Health Systems Spending Billions To Prepare for the 'Last War'

by Dave Chase


Health care IT departments have focused much of their attention on the $19 billion portion of the stimulus package allocated for the meaningful use of electronic health records. While this is logical given the available money, it is paying for health IT systems that are optimized for the "do more, bill more" model of reimbursement. However, that model is rapidly being replaced by a focus on value and outcomes -- a 180 degree shift.

On the one hand, it's hard to argue against modernizing health care systems. Thousands of lives are saved as a result of this modernization (e.g., avoiding deadly prescription errors). On the other hand, most companies benefiting from the stimulus package have two significant shortcomings that will need to be addressed for health systems to thrive in the new environment they are facing.

1. The core design principle of most systems being deployed was optimization for the rapidly fading "do more, bill more" reimbursement model, where the patient isn't much more than a vessel for billing codes. For entirely logical reasons, given the old reimbursement model, their success was measured by their ability to get as big of a bill as possible out as fast as possible. The shift to a value-based model of reimbursement flips provider incentives on their head.

2. These systems were designed for a health care system in stasis with a technology architecture not known for its nimbleness. If there is one area of consensus about the future of the U.S. health care system, it is that it's going through radical transformation. Nimble systems will be imperative to respond to a rapidly changing landscape. 

Already health care providers are realizing that what they thought was going to be their 100% solution is really best optimized for just 25% of where health care dollars are spent (hospital-based care). Indeed, 75% of health care spending is directed toward chronic disease. EHR systems from Epic and Cerner have their strength in automating internal workflows of hospitals and other clinical settings. In those high-intensity settings, health care providers make the decisions that drive the patient health outcomes. However, with chronic disease, it's an entirely different story. The decisions an individual makes drive the health outcomes (e.g., filling prescriptions, lifestyle choices, etc.).

Throwing Rocks or Birds at a Target: Manufacturing vs. Service Orientation

Providers who have demonstrated outstanding results with challenging patient populations recognize that there are two main care approaches. In a setting such as a hospital, many leading hospitals have successfully adopted a manufacturing-based model borrowed from Toyota. However, with chronic disease, a service-based approach is necessary to effect behavioral change. In a manufacturing setting, with the proper guidance a machine will do what it is intended to do and doesn't have a mind of its own. However, as anyone who has been in a service-based business knows, human interaction and a partnership-oriented approach lead to the best outcomes.

Let me draw an analogy. Consider throwing a rock at a target. Like a manufacturing scenario, with enough practice a well-trained professional can hone his or her craft and hit the target most of the time. Now imagine rather than throwing a rock, you are throwing a bird at a target. Perhaps you can influence 10%-20% of whether that bird hits the target. However, the other 80%-90% is going to be driven by understanding the motivations of the bird. Perhaps putting food or the bird's babies at the target would be necessary to drive the bird's behavior.

As with the bird example, doctors push patients toward a desired health target. However, only those organizations that have systems and processes optimized for engaging patients have had significant success with using IT to manage chronic conditions.

Disruptive Innovation in Care Delivery Requires Rapid Iteration

As highlighted in a previous article, the organizations driving breakthrough results aren't tweaking an existing model. Rather, they have developed new models that get rapidly iterated. As one who has implemented traditional health IT, the process is very involved with many months of planning before go-live. During that process, there is a ton of process planning and re-engineering before configuring the system. Roughly speaking, process is weighted 80%-90% toward pre go-live with 10%-20% focused on post go-live to deal with go-live issues and some further training.

In highly dynamic environments, the pre- and post-live weighting needs to be flipped on its head (i.e., 20% planning and 80% analyzing, refining, testing, etc.). While some areas of health care will be stable, the most critical area to manage is where the greatest costs reside -- chronic disease. Some best practices have begun to emerge; however, one can expect a high degree of iteration to address the various areas of chronic disease management. Some of the big EHR systems have the ability to address different workflows after significant customization. However, health care providers report that if they need to reorder workflow, the system has to be reconfigured with considerable time and expense involved. In this environment, it could prove to be even more costly than expected.

Eric Page of Amplify Health has shared his experience in doubling the national average for outcomes related to sleep disorders. He described the experience as one that involved constant testing and re-ordering of steps in the process. Sometimes, the changes were made day by day. I expect that rapid iteration will become the norm for the leaders of the next generation of health care delivery as they hone their craft. 

Human-Centered Design Will Trump Procedure Centric Design

Health systems have begun with modest efforts to weave in the individual into the care process. Even simple secure messaging has been held up as a big milestone. That a technology (email) that has been around for 40 years is held up as a breakthrough, in and of itself, is a statement. I liken the limited efforts to invite the patient into the process to seeing a muddy puddle of water in the Sahara Desert -- it's a welcome improvement but far from the promised land.

The health care organizations that will thrive are recognizing that a tweak to systems (both health IT and business process) that were designed around the patient as a billing vessel will fail. As we've seen in many areas, tweaks to an architecture designed around a different model never succeed in the new paradigm. If they did, AOL would be the leader in social media and Siebel would be the leader in CRM. Before long, you will see the equivalents of Facebook and Salesforce emerge in health care.

Deflationary Economics Will Drive Health Care

Underpinning virtually every business model in health care delivery has been an assumption of ever-increasing health care inflation. It's not hard to predict that deflationary economics will drive health care in the future given the local, state and federal budget situations that are largely driven by health care costs. While one expert warns of the health care bubble, another calls the upcoming period The End of the Third Bubble. It's worth noting that those who thrived after past bubble-bursts were those with lower cost structures and systems that were nimble.

Meanwhile, cost-cutting isn't limited to the government. After all, it is employers that foot most of the health care tab, and they are starting to flex their muscle. For example, IBM has shifted from thinking about health care as an employee benefit to a large cost driver that will affect its profitability. IBM recently decided where to locate 4,000 new hires based on its analysis of where the company received the best value from its health care spending. Consequently, IBM determined that Dubuque, Iowa, was the best location to expand. 

Many health systems operating with razor-thin margins may wonder how they can deal with these changes. Smart health care providers are taking lessons learned from the newspaper industry's colossal mistakes. The few newspaper organizations that have thrived realized that it can still be profitable to operate on a lower cost structure.

With 62% of hospitals operating as not-for-profit, mission-based organizations, they should have a relatively easier time making the transition. They don't have to explain to shareholders why a flat or declining top-line revenue figure can be a good thing (assuming they get costs optimized for the new normal).

Nonetheless, forward-looking organizations have already bucked conventional wisdom, such as thinking that they need to acquire practices to develop an accountable care organization. For example, I have seen both not-for-profit and for-profit health systems recognize that it is more capital efficient to create strong physician networks via open software solutions than to acquire practices and mandate a closed system. 

To sum it up, to support the array of new demands, systems will need to be nimble, affordable and person-centered. These aren't the adjectives typically applied to traditional health IT systems.

In the past, we have observed the military frequently spending money on capital built for the last war, such as aircraft carriers and other slow moving military tools. Over time, the military learned that it was as much or more important to focus on the hearts and minds of those they were trying to work with and that remote intelligence tools have been highly effective at winning battles.

When it comes to managing chronic disease, winning the "hearts and minds" of patients and remotely monitoring health are similar skills not factored into systems developed for the legacy reimbursement model.

Facing all the new requirements in a dynamic environment, health systems should leave some dry powder around in their health IT budgets. It's only logical that an entirely new reimbursement model and set of requirements will create new categories of software.

to share your thoughts on this article.