Outcomes for Trauma: Is There an End (Result) in Sight?

In the following address I will discuss the role that EAST has had in developing an evidence-based approach to the practice of trauma, why it is important, who has made it important, and what the future of trauma outcomes will likely resemble. This talk is the third in a series, a series that started in the early 1990s, and hopefully will continue into the future. It also reinforces and extends the thoughts that Dr. Russell left us with yesterday in his address regarding the future of quality assessment in surgery. In 1994, in his presidential address to the Seventh Scientific Assembly of EAST, Dr. Michael Rhodes introduced us to the concepts of evidence-based medicine and practice management guidelines.1 Evidence-based medicine is defined as a method of patient care, decision making, and teaching that integrates high-quality research evidence with physiologic reasoning, experience, and patient preference.2 Patient management guidelines are systematically developed statements designed to assist practitioners in making decisions about appropriate healthcare for specific clinical circumstances.3 The five major purposes of practice management guidelines are to assist clinical decision making by patients and practitioners, educate individuals or groups of individuals, assess quality of care, guide the allocation of resources, and reduce the risk for legal liability. In his discussion, Dr. Rhodes stressed that the idea was to base clinical decisions on the best available evidence and understand the power or quality of that evidence. Evidence-based practice involves, at its core, a fundamental acknowledgment that our clinical convictions can be wrong as well as the fact that we invariably underestimate the power of the placebo effect and assume that because most patients we treat in a certain way feel better, the treatment must be effective. Obviously, these concepts were not universally embraced when they were first introduced, and a great deal of work has been put forth to arrive at where we are today with evidence-based practice. In addition to patients and practitioners, it was felt that guidelines could be useful to purchasers of healthcare, legislators, and regulators. Further, it was felt that guidelines must be used to decrease unnecessary practice variation, formulate necessary research proposals, and most importantly, evaluate outcomes of care. Only through such measures can patient-specific guidelines be developed and improvements in care be realized. Dr. Rhodes challenged the organization to move forward in developing such guidelines and EAST responded. Subsequent to his address, a jointly sponsored conference by EAST and HRSA (Health Resources and Services Administration) was held to define the methodology for practice management guideline development, leading to the creation of the first subcommittees addressing specific trauma related topics for guideline creation. This ultimately led to the 1998 Journal of Trauma article in which practice management guidelines for screening for blunt cardiac injury, cervical spine clearance, prophylaxis for venous thromboembolic disease, and antibiotic prophylaxis for penetrating abdominal trauma were presented.4 At around the same time, EAST established its Website and began placing the guidelines on the site thereby making them available to the membership and others interested in trauma care. Today, there are 27 guidelines, as well as a primer describing the evidence-based methodology for guideline development available for viewing. To date, there have been 283,056 downloads of the various guidelines, and on average, there are 775 hits per day on the guideline section of the EAST Website. Importantly, EAST, recognizing the evolution of acute care surgery, has charged the Guidelines Committee to begin to develop guidelines on emergency and acute care surgery subjects, and they have responded by preparing several relevant guidelines that will be presented later at this meeting during the plenary session on Friday. In 1999, Dr. Timothy Fabian addressed the 12th Scientific Assembly, reminding us that EAST had made an early and profound impact on evidence-based medicine through the management guidelines project.5 In a Star Wars analogy, he noted that we, like the crew of the Millennium Falcon, had begun a journey not quite sure whether our ship would make it to wherever we were going, but knowing that we could adapt and fine-tune the process as we made our way. He also introduced us to Ernest Codman and his end result idea,6 which simply stated is that every hospital should follow every patient it treats long enough to determine whether the treatment rendered was successful and if not, why. Codman kept an end result card on each patient and developed a classification of errors for adverse outcomes. In this classification scheme, he noted that all results of surgical treatment that lack perfection may be explained by either errors in technique, judgment, care, or diagnosis, whereas other adverse outcomes could be explained by patient causes and still others by the calamities of surgery over which we have no control. He further thought that these results should be acknowledged by the surgeon and made available to the public. The concept of end result treatment outcome