The Elusive Promise of Electronic Health Records
Outcome-focused thinking does not focus exclusively on positive outcomes of a design or technology. In fact, predicting the potential negative outcomes is equally vital, yet is often ignored. This outcome-focused approach to technology is what we at Artefact are calling 21st Century Design. We believe 21st Century Design is the only way to mitigate the risk of ending up doing something bad. Equally importantly, it helps us maximize the good of what we do, not only at the individual level but also socially and globally.
The EHR outcome: The dream and the reality
The promised outcome of digitizing paper-based electronic health or medical records (commonly and interchangeably referred to as EHR and EMR) was that it would make health care cheaper and more efficient. By making records digital we would be able to share them with different caregivers in a patient’s care network, creating a holistic approach to care and better outcomes. Studies concluded EHR adoption would lead to better preventative care and chronic disease management. It did not hurt that the cost savings were also tantalizing – analysts at the RAND Corporation estimated that adoption of the new computerized systems could save about $80 billion over a decade.
In 2009, largely as a result of this kind of analysis and lobbying by EHR industry executives, the Health Information Technology for Economic and Clinical Health (HITECH) Act appropriated $35 billion to incentivize and promote the use of health information technology by physicians, hospitals, and other health providers. It’s estimated about $19Bn of that taxpayer money has been spent by care providers in the adoption of EHR systems.
Yet several years later, the projected benefits have failed to materialize. Earlier this year the RAND Corporation published a follow-up study concluding that the promised savings simply weren’t there. Neither were the promised efficiencies as many caregivers and institutions reported dramatically decreased efficiencies as they adopted these systems. On the patient side, the promised ability to share our records with our caregiver network for continued easy care remains a distant pipe dream. It begs examination of what has been the net benefit of this massive investment. What went wrong, could we have prevented it, and how can we fix it?
Short term focus comes at a cost to long-term outcomes
We can easily summarize the root of the problem as a perfect example of the right intentions but the wrong incentives. In order to make sure the investment in EHR was spent wisely, the government came up with some pretty comprehensive standards called ‘meaningful use,’ (MU) criteria. They directed software providers to focus on certain priorities that would make their customers, the hospitals and service providers, eligible for the massive amounts of subsidy money.
The MU criteria were divided into three important milestones, in Stage 1, targeted at 2011-12, the outcome to be measured was “Data capture and sharing.” Stage 2, slated for 2014 measures outcomes from the perspective of “Advancing clinical processes.” Phase 3, set to start in 2016, finally aims to focus on “Improved outcomes” for patients.
Like many well-intentioned government programs, the standards have had a massive effect on the priorities of the software companies in that industry. Handing out billions of dollars of taxpayer money with an outcome of “data capturing,” i.e. getting people’s medical data online, was a hugely wasted opportunity. Instead of pursuing the larger vision of measurably improving the healthcare of patients, everyone focused on compliance with the Stage 1 criteria in order to qualify for the subsidy money. The outcome: a sophisticated billing system.
The first generation of EHR products is nothing more than form-filling interfaces to convoluted patient databases, recording what drugs, treatments, and tests have been performed. These software systems provide a bunch of screen-based forms, organized around billing and insurance categorization standards and scenarios rather than patient outcomes. When designing complex systems and devices remaining focused on the ultimate goal is incredibly challenging. But when you incentivize providers for a lesser achievement than this ultimate goal, you are guaranteed to create a misalignment that is very hard to correct.
I am cautiously optimistic that Stage 2 and 3 MU criteria will drive the design of these systems back towards the originally stated clinical and patient outcomes. Every care must be taken to ensure that every single feature of a new system is relentlessly focused towards the long-term outcome of the product (healthcare). Yet my concern is that those outcomes haven’t been front and center during the last few years of frenzied adoption. It will take several years to undo the impression amongst caregivers (and patients) that these systems are anything more than a tax on their productivity.
This first stage of development and adoption may still have a silver lining. Over 70% of healthcare organizations have adopted EHRs – and for many that technology adoption is irreversible. While the promise of EHRs for better healthcare on the personal level is still elusive, the value of the data collected is starting to bear fruit as more and more big data analytics applications emerge that help with pattern recognition and prevention. Yet, unless we focus on improving individual outcomes for patients, the clear winners will be the big companies whose profits are soaring across the records industry, rather than patients or the caregivers themselves.
Re-shifting the focus to patient outcomes
Next year, hospitals and health organizations will initiate another wave of EHR upgrades that will make sure that they comply with the Stage 2 of the MU criteria. The temptation will be to focus on addressing the clinical advancement criteria, putting off patient outcomes until Stage 3 looms closer. However, a smarter and more forward-looking approach might be to aim for Stage 3 compatibility within Stage 2 implementation.
