Interview with Internist Roger Chou, MD, Lead Author of ACP/APS "Integrative" Pain Guideline
Written by John Weeks
Therapeutic Order: An Interview with Roger Chou, MD, Lead Author of the ACP/APS "Integrative" Pain Guideline on the Role of Non-Pharmacological Approaches
Summary: Analysis of the
new "integrative" pain guideline published by the American College of
Physicians and the American Pain Society provoked questions. The guideline notes spinal manipulation, acupuncture,
massage, exercise, yoga, cognitive-behavioral approaches. But they are
mentioned last, with an asterisk: only these were tagged as "weak
recommendation." What is the recommended therapeutic order, here? I contacted internist Roger Chou, MD, the lead author
of the guideline, at his base at Oregon Health and Science University. Chou commented candidly on the panel's thinking about these "non-pharmacological" approaches in the guideline's development.
Send your comments to
for inclusion in a future Your Comments forum.
Roger Chou, MD, lead writer for the ACP pain guideline
Roger Chou, MD, is a general medicine internist at Oregon Health & Science University. He is also director of clinical guidelines development for the American Pain Society(APS). In that capacity, Chou ran point on the October 2007 roll-out of a joint back pain guideline, endorsed by APS and the American College of Physicians (ACP).
http://www.annals.org/cgi/content/full/147/7/478
For those in the integrative medicine community, the guideline is an intriguing document. (See related Integrator article which
describes the guideline's components.) All of the licensed complementary and
alternative medicine disciplines claim value in addressing pain. When
mainstream players commence their initial exploration of integrative
approaches, they often choose to focus on pain conditions. In the last
month, the Integrator reported two of these. One was a unique acupuncture-pain initiative engaged by Ford Motor Company. The second is a Medicaid integrative therapies pilot project for chronic pain in Florida which involved licensed acupuncturists and licensed massage therapists.
Non-Pharmacological Approaches Noted
in the Guideline
An earlier Integrator article published details of the ACP/APS guideline, together with some analysis and commentary by Lou Sportelli, DC and Christine Goertz, DC, PhD. Both acknowledged that it was positive to see the non-pharmacological approaches integrated. But Sportelli suggested that the guideline represented a step backward and Goertz questioned why the non-pharmacological approaches (see sidebar) seemed
to be evaluated under a double evidentiary standard.
I called Chou to gain some
clarity. The perspectives Chou presented on the role of complementary
medicine merited more than a couple of quotes in the other article. So, here is the Integrator interview with Roger Chou, MD.
_______________________________
Integrator: A report on the guideline said that it was
the work of a multi-disciplinary team which began in 2006. Yet it was all medical doctors who were authors of the paper.
Chou: We had 23 people on the panel. It was a very diverse
group. We had a chiropractor, a nurse, an integrative medicine expert.*
The guideline we produced was for the American College of Physicians, so the
authors were all members and so they were all medical doctors.
Integrator: One issue
particularly raised a question for me. In all cases in the guideline, the
evidence on which your team made its call was presented as "moderate
quality evidence." Yet in all cases but one, this finding was coupled
with a "strong recommendation." The exception was the portion of the
guideline on non-pharmacological approaches, which only received a "weak recommendation." This read like
a double standard.
Chou: Making these determinations can be very tricky. We
want to be transparent. Each determination refers to multiple bodies of
evidence. In the non-pharmacological category, we are talking about
literature on many different types of therapies, from exercise to
manipulation to massage and yoga and acupuncture. In yoga, there was
only one study. With acupuncture, we felt there was good evidence, but
often the effect in the studies was no better than that with sham acupuncture. Was the effect only placebo? Even exercise was only associated with relatively small benefits. We only gave a "strong
"In the case
of the non- pharmacological approaches, we felt that one or more new,
high-quality trials with negative outcomes could shift the
recommendation for one or more of the interventions mentioned. So overall it was graded as a weak
recommendation."
- Roger Chou, MD
recommendation" where
we felt certain that the evidence would stand up over time and there were at least moderate benefits, which we defined pretty generously - 10 to 20 points on a 100 point pain or function scale. In the case
of the non-pharmacological approaches, we felt that one or more new, high-quality trials with negative outcomes could shift the recommendation for one or more of the interventions mentioned. So overall it was graded as a weak
recommendation. Integrator: I interviewed one of my advisers, Christine Choate, DC, PhD, about the guideline. She's a former program officer at theNIH National Center for complementary and Alternative Medicine. She commended your work, but raised the same question I did about the "weak recommendation." She referred to all the research on chiropractic and suggested that, evaluated alone, it would have met the bar. Might the guideline have been better if the individual approaches that are lumped together had been separated out and individually evaluated?
