Clinical Practice Guidelines in Psychotherapy Research Paper

1. The History And Context Of Practice Guidelines

The most commonly accepted definition of a practice guideline is ‘systematically developed statements to assist practitioners and patients make decisions about appropriate health care for specific clinical circumstances,’ (Noonan et al. 1998). In this sense, practice guidelines have been developed and implemented informally at the agency and clinic level for years. Names such as treatment algorithms, clinical protocols, and practice standards have been used to describe such efforts. In contrast to contemporary practice guidelines, these early efforts were highly localized, often relied on clinical lore or the influence of a few key opinion leaders, did not systematically incorporate scientific evidence about the efficacy of clinical interventions, and addressed only the most controversial problematic aspects of treating a patient with a particular condition. For example, many agencies and clinics have guidelines for assessing an acutely suicidal patient during the index crisis, primarily for the purpose of managing risk. At the same time, few guidelines have ever been developed for treating suicidal behavior when it is not of crisis proportions, even though the bulk of outpatient treatment occurs with less acutely suicidal patients. Due to their parochial nature, early practice guidelines—some developed only miles apart and for the same clinical condition—differed markedly in their emphasis and clinical recommendations.

The resurgence of interest in practice guideline stems from the growing influence of outcomes-based medicine and the associated philosophies of population and evidence-based care (Noonan et al. 1998, Strosahl 1994). Based on a public health model, population-based care is concerned with the health status of the entire population, not just those patients with current psychological disorders. The goal is to build consistent processes of care that reduce variation in clinical outcomes for the population served, thereby increasing the cost effectiveness of the overall system of care. A major area of concern with both medicine and behavioral health services is that systems of care vary wildly in the types of services available and the services rendered to treat a particular condition. The goal of population care is to narrow this variability across systems and providers, ensuring better access to, and quality of, care for consumers.

The related notion of evidence-based care is noteworthy because it suggests that the best outcomes across a population are achieved by the consistent use of interventions that have been shown to work in randomized clinical trials. Many clinical experts and the purchasers of health services argue that there is excessive variation among behavioral health providers in the types and amounts of treatment that are delivered for common psychological problems. In discussing the decision-making behavior of physicians, it has been suggested that only 25 percent of health services are objectively based upon the medical status of the patient. Primary nonmedical factors related to health services include availability of needed services, patient’s beliefs and preferences and the physician–patient relationship. Most providers are ill-equipped to meet the diverse array of psychosocial and medical needs posed by their patients. This ‘mismatch’ between what the provider can offer and what the patient expects is a major source of cost overruns in both health and behavioral health.

2. Types Of Guidelines

Broadly conceived, there are two types of practice guidelines. The system guideline is widely used (Noonan et al. 1998). It addresses the parameters of population-based care by specifying who is eligible to enter the system of care (i.e., medically necessary treatment), what setting of care is appropriate (i.e., inpatient, residential, partial hospital, out-patient) and the program to which a patient with a particular set of symptoms and functioning should be assigned (i.e., vocational rehabilitation, drugs and alcohol, dual diagnosis). Note that a system guideline functions to define who is eligible to receive services and what le el of care is appropriate to address the patient’s condition. It does not attempt to guide the decision making or practice patterns of providers within that level of care. However, system guidelines can and should be evidence based, evolving from the scientific literature concerning the type and intensity of treatment mostly likely to work for a patient with a given set of symptoms and specified functional status. System guidelines are currently the focus of a major developmental initiative by the Substance Abuse and Mental Health Services Administration of the United States Government. System guidelines can promote the goals of population-based care by specifying and ‘minimum services menu’ that is needed to properly address the diverse mental health needs of a population. System guidelines help identify gaps in programs or facilities that can then be brought to the attention of policy makers, funding agencies, and foundations. One of the chief criticisms of clinical practice guidelines by providers is that they recommend settings of care or types of programs that are not accessible. While system guidelines often receive less discussion, they are a critical component of a comprehensive population care system.

