Clinical practice guidelines: promise or illusion?

Clinical Practice Guidelines: Promise or Illusion?

Clinical decision-making is no longer the exclusive domain of the health care practitioner. Consumers, as patients and as businesses providing health benefits, feel entitled to more influence over the decision-making process and to greater access to information regarding health care practices and measures of quality. Hospitals have become the locus of quality management in response to cost containment pressures and regulatory initiatives. Payers are insisting that providers be held more accountable for their clinical decision-making and are scrutinizing the use of inpatient stays, procedures, tests, and technologies.

An optimal balance of cost containment and quality--health care value--is now the avowed goal of most major health care constituencies. Regrettably, the virtual absence of nationally recognized objective standards creates a significant barrier to achieving the goals of informed, collaborative decision-making; effective management of health care resources; accountability; and health care value.

The failure to achieve clinical consensus in health care has been highlighted by research in the areas of appropriateness and small area variations analysis. The 1980s RAND studies of appropriateness of procedures demonstrated that the clinical indications for several procedures were inappropriate or equivocal in a large percentage of cases. [1] The work of Wennberg and others demonstrates wide variations in practice patterns between small geographic regions as reflected by two- and threefold variations in the rates of hospitalization and in the use of certain procedures. [2-5] Others have shown that individual practice styles are a major factor in determining the rate of use of a procedure. [6] Such information has attracted the attention of a large number of health care constituencies who are clamoring for standards of appropriateness, cost-effectiveness, outcomes, and quality.

Brook has proposed 10 benefits that might accrue from the development of nationally recognized practice standards through the creation of a national institution "capable of developing and maintaining practice guidelines for both common diagnoses and common procedures." [1]

* Better direction for health care policy development.

* A stimulus for professional self-improvement.

* Comparative information for consumers to support a "buy right" strategy.

* New textbooks organized around guidelines.

* Development of performance-based education, certification, and recertification processes.

* Licensing based on compliance with standards.

* Selection of preferred providers.

* Reimbursement decisions based on guideline compliance.

* A basis for shared patient and physician decision making.

* A framework for malpractice decisions.

Four types of organizations have taken the lead in the development of more than 300 clinical practice guidelines: government research agencies, payers, individual specialty societies, and voluntary health associations. Government agencies developing practice guidelines or technology assessments include state health departments, the Centers for Disease Control (CDC), the U.S. Department of Health and Human Services (HHS), the National Cancer Institute (NCI), the National Center for Health Service Research and Health Care Technology Assessment, and the National Institutes of Health Office of Medical Applications of Research. [7] Private companies, such as Value Health Sciences, Inc., have adapted the results of government research to commercial products for sale to payers.

Blue Cross and Blue Shield Association's (BC/BS) Medical Necessity Program was started in 1976 to evaluate procedures to establish indicators of appropriateness. Along with the American College of Physicians ACP), BC/BS published a monograph in 1987 of 13 guidelines on the appropriate use of diagnostic tests. BC/BS has also developed the Center for Quality Healthcare to provide technical assistance to individual BC/BS plans, to advance the science of quality assessment, and to collaborate with other organizations on quality projects.

At least 40 medical specialty societies have expressed interest in developing practice parameters. [8] Twenty-four societies have already developed practice guidelines. The AMA is serving as a clearing-house for practice guidelines and appropriateness standards by publishing a compendium. The AMA's House of Delegates has adopted a resolution that mandated that the AMA review and endorse guidelines developed by specialty societies according to criteria under AMA development. The Hospital Association of New York State has recently compiled a comprehensive compendium that serves as a valuable reference source to make it easier for providers to obtain clinical practice guidelines. The Compendium of Clinical Protocols Criteria and Efficacy Research is "primarily an index of clinical practice guidelines on the appropriateness of a given procedure under certain conditions or on the appropriate treatment for a given diagnosis." [7]

Planned protocols and guidelines, along with current research, will add substantially to the growing body of resource materials. For example, the American College of Cardiology (ACC) is currently working on "Guidelines and Indications for Coronary Artery Bypass Graft Surgery." The ACP is continuing its Clinical Efficacy Assessment Project (CEAP), publishing the resulting guidelines in Annals of Internal Medicine. The AMA also continues its Diagnostic and Therapeutic Technology Assessment Program (DATTA). Established in 1982, DATTA reports examine the safety and efficacy of a drug, device, or procedure applied for specific indications. Most DATTA reports are published in JAMA.

