Supervised Exercise With And Without Spinal Manipulation
Performs Similarly And Better Than Home Exercise For
Chronic Neck Pain: A Randomized Controlled Trial

This section was compiled by Frank M. Painter, D.C.
Send all comments or additions to:   Frankp@chiro.org

FROM:   Spine (Phila Pa 1976). 2012 (May 15); 37 (11): 903–914 ~ FULL TEXT

Evans, Roni DC, MS; Bronfort, Gert DC, PhD; Schulz, Craig DC, MS;
Maiers, Michele DC, MPH; Bracha, Yiscah MS; Svendsen, Kenneth MS;
Grimm, Richard MD, MPH, PhD; Garvey, Timothy MD; Transfeldt, Ensor MD

Northwestern Health Sciences University,
Wolfe Harris Center for Clinical Studies,
Bloomington, MN 55431, USA.
revans@nwhealth.edu

---

STUDY DESIGN:   Randomized controlled trial using mixed methods.

OBJECTIVE:   To evaluate the relative effectiveness of high-dose supervised exercise with and without spinal manipulation and low-dose home exercise for chronic neck pain.

SUMMARY OF BACKGROUND DATA:   Neck pain is a common global health care complaint with considerable social and economic impact. Systematic reviews have found exercise therapy (ET) to be effective for neck pain, either alone or in combination with spinal manipulation. However, it is unclear to what extent spinal manipulation adds to supervised exercise or how supervised high-dose exercise compares with low-dose home exercise.

METHODS:   Two hundred and seventy patients with chronic neck pain were studied at an outpatient clinic. Patients were randomly assigned one of the following interventions: (1) high-dose supervised strengthening exercise with spinal manipulation (exercise therapy combined with spinal manipulation therapy [ET + SMT]), (2) high-dose supervised strengthening exercise (ET) alone, or (3) low-dose home exercise and advice (HEA). The primary outcome was patient-rated pain at baseline and at 4, 12, 26, and 52 weeks. Secondary measures were disability, health status, global perceived effect, medication use, and satisfaction.

RESULTS:   At 12 weeks, there was a significant difference in patient-rated pain between exercise therapy (ET) + spinal manipulation (SMT) and HEA (1.3 points, P < 0.001) and ET and home exercise and advice (HEA) (1.1 points, P = 0.001). Although there were smaller group differences in patient-rated pain at 52 weeks (ET + SMT vs. HEA, 0.2 points, P > 0.05; ET vs. HEA, 0.3 points, P > 0.05), linear mixed model analyses incorporating all time points yielded a significant advantage for the 2 supervised exercise groups (ET + SMT vs. HEA, P = 0.03; ET vs. HEA, P = 0.02). Similar results were observed for global perceived effect and satisfaction.

CONCLUSIONS:   Supervised strengthening exercise with and without spinal manipulation performed similarly, yielding better outcomes than home exercise particularly in the short term. Various stakeholders' perspectives should be considered carefully when making recommendations regarding these therapies, taking into account side effects, preferences, and costs.

TRIAL REGISTRATION:   ClinicalTrials.gov   NCT00269360

Key words:   neck pain , exercise , manipulation , chiropractic , orthopedic , randomized clinical trial.

---

From the FULL TEXT Article:

Background

Neck pain is a common, global health care complaint with considerable social and economic impact. Up to 3 quarters of individuals worldwide experience neck pain at some time in their lives. [1, 2] Although not life threatening, neck pain can limit work and activities of daily living and put significant burden on workers and employers in terms of work absenteeism. [1, 3] Although most individuals who experience neck pain do not seek care, it is still one of the most commonly reported symptoms in primary care settings. [6, 7] This has resulted in millions of ambulatory health care visits annually for neck pain conditions [9–10] and increasing health care expenditures. [11]

Systematic reviews have consistently found exercise therapy (ET) to be effective for neck pain, either alone or in combination with spinal manipulation or mobilization. [12 ] In a previous randomized clinical trial, we found that spinal manipulation combined with low-tech supervised ET and high-tech supervised exercise on its own resulted in significantly greater pain reduction 1 and 2 years after treatment than spinal manipulation alone. [16, 17]