Clinical practice guidelines are a second and more widely discussed entity. Ideally, they weave scientific evidence into clinical practice (Clinton et al. 1994). There has been an unprecedented proliferation of clinical practice guidelines in the beginning of the twenty-first century. Some are discipline specific such as the series of practice guidelines published by the American Psychiatric Association (Herz et al. 1997). Others, such as the Agency for Health Care Policy Research Guideline for Treating Depression in Primary Care, focus on treatment recommendations for a particular care setting, such as primary care (Depression Guideline Panel 1993). Some guidelines, such as those published by the Academy of Adolescent and Child Psychiatry are cross-disciplinary and designed to be applied across settings (Bernstein et al. 1997). Managed behavioral health care companies have also developed proprietary guidelines as a mechanism for supporting utilization reviewers. Finally, there are a growing number of ‘off the shelf ’ guideline packages that have been developed privately and marketed to the provider community. The promulgation of clinical practice guidelines is promising in that it testifies to the growing conviction among policy makers, administrators, and clinicians that using scientific evidence to guide clinical practice is the only way to assure consistent quality. However, there is growing concern that many existing practice guidelines are biased toward truncating care, promoting guild interests, or simply lacking scientific validity (Nathan 1998).

3. Functions Of A Clinical Practice Guideline

Properly constructed, a practice guideline should provide easily accessed, scientifically and clinically sound decision support that can aid the provider and the consumer at key decision points in the assessment and treatment process (Practice Guideline Coalition 1997), including:

(a) critical symptoms and/or functional attributes that make a patient eligible to be served under that guideline (i.e., diagnostic guidelines) and severity criteria that determine suitability for service setting (i.e., levels of care guidelines);

(b) conditions and/or syndromes that have similar symptom patterns and methods for distinguishing these conditions (differential diagnostic guidelines);

(c) relevant epidemiological and demographic in-formation that helps providers identify different forms of clinical presentation related to geographic, racial, ethnic, or cultural factors;

(d) clinical procedures shown to be effective with the condition, rates of remission when such procedures are properly administered, comparative effectiveness of different treatments, treatment side effects, and other known risks as well as the expected time frame for clinical functional response;

(e) commonly available and easily used self-report and or interview methods for assessing clinical response, criteria for determining whether a procedure(s) is working and, if not, guidance in selecting an alternative procedure or treatment; and

(f) relapse rates and maximum relapse risk periods, effective maintenance, and relapse prevention strategies.

Note that the concept of assisting the provider and consumer in selecting an appropriate intervention for a specific condition suggests that guidelines need to exist in a format that can support use by both the provider and consumer. In addition to the core clinical function of improving the quality and appropriateness of care to the individual consumer, the Institute of Medicine (Field and Lohr 1990) has identified other guideline functions, including:

(a) decrease service costs produced by the provision of ineffective and/or unnecessary services,

(b) assist consumers and their families in advocacy efforts,

(c) assist payers in establishing rates and mechanisms for reimbursement,

(d) encourage the collection of standardized data for the purpose of ‘validating’ a guideline, and

(e) improve formal and post-graduate clinical training.

4. Methods For Constructing Practice Guidelines

As mentioned previously, early methods for constructing practice guidelines used either an informal or formal consensus method. Whereas informal consensus might involve a group of providers agreeing to a care protocol at a lunchtime meeting, formal consensus methods require a structured development and ratification process (Woolf 1992). In the Delphi method, guideline development representatives determine rules about collecting and analyzing opinions from experts and distilling results into a set of recommendations. In iterative fashion, experts provide their opinions, then respond to the opinions of other experts until a convergence of opinion is obtained. In the Nominal Process method, a facilitator systematically elicits opinions from each group participant about the most important recommendations on each participant’s list. This process continues until all recommendations from all lists have been cataloged, at which point the recommendations are distilled into a composite list. Each participant then rates this list in terms of priority, areas of maximum agreement are identified, and the resulting ‘consensus’ is put back to the group for ratification. The primary advantage of consensus-based methods is that guidelines can be rapidly generated, often ahead of the existing science. Innovations in clinical practice that have not been scientifically tested are more likely to be incorporated. The major, and often fatal, drawbacks are that the quality of consensus guidelines are entirely tied to the professional skills, qualifications, and biases of the group, tend to omit controversial topics where clinicians need the most guidance, produce bland recommendations that provide no advancement in clinical practice, or institutionalize locally favored practices that in fact are scientifically unsound. One of the best-known consensus-based guideline projects is the Expert Consensus Consortium, which has used the Delphi method to develop a wide variety of practice guidelines that are available nationally.

