The Most Expensive Tool in Medicine: "THE CLICK"

In the article "Price transparency in electronic health records not linked to changes in physician ordering: The PRICE trial", a cost analysis was performed on giving feedback to the clinician on their test ordering. The study essentially found “that  electronic health record display of cost for laboratory studies was not linked to a change in clinician ordering habits.” This has been extrapolated to both radiological tests and medications.

In the past the “most expensive tool in medicine” was the pen but this has changed to the “click”. Today, one can order multiple tests, perform them hourly or daily, and prescribe endless medications, all in packages, and with just a click.

The goal and fantasy of the CPOE (Computerized Physician Order Entry) has been that it would lead to cost controls and critical thinking about diagnostic and treatment plans. That imagined goal would be protocol driven, evidence-based, and lead to financially sound actions.

But the devil is in the details. In healthcare IT, that devilish detail is in the implementation of the software design. As it stands, the burden of the various CPOE platforms which include pharmacy, lab, and radiology are so cumbersome that “Work-Arounds” are often taken to get the busy work out of the way. Unfortunately, the devil is winning.

In the past our blogs have suggested a need for an evidence-based national interchangeable CPOE platform that everyone learns and understands. The industrial world knows this by the term “standardization.“  But in the space of technological innovation, standardization is elusive. HL7 has made a valiant effort to standardize the use of electronic medical systems. But standardization has not happened. Why? Think of the credit card readers in supermarkets and stores. Do any work the same as the one down the street? How many times do even non-medical people scratch their heads and make mistakes when simply sliding a card instead of inserting the chip? Not to mention the frustration of the cashier toward the seemingly stupid customer who just cannot get it together.

Medicine should do better. Lives are at stake. Every provider, hospital, and vendor could adjust the content pertaining to the work environment and specialty.

CPOE with enhanced “Artificial Intelligence” could include:

1. Treatment protocols (especially for commonly encountered and high risk clinical presentations)
2. Pharmacy preferences (with one, instead of 18 clicks per Rx)
3. Work-up protocols (that cat an appropriate safety net for some presentations)
4. Financial data
5. Cost effectiveness data    
6. Elimination of boiler-plate, routine, tests
7. Rational for why a certain test is needed unless protocol driven or obvious
8. Feedback to the provider on cost per provider on 1. Diagnosis 2. Treatment compared to all users.
9. Easy or automatic access to up-to-date recommendations for the specific problem.

If all this happened the same way at hospital B as it does at hospital A, think of the overall benefit, efficiency, decreased error, and cost savings that would accrue toward healthcare overall. Familiarity, coupled with true interoperability, would lead to less clicks, less consternation, less mistakes, and even the possibility of more patient contact time.

Currently, practitioners are so overwhelmed with data input that it discernibly takes away from time needed to fully analyze important clinical decisions. As a result, costs go way up. Think though, how the tendency to prioritize getting paid, even if it is perhaps an unconscious tendency, could be replaced by the more important tendency to get “it right.” Costs could go significantly down if there were enough cognitive space to think thoroughly before ordering routine tests that in the end have minimal clinical benefit.

If there were a standardized national performance guide, a decrease in liability issues should naturally occur along with cost savings. Why? Non-essential tests are commonly ordered on the initial evaluation. Sometimes. And when they are not ordered, a malpractice suit can occur because physicians are held to the retrospective analysis of what might have been done when bad-luck cases arise, as they inevitably do.

(See Peter Rosen vs ACEP dispute).

Would it not be nice if the doc could say, “Well, I was just following the standard?”  Case closed. But, on its own, such standardization won’t happen.  Any more than the card readers in Walmart will work identically to the ones at Target.

Yet we in medicine have an obligation to make it happen. As an important first start, fix the CPOE and help providers intuitively and naturally change their behavior for everyone’s good. There are many articles on the cost of clicks in the business literature, sometimes they even work, and sometimes the lessons can even be transferred to medicine.