Clinical Decisions and Laboratory Use

Clinical Decisions and Laboratory Use

Donald P. Connelly
Ellis S. Benson
M. Desmond Burke
Douglas Fenderson
Copyright Date: 1982
Edition: NED - New edition
Pages: 376
https://www.jstor.org/stable/10.5749/j.ctttswd2
  • Cite this Item
  • Book Info
    Clinical Decisions and Laboratory Use
    Book Description:

    Clinical Decisions and Laboratory Use was first published in 1982. As the national expenditure for health care continues to mount, pressures from within and outside of medicine are encouraging more effective and economical use of medical resources. Though much of the current attention is focused on a few high-cost, technologically dramatic procedures, the rapidly growing use of laboratory tests makes up a substantial portion of heal care costs. Evidence suggests that a significant amount of laboratory use is ineffective and unnecessary and could be eliminated with no harmful effects on the quality of care. This book, which is based on the Conference on Clinical Decision Making and Laboratory Use held at the University of Minnesota in 1980, focuses on the problems faced by health care professionals as they attempt to control laboratory costs and suggests ways of making rational and effective decisions. The 32 chapters deal with the following topics: Medical Decisions, Technology, and Social Needs; The Process of Clinical Decision Making; Clinical Decisions and the Clinical Laboratory; Medical Education and Effective Laboratory Use; Tools for Supporting the Decision-Making Process; and Research Initiatives Toward Effective Decision Making and Laboratory Use. Physicians, clinic and hospital administrators, laboratory directors, and health care planners will find this book a useful source of information in establishing their own procedures for clinical decision making.

    eISBN: 978-0-8166-6197-8
    Subjects: Health Sciences

Table of Contents

  1. Front Matter
    (pp. i-iv)
  2. Table of Contents
    (pp. v-viii)
  3. Contributors
    (pp. ix-xii)
  4. Preface
    (pp. xiii-xvi)
  5. Section I. Medical Decisions, Technology, and Social Needs

    • CHAPTER ONE Medical Technology and Public Policy
      (pp. 3-10)
      John E. Kralewski and John H. Westerman

      One of the most important public policy issues in the health care field today centers on the appropriate use of state of the art technology. Since the National Institutes of Health were formally established in the 1930s, federal public policies favoring technological development have dominated the health field. In the mid 1960s, growing concern over the distribution of health services caused a slight but highly publicized shift in these policies as a number of governmental programs, such as the Regional Medical programs, were initiated to expedite the diffusion of technologies from research and teaching centers to the field of practice....

    • CHAPTER TWO Decision Making and the Evolution of Modern Medicine
      (pp. 11-15)
      Stanley Joel Reiser

      The physician’s search for evidence to provide clues to sources of the patient’s complaints has always occupied a preeminent place in the hierarchy of medical tasks. Discussion of how this process evolved is often cast as a succession of techniques, the better techniques thought, generally, to replace the inferior. Yet the change from one order of diagnostic procedure to another involves far more than a change of technique: it involves, basically, a transformation of the criteria we use to decide what to direct our attention to from the universe of possible observations that can be made on the person who...

    • CHAPTER THREE Medical Decisions and Society
      (pp. 16-21)
      Howard S. Frazier

      As autonomous practitioners, we share the belief that we have choices. We assume that, in the course of our professional activities, we make decisions that, if prudently arrived at, will not affect the care of patients adversely, and may significantly reduce the resources used in their care. We believe that we can do good (or at least as well) by using fewer resources and save some money.

      I subscribe to the first assumption; it has the ring of motherhood and the flag, and circularity as well. The validity and implications of the second assumption, that we can do as well...