The Usability Problem
A recent survey of clinicians by the American College of Physicians reported that user satisfaction with electronic health records has decreased since 2010. Nearly 40% of users expressed dissatisfaction with the ease of use of these systems, and that is not surprising given that other studies report physicians on average spend 17.5% more time during a shift documenting data in EHR. That is 17.5% less time spent with patients. In fact, the burden of data entry is so significant that The New York Times reported recently medical practices are reverting to hiring scribes, prompting the reporter to comment: “In most industries, automation leads to increased efficiency, even employee layoffs. In health care, it seems, the computer has created the need for an extra human in the exam room.”
It’s not hard to find anecdotal evidence from doctors and nurses – some even filed a class action lawsuit based on how bad the software is. The design of some of the EHR user interfaces has even been described as ‘hostile’ – as if designed to maliciously mislead the user and ignore the context of use. Our own conversations with clinicians confirmed these findings: one clinician admitted that physicians in his practice had to cut back the number of patients they see in a day to almost half while learning the new EHR system.
At the heart of the usability and design problem is that an EHR system is trying to serve two different masters: the clinical staff or caregivers (users) and the administrators (buyers). The needs of these two constituencies are rarely aligned –caregivers look for ease of use and integration with their very specific workflow, administrators optimize for cost savings and easy billing. The result is a disruption of process and workflow that not only slow clinicians down but introduces a higher risk of human error, as medical professionals admitted to us in interviews. Creating a user interface and product that satisfies the needs of clinicians and administrators, especially in light of the lack of alignment between MU criteria and patient outcomes, is one of the big design challenges in the space.
The healthcare industry is not the first one to experience such a conflict between users and buyers. Just a few years ago, the majority of enterprise IT decisions were made by CIOs and IT managers without direct input from users. Today, “consumerization of IT” or the idea that technologies must be easy-to-use by non-technologists, or that you can bring your own device or service and use it in your line of work is among the top criteria in selecting a technology solution. While the healthcare industry is bound to follow the same path, today’s EHRs are far from being “doctor-friendly.” Instead of user-centered approaches to the design of these systems, and user satisfaction-driven adoption, buying decisions are made by large boards and committees disconnected from the realities of caregiving. This process, especially in light of very specific MU incentives, has lead to the abstraction of functional and feature requirements into vague ‘checkboxes’. As a result, user experience design and measurable aspects of the usability have become ‘nice to haves’ rather than key criteria for purchasing. But usability is far from being a subjective requirement; a lack of usability in EHRs has not only caused frustration for caregivers and undermined adoption strategies, but they have also proven harmful when medical errors have occurred because a user could not find or did not see relevant information. Medical errors are only the most salient evidence. Coordinating between different clinicians on a care team, efficiently arranging discharge, choosing the best among several treatment options based on personal characteristics are more subtle examples that have a tremendous aggregate effect on outcomes and efficiency.
EHR evolution or survival of the most usable
While very profitable in the short term, the strategy of focusing on the government checkboxes rather than usability does not bode well in the long term for either the EHR providers or the hospitals that buy the clunky systems. We are already seeing innovative EHR providers like Practice Fusion who invest in usability and patient outcomes encroaching on the incumbents. As government subsidies shift to encourage advanced clinical processes and better healthcare outcomes, these user-focused EHR providers will be the companies that suddenly become the better choice for hospitals and service providers. And as clinicians get fed up by the ever-increasing workload that EHR systems force upon them, there will be internal pressure to reevaluate EHR choices and that process is so costly it may be practically impossible.
In order to future-proof EHR investments, we must build systems with sufficient flexibility to assist in the process of gathering information from various sources at the point of care and beyond. The wearable technology space and the Internet of Things will have a dramatic impact on the quantity of patient data being recorded. In addition, as services built around the Quantified Self movement evolve and begin driving long-term healthy behavior change and helping with chronic condition management, we would need to design systems that empower providers to easily synthesize and mine the data for insights, and that allow for different kinds of doctor-patient interactions. Finally, we must also consider the context of use, and focus on improving the patient-doctor relationship in the design, so that the technology supports better human communication – not more efficient human-computer interaction.
Better usability and outcome focused solutions are possible. We designed the Seattle Children’s Patient Information System prototype to support existing EHRs while focusing on better communication between members of hospital care teams and better patient care.
A treatment plan for EHRs: Outcomes + usability
While focusing on patient outcomes and usability will each help EHRs deliver on their original promise, working on both simultaneously is the only way to realize that promise effectively and immediately. Focusing simply on usability without improving patient outcomes will make adoption faster, but in a way, it will also help us get to the wrong destination faster. And with usability so poor, and doctors spending such a significant amount of their patient time taming the tool, it can be argued that positive healthcare outcomes without usability are impossible.
At $20B, the EHR space is approximately 10% of the US software industry. But the opportunity cost of developing EHRs with the best patient outcomes in mind is more than just money. It is our health. After all, isn’t it a universal wisdom that there is nothing dearer than our health?
This article originally appeared in MedCity News.