By Christopher Guadagnino, Ph.D.

Published September 2001

• React to this article in the Discussion Forum.

Gregg S. Meyer, M.D., is director of the Center for Quality Improvement and Patient Safety at the Agency for Health Care, Research and Quality (AHRQ), which commissioned an analysis of evidence-based, best safety practices to disseminate to health care provider organizations.

PND: Why did AHRQ sponsor a patient safety study?

GSM: We felt that it was important to provide decision-makers—including hospital directors, large purchasers and providers in groups and as individuals—the information that currently exists about what they can do to improve patient safety. We went to one of our evidence-based practice centers, one of 12 university-based sites that we sponsor around the country, to look systematically at the medical literature and other literature about what evidence exists concerning a wide array of potential patient safety interventions. We went beyond the usual medical literature and asked the evidence-based practice center very specifically to look at what we can learn from other industries such as aviation, manufacturing and nuclear energy about how they have improved safety and try to apply that to the health care context. Not surprisingly, the majority of the findings come from the clinical end because clinical interventions are much more often subject to greater risks, studies and evaluations. Some of the chapters, however, do focus on organizational interventions, things such as the role of leadership, the importance of setting a culture wherein the staff can talk openly about safety issues, learn from errors and learn about hazardous situations so that they can avoid them in the future. Other organizational issues deal with such things as staffing levels and worker fatigue.

PND: What sort of recommendations does the report make?

GSM: The report’s patient safety practices and targets are divided into eight sections: adverse drug events; infection control; surgery, anesthesia and perioperative medicine; safety practices for hospitalized or institutionalized elders; general clinical topics; organization, structure and culture; systems issues and human factors; and role of the patient. My recommendation to an outside reader is to go to the chart in the executive summary of the report and look at the list of interventions there. The web address where they could get the report is www.ahrq.gov. We focused in this study on 73 interventions that were largely clinical and six that were largely organizational. After looking at that chart and sizing up how the evidence was graded, I would then recommend for those who are interested in a particular area to go back and selectively review the chapter on that issue. For example, a reader who was interested in reducing infections may focus on evaluations and practices such as using maximum barrier precautions when inserting central venous catheters. If they’re thinking seriously about implementing that practice, they then should go to the full text in that chapter.

PND: Is it possible to sketch the study’s primary recommendations?

GSM: I don’t think that’s something that we necessarily could do or in fact should do. We have put the evidence out there and let others judge it through their lens as to what they see are the most appropriate places to start. We recognize that decisions about which practices to implement and how to prioritize them will be different among different stakeholders. For example, a small rural hospital may choose to focus on those measures that will deal with, perhaps a problem that they have with infection control, whereas a large urban academic health center may focus more on the issues related to information technology or to the care of patients with high risk procedures. We recognize that priorities for implementation need to be set with an eye toward local resources, constraints and contexts. So, there is no single list of priorities.

What you will note in the report is that we found that all of the 73 interventions showed some promise for improving patient safety. In some cases the evidence was much stronger than in others. For example, in the case of giving beta-blockers perioperatively to patients to avoid myocardial infarction, there was some very strong evidence. In others, that evidence was less strong or, in fact, there may have been some potential downside risks. An example there is giving patients in intensive care units prophylactic antibiotics to prevent hospital-acquired infections. On the one hand that should reduce hospital-acquired infections; on the other hand it raises the specter of increasing anti-microbial resistance. The report also points to areas in which we need to focus research that would allow us to take an intervention that looks promising and develop a critical mass of evidence that it works.

PND: How does this report differ from the one released two years ago by the Institute of Medicine?

GSM: The Institute of Medicine report basically gave a very broad outline of the problem and provided what we think was an important message that generated what has now become a national will to improve patient safety. What this new report does is provide information on the way to improve patient safety, the intervention that can be used by providers and by organizations that can help mitigate the problem that the Institute of Medicine’s report made so clear to all of us. So, the two reports are in many senses complementary.

PND: Why were eleven practices singled out by the study team as best candidates for more widespread implementation?

GSM: The report does include what the University of California at San Francisco-Stanford University team thought would be eleven best practices that would have the greatest impact and the greatest strength of evidence. We don’t agree or disagree with that list. We recognize that others could look at the evidence and develop their own list of priorities based on their own constraints and resources. We don’t endorse any one of those lists. Instead, we think we made a step forward by putting the evidence out there.

PND: How commonly are those eleven recommendations being put into practice currently?

GSM: It’s variable, but by and large there is room for improvement with all of them, and that was in fact one of the criteria that was applied to looking at any of the interventions that are included in the report. There are some things that are very important to improve patient safety, yet are pretty much universally done. For example, we recognize clearly that doing sponge counts in the operating room is important in order to prevent any gauzes being left behind in patients. Yet, that is essentially a universal practice right now, so it’s not a situation where implementing that practice is going to do a lot to improve patient safety. Across all of the interventions that were evaluated in the report, the study team did feel that there was some room for improvement.

PND: Is the report based on research that has already been reported?

GSM: It is a synthesis of research, although it includes not just the medical literature but other literature as well.

PND: Much of which may be well known, but not implemented?

GSM: Exactly.

PND: In that case, what reason is there to believe that it will be more widely implemented now?

GSM: The Institute of Medicine report and some of the activities which followed, I think, have done a very good job of raising the issue of patient safety to one where decision-makers at a number of levels are paying attention to the issues. It’s created a will to do something. What this report has done is pull together the information in a single source in a way that a decision-maker can go to it and say, "Here’s an array of options that I have for implementation," and then go into the chapters and see how strong the evidence is that that in fact will work in a setting like the one in which they happen to practice. The report also points out that the field of patient safety research is a relatively young one and that there clearly is a need for us to do more work in terms of developing new interventions.

PND: Are you doing anything specifically to foster implementation?

GSM: We are, this year, doing $50 million worth of patient safety research, some of which focuses on demonstration projects. Applications for current funding have been solicited through a series of requests for application, and you can find information on them on our website. Those opportunities have closed at this point and we are now in the process of evaluating those applications. But we do hope to be able to make a continued, substantial commitment to patient safety research. We also are very interested in funding research that examines what the barriers to implementation are, and how you can get best practices out there into practice. This is the science of studying the spread of innovation. We hope in the future to be able to make some investments through challenge grants or some other mechanisms to try to get these or other proven patient safety practices implemented. We are not yet able to make that commitment but that is a direction in which we hope to be able to move in the future.

PND: Do you think it would be appropriate for these recommendations to be required in any form?

GSM: I think that is a question that needs to be answered by an array of stakeholders. For example, we know that large purchasers are looking at these recommendations. We know that the Centers for Medicare and Medicaid Services are looking at them. It’s not our place to judge whether or not they should be required. We’re not a regulatory agency. What we’ve done is put the information out there so that they can make wise decisions about what they will or will not want to require in the future. The National Quality Forum, which represents a wide array of private sector stakeholders, is looking at this report and other information about patient safety interventions and is trying to make decisions about which should be shared with the public. So, in the future there may be public disclosure about whether or not a hospital or other health care institution has put this or that intervention into place.

PND: The JCAHO has added a new category to their accreditation, requiring health care organizations to put into place some patient safety improvement interventions. Do you think it would be appropriate for JCAHO to require some of the recommendations of this report for their accreditation?

GSM: I think that’s for the JCAHO to decide. This report provides a useful tool for hospitals and health care organizations because it provides some evidence as to what interventions could be put into place, which would help a hospital meet the JCAHO requirements.