MODULE 1: CASE ANALYSIS – CPOE at EMORY
HEALTHCARE
ID: 59255426

The differences between the types of physicians
Compared with the full-time and directly employed physicians in other settings at Emory, most physicians at EUHM were not directly employed. The positions in EUHM were more like “part-time contractors”. In addition to work for EUHM, these physicians also took jobs from other area hospitals. Many of them were already familiar with CPOE from other places but the CPOE systems they had used were quite different with the one
Emory implemented. Those physicians were very likely to react positively at the beginning because CPOE was not a new word to them, but would suffer from switching multiple CPOE system later when they realize the systems were totally not the same although they were all called the CPOE. However, Emory was not able to provide the customized education based on each individual. So these physicians had to adjust to that without the help. Indeed, I believe all the CPOE systems would be similar in general and might differ just about 20%. It seemed not a big problem to these physicians because the fact that they at least better than people don’t know CPOE at all before. However, the paper emphasized the challenge because of this situation. Why? This made me think about the find faults game. In the game, it would be harder and harder to find the faults when two pictures are getting more and more similar. It would be further harder for the pictures involving a large quantity of elements. Using multiple CPOE is like playing the hard level of find faults game. The physicians might expect same results happened for

entering an order based on the previous knowledge on the other CPOE system, but things might get totally wrong. The systems were so similar and so hard to find out where the differences were and where the problems were. They might fail on previous knowledge.
However, for the physicians in other settings at Emory, most of them had never used
CPOE before it was implemented in Emory. On the contrary, there were might many resistances at the beginning but would be mitigated later. People in general do not like jumping out the “confortable room” to adopt the new things, especially the thing is from different domain knowledge (COPE requires computer skills while many physicians do not have that). However, as the full-time salaried employees, they had no choice but follow the big plan. They might be forced or not forced to go though the designed training for the CPOE system. They learned the system from the same start point, and understood it in this CPOE’ way. Thus, for them there was just one picture and no other
“picture” to confuse them.

Another key difference between types of physicians at EUHM and the other settings is the independence that not just indicated independent from other physicians but also indicated independent from other employees. Although the paper didn’t straightly explain this, but I believe it influenced in many aspects. In other settings at Emory, physicians worked closely with all other employees at hospital. Before the CPOE, the physicians, nurses, unit clerks and all other employees set in their own positions at the same time and same location. It was very easy for them to communicate by face-to-face or to have an internal call to get the orders done. Everything works perfectly fine without CPOE if everyone would do the right thing. Many of them might feel CPOE was not necessary but

made everyone confused about what they should do. Physicians felt they would be overload, since all entries had to be made by them; nurse felt that an important peace of their job was taken; and there was no need for unit clerk at all in this circumstance. The team was broken that was not they wanted. From this perspective, they did not have the motivation to make the change. However, in the EUHM, the physicians are more independently. They worked at different time and in their own offices. It was not very likely that they could get instant response from others like in the other settings at Emory when they were not in the hospital. Before the CPOE, they might leave notes, send email, or schedule a time to get all things done. This would cause inconveniences and problems when something went wrong at some point during the information transition. The trust between the physicians and other employees would not as reliable as in the other setting at Emory. The physician might feel better if they could do it by themselves in a way that was not bounded by time and location. By leveraging a bunch of different technologies might be helpful, but they were not from the same platform and might have limitations on information exchanges because of the regulation on patient information. It would be wonderful if a single tool could help them solve the issue since no matter what they have to find the alternative ways to do so, while the CPOE system was the solution. Since the physicians at EUHM had the motivation to bring in the CPOE, and also they perhaps had used some substitutes tool for a while, they were more likely to accept the change compared with the physicians in other settings.

How should the approach to implementation differ at EUHM?

In order to implement CPOE successfully at a community hospital like EUHM, we have to know some key facts of community hospital and the physicians there:
•

EUHM did not have fully control on the physicians

•

The physicians were the leaders and decision-makers

•

EUHM was just small part of these physicians’ life

•

The physicians had short time to stay at each day

•

Engaging physicians was hard work

•

CPOE was not high on priority list

•

Wide range of knowledge in basic computer skills and CPOE system

•

The physicians might take their business to elsewhere

In these facts, we can see the overall environment was very different than the other settings at Emory, while considering all these facts were critical to make an implementation plan. Because of the previous implementation experience existing, it would not be to hard for implementation itself, but the challenge would become how to fit the EUHM environment.
The most challenging issue would be the training. To making the training classroom section that gathering the physicians together would not be even possible for open staff environment. Moreover, since the physicians would not stay at EUHM full day at each day and lack of time during community physician’s brief, they would be busy at every minute at hospital each day. Also, it was not effective to give universal training materials

for the physicians who had the wide range of computer skill and experience of other
CPOE.
Instead of putting the physicians into the classroom, a virtual and customized training class would be more efficient. The training materials could be made as short 20 minutes videos that each covered a small topic. Then, the physicians could access the training materials at anywhere, any time and finished it within 20 minutes without frequent interruptions. Even after CPOE was implemented, when they had difficulties to use some function, they could go back and watch the related topic.
The nurses could be the ACEs. Unit nurses could be as the first line of support in coaching. They are right at hand on the patient care unit and they know their physician’s personality and should have a closer working relationship with the physicians than anyone else. So instead of put the physicians to the classroom, it is more reasonable to put the nurses into the classroom and train them to be the “super users” to support the physicians on daily routine.
Physician peers were also important resources in helping physicians master heir electronic tasks. Involving enough physicians in decision-making is a particular challenge in the community hospital. Many physicians at EUHM divided only a part of time at the hospital and most hospital initiatives had small influence on their daily routine. So the champion movement could attract the physicians’ viewpoint in an effective way.
According to Dr. Donald Levick, “The role of the physician champion is to meet with people, paint a picture of where the hospital is going, find where their resistance points are and how best to overcome them.” That means the job of physician champion requires a good structure of environment and more importantly a good leadership to conduct these