and accountability was also extended to hospitals, and Codman argued that hospitals were responsible for the care given by their staff and thus, they should carefully note the results of each surgeon and all of this should be made public. It sounds eerily familiar to some of the issues that are confronting us today; clearly the surgeons and hospitals of 1914 were not ready for this concept, and Ernest Codman was forced into resignation and exclusion.6 Dr. Fabian thought that multidisciplinary development of trauma practice management guidelines was key and that the process should be physician-directed, nonthreatening, developed from evidence-based outcomes, and integrated into performance improvement programs. He also argued that collaboration between the various trauma organizations would be key to future advances in evidence-based trauma care and he challenged EAST to continue to attempt to resolve the issues of guideline consistency, continuity of the process, monitoring of utilization and validity, and coordination of clinical trials. He noted that the development, classification, and distribution of trauma and surgical critical care related evidence-based guidelines had progressed rapidly.5 However, the implementation of and outcome studies on the effect of guidelines were lagging. Staying with Dr. Fabian’s Star Wars analogy and in keeping with the trilogy concept, I would suggest that our evidence-based ideas have moved from the dark side and that the Force is now with us. The question at this point is where do we go and to answer that, we must answer several separate questions. The first is how do we assess quality of care? To answer this question, we must realize that there are three dimensions of health care that can be used to assess quality: structure, process, and outcome.7Structure refers to the attributes of how healthcare systems are organized to deliver care, whereas process refers to what we do to and for our patients, and outcomes are the results of that care. The standard outcome parameters of mortality and morbidity have been extended to include length of stay, cost, quality of life, patient satisfaction, and compliance with guidelines. Outcome tools used to measure these parameters have included audits, surveys, severity scoring, cost accounting, and research studies but until very recently, outcome research has been primarily comprised of descriptive or retrospective analyses of data that have been collected either retrospectively or prospectively.3 Using the “Algebra of Effectiveness” described by Lisa Iezzoni (Fig. 1), we see there are three major predictors of outcome of healthcare: a set of patient factors that constitute the patient’s prehospital, preoperative, or preinjury risk; the effectiveness of care, which encompasses the quality of the providing facility’s structure and processes of care as well as the competence of the providers and random events.8 Thus, theoretically, if one can account for patient risk factors by proper risk-adjustment models and for random variation by statistical models, one can then equate outcome to effectiveness of care and thus, make it possible to use outcome as a measure of quality of care.Fig. 1.: Algebra of effectiveness of major predictors of outcome of health care. 8As we discuss these quality measures, it is important to realize that there are multiple organizations involved in this process, i.e., Institute of Medicine (IOM), Agency for Healthcare Research and Quality, the Leapfrog Group, the Joint Commission (formerly the Joint Commission on Accreditation of Healthcare Organizations), American College of Surgeons-Committee on Trauma (ACS-COT), and National Institute of General Medical Sciences. Each of them are trying to assess quality, often in different ways and highlighting the need to collaborate and educate each other to avoid duplication of efforts and improve the efficiencies of each organization as we move forward. The next question to address is what are the problems that we are facing, as hopefully outcomes evaluations will help us find answers and resolve some of the significant issues. With the IOM publication of “To Err is Human”9 and “Crossing the Quality Chasm”,10 attention was focused on a number of preventable provider errors and iatrogenic injuries that were occurring in hospitals across the country. Conversations among healthcare policy makers emphasized the implementation of quality improvement processes and outcome measures across the healthcare system. The anesthesia and surgery communities, in part through the efforts of the Anesthesia Patient Safety Foundation, the ACS, and the Veterans Health Administration (VHA), put forth efforts to develop new structures and processes to enhance the safety of the surgical patient. The IOM report set as a goal for these national efforts a 50% reduction in error-related deaths during a period of 5 years. In addition, the report stressed the need for an evidence-based approach to patient care and evaluation, but more than 7 years later, there is little concrete evidence to suggest that any of these efforts have resulted in an overall reduction in error-related deaths. One of the important reasons for the failure to properly evaluate the impact and efficacy of these initiatives is the lack of a proper metric for assessment of patient