CPGs represent a compilation of the current best practices or standards in the treatment of various disorders. They are meant to be a supplement to physician education, not a "cookbook" for treatment. Every guideline that I have seen emphasizes the role of clinical judgment. Knowing when and how to apply the recommendations in the guidelines requires knowledge and experience. Most CPGs, even algorithms, contain broad outlines of care with multiple choices at each stage of treatment. These choices are under the clinician's control. Encouraging compliance with CPGs simply means encouraging psychiatrists to provide treatment that lies within the broad acceptable standards of care.
Applicability of CPGs
When psychiatrists argue that CPGs have "limited applicability" to their patients, they are usually referring to the inability of guidelines to effectively address the variations in patient responses to treatment. This is a valid and important issue. Although many patients respond to initial treatment with a single medication, a substantial number appear to require a more complex regimen of multiple medications. To be useful, CPGs must address these real-world clinical problems.
One of the readers who responded to my column correctly observed that this limitation is currently being addressed in studies such as the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) and the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD).5-7 The STAR*D, in particular, focuses on the clinical management of patients who have failed 1 or more treatment trials with a single medication. When the results from this and other new research are incorporated into CPGs, they should be able to address more of the clinical variability seen in everyday practice.
It may also be worth studying the types of variability and problems that clinicians face in the treatment of complex cases. Yet, even these types of studies cannot control for all the sources of patient variation. There will always be a role for clinical judgment in adjusting treatment for patients who do not quite fit the guidelines or who do not respond to treatment in a typical manner. CPGs are, as the name implies, guides to clinical care, not laws dictating how treatment should be provided.
CPGs and managed care
Some clinicians believe that managed care and insurance companies encourage clinicians to comply with CPGs as a way of reducing costs—often to the detriment of the patient. This assumption mistakenly equates CPGs with the medical necessity criteria used by managed care companies to control costs; these are 2 entirely different sets of guidelines with very different goals. CPGs can actually run counter to the cost-saving goals of managed care medical necessity criteria because CPGs are primarily focused on the provision of care that is defined to be appropriate by clinical experts in the field. CPGs can (but usually do not) explicitly take the cost of care into account in their recommendations.
Intrusion in practice
Some physicians believe that any attempt to convince them to comply with CPGs represents an intrusion into the doctor-patient relationship. Physicians are also concerned that increased emphasis on CPGs will lead to a loss of professional autonomy.8,9
Given these concerns, it is worth remembering why CPGs were developed. Investigators have repeatedly observed significant and unexplained variations in the type and quality of care that clinicians provide in different areas of the country. Moreover, the number of mistakes made in the provision of clinical care in the United States should be sobering to any clinician.10 The Institute of Medicine has referred to these and other problems in care as the "quality chasm."11 No well-intentioned physician would consciously offer poor care to patients, but we can all be misled by our biases and limited knowledge in some areas. CPGs attempt to address these problems by outlining the best medical practices and acceptable parameters of care.
One way to think of a CPG is as a set of treatment principles that fit a significant proportion of patients (perhaps 90% of those with a specific disorder). The remaining patients have a less typical response to treatment. Given this premise, the majority of a physician's patients should fall within the parameters outlined in a specific CPG. A smaller proportion will have to be treated in a manner that deviates from the guidelines. Most physicians will probably discover that they already treat the majority of their patients according to the relevant guidelines. If they find themselves consistently disagreeing with the guidelines, they might want to examine their reasons for doing so.