    • CHAPTER FOUR Use of Computers in a Clinical Library
      (pp. 22-26)
      Eugene A. Stead Jr

      Clinical practitioners of medicine are artists, with some training in methods of the more exact sciences. Patients are people who wish to stay well or who have diseases. In only a few of the diseases that are more or less completely understood can treatment be controlled by the scientific method. Doctors so far have little to offer adults in the way of disease prevention; they are at their best when caring for patients with acute illness. They are at a great disadvantage in caring for patients who have chronic illness because the methods of collecting data over time have not...

  6. Section II. The Process of Clinical Decision Making

    • CHAPTER FIVE The Clinical Decision-Making Process
      (pp. 29-38)
      Jerome P. Kassirer

      It is probably not merely an accident that the clinical decision-making process is the target of research efforts in so many institutions. The new interest in this area of research can be attributed in part to the development of new methods, but increased public awareness of medicine undoubtedly has contributed importantly. Decisions, particularly those involving the use of expensive diagnostic equipment, have come under public scrutiny largely as the result of the highly visible national expenditures for medical care and the widespread concern that these expenses are uncontrolled and possibly uncontrollable. Investigators are motivated by the desire to develop a...

    • CHAPTER SIX Cognitive Models of Medical Problem Solvers
      (pp. 39-51)
      Paul E. Johnson

      For years those who studied the problem-solving process hoped that individuals at different levels of training and experience would differ in characteristic, predictable ways. However, beginning with the classic work of DeGroot¹ and Chase and Simon² in chess, and continuing to present day work in a number of areas including medicine,3-5the dominant message seems to be that similarities outweigh differences in comparisons between individual problem solvers.

      In medicine as well as other areas, problem solving consists of an application of hypothetico-deductive reasoning in which the basic units are questions, cues, and diagnostic hypotheses. Clinical problem solving typically begins with...

    • CHAPTER SEVEN Decision Making in Primary Care: The Case of Obesity
      (pp. 52-58)
      Michael M. Ravitch, David R. Rovner, Penny A. Jennett, Margaret M. Holmes, Gerald B. Holzman and Arthur S. Elstein

      Most studies of clinical decision making fall into one of two classes. Retrospective reviews use chart audit techniques to assess the performance of physicians in the light of predetermined criteria.¹ The criteria may deal with the steps to be performed in the work-up of a target condition, as in Payne’s process index. Alternatively, evaluation may be concerned with frequency or rate of achieving desired outcomes of care, regardless of the process employed, as in the Performance Evaluation Program of the Joint Commission on Accreditation of Hospitals.² In either case, the record is evaluated retrospectively, in the light of some external...

    • CHAPTER EIGHT Evaluating Skills in Clinical Decision Making
      (pp. 59-66)
      Barbara J. Andrew

      We know that problem solving is a process that involves several interrelated stages, and that experienced problem solvers store, organize, and retrieve knowledge in a somewhat different fashion than do inexperienced problem solvers. Problem solving is certainly not the sole component of clinical competence, although it is, I believe, the principal one. Humanism and effective communication and psychomotor skills can enhance the quality and effectiveness of health care but they cannot compensate for a failure to reach a quick, accurate diagnosis or for an ineffective or inappropriate treatment plan.

      My task here is to discuss methods being used to assess...

  7. Section III. Clinical Decisions and the Clinical Laboratory

    • CHAPTER NINE The Role of the Laboratory in Clinical Decision Making
      (pp. 69-79)
      D. S. Young

      Except in unusual situations, physicians request clinical laboratory tests after they have taken the history of a patient’s illness and completed a physical examination. The laboratory tests are ordered by physicians to assist them in decision making—usually to seek support for impressions gleaned from personal interaction with the patient. The data reported by most clinical laboratories are usually in numerical form. In few laboratories do the staff interpret the data and provide it in such a form that it would reduce the number of decisions that a clinician has to make. Although laboratory staff traditionally provide detailed interpretations of...

