Publications
APS Bulletin • Volume 10, Number 6, November/December 2000
Pain Clinic Perspectives
Steven H. Sanders, PhD, Department Editor
Integrating Practice Guidelines for Chronic Pain: From the Tower of Babel to the Rosetta Stone
Steven H. Sanders, PhD
Over the last 5 years, at least partially because of increasing pressure for clinical accountability and effectiveness, six major practice guidelines have been introduced to address the treatment of chronic nonmalignant painful conditions. These included guidelines for migraine headaches (Campbell, Penzine, & Wall, 1999; McCrory, Matchar, Gray, Rosenberg, & Silberstein, 1999a; McCrory et al., 1999b), sickle-cell disease (American Pain Society [APS], 1999), complex regional pain syndromes (Stanton-Hicks et al., 1998), chronic pain in older persons (American Geriatrics Society [AGS], 1998), chronic pain syndrome patients (Sanders et al., 1995; Sanders, Harden, Benson, & Vicente, 1999), and chronic pain in general (American Society of Anesthesiologists [ASA], 1997). A review of these six guidelines indicates that they are a mixture of consensual and evidence-based recommendations directed toward rather specific (e.g., sickle-cell disease) to more general (e.g., nonmalignant chronic pain syndrome patients or older persons) types of pain patients. Most recommended using some type of interdisciplinary treatment team approach. Beyond this there are a number of differences and variations across the guidelines about patient selection and exclusion criteria, use of various types of medications, application of more invasive medical procedures, and the degree of treatment intensity needed to produce a certain level of improvement.
Indeed, trying to determine which practice guidelines to use in which situation with which patients can leave the practitioner confused and frustrated. Furthermore, institutions and agencies looking for established practice guidelines also can be left confused and frustrated due to the lack of integration across the existing ones. Such confusion sets the occasion for misdirected and sometimes contradictory treatment for the chronic pain patient. It also reinforces the insurance industry’s position that chronic pain management is inconsistent and to a large extent ineffective (Mitchell, 2000).
This article highlights some of the reasons for the differences seen across practice guidelines and offers suggestions for possible methods to integrate and use them in a more coordinated fashion. It is not the purpose, and is beyond the scope, of this article to provide an in-depth critique of each set of practice guidelines. For those interested in further comparative analysis, the National Guideline Clearinghouse (Agency for Healthcare Research and Quality, http://www.guideline.gov) offers more information.
Reasons For Differences
Targeted patient population. An obvious source of variation and difference between the six practice guidelines is embedded in the targeted patients for whom they are intended. As already noted, this ranges from more specific conditions like migraine headaches, sickle-cell disease, and complex regional pain syndromes, to much more general guidelines applicable across a variety of chronic painful conditions. Given the different clinical disorders and patient populations, one would expect some reasonable differences across the guidelines. Likewise, the migraine headache and sickle-cell disease guidelines dealt with acute and chronic pain management. This sets the occasion and need for further broadening of possible recommendations and treatment options.
In addition to the differences in pain patients targeted, the definition of chronicity varied across the practice guidelines. Most referenced the International Association for the Study of Pain Committee on Taxonomy report categorizing chronic pain states (Mersky & Bogduk, 1994); however, only those for nonmalignant chronic pain syndrome patients (Sanders et al., 1995; Sanders et al., 1999) indicated a specific 3-month duration for chronicity. The other guidelines were far less specific in defining which patients should be considered chronic. Thus, differences in targeted patients and definitions of chronicity led to variations in recommendations about proper clinical management.
The definition of evidence. In looking over the breadth and depth of literature reviewed across the six guidelines, it seems that thorough literature reviews were done by all. Likewise, the categorization of evidence strength was similar across guidelines. All used the guiding principles laid out by the Agency for Health Care Policy and Research for categorizing the strength of research findings (AHCPR, 1996). Evidence strength ranged from prospective, double-blind, randomized, controlled studies to uncontrolled case reports. Where no evidence could be found, most of the guidelines attempted to use expert consensus in formulating any recommendations. All of the guidelines attempted to grade the strength of their recommendations based on the quality of the research studies. The highest grade was seen for recommendations based on multiple well-designed, randomized, controlled clinical trials showing consistent effects, either reviewed individually or through quality meta-analyses.
