PORTs…represent the coming of age of health services research as a useful, clinically relevant discipline, whereby the evaluative sciences for the first time are, with the help and support of Congress, going to become directly relevant to clinical decision making in a way that they haven't before. They're going to have to carry the same burdens of scrutiny that clinical trials and other kinds of clinical research have carried.
We haven't considered epidemiology to be as important to regulate in the past as randomized clinical trials. We haven't considered decision-analytic papers as important to scrutinize, largely because we didn't expect them to have as much impact. Maybe that's changing, and maybe for that reason the increased authority carries with it increased responsibilities.
—David Blumenthal, Institute of Medicine Workshop, 1990
The term outcomes research—sometimes called effectiveness research, evaluative clinical science, or clinical evaluation—has come into common usage in the health care community to describe an approach to clinical research that has lately received increased emphasis. Outcomes research examines the treatment of clinical conditions rather than individual procedures or treatments. It is the systematic assessment of clinical practice, encompassing both outcomes that are relevant to patients—mortality, morbidity, complications, symptom reduction, and functional status improvement—as well as physiologic or biologic indicators; it involves all reasonably held theories and alternative clinical practices. Outcomes research emphasizes multidisciplinary assessment teams and new strategies and methods for making inferences from experimental and nonexperimental
data (Wennberg, 1990a). It has been embraced by health care professionals and policymakers who are deeply concerned about (1) two decades of studies showing wide variations in the use of health services in different locales but little understanding of the effect of those variations on the health of patients served, (2) evidence of frequently inappropriate use of a number of diagnostic and surgical procedures, (3) rapidly increasing health care costs, and (4) the possible effects of changing health care reimbursement schemes on quality of care.
Outcome—the effect of an intervention on patient health and well-being—is the greatest concern of patients who seek care and the most obvious criterion for measuring the effectiveness of care. However, Congress's Office of Technology Assessment has estimated that only 10 to 20 percent of what physicians do has clearly been shown to be of value as judged by well-designed randomized clinical trials (OTA, 1983). Even these studies, regarded as the best available, are limited in generalizability because they exclude large numbers of patients, such as those with other coexisting diseases or those not within the age or gender designations of the study protocol. More fundamentally, such clinical trials typically examine only a single and usually innovative modality and do not include all alternative treatments. The end points measured have often not included patient health and functioning status.
Why has outcomes research been slow in coming? It has proved difficult for clinical researchers to design short-and long-term end points that are reliable and valid for the dimensions of health and well-being valued by patients and to measure them in such a way as to draw conclusions about the care provided. Health status is determined in large part by environmental, personal, and social factors that are not related to clinical interventions. Outcomes research has been developed as a means of responding to the need for data that are useful and reliable despite these problems. Recently, researchers have demonstrated the practicality of using large administrative data bases to assess some outcomes, and they have developed and validated instruments for accurately assessing patient health status to augment administrative data bases. Decision analysis and meta-analysis (to be described in greater detail in Chapter 2) are two other methods that are used in clinical evaluative research to augment our understanding of the clinical effectiveness of treatments despite large areas of uncertainty and a dearth of randomized controlled clinical trials.
Recognizing the potential of these methods of health services research, the Department of Health and Human Services sought and received congressional appropriations beginning in fiscal year 1989 for outcomes research and particularly for the establishment of the patient outcomes research teams (PORTs) described in this report. These are important new initiatives that recognize the potential of improved methods
of health services research to assess the effectiveness of alternative medical practices.
THE INSTITUTE OF MEDICINE STUDY
The potential importance of the findings of PORTs generates the possibility of real or apparent conflicts of interest in the interaction of PORT members with professional groups and business firms affected by those findings. Recognizing the vulnerability of PORT findings to challenges on such grounds, the National Center for Health Services Research and Health Care Technology Assessment (now subsumed in the Agency for Health Care Policy and Research—AHCPR), the Pharmaceutical Manufacturers Association, and Dartmouth Medical School (site of the first PORT to be funded) sponsored an Institute of Medicine (IOM) study to provide points for the agency and PORTs to consider in anticipating and managing conflicts of interest.
The IOM appointed a nine-member committee representing expertise in medical center and research administration, health services and clinical research, law, medical sociology, ethics, pharmaceutical and device manufacturing, and regulation. The committee met three times from April to October 1990 and commissioned six background papers on topics that covered the structure and methods of PORTs and potential sources of bias; a comparison of PORT and clinical research; a review of conflict-of-interest regulation in biomedical science; the possible impact of PORTs on health services research, technology innovation, and payment policy; and PORT research from provider and industry perspectives.
In June 1990, the committee convened a one-and-a-half-day workshop with some 60 invited participants who also brought extensive expertise and offered wide-ranging views. In addition to informative presentations and discussion, the workshop featured discussions of three scenarios written for the conference.
STRUCTURE OF THE REPORT
This report presents the results of the deliberations of the IOM study committee, augmented by the background papers, and the presentations and discussions at the June workshop. Chapter 2 describes the Medical Treatment Effectiveness Program within AHCPR and the structure and methods of PORTs and then examines the ways in which PORT research differs from other biomedical and clinical research. Chapter 3 provides the committee's working definition of conflicts of interest,
describes financial and professional conflicts of interest, and gives a brief history of recent attempts to deal with conflicts of interest on the part of universities, medical publications, certain government officials, judges, lawyers, and other individuals with fiduciary responsibility.
Chapter 4 describes special concerns about conflicts of interest in PORTs, and Chapter 5 describes two models and some general, frequently proposed approaches for dealing with conflicts of interest. Chapter 6 outlines suggestions and other points for consideration by PORTs and their home institutions, AHCPR, the health services research community, industry, and Congress. Appendices to the report include the scenarios discussed at the June workshop, the rapporteur summaries of those discussions, and several of the background papers and presentations.