    • CHAPTER TEN Assessing the Effectiveness of Laboratory Test Selection and Use
      (pp. 80-84)
      William R. Fifer

      Presumably some laboratory tests are more useful than others. Given the immense number of tests available (all of which appear to have at leastsomevalue) and the increasing awareness of resource limitations, it becomes important to evaluate laboratory tests in terms of effectiveness, to guide their selection and use. This paper will present selected aspects of the problem of laboratory test evaluation.

      “Assessment” is defined as “rating,” “appraising,” or “evaluating.” Evaluation, which is the process of making a judgment about the worth of something, uses a referent or standard as the benchmark with which to compare the thing to...

    • CHAPTER ELEVEN Use or Misuse: Variations in Clinical Laboratory Testing
      (pp. 85-96)
      Werner A. Gliebe

      Increasing the effectiveness of laboratory test use is essential if we are to achieve maximal quality of medical care at the lowest possible cost. The significance of this component is readily apparent, since about $15 billion annually can be accounted for by clinical laboratory use. Focusing on increasing effectiveness does suggest, however, that substantial improvements could be made, an implication that may not be true ahistorically.

      If the pressures to contain high health care costs were not so intense, it is quite conceivable that the clinical laboratory would be seen for its positive effects and its limitations given little emphasis....

    • CHAPTER TWELVE Factors Leading to Appropriate Clinical Laboratory Workload Growth
      (pp. 97-107)
      Donald P. Connelly, Philip N. St. Louis and Lynn Neitz

      The rapid and continued expansion of health care costs has focused attention on the efficiency of use of our nation’s medical resources. Ten percent of the nation’s health care dollar is spent for clinical laboratory services, and by 1985, if present trends continue, the annual laboratory service cost will be nearly $50 billion.¹ It is becoming clear that “small ticket items,” in the aggregate, account for much more of health care costs than do the high technology, “large ticket items” such as the CT scanner.² A commonly held view among physicians, health insurance payers, and policy makers is that much...

    • CHAPTER THIRTEEN Reducing the Use of the Clinical Laboratory: How Much Can Be Saved?
      (pp. 108-118)
      Stan N. Finkelstein

      The recent increases in health care costs have led many to offer possible strategies for moderating the rate of growth. Ancillary services have been seen as a potential target for cost containment because some observers believe that a substantial portion of their utilization is unnecessary. Efforts directed toward achieving a more discerning use of ancillary services would also be consistent with an objective with which many clinicians would sympathize, a more reasoned practice of medicine.

      The clinical laboratory is one aspect of ancillary services in which some cost containment strategies are being proposed and tested. In general, two different kinds...

  8. Section IV. Medical Education and Effective Laboratory Use

    • CHAPTER FOURTEEN Teaching Effective Use of the Laboratory to Medical Students
      (pp. 121-130)
      M. Desmond Burke and Donald P. Connelly

      Physicians are said to order laboratory tests for the following reasons: to detect disease, to confirm or exclude diagnostic hypotheses, to estimate prognosis, and to monitor therapy.¹ The process is a simple one: Laboratories issue request forms complete with names of tests and designed to facilitate easy ordering. All the physician need do is make the appropriate check mark(s)—or have an intermediary do so—and in due time test results appear in a format designed for easy readability. Laboratories neither demand nor expect a statement of the clinical problem. By the same token, physicians neither expect nor seek interpretative...

    • CHAPTER FIFTEEN Training House Staff in Effective Laboratory Use
      (pp. 131-138)
      Paul F. Griner

      Educational strategies designed to meet the needs of residents concerning laboratory test use must recognize the different purposes for these tests and other diagnostic procedures. Strategies to influence physicians’ use of tests to monitor patients after diagnosis (i.e., the use of tests in patient management) may be quite different from those to influence use of tests in arriving at diagnoses. Both issues are equally important but the payoffs may be different. Strategies designed to reduce superfluous testing, which relates largely to test use in patient management, if successful, should result in some degree of cost savings. On the other hand,...

