Research Review By Dr. Daniel Avrahami©


Download MP3

Date Posted:

January 2012

Study Title:

Spinal Manipulation, Medication, or Home Exercise with Advice for Acute and Subacute Neck Pain: A Randomized Trial


Bronfort G, Evans R, Anderson AV et al.

Author's Affiliations:

Wolfe-Harris Center for Clinical Studies, Northwestern Health Sciences University, Minnesota, USA.

Publication Information:

Annals of Internal Medicine 2012;156:1-10.

Background Information:

Most manual medicine providers treat patients with neck pain on a daily basis. Historically, neck pain is treated differently among various health care providers: from passive manual therapy by chiropractors and osteopaths to medication by physicians and active exercise therapy by physiotherapists. Over recent years the lines between professionals is becoming blurred and many clinicians are incorporating various techniques in their practice.

As evidence informed clinicians we want to make sure that the therapies we offer our patients provide the best chance of achieving positive outcomes, while providing efficient service in a timely and cost-effective manner. The most efficacious treatment for neck pain is still a source of controversy. Cochrane Collaboration reviews and other scientific research have recently questioned the effectiveness of commonly employed therapies such as injections, exercise, mobilization or manipulation for patients with neck pain (1-3).

This study, widely discussed in the popular media, aimed to compare the effectiveness of three interventions for patients with acute and subacute neck pain: spinal manipulation therapy (SMT), medication, and home exercise with advice (HEA).

Pertinent Results:

Primary Outcome Measure – Participant-Rated Pain:
  • This significantly favored SMT compared with medication at 12 weeks (P < 0.01) and in longitudinal analyses every 2 weeks from baseline to 12 weeks (P = 0.017). SMT was also superior to medication at 26 and 52 weeks.
  • HEA was also effective. The differences in participant-rated pain between SMT and HEA were smaller than those between SMT and medication, and these differences were not statistically significant at any time point in the study. Further, there was no overall statistical difference noted between the HEA and medication groups throughout the course of the study.
  • Long-term analyses for participant-rated pain mirrored the shorter-term results indicating that at 26 and 52 weeks, SMT was still favored over medication, but was not favored over HEA. As mentioned above, HEA was not favored over medication in this outcome compared to baseline.
Absolute Proportion of Patients with 50% or 75% Pain Reduction:
  • At 12 weeks, a significantly higher proportion of patients in the SMT group experienced reductions of pain of at least 50% compared to the medication group. This was also noted at 26 weeks, but not 52 weeks.
  • HEA was superior to medication at 12 and 26 weeks, with a higher proportion of subjects achieving a 75% pain reduction. This advantage was no longer apparent at 52 weeks, however.
Secondary Outcomes:
  • Group differences in most secondary outcomes were similar to those of the primary outcomes.
  • SMT was superior to medication in terms of global improvement, participant satisfaction and SF-36-assessed physical function.
  • SMT and HEA groups performed similarly on most of the secondary outcomes, although SMT was better than HEA for satisfaction in short and long term.
  • HEA was found to be better than medication in the short and long term for satisfaction.
  • Cervical spine motion after 4 and 12 weeks was greatest in the HEA group. This was not elaborated any further in this study - results of the group differences in 3-dimensional cervical spine motion patterns will be reported in a future publication.
Adverse Reactions to Treatment:
  • Adverse reaction rates were similar among all groups in this study, with the SMT and HEA groups reporting more musculoskeletal side-effects (40% and 46%, respectively) and the medication group reporting more systemic side-effects (60%).

Clinical Application & Conclusions:

In this study SMT seemed more effective than medication according to various measures of neck pain and function. However, there were no demonstrated benefits of SMT over HEA, as subjects in both groups improved in similar manners on short- and long-term outcomes. Participants who received medication did not fare as well overall, but still improved.

It should be noted that the strong performance of the HEA group has implications for the potential cost savings over both SMT and medication interventions. One might argue that the use of HEA is a cost effective approach, with arguably less potential for adverse reactions (although as mentioned above more subjects in the HEA group reported adverse reactions compared to the SMT group in this particular study).

Another interesting finding was that participants in the medication group reported higher levels of medication use after the intervention. This is an important finding that clinicians should keep in the back of their mind when deciding on appropriate care for individual patients.

It should be noted that the researchers did not delve in their findings from their secondary measures. They did mention that HEA provided greater improvements in cervical spine range of motion but stated that these results would be published elsewhere.

The results of this study, like other trials on SMT for acute or subacute neck pain, suggest that SMT can provide reductions in pain and disability (4-6). Taking the existing body of literature into account, SMT and HEA both represent reasonable treatment options for managing acute and subacute mechanical neck pain. As always, clinical decisions should take studies like this into account, in conjunction with the preference of the patient, their response to prior treatment, and so on.

