Why Implement an Electronic Medical Record?

Tim Breaux, for HealthLeaders News, July 10, 2007

Fundamentally, any application system implementation should meet some sort of business objective. An Electronic Medical Record is no different. The standard framework for evaluation of a technical implementation effort typically focuses on one or more of three objectives: can we make a business process or objective faster, better or cheaper?

In healthcare, we often use different words to describe business objectives. We might discuss clinical outcomes ("better") or administrative efficiency ("faster"), but the framework still applies. Now for the key question: What is the business objective of most EMR implementations? Is it reasonable to suggest that "getting rid of all of the paper" is a business value that will improve the cost/quality profile of care? I believe the answer is a definitive "no".

In any other industry, we look at automation as a tool to improve a business process. When we evaluate the process of interest, we look first to (a) standardize the process in a manual environment, (b) simplify the process as much as possible, and finally to (c) automate the process to maximize efficiency. Any technology implementer knows that automation of a badly designed process results in a badly designed system. Now apply this process to an EMR implementation. What business processes are we automating? Do we really expect that improvements in these processes will result in a cost justification for an EMR, or more globally, an improvement in the cost/quality profile of care for the country? If the entire country implemented EMRs overnight, would we see significant overall improvement in healthcare productivity, efficiency, quality and customer service?

Let's take a step back for a moment. Let's imagine we bought cars the way we currently buy healthcare. If we did, we would go to an automobile consultant who would order a set of parts for a car. The consultant would commit to purchase that set of parts. The set of parts might end up as a functional vehicle, but there would be no guarantee that it would actually run. We, the ultimate consumer, would not know which parts were ordered, whether the parts were good or bad, and we would not know the price of any individual part. We would not know the price of the automobile consultant either. We might know what it would cost to see him/her once, but we would not know how many times we would have to see him/her to get a completed vehicle. The auto consultant would then hire a mechanic to assemble the vehicle. We would have no idea of the total cost of the car, but it would not matter much because the consultant, the mechanic and the parts suppliers all would bill our auto insurance directly. The cost of the car after the first several thousand dollars would be covered by our auto insurance. If the car failed, we would go back to the auto consultant and see if we need more parts, more service or both.

In this imaginary scenario,
1) An $80,000 luxury car would probably cost between $200,000 and $300,000
2) Everyone would own a luxury car, because they are "better" than economy cars and the driver did not pay for the vehicle
3) Automobile insurance would cost well over $5,000 per month
4) The cars would have an incredibly high failure rate, and
5) Warranties would be unavailable for any vehicle.

Does this sound familiar? Is the key issue that the auto consultant has inadequate information about the car requirements from the consumer? Would it be a really big help to get all of his information about car parts from an automated system and "get rid of the paper"? Does giving the ultimate consumer "more information" about car part or consultant costs help? Does automating this cumbersome process significantly change the output? Of course not.

Now, back to real world. Consumers, of course, do not hire a consultant to custom build them a car. They don't because they can buy a complete car for a known price, and they can compare prices between models and make an informed decision on the purchase. Why then, don't some providers take the lead and deliver a similar solution in healthcare? Simple: No one pays them to do it.

In our current reimbursement model, there are few incentives to deliver quality (although some payers or employers do have small incentives). Physicians generally do deliver quality, but it is mostly because they are professionally motivated to do so. If we really wanted to cost justify an EMR in a physician office, we would need to focus on using the EMR to increase the number of visits per physician per day, or to capture valid charges that may be missed in existing procedures. Neither of these values is particularly consistent with improving the cost/quality profile of care.

If we wanted to really shake up the quality/cost paradigm, would we implement an EMR? No. We would offer fixed priced products for things that consumers understand. We would not offer the price for an office visit: we would offer the price for a treatment. Imagine if a primary care physician offered a fixed price per year for care of a diabetic. Imagine if an orthopedic surgeon (in concert with an internist, a diagnostic imaging center, a hospital and a rehab therapist) offered a fixed price total-hip or total-knee surgery? Imagine if they offered outcomes guarantees on these services. Do you think that consumers would respond? If it was actually faster, better or cheaper, they certainly would. Providers could tune the processes (including the customer service processes) to deliver better quality for less money. Just like we have been doing with cars for the last 50 years.

Further, once the processes for care delivery were pretty well tuned, we would have something reasonable to automate. We actually could implement an EMR, and tune business process objectives consistent with reduced cost and increased quality. It would be cost effective because providers would get paid for an improved cost/quality profile, just like consumers would prefer.

Is all of this unrealistic? Maybe. But any set of providers in the United States could assemble a fixed price arrangement for a procedure. If it was actually cheaper (as higher quality care usually is) they could offer the fixed price to any insurer. Insurers usually have some limitations about how much they can motivate consumers, but any insurer could certainly decrease (or even waive entirely) the deductible for a clinical service package that is of known quality and guaranteed. How could providers offer guarantees? They could guarantee that any additional work after that initial package in completed is free. How could they afford that? Almost any reinsurer working in the United States could indemnify them. The cost of reinsurance would be trivial if the quantity of rework is low. Hmmm. Imagine that. Providers would be motivated to be successful.

Care delivery systems could give fixed rate prices for major procedures (e.g., total hip replacements, total knee replacements, coronary artery bypasses, coronary stents, etc) including all elements, not just the acute care components. The price could include the hospital, the surgeon (including pre- and post- discharge office visits), the anesthesiologist, the diagnostic imaging, the laboratory, the rehabilitative therapy and the pharmaceuticals. The care delivery system could broadcast the quality metrics for their procedures.

Non-surgical treatments (e.g., wellness, diabetes, congestive failure, blood pressure management) could be given annual pricing from primary care physicians and medical specialists. Consumers could then make informed decisions based on "products" they understand and on prices that are known.

Then we would have something to do with our EMRs.

---

Tim Breaux is CEO of Portland, OR-based InterCase and can be reached at tim.breaux@intercase.org.