The limitations in consensus models for building guidelines have created a corresponding interest in using more objective, data-based methods. In contrast to consensus models, empirically-based methods stress that the quality of a guideline is directly related to its scientific underpinnings. For many years, the US Preventive Services Task Force has used the evidence-based guideline method. This approach involves conducting a comprehensive review of all scientific studies and reports pertaining to the condition under consideration, rating the amount and quality of the evidence, and then producing graded recommendations based upon the strength of the underlying scientific evidence. A ‘Grade A’ recommendation is based upon a number of well-conducted studies that meet specific and rigorous methodological criteria (i.e., sample size, randomization, well-specified, replicable interventions). In contrast, a ‘Grade C’ recommendation would have only limited evidence to support it, but still be seen as clinically important. Strict adherence to evidence-based methods tends to produce many fewer recommendations that are based on ‘expert opinion.’ However, providers and con- sumers can be left with no guidance in very core areas because the science is not available, and this may limit the clinical usefulness of the resulting product. Another frequent criticism of evidence-based methods is that cutting-edge innovations in clinical practice are left out of guidelines because scientific validation of effective procedures may lag years behind the introduction of new procedures in a field.

Eddy (1996) proposed an Explicit Guideline Development method that involves conducting an analysis of condition base rates within a system of care, along with costs and outcomes associated with existing care processes. Using cost modeling data derived from a review of scientific data regarding alternative procedures, a balance sheet is created which outlines the projected clinical benefits, costs, and potential risks associated with implementing a guideline. Generally, this approach is designed to identify conditions that occur with a high enough frequency and/or associated cost to justify changes in a whole system of care. In essence, this method helps one determine whether a particular condition is appropriate for a clinical practice guideline. Normally, systems and agencies that employ the initial decision-making method will employ an evidence-based method for constructing the actual guideline.

5. Criticisms Of Practice Guidelines

Considering that guidelines vary greatly in both structure and content, it is interesting that a consistent set of criticisms have been forthcoming from key stakeholders (Noonan et al. 1998, Practice Guideline Coalition 1997).

Consumers have expressed concerns that guidelines are used to artificially limit care, are developed with little or no consumer involvement, ignore, and/or minimize consumer preferences and limit consumer access to information critical to the treatment decision-making process.

Clinicians complain that guidelines are rigidly applied to limit services and reimbursement, are too complicated to be useful in daily practice, recommend procedures or services that the provider is not trained to administer, neglect the important role of clinical judgment, function to suppress innovative clinical practices, and promote guild or discipline specific interests.

Scientists claim many guidelines incorporate unsubstantiated expert opinion, improperly characterize or ignore the scientific literature, recommend costoriented interventions that produce short-term benefits only, and are not kept current with recent advances in science.

Managed health care executives complain that guidelines cost too much and take too long to develop, exceed the skill and training level of most providers, are too guild oriented, are not ‘user friendly’ in daily practice, and ignore the issue of cost effectiveness.

Concerns about the varying quality of practice guidelines have led to several efforts to identify criteria for judging the quality of a guideline. The American Psychological Association (1995) has developed standards for evaluating the validity of guidelines. Generally, these standards require that guidelines incorporate valid and reproducible scientific data, be written clearly, recommend procedures that have proven clinical effectiveness in the field, allow for flexible decision making, provide clear and precise recommendations, and be ‘refreshed’ regularly to keep up with changes in the scientific literature. In addition, the guideline development process should be multidisciplinary and well documented.

6. Impact Of Practice Guidelines

At this point in time, there are few scientific studies concerning the impact of clinical practice guidelines for behavioral health conditions. Most reports are anecdotal and suggest that properly implemented, guidelines can alter the process of delivering care particularly if provider involvement is sought. Much less is known, even at the case report level, about the extent to which practice guidelines improve clinical outcomes.

Studies of practice guidelines for physicians have shown that dissemination of a guideline is not an end in itself. Training in use of the guideline is often required, as is the ‘translation’ of the guideline to fit local resource and practice realities. In addition, physicians may benefit from concurrent feedback and enforcement mechanisms. Even under the best of implementation circumstances, physicians may show enhanced awareness of the guideline, yet fail to modify their medical practices (Lomas et al. 1989). In essence, medical practice guidelines have not been consistently shown to alter the process of care, improve health outcomes, or improve consumer satisfaction (Grimshaw and Russell 1993). This may be due to variations in the quality of practice guidelines, differences in dissemination and training methods, or reflect the immediate financial incentives influencing physician behavior.

