Access to the PPM Journal and newsletters is FREE for clinicians.
14 Articles in Volume 18, Issue #7
A 2018 Update: The Federal Pain Research Strategy
A Commentary on Medical Cannabis
Are Abuse-Deterrent Opioids Appropriate for Your Pain Patient?
Behind the AHRQ Report
Challenges Facing Abuse-Deterrent Formulations
Demystifying Opioid Abuse-Deterrent Technologies
Editorial: Our Clinical Pain Neighborhood
Independent Pain Practice: A Case Example
Inside Performing Arts Medicine
Letters to the Editor: ACT Therapy; Compounded Topicals
Nerve Growth Factor and Targeting Chronic Pain
Pain Control for Athletes: What Works?
Quality Training: One Center’s Experience with Pain Assessment
The Importance of Developing Professional Relationships in Pain Practice

Behind the AHRQ Report

Understanding the limitations of “non-pharmacological, non-invasive” therapies for chronic pain.

A commentary

In the current restrictive regulatory climate that governs opioid analgesic therapy for chronic pain, there is much discussion of “alternative” therapies and “integrative medicine.” Unfortunately for proponents of such measures, the state of medical evidence in trials literature is very weak, reflecting weaknesses of trial design, execution, and size. This is perhaps unintentionally illustrated by a major systematic review released in June 2018 by the Agency for Healthcare Research and Quality (AHRQ), an agency of the US Health and Human Services Department.1

Scope of the AHRQ Review

In its review, AHRQ sought to “assess which noninvasive nonpharmacological treatments for common chronic pain conditions improve function and pain for at least one month after treatment.”1 The selected therapies focused on areas common to traditional “alternative therapies,” but did not include medical marijuana, electrostimulation devices, or corticosteroid injections. The review incorporated randomized controlled trials literature across the following categories of chronic pain, including interventions demonstrated to “[improve] function and/or pain for at least one month:”1

  • Chronic Low Back Pain: exercise, psychological therapies (primarily cognitive behavioral therapy, or CBT), spinal manipulation, low-level laser therapy, massage, mindfulness-based stress reduction, yoga, acupuncture, and multidisciplinary rehabilitation, or MDR)

  • Chronic Neck Pain: exercise, low-level laser therapy, the Alexander Technique, acupuncture

  • Chronic Tension Headache: spinal manipulation

  • Fibromyalgia: exercise, CBT, myofascial release massage, tai chi, qigong, acupuncture, and MDR

  • Osteoarthritis: exercise, ultrasound for the knee; exercise, manual manipulation and therapy for the hip.

Cross-roadsPrescribers are at a crossroads when it comes opioid therapy.

The researchers identified 4,996 candidate trial reports in medical literature, which they narrowed down to 1,193 reports for full-text review, and then further characterized 218 publications that met rigorous quality criteria. Many of the trials excluded from the final AHRQ report failed to follow patients for longer than one month after trials were completed. Interestingly, a shortage of long-term randomized trials was one of the factors said to imply a lack of long-term effectiveness in opioid analgesics by the writers of the CDC Guideline on Prescribing Opioids for Chronic Pain.2,3 A crucial difference between controlled trials of alternative treatments and trials of opioids is that there is nothing that intrinsically precludes long-term trials of the former, whereas extended controlled trials of opioids are infeasible due of high drop-out rates in patients randomized to placebo.

What the Review Really Focused On

As noted in the issued AHRQ report, “most effects were small. Long-term evidence was sparse.”1 In this context, “small” does not quite do justice to the details of the many tables in this 1,000-page report. For many therapy trials, the magnitude of changes in reported pain levels was less than one point on a scale of 0 to 10. The systematic review also characterized strength of medical evidence as “weak” in more than 150 reports of short-term, intermediate, or long-term outcomes.

“The majority of trials compared nonpharmacological interventions with usual care, waitlist, no treatment, attention control, or placebo/sham (93%); few trials employed pharmacological treatments (5%) or exercise (17%).”

However, buried deeply in the report, readers also learn that “it is assumed that most patients with chronic pain likely continued medications and other therapies or practices during the trials. These factors may have resulted in substantial mixing of effects of the intervention and co-interventions. These factors possibly attenuated observed effects.”

Four crucial points should, therefore, be borne in mind when considering the AHRQ analysis, as described below.

Pain Magnitude

Most of the reviewed trials enrolled patients with “moderate” pain intensity (> 5 on a 10-point numerical scale). However, the commonly used interval or analog pain scales, unlike the more complex McGill Pain Questionnaire, have demonstrated almost no correlation with a more objective measure of pain intensity.4 Thus, the validity of the endpoint measure is cause for serious concern. Further, how certain is it that a patient who rates their pain as 5 out of 10 would provide the same rating after having experienced 10/10 pain? These considerations call into doubt the comparability of the patients in trials of opioid therapy versus patients in trials of alternative therapies. The true test of the effectiveness of an alternative therapy would be its ability to achieve sustained reduction in opioid dosage. None of the indexed trials attempted this comparison.