safety. Clearly, another significant problem in the United States is the cost of healthcare, and compounding this is the fact that the increased costs of care have caused differences in how care is provided. This has resulted in differences in access of care, quality of care, and availability of adequate preventative efforts.11 Further, it is felt that these are directly linked to the method of payment. In short, our healthcare plan is dependent on a payment plan rather than a care plan.12 What about trauma? Do we have a problem? I would argue yes, and these problems are nicely articulated in the recently released IOM report on the future of emergency care.13 I would encourage all of you to read this document carefully and begin to educate your community representatives about the problems that trauma care faces. The problems in trauma care are similar to the ones I’ve mentioned: access, quality, prevention, and, of course, cost.13 What are the potential solutions and can they be applied to trauma? What I would like to do is look at some of things that have been performed in surgery and then see how those could be applied to trauma. Prompted by a 1987 congressional mandate that required the VHA to report outcomes of major surgery as per national average and risk-adjusted for severity of patient illness, the National Surgical Quality Improvement Program, or NSQIP, was born.14 The development of NSQIP has provided surgeons with new tools to assess and improve the quality of surgical care. NSQIP is the first national validated state-of-the-art system for the comparative measurement and continuous improvement of the quality of major surgery. The comparative metric employed is risk-adjusted outcome with focus on risk-adjusted 30-day morbidity and mortality. The program was initiated in the Veterans Administration Hospital in 1994, and because of its success in reducing postoperative morbidity and mortality, it became available to all eligible hospitals through the ACS in 2004. The distinguishing feature of NSQIP is that it applies to outcome-based quality measurement, the same scientific rigor that is normally applied to clinical trials in fundamental research. A trained and dedicated clinical nurse reviewer at each medical center prospectively collects prospectively preoperative, intraoperative, and 30-day outcome variables on patients undergoing major surgery. The data collection methodology is standardized and nurse competency and inter-rater reliability are periodically checked and validated. Data are subsequently transmitted to a national trauma data coordination center, where it is analyzed on an annual basis. These analyses identify the independent predictors of various 30-day outcomes for all operations in a hospital and then calculate a beta coefficient for each predicted variable. This coefficient is used to calculate the expected 30-day outcome of patients and an observed to expected ratio can then be calculated and used as a risk-adjusted metric for that outcome and appropriate benchmarking can then be performed (Fig. 2).Fig. 2.: General surgery O:E ratios for morbidity.Hence, to use outcome as a measure of effectiveness, quality, or competence, two basic ingredients are needed, a reliable clinical database of patient characteristics and outcomes and valid analytical risk-adjustment models that would account for random events. These two factors are at the core of NSQIP, and I think they can be applied to trauma15 (Fig. 3).Fig. 3.: Key components necessary to determine outcome.In the initial 10 years of NSQIP, the participating hospitals noted a 45% reduction in morbidity for all major operations and a 27% reduction in mortality.16 These results led to the expansion of the program into the private sector; today approximately 150 hospitals nationwide are enrolled in the program.17 In its current format, however, NSQIP is not easily transposable to trauma because of several limitations. Thirty-day follow-up is a barrier due to trauma patient noncompliance with scheduled follow-up visits and the variety of providers involved in the care of any one trauma patient. Additional full-time equivalent (FTE) support will be required to collect additional customized outcome data not currently tracked by our trauma registries which would add cost. There is also data collected by NSQIP that is irrelevant to trauma patients, and NSQIP clearly is better at assessing systems and hospitals, as opposed to surgeons and procedures. Despite these limitations, the core principles can be applied to trauma. Another potential solution argues that we need to change our focus from a system that cost and care to one that quality care, the for performance or as is to by our legislators, for Clearly, to follow the established by private health and to into a system where payment is on providing and The question as to who should on the effectiveness and of care, and I would argue that it has to be the trauma surgeons that the care. must the necessary to the future trauma system. Trauma surgeons must evidence-based processes and evidence-based outcome measures by which we will be we do and they will most likely not understand the of the trauma patient and the trauma surgeon as well as we there is a to which we can use to make such The for and Services are with and hospitals on a of quality and