By Christopher Guadagnino, Ph.D

Published December 2000

• React to this article in the Discussion Forum.

Nancy J. Wilson, M.D., MPH, is vice-president, Clinical Affairs, VHA Inc.

PND: Can you describe VHA’s Medical Error Reduction Initiative with community hospitals in western PA.?

NJW: VHA is a nationwide alliance of more than 2000 community-owned health care organizations and their affiliated physicians comprising 27 percent of the nation’s community hospitals. Our medication error reduction program is a clinical improvement initiative based on fundamental principles of quality improvement that include rapid cycle methodology, collaborative learning among participants and expert faculty, and sharing promising practices and lessons learned. The focus of the initiative is on reducing the errors associated with high risk medication and medication administration issues. Examples include reducing those adverse drugs events that occur with chemotherapy, insulin, heparin and warfarin, patient-controlled analgesia and other infusions pumps, concentrated electrolytes, as well as automated dispensing cabinets. These are high volume and/or high risk situations. The medications involved can cause a great deal of harm to patients when mistakes occur. A miscalculation on a chemotherapy dose can be lethal.

This program combines events where hospitals come together to identify their goals for improvement and to hear from experts about their practices on the scene nationally. Afterward, they go back into their environment where VHA local coaches support their ability to identify specific barriers to that particular hospital and come up with strategies that are customized for improvement. The local coaches are VHA employees who live in the geographic region that supports those hospitals. The team tends to be both physicians and nurses with expertise in quality improvement. VHA has 18 regional offices with ongoing staff that work with the hospitals within their region. The regional offices’ boards of directors are the members who are in that area. VHA’s Pennsylvania regional office is in Pittsburgh.

PND: Which Pennsylvania hospitals are participating in the program?

NJW: Altoona Hospital, Butler Health System, Community Medical Center in Scranton, Good Samaritan Hospital in Lebanon, Hamot Health Foundation in Erie, Heritage Valley Health System in Beaver, Latrobe Area Hospital, Pinnacle Health System in Harrisburg, Sharon Regional Health System, St. Clair Hospital in Pittsburgh, Uniontown Health Resources in Uniontown, and Wyoming Health Care System in Wilkes-Barre.

PND: How does the medical error reduction initiative apply models from industry?

NJW: The rapid cycle methodology has been used in health care for approximately 15 years and is based on action research principles that were utilized in industry. It emphasizes the need to identify, in short cycles, a plan of action, a testing of that action with a few patients and then building on the success of that to the next step of the process. One example is patients not being able to get their heparin monitored in a timely fashion. The team would look at the steps in the process of a patient being on heparin and getting blood tests at appropriate intervals, then try to improve each step itself, checking to see if implementing a change in the process works or not with a few patients and then moving on to the next step in the process. The kinds of interventions we use are things like checklists, reminders, standard operating procedures, standard protocols. Sometimes it’s a simple intervention such as postings on the walls of the hospital with clinical reminders before writing prescriptions, or handing out laminated checklists and reminders to all the physicians. The approach varies, but it’s introducing the kind of tools and techniques that have been used in other complex industries and applying those to health care to make the system safer.

We know from human factors work that human beings are very prone to making calculation errors. The way you would apply this to chemotherapy, for example, would be to see how many people are involved in transcribing and calculating a dose of the chemotherapy before it’s even drawn up into the medication infusion system. We know that the most common number of steps involved in medication administration for any inpatient administration system is 25 and the error rate is 22 percent. So, what we do is work with rapid cycle techniques to reduce the number of steps, reduce the number of hand-offs, and use tools even as simple as a calculator in order to calculate dosing. Physician order entry systems are a very good way of eliminating a lot of medication errors, but these are expensive systems and require strategic planning. What we’re helping our members do is to change the processes they can do today to make improvements, while buying them some time to be able to plan for more extensive solutions that involve technology.

An example of one intervention would be to have the physician type the order onto a screen that the pharmacist then reads, and not have handwriting that’s transcribed first by a nurse, then by a unit clerk, then given to the pharmacist. Another example would be to make sure the person who is calculating the dose of the medicine is not the same person who is checking their work. We know from human factors that, if you check your own work, you will continue to make the same mistake. It’s a normal, but faulty, cognitive process. We apply redundancies at critical points in the process to make it more difficult to do something that is irreversible. However, we eliminate steps that simply compound the possibility of error.

PND: Who performs error reduction interventions?