If all six guidelines used a similar process to review the evidence, then why was there not more consistency about recommendations? An answer to this question can be found if we examine the definition of adequate evidence across the guidelines. In doing so, it became clear that there were significant differences that led to differences in recommendations for treatment options. The more a given set of guidelines used lower quality evidence (e.g., case studies) and/or expert consensus, the more treatment methods were recommended (ASA, 1997).
The two practice guidelines that applied the most stringent definition for adequate evidence were those for migraine headaches (Campbell et al., 1999; McCrory et al., 1999a, 1999b) and chronic nonmalignant pain syndrome patients (Sanders et al., 1999). Such stringency led to noticeably more conservative treatment recommendations by these two sets of guidelines compared with the others. For example, the most recent chronic pain syndrome patient guidelines (Sanders et al., 1999) do not recommend the use of opioids or invasive procedures with this patient population, citing an obvious absence of evidence to do so. On the other hand, practice guidelines from the ASA (1997) and those of the AGS (1998) make far more liberal recommendations about opioid and invasive procedure usage based on weak evidence (e.g., poorly controlled outcome and case studies), as well as more reliance on expert consensus and common practice.
Thus, the definition of adequate evidence and the willingness to use expert consensual validation became critical factors in determining the formulation of actual recommendations. It is extremely important to define what constitutes adequate evidence and use of expert consensus in a set of guidelines to fully appreciate the empirical strength of recommendations within.
Author bias and special interest group influences. Regardless of the desire to achieve substantially impartial, scientifically based recommendations, guidelines can be flavored by the authors’ clinical and practice biases, as well as underlying special interest groups’ influences. At the commercial level (e.g., pharmaceuticals and medical procedures and devices), recommendations in a given set of guidelines can produce enormous financial gains for the suppliers and healthcare providers. Hence, the implicit and sometimes explicit pressure to endorse a given method, drug, or device can be extreme. Likewise, the authors’ clinical and practice biases are inescapable. As we have all heard, “to a hammer, all the world is a nail.” These factors need to be understood and appreciated when trying to interpret guideline recommendations. They can certainly add to differences seen in recommendations across guidelines.
The road to guideline integration
Given that some of the six guidelines reviewed were disease specific (e.g., headache, sickle cell, and complex regional pain syndromes), the need and feasibility of integration are more limited than for those dealing with general chronic pain across various conditions (AGS, 1998; ASA, 1997; Sanders et al., 1995; Sanders et al., 1999). Likewise, there are some inherent limits to total integration due to those very areas just discussed that produce differences.
In spite of recognized limits, more integration is now possible. For example, the chronic pain syndrome patient guidelines (Sanders et al., 1995; Sanders et al., 1999) could be integrated with the others using the well-documented observation of patient subgroups within the chronic pain patient population (Sanders & Brena, 1993; Turk & Okifuji, 1998). It is empirically and clinically obvious that not all chronic pain patients are chronic pain syndrome patients. Indeed, all of the reviewed practice guidelines, except those for the chronic pain syndrome patients, addressed a broader range of chronic pain patients. At least partial integration is possible if the chronic pain syndrome patient guidelines are viewed as a subset within the others; to be used when a chronic pain patient has been diagnosed with a chronic pain syndrome presentation. Otherwise, those recommendations made by the other guidelines would be applicable for patients not meeting the chronic pain syndrome diagnostic criteria.
Further integration and orchestrated application could be accomplished across the six guidelines if the following principles of guideline construction were consistently applied to current and future versions:
- Clearly define the patient population to which the guidelines are directed, including exclusions and chronicity.
- Use consistent standards for definition of acceptable evidence, with prospective, randomized, controlled outcome studies as the minimum standard.
- Minimize the expert consensus method for making recommendations.