When examining results from a study like this, simply looking at statistical significance may not suffice, as this may not necessarily provide the entire picture. Minimal Clinically Important Difference (MCID) is crucial parameter in determining the effectiveness of a given treatment. The MCID concept refers to the minimal change required on a particular outcome measure to confidently infer a meaningful clinical difference for the patient. This well-powered study might have produced statistically significant results, however it could be argued that the results do not suggest great clinical relevance. This issue has been raised in criticism of the positive press this study received. Readers should note that universally accepted MCID levels have not been established for many of the outcome measures utilized in this study.

Study Methods:

Participants were between 18-65 years of age with primary symptoms of mechanical, nonspecific neck pain, equivalent to grades I or II according to the Bone and Joint Decade 2000-2010 Task Force on Neck Pain (See Cervical Spine – Neck Pain section for these reviews), current neck pain of 2 to 12 weeks’ duration and a neck pain score of 3 or greater on a scale of 0 to 10.

Participants were randomized into SMT group (n = 91), medication group (n = 91) and HEA group (n = 90). Randomization was concealed from the investigators, treatment providers, and research staff. A pilot study was conducted for the SMT intervention prior to commencement of this study to ensure protocol adherence.

The SMT group was treated by six chiropractors with a minimum of 5 years’ experience. Visits lasted 15-20 minutes and included manipulation of hypomobile segments with diversified techniques using high-velocity manipulation and low-velocity mobilization. Specific treatment levels and the number of treatment sessions over the 12 weeks were left to the discretion of the chiropractor. Adjunct therapy including advice to stay active/activity modification, light soft-tissue massage, assisted stretching and hot and cold packs was also used at the discretion of the chiropractor.

In the medication group, physicians employed non-steroidal anti-inflammatory drugs, acetaminophen or both. Participants who did not respond or could not tolerate the first-line therapy received narcotic medications. Muscle relaxants were also used. Choice of medication and number of visits was decided by the physician.

Home exercise with advice was provided in two 1-hour sessions, 1 to 2 weeks apart by six therapists. The primary focus was self-mobilization exercise (gentle controlled movement) such as neck and scapular retraction and full range of motion movements. Patients were instructed to do 5 to 10 repetitions of each exercise up to 6-8 times per day. Supplemental material, education, advice and basic instructions for activities of daily living were provided for the participants.

The primary outcome measure was pain, measure with an 11-point numerical rating scale (administered 6 times during the 12-week treatment period and at weeks 26 and 52). Secondary outcomes included the Neck Disability Index, global improvement, medication use, satisfaction with care, the Short Form-36 Health Survey and cervical spine motion measured with a Spine Motion Analyzer (4 and 12 weeks).

Study Strengths / Weaknesses:

There are a few notable weaknesses of this study:
  • This study revealed small effect sizes, raising concern about the clinical significance.
  • Participants and providers could not be blinded because of the nature of the treatments.
  • There are no criteria are available to define clinically important group differences for the different outcomes, making inference about the clinical relevance of these findings difficult.
Some of the strengths of this study include rigorous concealed randomization procedure, use of recommended reliable outcome measures, masked objective outcomes assessors and long-term post-randomization follow-up (6 and 12 months).

It would be interesting if a future study added another treatment arm that includes both exercise and SMT just like many clinicians use in their everyday practices.

Additional References:

  1. Kay TM, Gross A, Goldsmith C, Santaguida PL, Hoving J, Bronfort G; Cervical Overview Group. Exercises for mechanical neck disorders. Cochrane Database Syst Rev. 2005:CD004250.
  2. Peloso P, Gross A, Haines T, Trinh K, Goldsmith CH, Burnie S; Cervical Overview Group. Medicinal and injection therapies for mechanical neck disorders. Cochrane Database Syst Rev. 2007:CD000319.
  3. Gross A, Miller J, D’Sylva J, Burnie SJ, Goldsmith CH, Graham N, et al. Manipulation or mobilisation for neck pain. Cochrane Database Syst Rev. 2010: CD004249.
  4. Hoving JL, Koes BW, de Vet HC, van der Windt DA, Assendelft WJ, van Mameren H, et al. Manual therapy, physical therapy, or continued care by a general practitioner for patients with neck pain. A randomized, controlled trial. Ann Intern Med. 2002;136:713-22.
  5. Hoving JL, de Vet HC, Koes BW, Mameren H, Deville´ WL, van der Windt DA, et al. Manual therapy, physical therapy, or continued care by the general practitioner for patients with neck pain: long-term results from a pragmatic randomized clinical trial. Clin J Pain. 2006;22:370-7.
  6. Pool JJ, Ostelo RW, Ko¨ke AJ, Bouter LM, de Vet HC. Comparison of the effectiveness of a behavioural graded activity program and manual therapy in patients with sub-acute neck pain: design of a randomized clinical trial. Man Ther. 2006;11:297-305.