Of Guidelines and Gridlines
By Arthur Croft, DC, MS, MPH, FACOCertainly we can all agree that peer review is an important form of checks and balances. In the course of their peer reviews and patient assessments, file reviewers and IME doctors have developed a number of strategies for reigning in or limiting treatment of cervical acceleration/deceleration (CAD or whiplash) patients. In my experience, these strategies can be roughly classified into two categories: 1) fair and reasonable; and 2) disingenuous and unreasonable. In the latter category, three documents have become veritable Swiss Army knives of the slash-and-deny technique. This article looks at the misuse of guidelines in the peer review process as it pertains to CAD injuries.
In my experience, three guidelines are the most commonly misquoted and/or misinterpreted: 1) the Mercy guidelines; 2) the AHCPR guidelines; and 3) the Quebec Task Force on WAD guidelines.
The Mercy Guidelines
On page IV of this document under the "General Disclaimer" heading is the following statement:
"These guidelines, which may need to be modified, are intended to be flexible. They are not standards of care. Adherence to them is voluntary. The Commission understands that alternative practices are possible and may be preferable under certain clinical conditions. The ultimate judgment regarding the propriety of any specific procedure must be made by the practitioner in light of the individual circumstances presented by each patient."
Just below that, it reads:
"This document may provide some assistance to third-party payers in the evaluation of care, but it is not itself a proper basis for evaluation. Many factors must be considered in determining clinical or medical necessity. Further, guidelines require constant re-evaluation as additional scientific and clinical information becomes available."
In chapter 8 ("Frequency and Duration of Care"), the area of these guidelines providing the basis for most of the more egregious misinterpretations I have seen, it reads (page 117):
"Guidelines concerning the treatment plan should be tempered with a balance of scientific information and systematic observation derived from clinical experience. Further, in order to be practical, they must be periodically upgraded to reflect advances in the ever-changing knowledge database. Their purpose is to assist the clinician in decision-making based on the expectation of outcome for the uncomplicated case. They are NOT [their emphasis] designed as a prescriptive or cookbook procedure for determining the absolute frequency and duration of treatment/care for any specific case."
They go on to note that:
"No attempt has been made to select for individual conditions by region of complaint or by diagnosis ... The majority of quantitative information available addresses the management of low-back [sic] and leg-pain complaints ... The references to low-back disorders in this section are used only as examples. There is no intent to imply that these conditions constitute the totality of chiropractic expertise or practice. Rather, since these recommendations were born from experience and from data on multivariate clinical circumstances, they may be extrapolated with appropriate case-specific modifications to most of the common complaints for which chiropractic care is sought."
On the same page, the authors go on to state:
"The approach to the development of guidelines for chiropractic quality assurance and standards of practice pertaining to the frequency and duration of treatment focuses on the uncomplicated case and logically includes the following considerations: 1) The natural history of common spinal disorders; 2) The characteristics and stages of tissue repair processes; and 3) Reasonable treatment/care outcome classified into short- and long-range goals."
On this page, and under the heading "Principles of Case Management," the authors note:
"The primary missions of health care delivery are to provide sufficient care to restore health, maintain it, and prevent the recurrence of injury or illness ... guidelines framing expectations of treatment outcome can be drawn from the literature and adapted by practical experience on a case-by-case basis."
Summary of the Mercy Document
As mentioned earlier, the literature review used by the authors was limited primarily to low-back research. For example, figure 8-1, which appears on page 128 of the guidelines, depicts the recovery of acute back-pain episodes as reported by John Triano, DC, MA. At the six-week mark, 100% recovery has occurred. Not surprisingly, this detail has spawned numerous misguided assumptions that CAD-injured patients will also achieve 100% recovery within six weeks.
In truth, the Mercy guidelines do not specifically consider CAD injuries in any section. Is it reasonable to extrapolate from acute, mostly industrial low-back strains all the way to a biomechanically and anatomically distinct structure such as the neck? And to simultaneously extrapolate from lifting and bending injuries to the very unique pathomechanics of CAD injury? No. Nor was this the intention of the authors. If this was possible, a logical assumption then would be that the recovery statistics of the two injuries are similar. In fact, they are not. Of the more than 35 outcome studies of CAD injury published in the past 32 years, the mean proportion of chronicity in these studies is 39% for mixed vector crashes and 56% for rear-impact vectors. Clearly, 100% recovery has never been reported in any published outcome study, although most such studies have not looked at patient populations treated solely by chiropractic paradigms.
Moreover, many risk factors for poor outcome in CAD have been verified in clinical and epidemiological studies. These include the following: 1) female gender; 2) non-awareness of impending collision; 3) head turned at impact; 4) pre-existing headaches, neck pain, or previous neck injury; 5) use of shoulder harness and seat belt; 6) being struck by a larger vehicle; 7) advanced age; 8) pre-existing spinal disease; and 9) poor head-restrain geometry. Any of these might lead us away from the uncomplicated category mentioned earlier and thus require longer duration care, as is allowed by the Mercy document. Multiple risk factors in particular will complicate the healing process.
If, as suggested by the Mercy authors, we consider the scientific literature on expected outcomes in the case of CAD, we find that the outcome is not particularly rosy. And if, as further suggested, we consider the healing time of tissues, we find that authors of animal studies of ligament and tendon injuries have reported a remodeling process that can exceed one year. Some would argue that chiropractic management on some regular basis would be prudent throughout this period when spinal ligaments are remodeling.
