By Cindy Atoji

August 5, 2008 | Nailing down ROI for electronic health records (EHRs) must be a deliberately thought-out process, says Michael H. Zaroukian, chief medical officer at Michigan State University (MSU), one of the early adopters of this technology. Zaroukian, who directed the EHR implementation at the MSU Internal Medicine Clinic, a community-based ambulatory care facility, says the 12-physician practice reaped major savings from a decline in paper chart pulls, staff costs, and transcription expenses.

“Simply putting an EHR system into place doesn’t give you return on investment (ROI), says Zaroukian. “EMRs that yield the most positive ROIs establish measurable financial outcomes at the start of the project.” Zaroukian spoke to Digital HealthCare & Productivity about the importance of tracking an implementation’s progress and avoiding the ROI-sapping effects of EMR customization.

DHP: You’ve said that “there is no free lunch for ROI.” What do you mean by this?

Zaroukian: If you take actions to improve quality, like putting in an EHR, you might expect cost savings, but it’s not automatic; it requires a cogent plan and striving to change processes in a way that decreases waste and increases efficiencies. The key is starting with the discipline to map your current processes in a paper world. Organizations, including ours, want to skip that—it’s hard and takes time, but if you don’t map out the current processes, you can’t get a handle on parts that are flawed or unnecessary. Be disciplined about process mapping and have a methodology for redesigning them.

If you focus on the few things that are measurable and then monitor the outcomes, your ability to achieve return on investment won’t be accidental, but deliberate. The good news is that if you look at improving quality of care, the natural consequence is that you will likely also see financial gains from that as well. For our ROI analysis we mainly looked at three things: decreasing chart pulls, eliminating transcription costs, and changing staff-to-physician ratio in a favorable manner.

DHP: What’s an example of a cost-benefit analysis that you performed?

Zaroukian: We needed a credible business case for this project, and when we looked at practice redesign and the various components of savings and cost, one of the things we looked at was paper chart pulls. We had a mathematically precise way of looking at this, because we have centralized medical records and allocated costs of $1.92 every time we wanted to pull a paper chart. We could predict on a basis of 3,500 chart pulls a month how much we could save by eliminating those.

After implementation, we monitored monthly paper chart pulls, which is one of the best indicators of adoption, and simply counting on a regular basis the number of paper chart pulls that are happening. By the time you’ve gotten down to 99 percent decrease in paper chart pulls, you must be [putting] everything in the EMR. We saw about a 90 percent decrease in first 8 or 9 months, which for faculty group practice is pretty fast. Over the course of the next few months, we were at 99 percent decrease in paper chart pulls, which for us was about an $87,000 savings per year.

We also monitored dictation cost and volume, as well as staff and labor costs, letting natural attrition take its place. The biggest cost for any clinic is staff costs. By improving staff-physician ratio, and not automatically replacing someone when they leave but evaluating whether a change is necessary, is an important discipline and a big difference between having a large return on investment versus not much of one at all.

DHP: One EHR expert says organizations are always trying to customize their EHR because they consider themselves “unique.” What do you do?

Zaroukian: We tried to spend time learning to be competent EMR users, more or less out of box, the way it comes, selecting from the options that are there, rather than focusing on customization or a lot of changes.

The phrase I tend to use is “customization is the enemy of implementation;” so you can customize or you can implement but you’ll have trouble doing both. Nothing is perfect, and you can personalize an EHR, which is different from customizing it. Personalization is the ability to say, “this is the way I like to document or work.” There are tools in most EHR systems that allow you to do that—narrative templates allow low-intensity personalization as well as quick text, or macros. Voice recognition is another way to personalize without having transcription costs.

DHP: Any other pitfalls of customization?

Zaroukian: The most expert person in leading customization should be a person highly experienced with the alternative. Because the notion that, “I know this form will work for me even though I’ve never used it” and “I want to design this new form, even though I’ve never used the new form,” in our experience has proven many times not to work.

Customizations come from experienced users, from a careful look at current process mapping, a deliberate storyboard approach to how the new customization would work [combined] with analysis of cost and benefit. We’ve spend many hundred of thousand of dollars on customizations at Michigan State University—and the same is true for many practices around the nation—for forms that have been never been used. That’s the definition of a complete waste of time.