Innovations, Technical Solutions, and Patient Safety: Pulse Oximetry, Health Care Checklists, and the International Classification for Patient Safety
Editor’s note: The following paper was submitted to JCInsight by Joint Commission International Editorial Advisory Board member Jorge César Martínez, M.D., chairman, Department of Pediatrics, Del Salvador University School of Medicine, Buenos Aires, Argentina. Dr. Martinez participated in the following presentation and discussion—with the consent of the other participating presenters—at The International Society for Quality in Health Care’s (ISQua’s) Paris 2010—Quality Outcomes: Achieving Patient Improvement on 13 October 2010. Dr. Martinez’s summary of both the presentations and subsequent discussion are below. Please direct any comments or queries to Dr. Martinez (jormar@intramed.net) or the World Health Organization’s Patient Safety Programme at (patientsafety@who.int).
Background
Any intervention to improve patient safety needs to offer relevant, user-friendly, and locally acceptable practical approaches. In this seminar, innovations and technical solutions to address patient safety problems at various levels were introduced and discussed.
Presentation 1: Gerald Dziekan, M.D.
Gerald Dziekan, M.D. is Programme Manager, World Health Organization (WHO), Patient Safety, Health Care Checklists, spoke about a number of projects, including the WHO Pulse Oximetry Project, the Safe Surgery Checklist, and the Safe Childbirth Checklist. Dr. Dziekan began his presentation by reminding the audience of the words of Lucien Leape, who said “Human beings make mistakes because the systems, tasks and process they work in are poorly designed.”(1) In describing learning from high-reliability organizations such as the airline industry, Dr. Dziekan spoke of the use of checklists, which he described as visual or oral aids that help overcome short-term memory deficits. The work of Peter Pronovost in the implementation of a central line insertion checklist in Michigan in the United States demonstrated that it was possible to decrease from 2.7/1000 catheter days to zero in 18 months.(2) This concept and checklist are in the process of being rolled out through WHO and others to other countries, including Peru, with great preliminary uptake. Dr. Dziekan emphasized that the implementation of a checklist alone is not enough; but when embedded in a quality improvement process, a checklist is an excellent tool to assist improvement. He spoke of checklists in general as a tool for overcoming the discrepancy between knowledge and practice and posed the question: Can a checklist be the link between knowledge and practice at the bedside? A checklist is a quality improvement tool to ensure consistency of practice to reduce provider variation. It standardizes care processes, aids memory of clinicians, and is a tool to improve communication and teamwork. However, Dr. Dziekan said a checklist is neither merely a piece of paper nor is it a panacea to all problems. Similarly, he said checklists are not regulatory tools nor should they be considered as set in stone, as they need to be adapted to local context to best aid implementation. In discussing the guideline development process, Dr. Dziekan said the WHO guidelines were developed on the best available evidence and through months of debate to reach consensus. They were also supported by an implementation guide, as a checklist alone cannot achieve improvement. Currently, most hospitals do most of the right things to most patients most of the time, Dr. Dziekan said. But he asked: "Can checklists actually move this to all hospitals doing the right thing for all patients all of the time?"
Safe Surgery Checklist
The WHO Safe Surgery Checklist has been tested in eight pilot hospitals around the world and was found to reduce the rate of post-operative complications and death by more than one-third.(3) A standard safety attitude questionnaire was administered in each of these safe surgery pilot sites before and after implementation of the checklist. The results of these surveys showed that 80% of staff found the checklist easy to use and 84% said that it improved communication within the team; 79% believed that the checklist prevented errors. Interestingly, the surgeons and other clinical staff surveyed commented that if they were to have surgery, 93% would want the checklist used to ensure their procedure was safe.(4) A safety attitude survey was used because it is known that safety attitude is associated with team function and that team performance and outcomes are inexorably linked.
Pulse Oximetry Project
Advanced Incident Management Study (AIMS) data show that 82% of 1,260 incidents studied of patients undergoing general anesthesia would have been detected by pulse oximetry.(5) Dr. Dziekan said WHO has looked at the global pulse oximetry gap and found that the lack of pulse oximetry in many hospitals poses an increased risk to patients. For example, the number of operating rooms (ORs) per patient ranges from 25 per 100,000 persons in eastern Europe to only 1 per 100,000 persons in western sub-Saharan African nations.6 The number of ORs lacking pulse oximetry technology globally has been estimated as greater than 70,000, resulting in 31.5 million operations performed annually without pulse oximetry.(6)
In response to this safety issue, WHO has established pulse oximetry as a standard in operative care and is partnering with the World Federation of Societies of Anaesthesiologists (WFSA) and others in improving the safety of anesthesia care in operating rooms in low- and middle-income countries, where pulse oximetry technology is lacking. WHO Patient Safety is testing the effect on patient outcome by providing a bundle consisting of the Surgical Safety Checklist, Pulse Oximeters, and anaesthesia training to providers in a number of pilot hospitals globally via the WHO Pulse Oximetry Project. The pilot testing is ongoing to demonstrate a reduction in low oxygenation time and reduced morbidity and mortality. Starting from here, WFSA, the Harvard School of Public Health and many others have developed a charity, Lifebox, to address the pulse oximeter gap by making high-quality, low-cost pulse oximeters available to low and middle income countries.
