooted in convoluted regulatory processes, the nature of the free market, and the fact that not even the top executives in the industry can actually predict the future.

At the start of the EHR Incentive Programs in 2011, interoperability wasn’t even on the radar.  There was no such thing as the app-based ecosystem that has become so familiar to consumers today.  There were no industry-wide data standards, and there wasn’t yet a strong financial incentive for developing them.

“You had access to 5000 pages of information.  How come you didn’t read them in the ten minutes that you had with the patient?’”

“We've gone from 30 percent EHR adoption to 80 percent or 90 percent adoption in the blink of an eye, in the context of technology,” said Arien Malec, Vice President of Data Platform and Acquisition Tools at RelayHealth.

“I remember using the internet in 1994 or 1995, and I can tell you that it wasn't super simple or super easy to use with a great customer experience.  I remember plugging in a USB drive and first having to go download drivers before I could use it.”

“If we expected the internet of 1995 to work the way it does today, we might be pretty frustrated.  That’s where we are right now.  When you go from low technology adoption to high technology adoption, you're going to have some issues.  It takes time to work through that and create an ideal experience.” 

READ: The Difference Between Big Data and Smart Data in Healthcare

And to their credit, vendors are generally keenly aware that they have a lot of ground to make up in that department before the gulf widens any further.

“If nothing else, there is a willingness to have the conversation now, and that wasn’t always there,” observed Bharat Rao, Principal in KPMG’s Advisory Services.

“There’s an increased willingness from providers to stand up and to say, ‘I’m tired of not having data.  Let’s do something about it.’”

Vendors are indeed working diligently to respond appropriately, asserted Cerner’s Meg Marshall.

“If you ever hear a vendor saying they’ve done everything they possibly can but their customers are still seeing friction on the other side of that – well, then it becomes pretty clear that the vendor is not delivering services to them in a way that’s useful or the way they would expect to see,” she said.

“If you have a physician who’s spending way too much time on a given function, we want you to use the built-in timers and other mechanisms that can feed information back to us and let us know there’s a pain point there.  And we want you to do that long before that friction results in a patient safety event or starts becoming a major part of someone’s burnout.”

“There’s an increased willingness from providers to stand up and to say, ‘I’m tired of not having data.  Let’s do something about it.’”

Tina Joros, Allscripts VP and Open Business Unit General Manager, agrees that vendors must keep pushing forward with optimization and interoperability until providers start taking a kindlier view of their health IT products. 

“Until everyone has all the data they need, our job isn’t done yet,” she said. 

“I think every vendor is looking forward to the day where integration itself isn’t the selling point.  Once we’re all integrated, we can put even more focus on the features and functionality, usability, user interfaces, and all those things that make applications great.”

App stores, APIs, and the little guys

In those early days of EHR implementation, providers who cobbled together workable infrastructure from bits and pieces of best-of-breed systems quickly found themselves in a pretty tight spot when it came to data integrity and interoperability. 

Many organizations are still trying unwind their convoluted data systems after some early missteps.  Rip-and-replace projects remain common, and EHR vendors are still actively fighting each other for new customers as providers muse over how to overhaul their first-generation attempts at health IT maturity.

Their options in 2017 look very different than they did in 2011. 

Picking the right core “operating system” is still a critical decision, but just like the choice between current-day Android and iOS smartphones, it’s become less about losing or gaining functionality and more about how the user wants their system to look and feel. 

Tina Joros, Allscripts VP and Open Business Unit General Manager

Tina Joros, Allscripts VP and Open Business Unit General Manager

Source: Xtelligent Media

That is largely due to the rise of the application programming interface (API), the new darling of the tech world in both the consumer and enterprise markets.

“Our clients are very excited about APIs,” said Joros.  “They have a lot of questions about how they can use them, and we’re working with a large number of developers who are now able to connect using those solutions.”

Allscripts, along the vast majority of its competitors, is actively working to expand its relationships with third-party developers experienced with APIs and especially with FHIR.  

The company is rolling out FHIR connectivity to all its clients in its Stage 3 Meaningful Use release, and has streamlined its development partnership program into a click-through agreement that allows third parties to speedily access tools.

“We made a ton of improvements to our developer portal to help improve the experience of them working with our APIs, and so we’re really excited about that,” said Joros. “We think it makes us more open than ever.”

Cerner is also an old hand at API-based interoperability.  In 2015, the company started testing SMART on FHIR applications in its Millennium product, and launched a developer portal in early 2016.  

READ: Why Application Programming Interfaces Are Key for Healthcare

Later that year, McKesson Health Solutions announced its Intelligence Hub, an interoperability commons that included options for third-party use. 

Epic Systems waited until just last month at HIMSS17 to launch its App Orchard, but foresight – and perhaps even a touch of that elusive sudden inspiration – was on athenahealth’s side way back in 2013.

The idea of an app store may have been somewhat novel back then, but the modern technology user is perfectly comfortable with the concept, giving smaller external developers a chance to leap into action.

“There are a lot of small, nimble companies with great ideas who are going to drive the industry forward in a really important way,” said Jason Burke. “Everyone is talking about using machine learning, natural language processing, and artificial intelligence for their breakthrough products, but not all of them are going to succeed.”

“The challenge for provider organizations and for the bigger, established vendors is going to be how to pick the best partners that will help them achieve tangible results.  A lot of the EHR vendors are rushing to form good partnerships, and it’s going to be very interesting to see how that all plays out.”

 “You don’t even have to spell out the acronyms anymore.  People get it.”

The market is already jam packed with companies looking for their breakthrough into the big time, and the trend is likely to continue as providers turn their attention to what workflow, analytics, and interoperability enhancements third-party developers have to offer. 

“There is an explosion of platforms – small companies, medium sized ones – creating connectors and apps and pathways for data,” said Rao. 

“We will probably still be talking about the challenges of interoperability twenty years from now – it won’t ever really go away – but we are making progress.  I see a lot of progress being made through APIs and putting data to new uses.  That is especially prevalent in organizations that start with good data integrity practices and build out from there.”

If established EHR vendors continue to embrace third-party developers as innovation partners, perhaps they will be able to bring back a little bit of the storybook lightbulb moment to disillusioned providers by turning a drowsy Friday afternoon browse through the app store into an opportunity to make a game-changing discovery for their organization.

The more open the industry can become, the easier it will be to enact exponential improvements that make a tangible difference to providers and their patients, said Joros.

“We’re actually seeing the results of what happens when we all work together and collaborate,” she said.  “You don’t even have to spell out the acronyms anymore.  People get it.  They want it.  And they’re starting to use these new pathways to do some really great things for their organizations and for their patients.”

“It’s very exciting to be in the industry right now.  We’re heading for some beautiful things that can only come from listening to what our clients need and giving them those things as quickly and easily as we can.  APIs, better workflows, and better governance are going to be a big part of that, and they’re going to make a huge difference if we can get them right.”

This article was originally published on March 17, 2017.


Join 25,000 of your peers

Register for free to get access to all our articles, webcasts, white papers and exclusive interviews.

Our privacy policy

no, thanks

Continue to site...