Research Review by Dr. Shawn Thistle©


Oct. 2007

Study Title:

Comparison of general exercise, motor control exercise and spinal manipulative therapy for chronic low back pain: A randomized trial


Ferriera ML et al.

Publication Information:

Pain 2007; 131: 31-37.


It is well known that patients with chronic low back pain (CLBP = persisting for more than 3 months) consume over 80% of health care resources for back pain. These patients also represent a large percentage of disability and work absenteeism from all causes in industrialized societies. Unfortunately, the most effective way to clinically manage this common condition remains unclear based on existing literature. There is evidence supporting various interventions for CLBP including different types exercise, spinal manipulation, and acupuncture (the list could go on…), but no clear winner emerges. Practice guidelines in many countries generally recommend various types of exercise and a trial of spinal manipulation, but head-to-head studies involving these two interventions are sparse.

This randomized controlled trial compared the clinical effectiveness and patient-perceived effect of general exercise, motor control exercise, and spinal manipulative therapy (SMT) in patients with chronic LBP. 240 patients seeking treatment for CLBP in physical therapy departments of three Australian teaching hospitals participated in this study. Subjects were between the ages of 18-80, and had to have LBP for > 3 months with or without leg pain (note: those reporting osteoarthritis or disc lesions without neurological deficit were allowed to participate). Exclusion criteria included:
  • neurological signs
  • specific spinal pathology – malignancy, inflammatory joint or bone disease
  • previous back surgery
  • contraindications to exercise or SMT
Each subject attended 12 treatment sessions over an eight week period, in one of the three following intervention groups:
  1. General Exercise: Participants were initially assessed by a physical therapist – who then instructed and supervised a progressive exercise program in a group format (up to 8 people). The main aim of the program was to improve physical function and confidence in using the spine. It was modeled on the “Back to Fitness” program and included strengthening and stretching exercises for the main muscle groups of the body as well as general cardiovascular exercise. Participants were also encouraged to engage in daily activity and avoid unaccustomed rest.
  2. Motor Control Exercise: These participants were prescribed exercises aimed at improving the function of specific trunk muscles, and was modeled after the commonly known approach originally described by Jull, Richardson, Hodges etc., focusing mainly on the function of the transverse abdominus, pelvic floor, and other deep stabilizing muscles of the trunk. Ultrasonography was used in accordance with this approach to assist in patient visualization of the muscles being used.
  3. Spinal Manipulative Therapy: Participants were treated with joint mobilization or manipulation techniques directed at the spine and pelvis. The dose and specific techniques utilized were at the discretion of the treating therapist (a pragmatic model). Participants in this group were NOT prescribed home exercises, but were advised to avoid pain-inducing activities. This therapy was discontinued if the participant reported complete recovery.
In both exercise groups, a cognitive-behavioral approach was employed, including encouragement of skill acquisition, goal setting, progression, pacing, self-monitoring, and positive reinforcement. Further, all exercise participants were encouraged to attend all 12 treatment sessions regardless of the extent of their recovery.

Outcome measures were obtained at study inception, at 8 weeks (end of treatment period), and at 6 and 12 months.

Pertinent Results of this study include:
  • Primary Om: Patient-Specific Functional Scale (10-point scale rating difficulty in performing 3 daily tasks that the patient chooses – scores summed and graded out of 30…higher score indicates less difficulty); Global Perceived Effect (11-point scale from -5 [vastly worse] to +5 [completely recovered]).
  • Secondary Om: pain measured on a 10cm Visual Analogue Scale (VAS), and disability measure by the Roland-Morris Disability Questionnaire.
Pertinent results of this study include:
  • follow-up data was obtained for 93% of participants at 8 weeks, and 88% at 6 and 12 months
  • the groups were similar in terms of baseline characteristics – except that the motor control group had a lower mean duration of symptoms versus the other groups (36 vs. 60 months)
  • participants were generally moderately or severely disabled, according to baseline measurements
  • all three interventions had a high level of adherence – all had an average of over 9 of 12 possible sessions attended
  • all three groups demonstrated improvement on average over the 12 month follow-up period
  • in the short term (8 weeks) – those receiving motor control exercise or SMT improved more than the general exercise group: motor control had better function (average 2.9, p = 0.004) and perceived effect (average 1.7, p < 0.001); and SMT had better function (average 2.3, p = 0.016) and perceived effect (average 1.2, p = 0.004) when compared to the general exercise group
  • differences between the motor control and SMT groups were small
  • no significant differences in primary or secondary Om were noted among groups at 6 and 12 months

Conclusions & Practical Application:

In general, this study suggests that motor control exercise (in the Queensland physiotherapy model of Richardson, Hides, Jull et al.) and spinal manipulation both produce slightly better short-term functional and patient-perceived benefit compared to a general exercise program in patients with chronic LBP. Medium and longer-term outcomes among all three groups were equivalent (but as mentioned, all three groups showed significant improvement). It is significant to note that due to the baseline level of functional deficit (noted as moderate to severe on average), this study population represents a group of patients that is generally harder to treat. Therefore, it is encouraging to see positive results in all groups.

This study had a sufficient sample size, employed a pragmatic approach to the manual therapy intervention, and used appropriate outcome measures – all strengths of this study. It should be noted that since all interventions were active, no inference can be made about the effectiveness of these interventions compared to no treatment. Further, the effectiveness of the motor control intervention (and SMT to some degree) versus the general exercise may in part be due to the fact that the general exercise was administered in a group, whereas the other interventions were on an individual basis. Regarding the motor control intervention – ultrasonography has been recommended to assist in this intervention, despite an absence of evidence suggesting it influences outcomes. Realistically, most clinicians would not utilize this method in practice, and until the evidence dictates otherwise, one should assume implementing this method of rehabilitation without ultrasound is acceptable.

The other potential detractor from these results is that the authors did not control for co-interventions between the 8-week and 12-month follow-up. It is possible that participants sought additional therapies that influenced their outcomes.

This study supports an active approach in treating CLBP, and provides further evidence that SMT can have beneficial short-term effects. It stands to reason that a combination of active therapies and SMT may be the most effective, but future studies will have to clarify this issue.