Other Checklists
Another checklist described by Dr. Dziekan was the H1N1 Patient Care Checklist, which has been tested in emergency situations during the 2009 global pandemic. No full-scale pilot study has been conducted, but the process shows that if common design criteria were applied, it would be possible to produce an effective checklist in a short period of time.
Another checklist Dr. Dziekan discussed was the Safe Childbirth Checklist. This is a simple tool that front-line care givers can use to ensure that essential care practices are performed during any childbirth. WHO is currently doing a single-center trial of this checklist in India, Dr. Dziekan said. Three months of pre-intervention data collection, followed by 2 weeks of post-checklist intervention data collection have produced complete data sets from over 400 childbirth cases. Preliminary results suggest that the introduction of the Safe Childbirth Checklist resulted in a marked increase in adherence to best practices, measured by selected quality of care process indicators. These are preliminary unpublished findings, Dr. Dziekan mentioned, and the ongoing study needs to verify these preliminary findings.
WHO has also developed a Trauma Care Checklist, which is currently being pilot tested in 13 hospitals around the world.
In summary, Dr. Dziekan noted that checklist interventions have the potential to improve patient outcomes by:
• Standardizing care processes
• Ensuring consistency of care
• Structuring communication, and
• Improving teamwork
Presentation 2: William Runciman, Ph.D.
Professor William Runciman is Professor, Patient Safety & Healthcare Human Factors, School of Psychology, Social Work & Social Policy at the University of South Australia; Professorial Research Fellow—Patient Safety, at the University of Adelaide (Australia); and Content Expert for the WHO International Classification for Patient Safety (ICPS)
Professor Runciman is part of the WHO ICPS development network, and he is renowned for the development of the Advanced Incident Management Study (AIMS; see above) in Australia. Although there are large bodies of sophisticated work in regard to incident monitoring across the world, the data cannot be compared due to the different terminologies and definitions applied, making it impossible to conduct a meta analysis of the data they contain. Professor Runciman presented the thesis that there is a need for common understanding. He noted that one of the strengths of the WHO is that their work can be disseminated all over the world that health care is a global business, and that patient safety is everyone’s business.
Professor Runciman noted that countries such as Brazil have had some spectacular public health successes, particularly in the area of HIV/AIDs, providing encouragement for "cross fertilization" across different countries.
Professor Runciman also said that the percentage of adults receiving recommended care in United States hospitals was assessed by the RAND organization as between 30% to 87%— (average 56% of adults and 46% of children) indicating that less-than-recommended care is being delivered with remarkable consistency across the country.(7) However, Professor Runciman noted that the rate of change in improvement of the proportion of patients who receive recommended care is slow. The question then is why this is so, Professor Runciman asked, and what can we do about it?
Professor Runciman presented the conceptual framework for the international classification for patient safety (ICPS) which includes the definitions of 48 key terms, which was developed by the WHO ICPS Conceptual Framework Drafting Group and was published in February 2009.(8) In the current phase of the ICPS development, the intent is to populate this framework with over 20,000 concepts and translate them into a variety of languages. This is the essential first step in allowing basic patient safety to become global, said Professor Runciman.
Professor Runciman said the ICPS will become a framework for the following:(8)
• Deconstructing incidents
• Eliciting and capturing information
• Storage analysis and strategy development
• Compendium of evidence-based actions
Professor Runciman said the ICPS conceptual framework is now being populated and undergoing peer review. A Web-based collaborative authoring tool is being developed in line with the International Statistical Classification of Diseases and Related Health Problems (ICD)—the Initial ICD-11 Collaborative Authoring Tool (iCAT)—and ontology development is underway to pave the way for developments such as automated reading and translation to different languages. Professor Runciman emphasized that we cannot make progress on delivering appropriate care unless caregivers speak a common language.
In regard to the question of how we engage the health care community to change their practices and use these tools, Professor Runciman spoke about the failure of self-regulation of health care and the need for some “small r” regulation. To explain, he referred to the “pyramid of regulation” described by John Braithwaite and colleagues, which is a bottom-up model in which the following occurs:
• Level 1 refers to voluntary self-regulation
• Level 2 refers to professionally led self-regulation
• Level 3 refers to meta-regulation
• Level 4 refers to command and control(9)
Professor Runciman said there are currently no realistic prospects of sanctions in any system for people who consistently violate protocols or standards, and there is no way to regulate in areas when there are no standards. He also postulated the following:
• Protocols, guidelines, and standards for common medical conditions need to be introduced.
