Putting a Price Tag on Failure | Jim Feldbaum | Healthcare Blogs Skip to content Skip to navigation

Putting a Price Tag on Failure

June 2, 2009
| Reprints

Implementing an EMR, whether in the hospital or in physician’s offices, is tricky business. Although there are predictable pitfalls we seem to always find new ways (or combination of ways) to stall or fail altogether. I was wondering if we have been including the cost of failure in our calculations of the predicted cost of an average implementation. If one considers the concept of “meaningful use” what percentage of our “successful” implementations fall short of “meaningful success” and what is the cost of meeting all new requirements? My gut tells me we are being low-balled, but I have to confess that statistics was never my forte. Any thoughts?



I have always found that the best way to assure a "successful" implementation was not hard work, but to lower expectations. Yes, I am being a bit glib, but as a consultant I often find that stated goals are often impractical and naively underestimate the limitations of timetables, software, hardware, budget and manpower. Now, we all have a label for a concept that we have always understood in our gut: "meaningful use." It is about time that we target success not by the number of modules implemented or percentage of people using the system, but by whether we have facilitated the delivery of better care in a "usable state." I agree Joe, that is the price tag we should be measuring.

I think you're correct. Our cost accounting is inadequate.

When we look at the millions of dollars per hospital spent on EMR in the last decade, we have to look at what this has achieved. That's why I started my recent post, Phasers on Stun, with that accounting (distribution of the stages of achievement of the hospitals in the US.)

I'm not sure that putting the price tag on failure is that right framework. It will probably be much more useful to tally up what it costs to be successful. Part of that is hardware, software, networks, implementation services, training, etc. Those "prices" can be relatively straight-forward to collect and add up.

The prices and costs to make EMRs work, in the hospital and physician offices, include impacts on practices. Changing mindsets, training time, learning time, evolution of practices, and the negative stuff, like software that almost works but not quite ... either because the software is way too slow, or it freezes up, loses work or is otherwise incompetent for the task.

In the fascinating book "The Inmates Are Running the Asylum: Why High Tech Products Drive Us Crazy and How to Restore the Sanity " by Alan Cooper, Cooper describes ten attributes of "Polite Software."

We're far from having Polite Software. The simplest subset of this is "usability." We really want and need software that looks one or more steps ahead, factoring in patient's co-morbidity and workflow. As we all know, we're all still using code that's functioning at the task-level.

(There's a catch-22 here - software has to be well enough designed to allow collection and maintainable of problem list to support that. We all know the ICD-9 issues, the costs and other realities of coding, and current time pressures of delivering care.)

What would be the Price Tag of Success against Meaningful Use for specific care settings? I'm more interested in that number, than the Failure number. For planning purposes, we probably need both numbers.  Otherwise, we'll continue to confuse the price of getting it done wrong with the price of getting it done right.  (The former, getting it done wrong, is often referred to as "getting it done.")  Therein lies the issue of the price tag on failure!


Although you say you are not a mathematician, you raise the valid point that when calculating expected return on any investment, you need to account for the likelihood of achieving that return—or "risk."

Everyone who purchases an EMR enters into the process with the expectation that they will be successful. Unfortunately, history has shown that this is not the case the chances of failure are relatively high. Depending on the physician's practice specialty, traditional EMRs have experienced a 50%-90% failure rate when you include implementation failures and failures to meet the expectations of the purchasing physicians.

If the motivation for adoption is the government's EHR incentives, you need to add the risk of failing to successfully meet (the yet to be determined) "meaningful use" requirements.

The issue of risk is too often overlooked in evaluating EMRs. Because it is such an important factor, it will be the subject of my "Straight Talk" next week. I hope you will read it and add your thoughts.

Evan Steele
CEO, SRSsoft