American Journal of Law & Medicine

Outcomes assessment in health care reform: promise and limitations. (Quality of Care and Health Reform: Complementary or Conflicting)


If the fundamental goals of the health care reform effort are to ensure universal access to an acceptable quality of health care at an affordable cost, then the threshold question for reform is: What health care services should be provided in an efficient, equitable system?

Answering this question requires weighing a complex mix of medical and social policy factors, a process not attempted in this article. But the starting point for that process should be determining what health care services "work" and what they cost. Outcomes assessment holds considerable promise in finding answers to these subsidiary questions, because it is intended to assess the effectiveness of health care services, that is, what works and what does not work for patients. It holds promise, not only for improving the quality of care, but also for identifying unnecessary or wasteful practices that increase health care costs.

This article outlines the possible uses of outcomes assessment in creating and operating a reformed health care system, and key legal implications. But, first, it attempts to define what outcomes assessment is, and to summarize its strengths and limitations.


Outcomes assessment is one term used to describe research that evaluates medical and health services and procedures. Such research is primarily designed to determine whether particular procedures or services, such as surgical techniques, diagnostic tests, or therapeutic interventions, work as intended; that is, whether they are effective in identifying a source of illness or disability, relieving symptoms of illness, curing a disease, or preventing illness. The term "outcomes" denotes that the focus is on the service's success or failure in producing a desired outcome for the patient. Several other terms are used to mean the same or similar things, "outcomes research" and "technology" assessment being the most common.

Quality assessment or assurance has a broader connotation.(1) It is used in hospital and other health care facility programs to monitor the quality of care provided to patients and ordinarily includes measures of available resources and the process of care, as well as patient outcomes.(2)

Medical practice standards, policies, parameters, or guidelines are more specific statements of what should or should not be done in particular clinical circumstances.(3) Most have been developed by professional specialty organizations for use by practitioners. Increasingly, practice standards use the results of outcomes research in establishing general guidelines for preventing, diagnosing, or treating certain clinical conditions. In fact, many believe that the primary value of outcomes assessment is its potential to produce the data necessary to develop medical practice guidelines that will improve the practice of medicine and, consequently, patient health outcomes.(4)

Theoretically, outcomes assessment can offer a data base for deciding what works and how and when to use it. This potential has not yet been realized fully, however. First, the study of effective medical care is still in its infancy. Only a small proportion of medical services have been subjected to rigorous testing to determine their actual effectiveness or relative merit compared with alternatives.(5) Although new drugs and some medical devices must generally be shown to be both safe and effective to be approved by the Food and Drug Administration before being distributed in interstate commerce,(6) medical procedures and services are largely unregulated.

Second, appropriate methodologies have not been developed for assessing the many different types of medical services.(7) It is often difficult to define what a desired outcome should be: longer life, better physical function, or some other status? Should it be objectively observable or a subjective impression of the patient? What does it mean to say that a medical service is "effective" or "works?" In some cases, it is difficult to determine whether a desired outcome is the result of a medical intervention or some other cause. The way studies are designed and carried out also affects the validity of their results. Although the science of technology assessment has expanded rapidly, much remains to be done.

Third, the study of outcomes alone tells us little about what works to achieve desired outcomes. A good outcome may result in spite of inappropriate care, and a poor outcome may occur even if stellar care is provided.(8) Patients do not enter the health care system neatly labeled with instructions. Health care practitioners must select the relevant diagnostic services that can pinpoint the problem and its cause, identify the best means of treating the problem (and decide by what criteria to judge what is "best"), and provide the treatment correctly. In the parlance of health services research, these steps are matters of procedure or process, rather than outcomes assessment. Yet they are essential to deciding what works.

Finally, translating the results of outcomes research into effective medical care can be problematic. Even assuming that concepts like "effective," "desired outcome," and "benefit" can be usefully defined and measured for research purposes, the results may mean different things to policy makers, health care practitioners, payors, and patients. Much depends on how the research is used.