safety outcomes and in the new the will be to the established by the Surgical Improvement The goal of is to use evidence-based processes to reduce surgical by during the next years. are the focus as the available data would suggest that not our cost of care and length of stay, but they also the mortality risk in is on cardiac the and treatment of reduction of postoperative through the appropriate use of of and in the of venous and by appropriate use of measures, because of the fact that major operations are by and in and of and and treatment of which has been to the hospital mortality by as as to This focus on process is at and it that trauma could from such a when you look at focus and have problems and we have potential but we these solutions to trauma, we must the question as to whether or not trauma is a of the and the cost it would to do do trauma trauma and trauma systems make a The answer is a The by at the of Health and the of of Medicine that care at a trauma center by the risk of for patients with the treatment at This the of work that we must do that as trauma surgeons we can continue to our effectiveness and efficiencies of care. it is how should we go about that and at the same improve our that the number one to performance measurement systems is the lack of evidence-based process and outcome measures and from the work that I have the has the of Trauma and to develop a system for trauma. The work put forth by this is now to as the Trauma Quality Improvement the by Dr. in his trauma outcome research at the of at the key factors involved in NSQIP, data for and of data and of inter-rater reliability audits, and the development of risk-adjusted for you can see that trauma is the to the NSQIP The Improvement and Patient Safety and of the are currently developing for appropriate and the our our trauma and with the National Trauma Data also have a to inter-rater reliability audits, as the appropriate risk-adjusted analyses are performed they can be into our current The overall of are to a customized for collection of and this with the to one database for trauma patients, risk-adjusted analyses and to identify best with and, in improve the quality of care to the at outcomes, we have to realize that there are problems and one of these is that there are a number of different trauma and surgical organizations that are at outcomes and them The outcomes that are on what is This problem can be by providing for those things that we think are important to improve our trauma care. To that the following have been as outcomes that we can begin to collect and mortality with and of and of patients who arrive to our on length of stay, and is currently performed to define or better define how we how we deal with of care and the of a do not and of we must look at the more those with multiple and The next in of a of outcomes with to the work on and outcomes, of data on appropriate data collection and and the development of risk-adjusted for assessment and to an O:E ratio for trauma mortality by a risk-adjusted by one of the and of the and are forth efforts on of this O:E ratio for trauma mortality by risk-adjusted of the problems with our current system is the lack of adequate risk which benchmarking and quality improvement an I would you to look at presented at the of the regarding mortality at through it can be that by of life, increased the risk of when with patients the of In addition, a of cardiac or and also increased the risk of in these patients, factors that our current systems simply do not in a way. It was that and have a significant impact on mortality for injuries and appropriate risk and compliance with are of on of on of on other important in mortality is the of and I would like to with you some of the results from a that we in to the Trauma Foundation, and Joint Trauma of the Institute of Surgical for their help in us to this and were by and by a of trauma surgeons in the Each was by at 10 trauma surgeons and a was made between the and of The used to make the was the established and Improvement At first it as there was between the and of mortality however, a look that there was a significant in the classification when with the classification 50% of the was there between the of and the of of Another was that our current systems and not to be in and this was most when at preventable and preventable of the from this were that provided different mortality than the and and not to be in making a The question that as we move forward is whether or not classification is the best way to look at mortality. with the creation of better risk-adjusted measures, the focus will change from classification to processes of care by outcomes and by from or or think that we all that there are problems in trauma and we must address are responsible for effectiveness and processes of care, as well as our outcomes of care. There are models available and we should begin to these to trauma. I would argue that we are by outcomes and our reliable clinical the and to develop the appropriate risk-adjusted that we can use outcome as our measure of quality and to improve our processes and quality that trauma care in this will continue to be I would like to by that it has been a to be part of this organization and an to as your I have more than I have and for that, I To EAST and to you I you very