7. Future Developments

It is clear that the next generation of practice guidelines will need to be less cumbersome to use, less expensive and time consuming to develop, and reflect a consensus of all the key stakeholders as to what constitutes scientifically sound, feasible, cost-effective treatment. The Practice Guideline Coalition (PGC) represents one such effort to bring the key stakeholders together to form scientifically sound, user-friendly practice guidelines. The PGC is a coalition of approximately 80 scientific, clinical, managed care, and consumer groups, who have developed a common framework for the development, evaluation, and dissemination of practice guidelines. PGC guidelines have a very short development cycle, incorporate the views of scientists, clinicians, managed care representatives, and consumers, focus on core procedures that can be delivered by any qualified mental health provider and are designed for easy access in the context of daily practice (PGC 1997).

A second future theme will be the development of empirically validated methods for disseminating guidelines to the practitioner community. This very likely will involve the creation of training methods that are affordable and incorporate the results of early studies of guideline dissemination. The existing literature clearly suggests that it is one thing to put a guideline in front of a practitioner. It is quite another to have it impact the process or outcome of care. Dissemination and training represents perhaps the greatest challenge to the success of practice guidelines as a quality improvement tool.

Finally, there will be increasing pressure to scientifically validate practice guidelines in the field. The assumption underpinning interest in practice guidelines is that they will reduce unexplainable variations in the type of behavioral health care that is delivered, thus improving the overall quality. As suggested in the previous section, the evidence supporting this assumption is mixed, although the conceptual stance is intuitively appealing. Nevertheless, guidelines must be evaluated in field-based randomized studies that examine their impact on both processes and outcomes of behavioral health care. There is a need to demonstrate that guidelines produce their intended impact. Otherwise, the behavioral health community may be compelled to look for other methods for improving the quality of their care.

References:

1. American Psychological Association Task Force 1995 Template for Developing Practice Guidelines: Interventions for Mental Disorders and Psychological Aspects of Physical Disorders. American Psychological Association, Washington, DC
2. Bernstein G A, Shaw K, Dunne J E 1997 Practice parameters for the assessment and treatment of children and adolescents with anxiety disorders. Journal of The American Academy of Child Psychology 36: S49–S84
3. Clinton J J, McCormick K, Besteman J 1994 Enhancing clinical practice: The role of practice guidelines. American Psychologist 49: 30–3
4. Depression Guideline Panel 1993 Depression in Primary Care: Detection, diagnosis and treatment. (Clinical practice guideline No. 5, AHCPR publication No. 93–0553.) US Department of Health and Human Services. US Public Health Service. Agency for Health Care Policy and Research, Rockville, MD
5. Field M, Lohr K (eds.) 1990 Guidelines for Clinical Practice: From Development to Use. Institute of Medicine Committee on Clinical Practice Guidelines. National Academy Press, Washington, DC
6. Eddy D 1996 Clinical Decision Making: From Theory to Practice. A collection of Essays from SAMA. Jones & Bartlett, Boston
7. Grimshaw J M, Russell I T 1993 Effect of clinical guidelines on medical practice: A systematic review of rigorous evaluations. Lancet 342: 1317–22
8. Herz M I, Liberman R P, Liberman J A 1997 Practice guideline for the treatment of patients with schizophrenia. American Journal of Psychiatry 154(4): 1–63
9. Lomas J, Anderson G M, Dominick-Pierre K, Vayda E, Enkin M, Hannah W 1989 Do practice guidelines guide practice? The effect of a consensus statement on the practice of physicians. New England Journal of Medicine 321: 1306–11
10. Nathan P E 1998 Practice guidelines: Not yet ideal. American Psychologist 55: 290–9
11. Noonan D, Coursey R, Edwards J, Frances A, Fritz T, Henderson M, Krauss A, Leibfried T, Manderscheid R, Minden S, Strosahl K 1998 Clinical practice guidelines. Journal of the Washington Academy of Sciences 85: 114–12
12. Practice Guideline Coalition 1997 Mission Statement and Framework for Developing Practice Guidelines. Available from Practice Guideline Coalition, Department of Psychology 296, University of Nevada-Reno, Reno, NV. 89557
13. Strosahl K 1994 Entering the new frontier of managed health care: Gold mines and land mines. Cogniti e and Behavioral Practice 1: 5–23
14. Strosahl K 1995 Three ‘gold-mine land-mine’ themes in generation of managed care. Behavior Therapist 18: 52–54
15. Woolf S H 1992 Practice guidelines: A new reality in medicine, Part II. Methods of developing guidelines. Archives of Internal Medicine 152: 946–52