A variety of research projects are under way that should yield a broad range of guidelines and appropriateness criteria. The American Heart Association (AHA) and the ACC have bormed a joint Task Force on Assessment of Cardiovascular Procedures. Its mission is to define the role of invasive and noninvasive procedures in the diagnosis and management of cardiovascular disease. [7]

The Department of Health and Human Services has undertaken two initiatives: The National Center for Health Services Research and Health Care Technology Assessment (NCHSR) and the Medical Treatment Effectiveness Initiative. NCHSR is supporting 13 multiyear studies on patient outcomes, with results due in two years. The Effectiveness Initiative Project was begun in 1988 under the auspices of the Health Care Financing Administration (HCFA) and was transferred to the Public Health Service (PHS) in 1989. The initiative has focused on combining large national databases, making the information available to researchers, and developing a Uniform Clinical Data Set. Outside researchers funded by HHS to work on the initiative include the Institute of Medicine, which has been conducting outcomes research, and the American Medical Review Research Center, which has been studying small-area variations in care. [7]

The Institute of Medicine, an arm of the National Academy of Sciences, has been conducting Effectiveness Workshops under the auspices of the PHS Medical Treatment Effectiveness Initiative. The workshops bring together experts to identify key research issues relative to selected conditions. Workshop reports on three conditions--acute myocardial infarction, breast cancer, and hip fracture--will be available shortly. Appropriateness of care guidelines are being established for Vermont hospitals by the Vermont Program for Quality in Health Care. Patient outcome and other risk-adjusted information from Vermont and other states will be included in the databank. [7]

Physicians and hospitals in Minnesota have initiated a program called the "Minnesota Clinical Comparison and Assessment Project" to develop guidelines for diseases and procedures. Existing guidelines from medical specialty societies and others serve as a template for appropriateness and management guidelines to be developed by local consensus panels. The guidelines are distributed through the existing hospital medical staff structure and teaching conferences. Hospital practice data are collected on individual physicians along with performance information, including functional outcome, and then are fed back to individual physicians to permit comparison with peers and establishment of clinical guidelines. The impact of the educational feedback loop is being studied for its effectiveness in modifying highly variable practice patterns. [9]

The traditional concept that physicians can assimilate enough information from journals, textbooks, and teaching conferences to make appropriate use of past and present medical advances is proving to be invalid. Current information is overwhelming; cost benefit analyses have yet to be done for most advances; and few physicians can apply the techniques of decision analysis to existing information. When asked how they would like to receive medical information, a majority of physicians in a national survey preferred a concise format such as practice guidelines. [10]

Thus it would appear that guidelines are increasingly available, federal funds are likely to accelerate the introduction of new guidelines, and clinicians are receptive to the concept of physician-generated practice guidelines. Unfortunately, there is little evidence to support the contention of some advocates that simply making guidelines available will improve the appropriateness, quality, and effectiveness of medical practice. The ACP has developed more than 160 sets of clinical guidelines during the past 12 years through its clinical efficay project, but there is no evidence that the CEAP has had a salutary impact on clinical practice. The impact on clinical practice of more than 200 sets of clinical guidelines produced during the 1980s by 24 medical specialty societies is virtually unknown.

The modest literature that exists to document the effectiveness of practice guidelines is found in the field of obstetrics. The American College of Obstetrics and Gynecologists has issued clinical guidelines promoting vaginal birth after cesarean section. [11, 12] Evaluations of the clinical guidelines have noted their failure to decrease repeat cesarean section rates. [13, 14] However, several hospitals have implemented successful clinical guidelines to reduce cesarean section rates. [15-19] These teaching hospital-based strategies featured uniform criteria for diagnosing fetal distress and dystocia, second opinion requirement, and detailed peer review when a cesarean was performed. It is unclear, however, whether such programs can be successfully implemented in community hospitals, with their less formal hierarchies and possibly greater vulnerability to nonclinical pressures.

Another example of a positive impact of practice guidelines comes from Massachusetts, where the introduction of newly developed standards for basic intraoperative monitoring by anesthetists was associated with a marked reduction in hypoxic injuries. [11] It is noteworthy that physicians had a monetary incentive to use the standards in the form of a premium reduction and that the standards required the installation of a monitoring technology not in common use. Thus, direct financial incentives linked to the implementation of a particular technology distinguished these standards from those issued passively.

The limited number of successful applications of practice guidelines suggests that they be developed through a physician consensus process, be tailored to the needs of a region or a specific institution, and include feedback and/or incentives to individual physicians. The current Minnesota Clinical Comparison Program should help to assess the success of the latter approach, as it uses a local physicians consensus process based on nationally developed standards and provides direct feedback to practitioners on performance relative to peers and practice guidelines. [9]

Further evidence that the mere dissemination of practice guidelines does not materially affect actual practice patterns comes from a recent Canadian study by Lomas et al. They examined the effect of a consensus statement regarding guidelines for cesarean section on the practice of physicians. [20] Post-release surveys revealed that most obstetricians were aware of the guidelines and agreed with them. Moreover, one-third of obstetricians reported changing their practice as a consequence of the guidelines, with significant reduction in self-reported rates of cesarean section in women with previous c-section. However, data on actual practice showed rates of c-section significantly higher than the self-reported rates. Moreover, testing of the physician's knowledge of the specific content of the recommendations showed an average overall score of 67 percent, with only three percent of respondents correctly identifying all eight recommended strategies. The authors note the various barriers tof the successful implementation of practice guidelines, including perceived threats of malpractice litigation, inadequate skills, economic and socioeconomic incentives, and pressure from patients. The authors conclude that, "In the absence of any accompanying strategies to overcome these other influences, the dissemination of practice guidelines issued by a national body is unlikely to have much effect on inappropriate practices that are sustained by powerful nonscientific forces." [20]