• There is a need for tools that
— are endorsed by peers;
— with attention paid to clinician workflow;
— with a standard either implicit or explicit on these; and
— which documents the process so that it is easy to audit with a database.
Professor Runciman asserted that this practice of establishing standards and evaluating compliance is part of professional life and that there should be a standard for communication between clinicians, recording information, and benchmarking activities. He concluded by stating that the basis for these activities should be an international classification.
Panel Discussion
A panel discussion followed with the following panelists:
• Carolyn Hoffman, Executive Director, Alberta Health Services, Canada
• Philippe Michel, M.D., Ph.D., Director, Regional Center for Quality and Safety, Hôpital Xavier Arnozan, Bordeaux University Hospital, France
• Jorge César Martínez, M.D., Chairman, Department of Pediatrics, Del Salvador University School of Medicine, Buenos Aires, Argentina
• William Runciman, Ph.D., Professorial Research Fellow—Patient Safety, University of Adelaide, Australia
• Gerald Dziekan, M.D. Programme Manager, World Health Organization (WHO), Patient Safety Health Care Checklists
An audience member asked the panel about culture change in the implementation of the central line checklist and to comment on what were the other contributors to change in behavior.
In response, Dr. Dziekan spoke about the WHO Safe Surgery Checklist pilot as an example of behavior change not being only about applying the checklist but including such a tool in a package supporting and sustaining change, including also staff training. He said that this would lead over time to gradual changes in policy and care practices to include adoption by all health care workers, emphasizing that interventions need to be applied as a package, not individually.
A concern was raised from the floor about interinvestigator reliability of the ICPS.
Professor Runciman responded by saying that at present the conceptual framework needs populating. While reliability will be increased by imposing strict rules on data entry, users will get a more complete picture of an incident by looking at different versions of the incident as reported by different people. Analysis of incident reports need to be carefully made as there are no numerators and no denominators—the data is more useful to provide an understanding of what went wrong rather than counting events. In this case, interrater reliability is less important than the richness of data the system collects. Dr. Martinez added that communication is important, as caregivers are trying to communicate across nations and improve understanding about incidents and patient safety.
The panel was asked a series of questions about culture and implementation—Why are the clinical professions not systematized? What have been the inhibiting factors in cultures that have led to this state? What in the culture will allow implementation of checklists now when we are more puzzled about what works?
Philippe Michel said and the panel agreed that we are all working to drive change and that change is not linked to the tools we use. The change must make sense to the professionals, because if they don’t understand why they are being asked to do things, they simply won’t do them. Further, they said one of the barriers to checklists is that they are by some interpreted as an attempt to reduce professional autonomy..
An audience member asked another series of questions of the panel: How do we regulate? Do we want clinicians to self-regulate and give ourselves the tools to implement? Or do we go to the next level to regulation by the professional societies? Or do we want to implement at a higher level, make regulations mandatory, and include the requirement for inspections and reporting?
Carolyn Hoffman spoke of her experience in the implementation of the Central Line Checklist. Using the checklist was mandated in the Intensive Care Unit (ICU) at her hospital and additional resources were provided, so the intervention was implemented with little resistance, which allowed her to implement the tool with little resistance. The result has been the ICU at her hospital has gone 15 months without a central line infection.
The panel agreed that the important components for successful implementation of a checklist include:
• the application of appropriate resources;
• physician leadership;
• data collection tools; and
• feedback to the clinical team of the outcomes of implementation.
The discussion went on how hospitals are now trying to implement the Safe Surgery Checklist but without the same investment they applied to the Central Line Checklist. Hospitals are now asking for a checklist to implement a checklist. Clinicians are saying “if you want us to implement this, you need to give us the power to do it!” Concern was also raised from the audience that perhaps we are confusing health care workers by offering them too many checklists and too many options. The question was asked, “How can you standardize a standardizing tool?”
The discussion moved on to what the key factors in producing a checklist are. The panel concluded that it is clear the caregivers must go through a rigorous process of debate to select items to go on the checklist, to plan how to implement the checklist, and how the checklist fits with care processes. The panel offered that a tool that addresses both local practices and cultural aspects of how to “do” care is important. Then, the panel concluded, the field can move away from a position where health care workers don’t understand why they need to use a checklist to a place where health care workers don’t feel comfortable if they don’t use the checklist. An analogy with seat belt use was raised by Drs. Dziekan and Michel—whereas seat belts were once considered a nuisance or uncomfortable, most drivers no longer feel comfortable driving without a seat belt. The panel concluded that the goal must be for clinicians to feel the same about the use of checklists as drivers do about seat belts, with the ultimate aim to ensure that the checklist becomes part of everyday behavior.