It remains unclear, however, to what extent spinal manipulation adds to supervised exercise for chronic neck pain or how more intensive supervised high-dose exercise compares with low-dose home exercise programs. [16, 17] Given the differences in effort and costs, resolution of these questions has consequence for patients, providers, and policy makers. Finally, although the methodological quality of neck pain studies continues to improve, there is a need for rigorous trials that take into account patient preferences and views. [18–20]

The purpose of this mixed-methods randomized trial was to address these issues by evaluating the relative effectiveness of 3 treatment approaches for chronic neck pain: (1) high-dose supervised strengthening ET combined with spinal manipulation therapy (ET + SMT); (2) high-dose supervised strengthening ET alone; and (3) low-dose home exercise and advice (HEA). To assist with the interpretation of trial results, we explored patients’ perspectives, specifically the issues they considered when determining their satisfaction with care and the outcomes that were most important to them.

---

DISCUSSION

      Statement of Principal Findings

This study suggests that high-dose supervised strengthening exercise with or without manipulation results in greater pain reduction, global perceived effect, and satisfaction than lowdose home mobilization exercise and advice for chronic neck pain, particularly in the short term. The 2 supervised exercise groups were not significantly different from one another in terms of any of the patient-rated outcomes, suggesting that spinal manipulation confers little additional benefit when added to supervised exercise for chronic neck pain.

      Strengths and Weaknesses of the Study

Strengths of this study include a high level of adherence to the study interventions and no observed group differences in cointerventions, which enhances our confidence in the study results. Also, to aid future systematic review efforts and clinical interpretation, we have described the exercise interventions, using a standardized classification format (i.e ., type, program design, delivery, and dose). [35]

A limitation of this study is that it was not designed to differentiate between the specific effects of the exercise and spinal manipulation treatments and the contextual, or nonspecific, effects, including patient-provider interactions and expectations. Rather, this study was intended to be pragmatic in nature, answering clinical questions regarding treatments offered in health care practice for which patients have varying degrees of experience and expectations. Also, the home exercise group was intentionally minimal in its approach in terms of time and resources and, as such, served as a control. Indeed, patients in all 3 groups had greater expectations of improvement for supervised exercise than for home exercise; this was likely due to obvious differences in dose and supervision. The observed between-group differences in some of the blinded strength measures in favor of the 2 supervised exercise groups (consistent with patient self-report measures) suggest that at least some of the demonstrated effects may be attributable to the high-dose strengthening exercise program (i.e ., number of sessions, repetitions, and load on the cervical musculature). We did not measure patients’ long-term adherence with exercise and thus do not know whether that affected outcomes. However, an earlier study conducted by our group found no difference in outcomes at 1–year follow-up between those who complied with exercise and those who did not. [16]

Side effects were more frequently reported in the 2 supervised exercise groups; this was expected because of the dose and intensity of the exercise treatment; however, it is possible that side effects in the home exercise group were underreported because of our data collection methods (i.e ., side effects were queried at treatment visits, of which there were fewer for home exercise).

Another limitation of our study is that, like all research on exercise, we were unable to blind study participants to treatment group. This limitation was minimized by measuring expectations at baseline and factoring them into the statistical analyses. [19]

      Strengths and Weaknesses of the Study in Relation to Other Studies

The clinical and baseline characteristics of our study population are similar to those observed in other studies (including primary care settings), which enhances the generalizability of our findings [18, 51]; however, the growing variety of exercise types, program designs, delivery methods, and dosages (e.g ., repetitions, load, number of sessions) evaluated for chronic neck pain makes it difficult to compare our results with other studies. [15, 35] The most comparable study is an earlier trial performed by our group, in which supervised high-dose, lowtech exercise with spinal manipulation was compared with supervised high-dose, high-tech exercise alone and spinal manipulation alone. [16, 17] That study found an advantage for the 2 high-dose supervised exercise groups, with the magnitude of effects similar to what was observed in this study. Similar results were also reported by Walker et al, [52] who demonstrated a combination of manual therapy and exercise to be superior to minimal intervention (advice and home exercise), both in the short and the long term.