And of course, it is the individual patient that always determines the need for care and its frequency and duration. The authors of the Mercy document are to be congratulated for their monumental and thankless hours devoted to the betterment of this profession, despite the undeserved upbraiding they have occasionally received at the hands of those lacking vision and insight. Nevertheless, when it comes to CAD, the Mercy document cannot be applied in anything more than very general terms. It does not provide treatment recommendations for frequency and duration of care. It does, however, allow for treatment as long as it can be shown to "provide sufficient care to restore health, maintain it, and prevent the recurrence of injury or illness." Needles to say, medical records must reflect this need.
AHCPR Guidelines on Low-Back Pain
On several occasions, I've read depositions in which chiropractors, acting as IMEs or expert witnesses, referred to "federal guidelines" which allegedly applied to CAD injuries, noting that the guidelines had been exceeded by the treating doctor. The "federal guidelines" they were referring to were the AHCPR (Agency for Health Care Policy and Research) Guidelines on Low Back Pain. This misrepresentation is even more disingenuous than those suggesting that one can determine a treatment protocol from the Mercy document. The AHCPR guidelines, after all, are titled, Acute Low Back Problems in Adults. The authors point out that they are not intended for use in children or in adults with chronic low-back pain. Needless to say, they are also not intended to be used as guidelines for the treatment of CAD injuries.
Summary of the AHCPR Guidelines
The AHCPR guidelines do not attempt to define treatment parameters or guidelines for CAD injuries. Extrapolation is not possible in any meaningful or scientific way.
The Quebec Task Force on Whiplash-Associated Disorders (WAD)
Published in April of 1995 in the journal Spine, these guidelines have gained wide currency (and some criticism) on the international scene. Readers who have followed my work know that my colleagues and I have been critical of these guidelines for several reasons. Our work on this subject is scheduled to be published in Spine in May 1998, and I will not prorogue the debate other than to deal specifically with two issues: 1) the QTF cohort; and 2) the actual guidelines for treatment as recommended by the QTF-WAD.
The cohort, which suffered from selection bias, was followed only to collect data on a return to usual activities. The QTF referred to this return as "recovered." No information was collected about whether patients continued to be symptomatic or whether they continued to need care. Thus, this study, which reported a 97% recovery after one year, actually only reported the proportion who had returned to work and cannot be used as a true outcome study.
The guidelines for treatment are not specific, but allow "short-term" chiropractic care in CAD grades II and above. No operational definition of "short-term" was provided. Moreover, these guidelines only apply to patients who have not yet "recovered," i.e., returned to work or school.
Summary of the QTF-WAD Guidelines
If the patient is back at work or school, one must rely on another guideline to determine whether his or her treatment is reasonable. The QTF-WAD guidelines offer no help to experts or peer reviewers struggling with the problems of frequency and duration of care. However, there is such a guideline.1 These guidelines are based on a grading system published earlier.2 Fortunately, the QTF's grading system describes the same pathology and is thus interchangeable. Maxima of treatment durations are as follows: Grade I -- <10 weeks; Grade II-- <29 weeks; Grade III -- <56 weeks; Grade IV -- may require permanent care on a prn basis. The maxima for total number of treatments are <21, <33, <76, and possible permanent prn care, respectively. While these guidelines have been adopted by some organizations and associations and criticized by others, they are, nevertheless, the only widely published CAD guidelines and are based on actual patient data and experience. Moreover, they more accurately reflect the scientific literature currently available and the stated spirit of both the Mercy and AHCPR guidelines.
As with any guidelines, these assume that the patient's response to care is the best measure of the need for care, and that complicating factors may increase the need for care. These guidelines are also not intended as recommended treatment plans or prescriptions for care; many patients will recover earlier than allowed by the guidelines. They also allow clinicians to determine their own clinical efficiency and, in some cases, to suspect that occult lesions may be present. Some patients may require upgrading or downgrading as more clinical or laboratory information becomes available.
Inveterate defense experts (IMEs and file reviewers) have become adroit in the manipulation of more than just spines. Their practice of disinformation and misuse of existing guidelines has resulted in a devolution in the credibility of and reliance upon these well-intentioned guidelines: a high price to pay for providing the mere pretext for some rather unsavory and doctrinaire methods of care cutting. As an example, it is a common practice to arbitrarily allow 12 weeks of active treatment for all CAD injuries, regardless of the clinical circumstances. After the precise calendar date has been reached, no more care is authorized by the file reviewer. This technique, as ethically dubious and unscientific as it is, allows these experts to burnish their reputation with their employers. It will, I suspect, become the epiphany for these reviewers when these employers finally realize that any high school graduate can count off 12 weeks on a calendar, or when another expert points out this mischief in a more public forum, such as a court case.
From my experiences in teaching postgraduate courses, I would estimate that only about two percent of the profession has actually carefully read these guidelines. Thus, they fall easy prey to these unscrupulous "experts" who seem to carry on with virtual impunity.
Reasonable and fair peer review requires a serious look at an individual patient, his/her complaints, and the physical and laboratory findings, along with a consideration of risk factors and complicating features. The consanguineous marriage of statistics and guidelines -- in the vacuum of clinical information -- provides nothing more than an example of a wrong question inviting an irrelevant answer. In truth, we are not likely to know the right question until we are closer to knowing the answer. In the meantime, we do have guidelines which, like science, are thankfully self-correcting over time.
Arthur C. Croft, DC, MS, FACO
Director, Spine Research Institute of San Diego
San Diego, California
Click here for more information about Arthur Croft, DC, MS, MPH, FACO.