    • CHAPTER SIXTEEN Continuing Medical Education and Effective Laboratory Use
      (pp. 139-144)
      Stephen E. Goldfinger

      It is quite possible that the cause of effective laboratory use could be best served by abolishing all continuing medical education offerings for the next five years. Then, oblivious to what new round of costly and possibly dangerous technology was needed to remain up-to-date, physicians might start looking more sharply at their current testing practices to find out which ones really made a difference. A facetious suggestion, perhaps, but one that arises from deep concern that traditional programs of “show-and-tell” continuing medical education may give rise to the uncritical proliferation of newer techniques rather than to their rational use.

      Dissatisfaction...

    • CHAPTER SEVENTEEN Modifying Physician Patterns of Laboratory Use
      (pp. 145-158)
      John M. Eisenberg

      Efforts to contain medical care costs often emphasize the role of the physician in generating the demand for medical services. It follows that cost containment programs include attempts to decrease physicians’ prescription of hospitalization, drugs, x-ray examinations, and laboratory tests. Therefore, implicit in those programs designed to change physicians’ use of laboratory tests is the assumption that physicians currently use the laboratory inappropriately. Before reviewing methods of changing physician behavior, we need to ask the prerequisite question: “Are physicians misbehaving?” Before turning the hospital into a huge Skinner box to change physician behavior, we must ask “Do we have a...

    • CHAPTER EIGHTEEN Problem Solving in Medicine: Can We Teach It?
      (pp. 159-182)
      Lawrence L. Weed

      In the last analysis, we teach by example. People will do what is done in the real world of everyday practice—regardless of what they were “taught” in an “educational situation.” High quality, efficient problem solving for most of the patients most of the time is not occurring. This is because in all types of medical settings, physicians and other medical personnel hear new data, recall memorized knowledge, formulate problems, generate plans, and record all the data in rapid sequence, without systematic use of appropriate aids to their senses and their intellectual capacities. So to varying degrees they may falter...

  9. Section V. Tools for Supporting the Decision-Making Process

    • CHAPTER NINETEEN Decision Making in Radiology: ROC Curves
      (pp. 185-198)
      Barbara J. McNeil

      The evaluation of new diagnostic procedures is becoming as commonplace as is the evaluation of new therapeutic procedures. And, just as the need to evaluate therapies led to the generation of a variety of new methodologic techniques, the randomized clinical trial being the most common, so too has the need to evaluate diagnostic procedures led to the introduction of a variety of other methodological techniques.1-3Some of these are quite familiar to most physicians by now. For example, there has been a flurry of articles recently on Bayes’ theorem, decision matrices, and predictive values. Other techniques, borrowed heavily from signal...

    • CHAPTER TWENTY The Timing of Surgery for Resection of an Abdominal Aortic Aneurysm: Decision Analysis in Clinical Practice
      (pp. 199-213)
      Stephen G. Pauker

      Clinical decision analysis is a relatively new discipline that is increasingly infiltrating the medical literature and beginning to form the basis for health policy decisions.1-12It applies the established tools of decision analysis13to a new domain, medicine. Even more recently, decision analysis has begun to be applied to patient care, and the technique is being slowly adapted for bedside use.14-17This paper presents the clinical decision analysis of an actual clinical case seen in consultation at the New England Medical Center Hospital and demonstrates the feasibility of applying these tools to selected clinical cases. Since this analysis was developed...

    • CHAPTER TWENTY-ONE Biostatistical Tools: Promises and Accomplishments
      (pp. 214-218)
      E. A. Johnson

      Dramatic promises have been made for the biostatistical tools for classification. Those promises are well-founded. But practically no reports have come out about successful utilization of biostatistical tools for classification in realistic clinical settings, and there probably will not be for some time to come. These two concepts—confident promise and no record of accomplishment—are inconsistent. I will try to explain.

      This is not an appropriate setting for a technical description of classification rules. In most cases, the mathematics involved with describing the rules and proving their behavior is cumbersome. An important review by Solberg¹ summarizes most of the...