works. The paper mentioned that the leadership at EUHM contributed significantly to the success, so it gave a good foundation that physician champion would work well in such leadership. From problematic modules perspective, I believe that it should be turned off. The community physicians in general were less likely to contribute effort on improving the
CPOE community. They practiced more independently and were busy on different things at different hospitals. To them, the CPOE was not on the high priority list. The motivation that they wanted to bring in CPOE was that they wanted to get things done efficiently. So it would be better not use it if it has significant flaws and let other academic or third party to figure out the solution and then bring the working function back. For applying the power plan - ordering sets at EUHM, the debate of standardization versus customization was similar as in the other settings at Emory. The ordering sets would be useful for frequent orders that saved times for typing in detail and repeatedly.
However, formulating the ordering sets would cause problem for individual treatments due to the differences of cases. In EUHM environment, the high-standardized order sets were not as feasible as the other settings, because the physicians and perhaps their treatment were more diverse than the physicians and treatments in the other setting.
Recall the facts that in EUHM the physicians were independent to practice and the decisions were and should be made by themselves. So perhaps a compromise should be made that allows the customization on suggested ordering sets. Whenever the physicians started to make an order, the suggested ordering sets would pop up as keys stroke, and the physicians could create their own sets if it was not the sets they want.

Broader strategies

Preparation of CPOE implementation
The effort of the preparation of CPOE implementation was primarily invested in two critical areas, 1) Sorting out the order management workflow and 2) promoting the physicians’ commitments. However, these things were not easy to do. Before actually implementing the CPOE system at EUHM, we had to recognize that there are many different audiences and constituencies involved. The first thing Dr. Bornstein should do was engaging them, and made them come to share the understanding of why it was necessary and the right time to implement CPOE; what would be changed after CPOE; and who would be involved in what perspectives in this implementation. The goal was to put everyone on the same page and gave an open environment to address doubts, attitudes and ideas. Later, design a working schedule and setup timeline to figure out the issues and have a weekly report on progress until it was ready to implement. As we know, community practice physicians were busy and difficult to reach. So multiple modes of communication should be used to reach individuals and groups of physicians because no single channel was sufficient (email, memos, newsletters, system notifications, meetings, and so on). “Put the plan and the progress everywhere they could possible to see.”
Reinforcing the plan via multiple dimensional appearances and creating the “hot” environment surround the stakeholders at EUHM and make them talk and think about the
CPOE, and eventually get know about it.
Setting expectations

An expectation was very necessary to accomplish such a big project. Dr. Bornstein could set the expectation into relative small steps so that people could achieve a quick accomplishment and get reward on the performance and make quick adjustment if thing went other ways. It would also reinforce the big vision in a forwarding cycle – accomplish the current step, see how far we made and how far is the final goal, then introduce the next step until the next step is accomplished again. Also, a reasonable high expectation drives to a high performance. Given a high expectation, people often would deliver the best solution in a more efficient way. In order to encourage people to do so, champion selection with financial incentive could be a good approach. Creating a championship environment, make celebration on every stage and let champions to share their experience on achieving the success that not only facilitate the single step success but also attract the attentions of the whole hospital.

Building high-performance teams
Physician was the center of the CPOE implementation. Ideally, a practice physician should spend any possible time to do good practices and all other thing should be done by others. If we consider that is the highest performance team, then the physicians should not use CPOE at all, because in this case entering orders is not that directly relevant to the practice. However, in the real world, things are much complicated than that. Ordering entries highly influence the physicians to give the good practices. So it might be better to be done by the physicians in an efficient way. Then, how to support physicians becomes the goal of high-performance team. We repeatedly talked about some key facts that would be challenging to the physicians at EUMH to master on CPOE in short time, but

the nurses could be the ACEs to help the physicians to use CPOE, because they are always at side the physicians and they have the closest working relationship with physicians. Besides, since many physicians would have the need to make the order remotely, as many as possible resources should be accessible by the physicians at anytime and anywhere. To make this possible, the tech team and information desk team should provide their helps on that. So to build such team, we should bring all these people on the same table. Communication would be essential in this team building. Tech team and information team should know what are the common problems of the physicians have, and how to response in a more efficient way when the problems occur. Over the time, summarize the solutions on all problems to provide a question and answer book to the physicians.

Framing from learning
At last but not least, the execution and long-term commitment might be the most important for the “real” success of the CPOE. The problem was how to ensure continuances and improvement over time after CPOE implemented in EUHM. What could be done when for example people started to feel uncomfortable with switching different CPOE systems. The issue might be omitted at the beginning but when it happened it might even cause the ultimate fare – the physicians take business to elsewhere. A good strategy could be built at the early stage to ease the sudden like that.
Frame from learning at the beginning; and form a core implementation team that had experience with CPOE implementation from different aspects; or consult the experts who experienced similar situation before. Then develop a perceivable strategy on top of each

phase. In addition, we could frame the implementation as a learning opportunity for everyone, and give promotion and incentive to the structural change in order to facilitate the overall adoption of CPOE.