Quality Improvement initiatives are not new. In 1942, W. Edward Deming was already lecturing on methods of statistical quality control for the benefit of our war effort. In post-World War II Japan, he taught those principles as management techniques to industrial leaders. Similar concepts were applied to medicine as part of the 1988 National Demonstration Project for Quality Improvement on Health Care (NDP). That same year, Arnold S. Relman, MD, (NEJM, 11-3-88, 319, No. 18), predicted the coming of a Third Revolution in Medical Care, an Era of Assessment and Accountabil ity. That era is already upon us. We all need a better information base in order to knowl edgeably act rather than passively react within this new environment.
Our AANS Membership should be aware of the possibility that their practice behavior resulting from the application of various types of monitoring methods is being assessed. They should also become better informed as to how that evaluation occurs and for what purposes. It has led, as a natural consequence, to both qualitative (patient satisfaction) and quantitative (medical or clinical result) outcomes research. In one way or another, it is going on all around us.
Changing Physician Behavior
The large medical organization with which my group is affiliated, Intermountain Health Care, Inc. (IHC), has, for some time, been employing Quality Improvement (QI) methodology. In 1987, they corroborated the work of John E. Wennberg, MD, ( Science, 1973:142, 1102-8) in surgical variation analysis as it applied to TURP operations and have been performing similar studies in different clinical areas ever since. Demonstrated changes in physician behavior have convinced them that the considerable workforce, time and money they have invested in these studies has resulted in significant clinical quality improvement.
A recent IHC initiative, Patient Perception of Quality Monitoring System (PPQMS), studied, by telephone follow-up, all patients discharged from three IHC hospitals between September and December 1995. To any survey evaluation response of fair or poor, follow-up questions by theinterviewer focused carefully on problem areasincluding physician services.
Why does physician behavior change? Suppose that you are in a departmental meeting and a slide is projected onto the screen entitled: Caring and Concern of PhysiciansSurgical Inpatients, Fall 1995. The slide compares you with your peers on the basis of detailed encounters between your patients and well-trained, experienced interviewers. You note, having been informed only of your individual bar-graph number, that about 50 percent of your patients rated you as having shown very good or excellent caring and concern for them. One of your colleaguesidentity unknown to youhas a bar-graph result signifying that more than 90 percent of his or her patients rated the care and concern rendered to them as very good or excellent. The patients whose responses are being compared were in the same hospitals, treated during the same 90 days, and underwent the same types of operations.
To IHC’s credit, the data that has been assembled and the results of the comparisons have not been used for punitive purposes. Rather, they intend to change behaviorand have found a rather effective way to do that. The PPQMS study, therefore, exemplifies another effective application of QI and also the increasingly evident link between assessment and accountability.
Quality Clinical Outcomes
Patient encounters with hospitals and physicians continue to provide large amounts of data obtained by increasingly sophisticated means. It is generally divided into three categories: patient satisfaction, cost comparisons, and clinical or medical outcomes. There are obvious applications presently feasible for the first two informa tional categories. Clinical outcomes, however, when subjected to rigorous scientific scrutiny, are difficult to accurately delineate. After several years of intense study and the expenditure of staggering amounts of money, the precise definition of what constitutes a quality medical outcome remains elusive.
Scientifically convincing outcomes data are of great potential value to payers. Physiatrists, orthopedic surgeons, and neurological surgeons, for instance, sometimes differ in their approach to the diagnosis and treatment of cervical and lumbar radiculopathic syndromes. Since a given Health Plan Physician Board is likely to increasingly employ the reasonable criteria of value = quality of outcome divided by cost , in deciding about referral decisions, who will end up providing an accurate definition for the numerator on the right side of that equation and to whose advantage will that definition derive? The determination will be madeoptimally with our collaboration and the input of our expertise, but no doubt it will be made by someone.
Coordinated Effort Needed
In concert with our sister organization, the Congress of Neurological Surgeons, we have assembled a highly competent Quality Assessment committee. The potential scope of their agenda, however, in both the breadth and depth of the information base they confront, is daunting. Unlike IHC, the AANS and CNS have neither the staff workforce nor financial resources necessary to employ the optimal methodology required to evaluate very large numbers of patients. A coordinated effort, mobilizing expertise from the various Joint Sections, will be necessary if we are to have any hope of success. Inter -organizational and inter-specialty cooperation will also be essential. Duplication of the enormous amount of work already done by others would constitute a luxury that none of the involved organizations can afford.
Our membership represents only 4,700 out of more than one-half million health care providers in the United States. Though our numbers are small, the role we play in our local medical environment is usually one of crucial importance. That makes it all the more important that we become an integral part of this processalready initiated by othersof clinical outcomes assessment. Neurosurgeons, after all, understand the context of what is being evaluated better than those who, heretofore, have been performing that assessment.
It is imperative that we vigorously define — with as scientifically valid and methodologically convincing outcomes data as possible — why our neurosurgical treatment modalities are efficacious and cost-efficient. By doing so, we, North American Neurosurgery, will have discharged our responsibility and been truly accountable to our patients. Ultimately, it is to them that we owe our allegiance and it isto them that we are accountable.
The dimensions of this Herculean task are hard to comprehend. Accordingly, it is very likely that each and every member of this organization will be asked to assist in contributing to the database necessary for a scientifically valid assessment of clinical outcomes in neurosurgery. For the reasons outlined above, it represents our most important and challenging assignment.