Meaningfulness of Pain Improvement

Treatment-associated improvements in objective outcomes speak for themselves. For instance, treatment with new oral anticoagulants reduces stroke risk by 70% in patients with atrial fibrillation. However, the same cannot be said of subjective outcome measures. What does it mean to say that an alternative treatment reduced scores on a numerical outcome scale by 0.5 – 0.79 standard deviations (a moderate effect size)? A more meaningful outcome measure would be whether or not pain was adequately controlled. Alternatively, meaningfulness could be measured in pragmatic terms as a quantitative reduction in a clinically objectifiable outcome such as opioid dosage, enabled by alternative therapy.

Hawthorne Effect

The Hawthorne effect, simply stated, is that human beings do better when close attention is paid to them and they feel valued. Thus, it is quite possible that some or all of the beneficial effects documented in trials of alternative therapies accrued substantially or entirely to the Hawthorne effect. Hawthorne effect was not mentioned in the AHRQ document and no efforts were made to control for it.

Trial Progression

For many decades, the field of medicine has agreed upon a systematic clinical trial progression. Phase I trials determine toxicity and, in some trials, incipient evidence of benefit. Phase II trials, which are larger and invariably involve a randomized controlled design, seek evidence of efficacy and dosing effects. Phase III trials (all randomized controlled trials, usually multi-center) marshal large numbers of patients in order to demonstrate effectiveness in a population sufficiently large to assure that it is representative of a general population. Only when more than one Phase III trial demonstrates effectiveness (and safety) do we accept a treatment as being ready for translation into clinical practice. A treatment remains experimental until such Phase III trial evidence is available. Alternative therapies are therefore at this stage “experimental.”

The Argument for “Usual” Therapy

In treating chronic pain, usual therapy frequently includes anti-inflammatory medications, opioids, anti-depressants, and anticonvulsants. So long as customary and usual therapies provided to the two groups in a randomized clinical trial of an alternative therapy are adequately detailed and shown to be statistically equivalent, there is statistical power to detect meaningful differences. It should not matter what usual therapies are employed. However, AHRQ reviewers were frequently unable to determine what constituted usual therapy.

Also of deep concern within the review were the report’s offered “Implications for Clinical and Policy Decision-Making,” as stated:

“Recent guidelines from the Centers for Disease Control and Prevention (CDC) in the United States and the Canadian Guideline for Opioid Use in Chronic Non-Cancer Pain recommend nonopioid treatment as the preferred treatment for chronic pain. Further, guidelines from the American College of Physicians recommend non-pharmacological therapies over medications for chronic back pain. Our findings support the feasibility of implementing these guideline recommendations by showing that there are some nonpharmacological treatments for chronic pain that have evidence of sustained effectiveness after the completion of therapy. Importantly, some interventions, such as exercise, CBT, multidisciplinary rehabilitation, mind-body interventions, and some complementary and integrative medicine therapies, such as acupuncture and spinal manipulation, also were associated with some sustained effects on function, although evidence beyond 12 months is sparse…”1

In light of the significant weaknesses in medical evidence and methodology described in the AHRQ report, we must suggest that the AHRQ review findings clearly do not support the substitution of non-pharmacological treatments for analgesic therapy. AHRQ findings instead place the CDC in the extraordinary position of advocating use of experimental treatments over treatments of demonstrated effectiveness. At most, the evidence might suggest expanded use of alternative treatments as adjuncts to usual therapy, given their apparent safety.

Suggested Next Steps

The strongest AHRQ recommendation is that alternative treatments be subjected to further scientific study. Under prevailing scientific standards, such study should center upon trial designs that employ valid, quantifiable outcome measures that translate meaningfully into practice – and ultimately, non-pharmacological therapies should enter Phase III trials, to provide assurance of the validity of trial findings.

In the meantime, use of such therapies as adjuncts to treatment with analgesics or anti-inflammatory drugs should be approached with the same care as all experimental protocols. Treating physicians should carefully document both the existing therapy protocols and the outcomes – both positive and negative – of adding adjuncts to the therapy plan. When opioid therapy is to be tapered, the taper should be gradual to avoid withdrawal symptoms in opioid-dependent patients. If breakthrough pain is observed, the taper should be abandoned. Pending stronger medical evidence of efficacy, so-called “integrative medicine” should not become a mandated replacement for opioid therapy in either practice standards or state regulations. If insurance companies are serious about efforts to identify non-opioid alternatives, then adjunct therapies and collection of outcomes data should be funded even though they are experimental. 



Last updated on: October 3, 2018
Continue Reading:
Alliance Offers Patient-Centered Opioid Prescribing Recommendations
close X