Chou: The guidelines have to be succinct and user-friendly, otherwise they won't be used. When writing guidelines, we constantly struggle with the issue of providing enough information without having a product that will be too long for the average, very busy clinician to use. The method used by ACP's guideline group is not to have separate grades for every component of a recommendation. However, if you look at the articles published in Annals of Internal Medicine, there are tables showing how we graded evidence and magnitude of benefit for each intervention we evaluated. It's there for people who are interested in looking into the evidence a bit more. Integrator: I was surprised to discover in your separate paper
that specifically focused on non-pharmacological approaches that the
conclusion appeared to be less supportive than the guideline. You
and your co-author concluded that there is good evidence of moderate
value for chronic low back pain with cognitive-behavioral therapy,
exercise, spinal manipulation, and interdisciplinary rehabilitation,
and that for acute the only good evidence of efficacy is superficial
heat. No mention here of acupuncture or massage or yoga.
Chou: The evidence-based process evaluates evidence for all interventions with an open mind. The evidence for acute pain is pretty weak
for nearly all interventions other than medications, which provide symptomatic relief. This isn't surprising because the natural history of acute low back pain is for the vast majority of patients to improve substantially over the first few weeks no matter what you do. For chronic low back pain, we highlight areas with good quality evidence in the conclusion of the article. But if you read the results section of the abstract, and the article itself, you will see that acupuncture, massage and yoga all received "fair" grades. This is because the evidence is not as strong for those interventions. For yoga there is really only one high quality trial. For acupuncture there is very inconsistent evidence about whether acupuncture is more effective than sham acupuncture. Integrator: The surprises me. Usually the conclusions on
chiropractic are more positive for acute. The 1994 guideline from the
Agency for Health Care Policy and Research had spinal manipulation and
watchful waiting as the top recommendations and this was specifically
for acute low back pain.
"There aren't any trials directly comparing
non- pharmacologic therapies to pharmacologic therapies that
consistently show that people do better with the non-pharmacologic
therapy."
- Chou.
-
Chou: I am not sure how they made that finding at that time. Even now there are only a handful of trials suggesting that spinal manipulation is effective for acute low back pain, and the benefit is pretty marginal - about 10 points on a 100 point scale. Methods for developing guidelines have changed a lot since 1994 and take into account many more factors including consistency between studies and magnitude of benefits. People didn't pay much attention to these factors in the past, but all guideline development groups today agree that they are important. Integrator: The media has been full of concerns with the adverse effects for pharmaceuticals, especially since the Institute of Medicine studya half-decade ago. There is mounting evidence that we are seeing
increasing abuse and harm from pain medications, especially since the
Joint Commission pain guideline in 2001. Did the panel factor in these
adverse effects?
Chou: Whenever we evaluated a therapy, we performed a benefit-risk assessment. What we want in order to recommend an intervention is for the benefits to outweigh the risks. A lot of these adverse effects are hard to get
concrete numbers on. They're hard to quantify. Even bleeding rates for
such a common therapy as aspirin used at doses to treat pain. People conducting trials just don't pay all that much attention to harm. Also, almost all of the
drug studies are short term. Long-term harms might not always be evident. However, in the case of the pharmaceuticals we evaluated, in all cases there are clear benefits that appear to outweigh harms. Obviously this is the case, otherwise the medications probably wouldn't be approved by the FDA. We believe it's up to the individual patient and providers to weigh the trade-offs between benefits and harms for different therapies given the circumstances and preferences of individual patients. There aren't any trials directly comparing non-pharmacologic therapies to pharmacologic therapies that consistently show that people do better with the non-pharmacologic therapy. Integrator: Interesting that you didn't observe significant attention to harm.
The potential for limiting harm, the lack of adverse effects, or at least their relative rarity, is what has drawn many people to these integrative and non-pharmacological approaches.
"It's not just the pharmaceutical
studies
that don't pay adequate
attention to harms. It's also the
studies of
non-pharmaceutical
therapies. Most acupuncture trials
don't report
harms at all."