- Minimize author bias and special interest group influence by employing the two previous principles as well as maintaining broad-based interdisciplinary author participation and review, maintaining full public disclosure of guideline preparation funding, and establishing more federally based funding for such guideline preparation.
- Promote and support quality research to clearly establish the efficacy of evaluation and treatment methods delineated in the various guidelines.
- Follow the science more and the money less.
The preceding recommendations to improve the consistency and possible integration of guidelines are far from definitive and should be considered a starting point. The movement towards more evidence-based practice is an important and positive step to improve the overall care for chronic pain patients. Obviously, this must be tempered by compassion and clinical wisdom to maximize the potency of guideline application. I hope this article highlighted and raised the consciousness level of all of us about the critical issues and factors involved in the development of practice guidelines for chronic pain patients. We must focus our attention on maximizing the application and integration of existing guidelines as well as improving future versions by using the concepts outlined herein. Otherwise, the destructive Tower of Babel effect will significantly diminish the value of such practice guidelines for the patients we serve.
References
American Geriatrics Society. (1998). The management of chronic pain in older persons: New guidelines from the American Geriatrics Society. Journal of the American Geriatrics Society, 46, 128–150.
American Pain Society. (1999). Guideline for the management of acute and chronic pain in sickle-cell disease. Glenview, IL: American Pain Society.
American Society of Anesthesiologists. (1997). Practice guidelines for chronic pain management. Anesthesiology, 87, 995–1004.
Campbell, J.K., Penzine, D.B., & Wall, E.M. (1999). Evidence-based guidelines for migraine headache: Behavioral and physical treatments. The US Headache Consortium, American Academy of Neurology. [Online]. Available: www.aan.com.
McCrory, D.C., Matchar, D.B., Gray, R.N., Rosenberg, J.H., & Silberstein, S.D. (1999a). Evidence-based guidelines for migraine headache: Overview of program description and methodology. The US Headache Consortium, American Academy of Neurology. [Online]. Available: www.aan.com.
McCrory, D.C., Matchar, D.B., Gray, R.N., Rosenberg, J.H., & Silberstein, S.D. (1999b). Evidence-based guidelines for migraine headache: Pharmacological management for acute attacks and prevention of migraine. The US Headache Consortium, American Academy of Neurology. [Online]. Available: www.aan.com.
Mersky, H., & Bogduk, N. (1994). Classification of chronic pain: Descriptions of chronic pain syndromes and definitions of pain terms. Seattle: IASP Press.
Mitchell, K. (2000). The dance of the invisible impairments: Chronic pain syndrome and the disability insurer. APS Bulletin, 10, 1–5.
Sanders, S.H., & Brena, S.F. (1993). Empirically derived chronic pain patient subgroups: The utility of multidimensional clustering to identify differential treatment effects. Pain, 54, 51–56.
Sanders, S.H., Harden, R.N., Benson, S.E., & Vicente, P.J. (1999). Clinical practice guidelines for chronic non-malignant pain syndrome patients II: An evidence-based approach. Journal of Back and Musculoskeletal Rehabilitation, 13, 47–58.
Sanders, S.H., Rucker, K.S., Anderson, K.O., Harden, R.N., Jackson, K.W., Vicente, P.J., & Gallagher, R.M. (1995). Clinical practice guidelines for chronic non-malignant pain syndrome patients. Journal of Back and Musculoskeletal Rehabilitation, 5, 115–120.
Stanton-Hicks, M., Baron, R., Boas, R., Gordh, T., Harden, R.N., Hendler, N., Koltzenburg, M., Raj, P., & Wilder, R. (1998). Complex regional pain syndromes: Guidelines for therapy. The Clinical Journal of Pain, 14, 115–116.
Turk, D.C., & Okifuji, A. (1998). Directions and prescriptive chronic pain management based on diagnostic categories of the patient. APS Bulletin, 8, 5–11.
Steven Sanders is director of Siskin Hospital’s Center for Pain Rehabilitation in Chattanooga, TN.