Furthermore, our study demonstrated that spinal manipulation conferred little additional benefit to supervised exercise. This seems consistent with the findings of Dziedzic et al, [54] who found that manual therapy in addition to a home exercise program and advice did not result in improved outcomes when compared with HEA alone. Our findings differ from the conclusions of the Task Force on Neck Pain and Its Associated Disorders [12] and systematic reviews, [14, 15] which found an advantage for exercise combined with manual therapy for chronic neck pain. Contrary to the trials that were the basis for these reviews, our study design allowed us to evaluate the added benefit of spinal manipulation to highdose supervised exercise. Importantly, our study was not designed to assess the effect of spinal manipulation alone. A recent Cochrane systematic review has found limited evidence to support spinal manipulation alone for the short-term relief of chronic neck pain. [19]

      Meaning of the Study: Possible Explanations and Implications
      for Clinicians and Policymakers

There remains no standard method for interpreting the clinical importance of study results for patient-rated outcomes in neck and back pain studies. [55] One approach is to calculate standardized between-group effect sizes (betweengroup mean difference divided by the baseline standard deviation). [56] In our study, the between-group differences for pain between the 2 supervised exercise groups and home exercise were 11 to 13 percentage points at week 12, which translated into large effect size differences (0.8–0.9); however, these group differences diminished to 3 to 6 percentage points by week 26 and 2 to 3 percentage points at week 52, which translate to small effect sizes (0.2–0.4). Although some have argued that even small between-group effect size differences are meaningful at the population level, others remain skeptical. [57] A complementary method to aid with study interpretation is the calculation of proportions of patients in each group who experience a prespecified clinical improvement. [51] We used a 2.5–point reduction for the primary outcome, patient-rated pain, to calculate relative risk and absolute risk reduction (Table 5). [49–51] Overall, similar proportions of patients in the 2 supervised exercise groups reported clinically meaningful improvements (ET + SMT = 74% and ET = 65% at 12 wk; ET + SMT = 51% and ET = 57% at 52 wk). Noteworthy, however, is the sizeable proportion of the home exercise group (41%–42%) who experienced meaningful improvements in pain in both the short term and the long term (Table 5). From a societal or payer’s perspective, the benefits of frequent, supervised exercise, with or without manipulation, may not outweigh the associated time, effort, side effects, and costs when compared with a home exercise program. [58] Consequently, a low-dose home exercise program may be a prudent first line of therapy for people with chronic neck pain, which, if unsuccessful, could be followed by more aggressive, high-dose supervised exercise.

Careful consideration should be given to choosing the most appropriate exercise program for individual patients. [59] The time commitment, physical effort, and side effects associated with high-dose supervised exercise versus low-dose home exercise may be important factors in terms of patient willingness and compliance. Furthermore, the amount of supervision necessary to motivate patients is likely to vary among individuals. Future studies are needed to investigate individual preferences related to supervised and home exercise programs and their relationship to outcomes and program adherence for people with chronic neck pain.

---

CONCLUSION

Our study found that groups receiving high-dose supervised ET with and without spinal manipulation performed similarly, reporting less pain, greater global perceived effect, and more satisfaction than the low-dose home exercise group, particularly in the short term. The supervised exercise groups also demonstrated greater gains in blinded assessment of neck endurance and strength, supporting the patient-self report measures. The results of qualitative interviews suggest that personal attention played an important role in the supervised exercise groups. Various stakeholders’ perspectives should be considered carefully when making recommendations regarding these therapies for chronic neck pain patients, taking into account side effects, preferences, and costs.

---

Key Points

• ET, with or without spinal manipulation, has previously been shown
  to be more eff ective than other noninvasive treatments for
  nonspecific chronic neck pain. Little is known to what extent
  spinal manipulation contributes to clinical benefits.

• There has been little research comparing high-dose supervised exercise
  with low-dose home exercise programs.

• High-dose supervised exercise (with or without spinal manipulation)
  resulted in greater short-term pain reduction, global perceived effect,
  and satisfaction than low-dose home exercise for people with
  nonspecific chronic neck pain.

• No significant differences were found between supervised exercise with
  or without spinal manipulation, suggesting that spinal manipulation
  confers little additional benefit.

• A sizeable proportion of the home exercise group experienced clinically
  meaningful improvements in pain in both the short term and the long
  term. This suggests that home exercise may be a prudent first line of
  therapy for people with chronic neck pain, which, if unsuccessful,
  could be followed by more aggressive, high-dose supervised
  exercise programs.

Return to CHRONIC NECK PAIN

Return to EXERCISE AND CHIROPRACTIC

Since 7–06–2014