Dr. Martínez said that those who are working at the front line of health care are frequently overwhelmed, exhausted, stressed, and sleepless—they are asking for help. A checklist, Dr. Martinez said, is an excellent instrument to make health providers feel more secure and supported. Of course, the proper implementation needs leadership, commitment and cultural changes, Dr. Martinez said, and went on to say that the field is “moving in that direction, slowly, but we are moving. In my experience, medical students love checklists.”
The audience asked if there are arguments against the use of checklists. As with anything, the panelists warned, rigid adherence without thought of circumstance can be as dangerous as not using a checklist. Dr. Runciman made the point that guidelines are always qualified as “modified according to local needs.” However, the ritualistic defense mechanisms of clinicians still need to be overcome as exemplified by statements such as “my patients are different.” Moreover, panelists said that some caregivers think that guidelines are optional, as compared with a tool which is something that has been designed for a purpose, can be used to document what you are doing, can be audited, is based on a standard which is something you should follow, and is therefore “mandatory.” The need to develop guidelines for patients and not for diseases was emphasized in the discussion as well as the need for them to be easily translated.
Dr. Dziekan said—and the panelists agreed—that caregivers need to be selective about what is included on a checklist, asking these questions:
• What are the “killer” items—the items if omitted, that would lead to harm?
• What is the evidence to select this rather than another item?
The panelists said the inclusion of pulse oximetry on the Safe Surgery Checklist was one such “killer” item and is clearly nonnegotiable.
Another aspect of the Safe Surgery Checklist that has created some controversy has been the requirement to introduce one’s self to other staff in the operating theatre. The panel justified this as very important to effective communication and it was noted that it is very important to address some of the “soft” issues which contribute to improvements in care. The current Safe Surgery Checklist includes 19 items, but could easily have included many more, Dr. Dziekan said. The checklist requires a highly selective presurgical process to make it a tool which is more likely to be accepted around the world. The ability to be adapted to local settings is important for clinician buy-in, according to Dr, Dziekan.
The panel asked the audience if they thought that if patients knew that health care were so unsafe, they not come to a hospital or other health care organizations any more. The analogy was made that if planes were crashing at the rate of errors in hospitals, nobody would fly. So then why aren’t patients and their families more aggressive when things go wrong?, the panelists asked. Clearly, patients have to be more involved, to join the care “team,” to make things better, and one of the ways is by being engaged in the use of checklists.
Professor Hughes made the point that surgeons have their own internal protocol for the procedure that they follow each time they perform a procedure, so why then are they so resistant to an externally derived protocol such as a checklist? The answer, according to Professor Runciman, lies in the fact that caregivers are dealing with highly motivated, highly intelligent people who have been given amazing power and autonomy when they graduate from medical school, and so they resist checklists as the thin end of the wedge of losing their autonomy in clinical practice. Professor Runciman went on to say that caregivers are instinctively resistant to their “patch” (domains) being invaded by people who haven’t been through the same process of initiation that they have.
Workshop facilitated by Professor Clifford Hughes, Chief Executive Officer, the Clinical Excellence Commission, New South Wales, Australia
Special acknowledgment to Workshop scribe Dr. Annette Pantle, MBBS (Syd), Dip Obs RACOG, MPH, FRACMA, Director Clinical Practice Improvement, the Clinical Excellence Commission, New South Wales, Australia.
References
1. Leape, L.L.: Written testimony to the Advisory Commission on Consumer Protection and Quality in the Health Care Industry, November 19, 1997.
2. Pronovost P., et al.: An intervention to decrease catheter-related bloodstream infections in the ICU. N Engl J Med 355:2725-32, Dec. 2006.
3. Haynes A et al. A Surgical Safety Checklist to Reduce Morbidity and Mortality in a Global Population. N Engl J Med 360:491-499, Jan. 2009.
4. Haynes A.B. et al. Changes in safety attitude and relationship to decreased postoperative morbidity and mortality following implementation of a checklist-based surgical safety intervention. Qual Saf Health Care 20(1):102-107, Jan. 2011.
5. Webb R.K. et al.: The Australian Incident Monitoring Study: An analysis of 2000 incident reports. Anaesth Intensive Care 21(5):520-8, Oct. 1993.
6. Funk L.M. et al.: Global operating theatre distribution and pulse oximetry supply: An estimation from reported data. Lancet 376(9746):1055-61, Sep. 2010.
7. McGlenn, E. A. et al.: The quality of health care delivered to adults in the United States, N Engl J Med, 348: 2635-45, 2003.
8. Runciman W.: Towards an International Classification for Patient Safety: Key concepts and terms. Int J Qual Health Care 21(1): 18–26, Feb. 2009.
9. Braithwaite J., Healy J., Dwan K.: The Governance of Health Safety and Quality. Canberra: Commonwealth of Australia, 2005.