Outcomes assessment can be used in at least the following ways:

Quality Assurance: To establish professional practice policies incorporating appropriate standards of care, educate practitioners, and review and monitor practitioners' performance to improve the quality of care;

Cost Control: To reduce unnecessary expenditures by identifying and eliminating inappropriate or ineffective services;

Patient Decision Making: To educate individual patients and the public about effective and ineffective health services, and to improve patients' ability to choose the care they want to receive;

Defining Benefits: To determine what types of services should be offered or covered in a national universal health care system or in a private health insurance program;

Individual Care or Payment Determinations: To decide whether a particular service should be provided to an individual patient as necessary care or as part of the benefits covered and paid for by a national health care system or a private insurance program;

Dispute Resolution: To determine whether the care provided to an individual patient was appropriate or negligent for purposes of medical malpractice claims.

The results of outcomes assessment are likely to have the most positive effects and raise the fewest legal questions when used to provide information, as in quality assurance programs and patient education efforts. More complicated legal questions arise when the data are used to make binding decisions about what specific services will be paid for or provided to an individual patient, as in insurers' utilization review systems, or who should be responsible for a poor outcome, as in medical malpractice cases. Here, the concern is that practice standards could be used to lower the quality of care or to deny remedies to injured patients.


The quality of health care has traditionally been defined by the health professions on the basis of cumulative experience in clinical practice and research published in peer reviewed journals. But much of medical practice has no scientific basis at all. When properly studied, some commonly accepted medical procedures are found to be of no signficant benefit.(9) Others have been found to be used without appropriate indications.(10)


Outcomes research can generate valuable information about whether current health care practices are effective. Indeed, professional societies such as the American Medical Association, the American Academy of Pediatrics, the American Nurses Association, the American College of Obstetrics and Gynecology, and the American College of Emergency Physicians have sometimes drawn upon such research to prepare practice policies that express a consensus on the best way to diagnose or treat specific problems. Similarly, the National Institutes of Health use outcomes assessment to develop policy statements in consensus conferences.

Outcomes research may also shed light on why different physicians use different approaches to diagnosing and treating the same illnesses in similar populations. This variation in practice styles has been described for several conditions.(11) The reasons for the variations are not well understood. Researchers have suggested that lack of relevant information among physicians may account for some of the difference.(12) The more widely disseminated the results of outcomes research, the better the information available to practitioners as they make clinical decisions.(13) This should help practitioners achieve a more consistent, better quality of care.

The federal Agency for Health Care Policy and Research (AHCPR) was created in 1989 to help provide just such information.(14) Its Forum for Quality and Effectiveness in Health Care is to develop and disseminate "practice guidelines, quality standards, performance measures, and medical review criteria" as part of a substantial new health services research effort.(15)


Hospitals and other health care facilities anticipate using practice policies to monitor and improve the quality of care provided by health care practitioners employed by or affiliated with them. State common law principles hold such facilities legally responsible for ensuring that their staffs are qualified to treat patients and performs in a satisfactory manner.(16) Institutional peer review programs, in which staff physicians review the medical decisions and performance of their colleagues, may refer to such policies in determining whether physicians have treated patients appropriately. Those who fall below acceptable standards may be subject to internal supervision, discipline, or loss of employment, panel membership, or hospital privileges.(17) In addition, the federal Health Care Quality Improvement Act of 1986 requires that most disciplinary action taken by a hospital be reported to the National Practitioners Data Bank. Hospitals must consult the Data Bank before granting or renewing privileges in order to prevent unqualified practitioners from continuing to practice.(18) Legislation, however, generally does not dictate any standards that must be applied in deciding whether a properly licensed practitioner has performed appropriate procedures.

The quality of care in hospitals that receive Medicare funds (most acute care hospitals) is subject to review by peer review organizations (PROs). PROs, which replaced Professional Standards Review Organizations, are intended to counteract the incentives to underserve patients that can arise from limits on Medicare payments under the Prospective Payment System. But the limited reported experience of PROs suggests that they may not always live up to expectations.(19)

It would be possible to impose tighter controls on peer review by requiring review committees to judge physician performance strictly against practice standards, where they exist. If a practice standard is the sole determinant of acceptable care, however, it must have the confidence of those judged by it as well as those who apply it. This requires that it be issued by a legitimate authority and contain standards based on valid data.(20) Moreover, it must be sufficiently specific to serve as a useful standard against which performance in individual cases can be measured.

It must be stressed, however, that it is difficult to develop practice policies that, by themselves, are sufficiently definite and specific to judge all medical practice decisions. …

Log in to your account to read this article – and millions more.