Assume for a moment, there will be enough stimulus money to cover the cost of your electronic medical record system. Also assume that you receive the funds prior to your investment, so cost is not an issue. Will your EMR project succeed?
The healthcare industry, has been discussing, testing, and building electronic medical records for 20 years. And yet, the latest study by HHS shows that less than 2% of hospitals have a fully functioning EMR and only 8% to 11% have at least one area fully digital. So, when I see surveys stating 75% of hospitals think cost is the biggest obstacle to implementing an EMR, something doesn't ring true. Until hospitals address all of the barriers, they won't be successful.
Why has success been so elusive? There are six stages to implementing an EMR and each has its own pitfalls. Hospitals can improve the success rate if they recognize the risks at each stage and plan accordingly—especially if the stimulus money reduces the cost barrier.
1. Determine readiness to tackle the EMR. This is a basic planning step, but in practice it's often overlooked. So much time is spent justifying the investment and getting it approved in the budget, no one wants to look back and see if the organization is really ready. Basic issues like prerequisite applications and network capacity will come out in the wash, but other issues lurk in the background. The two biggest are inaccurate or incomplete data and a resistant organizational culture. Both will kill an EMR project if they are not spotted and addressed prior to the implementation.
2. Identify the needed components of an EMR. Everyone has a sense for what a fully digital hospital might look like with every record, test, procedure, image, and file online. But reality is not so clear. The EMR is really a poorly defined term. Vendors and analysts have different definitions for what constitutes an EMR. Too often, the buyer and seller have different definitions and therefore different expectations of what the new system will be able to do. In addition, no one can implement a full EMR at once; the big bang doesn't work. So, the organization must define what components of an EMR have the biggest impact on the hospital and prioritize what will be done first.
3. Select a vendor. There is nothing more basic than contacting vendors, arranging demos, and soliciting user feedback. But how often does the decision come down to vague statements like "either one of these systems will meet our basic needs, but this one seems more cumbersome to use?" This is such a common response that it could be a symptom of a problem—not a conclusion. The evaluation team often sits through canned demonstrations by the vendors, has limited time to devote to the process, and feels ill-equipped to evaluate the systems. In the end, the demonstrations are not easily compared one to the other and do not represent how the system will be used by the hospital. So the team draws conclusions in a very subjective manner.
The solutions to the three barriers above are basic pre-project blocking and tackling. The fundamental problem is that most organizations rush through these steps, which plants the seeds of future failure. In my opinion, 25% of EMR projects fail because of one or more of these three issues.
4. Implementation and training. This is, of course, where many projects fail. In some cases, mistakes in the first three steps come to roost and show themselves here. In others, failures in this stage cause the project to break down. One common pitfall is letting the vendor build the project plan. For example, timelines are governed by the vendor's resource constraints and the project steps are very general or written in complex technical terms. As a result, it is difficult to really know how the project is progressing until it is well behind schedule.
The other pitfall occurs at the training stage—especially when the project is behind schedule. Management presses for the go-live date and as a result, training schedules are compromised. Organizations give training less time than is needed and assume once users are trained, they will recall the functions they were taught. In fact, most students go into information overload within a few hours because they are taught aspects of the system they may never use. In addition, the class may be held hostage by tangent discussions and business process questions. These slow the class to a crawl, time slips away, and the instructor rushes through the remaining material to finish on time.
In my opinion, another 25% of EMR projects will fail because of these issues. If the project makes it through this stage without too many delays, the organization is usually ready to celebrate success. The project may be a technical success, but it is not an operational one at this point. There is still a 50% failure rate that occurs after go-live.
5. Manage adoption rates of the EMR. This is one of the toughest and most overlooked aspects of the project. In it, hospitals have to confront the natural inertia of an organization to change. The system has been installed properly and users trained, but we just can't get widespread use of the system. There is a vague passive resistance, as well as, users who think "If I just keep my head down, this too will pass." Poor adoption rates ruin the return on investment, and the longer it goes on, the more it erodes the enthusiasm of the early adopters.
Before organizations' celebrate success, they should implement a measurement system that will track usage and publish the results. Build an incentive program that rewards the behavior you want and be sure all members of senior management are visibly behind the effort. This is the single biggest pitfall to a successful implementation—35% of EMR projects will fail because the organization does not effectively manage adoption.
6. Ongoing maintenance and support. Organizational knowledge about information systems has a half life, like a radioactive material. It is most potent within the first few months of implementation and then slowly deteriorates if not renewed. This occurs for two reasons. First, employee turnover slowly reduces knowledge of the systems, especially if the departing employee is expected to train his or her replacement. The second driver involves new releases of the software. Sometimes new releases are not installed so new features are not available. Other times, the organization does not utilize the new features and continues to use the software as it was originally implemented.
Adequate staffing, funding, and new hire training can alleviate the risk of this barrier. In my opinion, the final 15% of EMR projects will fail because of these issues.
So, while the experts weigh in on whether $19 billion is "enough of a stimulus" to get the healthcare industry to adopt electronic medical records, an equally important question is whether the stimulus money is enough. Clearly, it is not. Unless healthcare organizations address some of these basic issues that have plagued them for the past 20 years, EMR adoption will still be spotty at best.
Bill Mitchell is an IT professional with 15 years experience in healthcare. He can be reached via email at mitchellnvl@mac.com.
For information on how you can contribute to HealthLeaders Media online, please read our Editorial Guidelines.