NJW: The teams that come to the workshops to learn what other hospitals and national experts are saying tend to include a pharmacist, a nurse, a physician champion and frequently there’s someone from the operational side who is able to eliminate any kind of organizational barriers. They then go back and champion this work with other team members who are working on the initiative. It tends to be folks who actually understand the process the best and therefore are able to come up with the best solutions for improving the process. We have the workshops approximately every four months, and the regional VHA coaches provide ongoing support between the workshops. They call and visit the hospitals and coach the team directly. All participation in the program is voluntary. The hospitals are delighted to have the help.

In addition to workshops and local coaching, we also have a web-based interactive medium that we call Clinical Knowledge Management for ongoing communication among the hospitals, faculty and regional coaches throughout the collaborative. Every member in VHA has access to an on-line Intranet, of which the Clinical Knowledge Management system is part. It’s a safe place for hospitals to learn from other hospitals that are involved in clinical improvement work. The system has a private workspace where the hospitals that have agreed to share data and information can communicate with each other in any way they see fit. There’s a slightly more public layer that is available for anyone who’s involved in any of our clinical improvement initiatives, and there we try to post the best practices, highlight headlines that are in the literature, and success stories. The value over some other web-based resources is the additional ability to ask a question of all these folks and get an answer.

PND: How was the medical error reduction program developed?

NJW: The program was developed in the fall of 1999 as part of a larger VHA clinical improvement initiative called Clinical Advantage. Medication error reduction was thought to be the first phase in a larger safety initiative that VHA kicked off last June with a national symposium that VHA co-sponsored with its business competitors. The focus on medication errors first was due to the volume of medications that are actually used in this country and the belief that this was an opportunity for rapid improvement. We’ve known for many years, for example, that concentrated potassium chloride should be taken off of the units in hospitals so that the harried professional doesn’t inadvertently grab the concentrated potassium chloride when they really needed a more dilute vial. It’s a simple measure to just remove it and have the pharmacy create the concentration. But we still have hospitals throughout this country that have concentrated potassium chloride on the ward.

VHA partnered with the Institute for Safe Medication Practice and enlisted the president of that institute, Michael Cohen, as our national chair. In addition to our safety initiative, we have hospitals participating in improving the care of stroke patients, acute myocardial infarction, congestive heart failure and breast cancer. We are also planning two new initiatives that will be offered in first quarter of 2001: one on pain management and the other on spreading and sustaining the improvements. Our institutional participants register for each disease-specific initiative separately, but for our new topic of "spread and sustain," we’re encouraging members who have already achieved a certain amount of success to move into that initiative, which encompasses all topics and all diseases.

PND: Is this program offered to all of your member hospitals?

NJW: We are offering it to all our hospitals, although our capacity to deliver has been one issue for us in this past year. We’ve had such overwhelming interest in this program, we’ve found we’ve only been able to deliver the program in one region at a time. We fully expect over the next couple of years that we will be engaging more and more of our hospitals. Right now we have over 550 engagements out of 2000 members nationwide.

PND: Why are there no southeastern Pa. hospitals participating in the medical error reduction initiative?

NJW: We’ve found that some of our large health care systems like to collaborate with others that are of similar size. Sometimes we have folks who are collaborating with health care systems across the country. Some folks want to look at the big picture, look at the organization-wide needs and then drill down to something more specific. In addition, our initiative work has just begun and we’re spreading it as quickly as we can.

PND: Would a non-VHA member be able to use this program?

NJW: To-date, our capacity has not been able to extend it to non-VHA hospitals, except in the area of sharing practices at our annual national congress on patient safety. The first one was in June 2000 and we will be doing that again in October 2001. That’s an opportunity for sharing across all health systems.

PND: How do you measure the program’s effectiveness?

NJW: There’s a three-fold approach. We use pre/post measurement looking at the organizational vulnerabilities and strengths which are scored in a quantitative fashion and reviewed at the end of the program to look for changes. We also use satisfaction of the participants and we work with our members to identify the financial impact of doing these programs. The tool we’re using on an annual basis in medication error reduction is the Institute for Safe Medication Practice tool.

PND: One medical error reduction issue that hospitals have found controversial is the prospect of having mandatory reporting of errors. To what extent will hospitals’ error experiences be made public through this initiative?

NJW: The hospitals share what they feel comfortable sharing, and success stories are usually something that they want to share. The issue of discoverability is an interesting and tricky one. Some folks would say that all reporting is voluntary because in today’s health care culture we only report what we cannot hide. That is a direct result of the climate that tends to focus on individual blame. The information that folks share with each other is voluntary, but we’ve found that the participants recognize that most information is discoverable anyway, that the risk is very low, and that the yield for learning how to improve is very high.