    • CHAPTER TWENTY-TWO Test Selection and Early Risk Scale in Acute Myocardial Infarction
      (pp. 219-224)
      Adelin Albert, Jean-Paul Chapelle, Camille Heusghem, Gérard Siest and Joseph Henny

      In countries with a high level of medical technology, the costs of clinical laboratory tests have reached a critical point. Faced with such an acute financial situation, health authorities have taken measures to reduce laboratory expenditures. Unfortunately, such decisions are more administrative than medical in nature.

      From a medical standpoint the main problem is to know whether this situation can influence the clinical decision-making process. The time has come to take advantage of these difficult economic circumstances to modify the attitude of both biologists and clinicians. Amazingly, they are faced with a paradox: the overproduction and the insufficient use of...

    • CHAPTER TWENTY-THREE Clinical Algorithms and Patient Care
      (pp. 225-237)
      Harold C. Sox Jr

      The “health manpower crisis” of the mid-1960s led to the deployment of nonphysicians in roles that have been traditionally reserved for physicians: the diagnosis and treatment of illness. Teaching these skills to nonphysicians brought about a revolution in clinical teaching methods. Heretofore, medical diagnosis had always been less taught than learned. Physicians-in-training started their clinical work with an extensive knowledge of normal and abnormal human biology. Medical diagnosis was learned, somewhat haphazardly, through caring for a random selection of patients. Nurse practitioners and physician’s assistants lack detailed understanding of disease mechanisms. Furthermore, their training is compressed into a much shorter...

    • CHAPTER TWENTY-FOUR Improved Clinical Strategies for the Management of Common Problems
      (pp. 238-243)
      Anthony L. Komaroff, Theodore M. Pass and Herbert Sherman

      Studies of clinical decision making and laboratory use have concentrated mainly on “expensive” technologies (high unit charges) and/or problems of the hospitalized patient with high-risk illnesses.

      Several years ago, we began to investigate these questions with regard to “inexpensive” technologies (low unit charges), primarily in the ambulatory patient with low-risk illness. There were several reasons for doing so:

      1. The enormous volume of such technologies means that the total charges are substantial, despite the low unit charge.

      2. Ambulatory medical practice may be more susceptible than hospital practice to the inflationary pressures of an expanded health insurance system.

      3. There...

    • CHAPTER TWENTY-FIVE Computer-Based Support for Medical Decision Makers
      (pp. 244-250)
      Homer R. Warner

      Although diagnosis has largely monopolized the attention of people working in the field of computer algorithms for medical decision making, diagnosis is but one of a wide variety of challenging decisions that face the clinician. Listed in the accompanying tabulation are nine kinds of decisions currently being made by HELP, a computer-based system operational at the LDS Hospital in Salt Lake City.¹ In this presentation I would like to discuss examples of each of these nine types of decisions, present a description of the decision tools needed to perform these functions, and discuss the tools we have implemented to initiate...

    • CHAPTER TWENTY-SIX INTERNIST: Can Artificial Intelligence Help?
      (pp. 251-270)
      J. D. Myers, Harry E. Pople Jr and Randolph A. Miller

      On the basis of the work that we have been engaged in over the past eight years, the question posed in the title can be answered in the affirmative.

      Many computer porgrams for diagnosis in medicine have been devised utilizing (a) branching logic techniques, (b) Bayesian analysis, and (c) to a lesser degree, artificial intelligence. These programs have all dealt with circumscribed problems in internal medicine and other clinical fields; none other than INTERNIST has attempted to encompass as broad an area as general internal medicine.

      To distinguish clearly the INTERNIST approach from these other approaches to computer-based diagnosis, it...