- Chou
Chou: Yes, even spinal manipulation, which people worry about, has extremely rare side effects, at least when done for the lower back. But it's not just the pharmaceutical studies that don't pay adequate attention to harms. It's also the studies of non-pharmaceutical therapies. Most acupuncture trials don't report harms at all. Integrator: So this brings us to in interesting question from the integrative medicine perspective. There is a concept that is strong in integrative medicine thinking, in
naturopathic medicine thinking, of the "therapeutic order." Basically,
the ideas is to use the least invasive things first. Arecent focus group study of integrative medicine clients found that these integrative medicine patients have this same approach. They want their doctors to use drugs as a last resort. Did this concept factor into the
panel's thinking?
Chou: We try to base guidelines on evidence. The evidence was not there to make a warrant recommendation of non-pharmacological approaches as first-line therapy. To have done so would not have been defensible based on the currently available evidence. In addition, just because some people prefer a non-pharmacologic approach doesn't mean that everyone prefers a non-pharmacologic approach. In the absence of evidence showing that one approach is better or less harmful than the other, our approach is not to impose one approach to medicine over another. But we lay out the options and try to provide information for people to make informed choices. Integrator: It couldn't be defended based on known harm?
"You might be surprised that a lot of people in
the allopathic world will be kind of surprised by how much presence the
non-pharmacological and alternative approaches have in the guideline."
- Chou
Chou:There really is not much harm at over-the-counter strengths of pharmaceuticals such as acetaminophen. Part of the issue is convenience. For a lot of people, the convenience of medications, just stopping in to pick up an OTC or even a prescription medication is a lot easier than scheduling appointments with an acupuncturist or massage therapist or chiropractor. We did set up the guideline so that medicine didn't have to be the first choice. If you look at the algorithm, Box 9 suggests either pharmacologic or non-pharmacologic therapies, and doesn't place a higher value on one over the other. Patients who prefer not to take medication can benefit from
non-drug
treatments. It has a lot to do with patient choice or preference. A person might want to take a pill instead of traipsing over to the chiropractor, since there is no strong evidence of one being better than the other. Integrator: It's my own bias that it would be good for physicians to be urging people toward the non-pharmacological approaches.
Chou: The guidelines can't recommend that approach unless the evidence is there. You might want more of an emphasis on non-pharmacologic approaches but you might be surprised that a lot of people in the allopathic world will be kind of surprised by how much presence the non-pharmacological and alternative approaches have in the guideline. * Dan Cherkin,
PhD, was on the panel. Cherkin, based out of the Center for Health
Studies at Group Health Cooperative of Puget Sound, is the chair of the
program committee for the May 2009 research conference on complementary
and integrative medicine sponsored by the Consortium of Academic Health
Centers for Integrative Medicine.
_______________________________
Comment: A constitutional problem with guideline writing is that they represent truth constrained by past research investment. The past and present economic and cultural biases toward research investment on pills do, and will, cast a long-shadow over any guidelines for well into the future.
A second, present problem, is that they, like most of the rest of public policy in the United States, do typically not take a global cost accounting approach. What does it mean to a population to place power in a pill when pain arises, versus to place power in the hands of acupuncturists, massage therapists, acupuncturists, psychologists and inter-disciplinary teams? How is a community shaped by the former versus the later?
Lou Sportelli, DC, touches obliquely on these global issues in his comments on the guideline in the first Integrator article on the subject: "Relegating (decision on a) therapeutic intervention to one issue is
reducing the complexity of the care process. Interventions are complex
encounters. They involve the human communication, trust, understanding,
confidence, belief and culture surrounding the experience."
I thank Chou for his time and his direct talk. After the interview, the guideline seemed friendlier to integrative, non-pharmacological inventions than it originally read to me. Medication, in my reading, seemed the first choice, with their non-pharmacological approaches a weak second. Chou's statement that the team didn't put higher value on one over the other was surprising.
One remedial step for future guideline writers might be to consider that if non-pharmacological approaches are not viewed as a last resort in the therapeutic order, it might be nice to not position them at the bottom of the guideline. Just after the self-care bullet, and before pharmacotherapy, is a happy home in a therapeutic order for non-pharmacological approaches.
Send your comments to
for inclusion in a future Your Comments forum.
Integrator: I was surprised to discover 
