Current issues of ACP Journal Club are published in Annals of Internal Medicine


Spinal manipulation for back and neck pain: a review

ACP J Club. 1993 Mar-April;118:61-2. doi:10.7326/ACPJC-1993-118-2-061

To the Editor

In evaluating manipulation therapy for back and neck pain with subjective responses, patient expectation and the transmission of expectation from therapist to patient are important. In the review by Koes and colleagues (Manual Therapy, Physiotherapy, or Usual Care for Back and Neck Pain?)(1), no control was reported on the measurement of either. In addition, the inclusion of patients who had chronic pain only does not match clinical circumstances in which manipulative therapy is given—for acute and subacute pain.

The reviewer also missed other points. The benefits of standard physical therapy, manipulative therapy, and placebo treatment did not differ. The curves of treatment benefit, including patient evaluation and performance, were superimposable. Although patients under general practice care fared the worst, all persons accepted for this study had already been under general practice care for a minimum of 6 weeks. Those who were switched to physical therapy, manipulation, or placebo did slightly better, but this improvement may have been caused by the change in therapy rather than by the specifics of any particular therapy. Several patients also switched therapy at various times in the 12-week period: 15 (about 25%) in the placebo therapy switched to physical therapy, and 21 (about 30%) in the general practitioner group switched to various other specialties, including manual therapy and physiotherapy. Finally, known adverse effects of manipulation were not exposed by this small study.

These results suggest that it is not the specific method that counts but the therapist's attention and the change in treatment that make the difference. In fact, the authors (1) state this. Dr. Bloch's conclusion that physicians should "no longer feel bashful about sending a patient for manipulation" is inconsistent with the lack of evidence of effect by those who do most manipulative therapy—chiropractors. Cost data analysis in most insurance investigations (2, 3, 4) shows higher cost for manipulative therapy than for standard medical care.

The appropriate conclusion of the paper is that manipulative therapy fares no better than placebo or physical therapy. The authors did not conclude that manipulative therapy was a reasonable option. The latter was the conclusion of the reviewer and is inappropriate.

Wallace Sampson, MD
Santa Clara Valley Medical Center
San Jose, California

In response

Dr. Sampson indicates that the patients with chronic pain included in the studies are not representative of those seen in practice. Any practitioner sees a mix of patients with acute and chronic pain, with chronic pain visits predominating. Also, other studies do, in fact, support the short-term benefit of manual therapy in patients with acute pain (5).

Dr. Sampson feels that the effects of manipulation may simply be the result of laying on of hands. Both the authors of the article and I stressed that "manual therapy" does not constitute a pure therapeutic modality, but one whose individual ingredients may be active separately, in concert, or not at all. The study examines the global effect. As long as the composite studied is representative of what is available to patients, the conclusions remain valid. Similarly, Dr. Sampson points out the lack of control of therapist-patient interaction. Attempts to control, observe, or standardize the psychosocial interactions between therapist and patient would constitute a major intervention and would unnecessarily complicate the experimental design.

The concern that general practice patients simply continue, while the other three groups change management, is a valid criticism that was considered in the article. I presume the investigators included the placebo group as a true control group for this reason. In the commentary that I submitted to ACP Journal Club, I did not, therefore, stress the relatively poorer outcomes of general practice patients, something that was added by the editors.

As for the adverse effects of chiropractic therapy, I am not aware of a carefully controlled study comparing adverse effects of manual therapy with those of heat therapy, nonsteroidal anti-inflammatory drug therapy, surgery, and other treatments. With respect to the effect of health economics, I have not carefully reviewed cost-effectiveness studies comparing prevalent ways of managing neck and back pain.

As for the "high" proportion of crossovers, given the complexity of running such trials, the number of crossovers is astonishingly small. Because the majority of crossovers were away from general practice and placebo management, a worst-case analysis, treating crossovers as failures, would even strengthen the conclusions.

Finally, Dr. Sampson indicates that my conclusions differed from those of the authors. On page 35 of their article, the authors state: "We conclude that it seems useful to refer patients with non-specific back and neck complaints lasting for at least 6 weeks for treatment with physiotherapy or manual therapy."

Manual therapy has a long history. It had, until recently, not been studied carefully for its benefits, risks, and costs. Any attempt to do so would be welcome. At the same time, we should not apply stricter scientific criteria to studies of manual therapy than to those of conventional therapies of musculoskeletal disorders simply because what we believe should not be, cannot be!

Ralph F. Bloch, MD, PhD
University of Berne
Berne, Switzerland


1. Manual Therapy, Physiotherapy, or Usual Care for Back and Neck Pain?. ACP J Club. 1992 Jul-Aug:18. (Ann Intern Med. vol 117, suppl 1). Abstract of: Koes BW, Bouter LM, van Mameran H, et al. The effectiveness of manual therapy, physiotherapy, and continued treatment by a general practitioner for nonspecific back and neck complaints. A randomized clinical trial. Spine.

2. Marter S, Weisberg H.Medical expenses and the Massachusetts Automobile Tort Reform Law, First Review of the Automobile Insurance Bureau Study of 1989. Bodily injury liability claims. Correlation Research, Inc. Needham, Massachusetts; 30 July 1991.

3. Greenwood JG. Report on work related back and neck injury cases in West Virginia: insurance related to chiropractic and medical costs. West Virginia Worker's Compensation Fund, Charleston, West Virginia; November 1983.

4. Linkner LM. The financial impact of chiropractic benefits on health care in Arizona (the Blue Cross and Blue Shield experience, 1983-1986), Phoenix, Arizona; February 1987.

5. Shekelle PG, Adams AH, Chassin MR, Hurwitz EL, Brook RH. Spinal manipulation for low-back pain. Ann Intern Med. 1992;117:590-8.