  10. Section VI. Research Initiatives Toward Effective Decision Making and Laboratory Use

    • CHAPTER TWENTY-SEVEN Research and Operational Strategies: An Overview
      (pp. 273-278)
      Ellis S. Benson

      One of the most important developments in recent years with respect to the clinical laboratory has been the dawning realization that more is required of a laboratory test than the assurance of technical validity. We in laboratory medicine now know that we have responsibilities that go beyond the assurance of the reliability of test results. The concepts “diagnostic sensitivity” and “diagnostic specificity” have been with us for many years but it was not until Galen and Gambino¹ rediscovered and popularized these concepts in 1975 that we paid much attention to them. Use of these concepts, and that of “predictive values,”...

    • CHAPTER TWENTY-EIGHT The Predictive Value Model and Patient Care Decisions
      (pp. 279-286)
      Robert S. Galen

      The predictive value model has been applied by a number of investigators to the evaluation of a single test to optimize its screening or diagnostic function. In fact, however, it is rare to use the result of a single test as the final arbiter of a medical decision. The predictive value model can be expanded to deal with multiple tests. We have applied the predictive value model in evaluating multiple tests used either in series or parallel fashion. These two approaches to laboratory testing, series and parallel, will be explored relative to the sensitivity, specificity, and predictive value of each...

    • CHAPTER TWENTY-NINE Use of a Laboratory Database to Monitor Medical Care
      (pp. 287-293)
      Charles H. Altshuler

      A database composed primarily of comprehensive laboratory information may be used as a clinical decision support system. In this presentation, I hope to show how the system may be used to identify and correct some deficiencies in practice, and how it can be used to study some of the more important factors affecting hospital care costs.

      Since 1968 the laboratory at St. Joseph’s Hospital—a 580-bed general acute care general hospital—has offered a programmed accelerated laboratory investigation (the PALI). This system was designed to exploit developments in laboratory automation and information handling; during the time in which it has...

    • CHAPTER THIRTY Doctor Billing in a Federal Institution
      (pp. 294-299)
      Joyce A. Campbell and Michael T. Makler

      This study was designed to determine whether a weekly bill delivered to interns containing the total number and cost of tests would affect laboratory use at a federal institution. Previous studies have suggested that lack of awareness of laboratory costs1-3and overuse of laboratory tests4-6-account for a substantial amount of the costs generated by house staff at private and teaching hospitals. Various educational programs have been described to reduce overutilization of the laboratory.7-10This study was also designed to examine the questions of the level of cost awareness and the priorities interns use in ordering laboratory tests.

      The subjects...

    • CHAPTER THIRTY-ONE The Use of Reference Values and the Concept of Reference State: A Contribution to Improved Laboratory Use
      (pp. 300-307)
      Gérard Siest, Joseph Henny, Camille Heusghem and Adelin Albert

      The concept of “normal values” is beset by many ambiguities, mostly related to definition and selection of “normals.” To avoid these ambiguities and their attendant confusion, Dybkaer and Gräsbeck¹ in 1969 introduced the concept of “reference values.” These authors recommended that the term “normal value,” with all its ambiguities and misconceptions, be discarded. They defined reference values as a set of values of a certain kind of quantity attainable from a single individual or a group of individuals corresponding to a stated description. This description must be spelled out in detail and available if others are to use the reference...

    • CHAPTER THIRTY-TWO The Nominal Group Process Reports
      (pp. 308-324)

      To promote a synthesis of ideas and experience related to medical decision-making and laboratory use, the contributors of each major section joined in two nominal group process (NGP) sessions.

      NGP is a structured group process strategy, developed in 1968 by André Delbecq and his associated at the University of Wisconsin.¹ It has been widely used for program planning in health, social service, education, industry, and government organizations. The term “nominal” denotes a key feature of the strategy. During thefirst phase, participants are in each other’s presence, but do not interact verbally. Instead, they generateideas in writingin response...

  11. Appendix. Discussion
    (pp. 325-348)
  12. Index
    (pp. 351-356)
  13. Back Matter
    (pp. 357-357)