In spite of the large number of practice guidelines currently available and the increasing support for standards, there is little evidence that existing guidelines will affect practice patters. Passive distribution of guidelines has not altered inappropriate practice. Those strategies tailored to local conditions, with feedback or incentives for individual practitioners, are most likely to meet the objective of altering inappropriate practice patters. It may also be necessary, as suggested by Brook, to create a national institute whose purpose will be to develop, maintain, update, and test practice guidelines. Such an institute will require the coordination of a full spectrum of disciplines associated with health services research. [1]

References

[1] Brook, R. "Practice Guidelines and Practicing }e icine. Are They Compatible?" JAMA 262(21):3027-30, Dec. 1, 1989.

[2] Wennberg, J., and others. "Use of Claims Data Systems To Evaluate Health Care Outcomes: Mortaility and Reoperation Following Prostatectomy." JAMA 257(7):933-6, Feb. 20, 1987.

[3] Wennberg, J., and Gittelsohn, A. "Variations in Medical Care among Small Areas." Scientific American 246(4):120-9, April 1982.

[4] "Small Area Analysis Final Report." Rutland, Vt.: Rutland Regional Medical Center, Nov. 1986, and Codman Research Group Small-Area Analysis of Hospital Use in Vermont, 1985 and 1986.

[5] "Bistate Small Area Analysis of Hospital Use in New Hampshire and Vermont." Rutland, Vt.: Codman Research Group, 1985 and 1986.

[6] Goyert, G., and others. "The Physician Factor in Cesarean Birth Rates." New England Journal of Medicine 320(11):706-9, March 16, 1989.

[7] Hospital Association of the State of New York. Compendium of Clinical Protocols, Criteria, and Efficacy Research. Guidelines and Ongoing Research. Albany, N.Y.:HANYS, Oct. 1989.

[8] Koska, M. "Moves To Standardize, Monitor Practice Work '89." Hospitals 63(24):48, Dec. 20, 1989.

[9] Schultz, A. "The Role of Medical Societies in Technology Assessment and Development of Clinical Practice Guidelines. Presented to the ACPE Forum on Quality Health Care, Tucson, Ariz., Nov. 15, 1989.

[10] Kanouse, D., and others. Changing Medical Practice through Technology Assessment: An Evaluation of the NIA Consensus Development Program. Ann Arbor, Mich.:Health Administration Press, 1989.

[11] Guidelines for Vaginal Delivery after a Previous Cesarean Birth. Statement of the Committee on Obstetrics: Maternal and Fetal Medicine. Washington, D.C.:AMerican College of Obstetricians and Gynecologists, 1984.

[12] New Guidelines for VBAC: Statement of the Committee on Obstetrics: Maternal and Fetal Medicine. Washington, D.C.: American College of Obstetricians and Gynecologists, 1988.

[13] Gleicher, N. "Cesarean Section Rates in the United States: The Short Term Failure of the National Consensus Development Conference in 1980." JAMA 252(23):3273-6, Dec. 21, 1984.

[14] Shiono, P., and others. "Recent Trends in Cesarean Birth and Trial of Labor in the United States." JAMA 257(4):494-7, Jan. 23-30, 1987.

[15] Porreco, R. "High Cesarean Section Rate: A New Perspective." Obstetrics & Gynecology 65(3):307-11, March 1985.

[16] Myers, S., and Gleicher, N. "A Successful Program To Lower Cesarean Section Rates." New England Journal of Medicine 319(23):1511-6, Dec. 8, 1988.

[17] Peter, J., and others. "Reversing the Upward Trend in the Cesarean Section Rate." European Journals of Obstetrics, Gynecology and Reproductive Biology 25(2):105-13, June 1987.

[18] Gilstrap, L., and others. "Cesarean Section: Changing Incidence and Indications." Obstetrics & Gynecology 63(2):205-8, Feb. 1984.

[19] Meier, P., and Porreco, R. "Trial of Labor Following Cesarean Section: A Two Year Experience." American Journal of Obstetrics & Gynecology 144(6):671-8, Nov. 1982.

[20] Lomas, J., and others. "Do Practice Guidelines Guide Practice? The Effect of a Consensus Statement on the Practice of Physicians." New England Journal of Medicine 321(19):1306-11, Nov. 9, 1989.

Edward C. Geehr, MD, is Senior Vice President for Center Operations, Albany Medical Center, Albany, N.Y., and Richard F. Salluzzo, MD, is Chairman of the Quality Assurance Committee, Albany Medical Center, and Acting Chairman, Department of Emergency Medicine, Albany Medical College. Dr. Geehr is Chair of the College's Forum on Quality Health Care.