Deep tissue massage therapy and/or strengthening and stretching exercises for disabling subacute or chronic neck pain : a randomized controlled trial

Abstract

Background: Neck pain is a common condition responsible for a significant amount of disability worldwide. Various treatment modalities are used to manage neck pain, but evidence supporting their use is scarce, conflicting or of low quality.

Objectives: The aim of this thesis is to present the results of the Stockholm Neck (STONE) trial, a four-arm randomized controlled trial of 619 participants with disabling subacute or chronic neck pain who were followed up to one year. The objectives of the STONE trial were to determine the effectiveness, safety profile and cost-effectiveness of deep tissue massage, strengthening and stretching exercises and a combined therapy including both components, in comparison to advice to stay active. Moreover, additional information was collected with the objective of describing the course of the condition over time.

Methods: In Study I, different trajectories of the course of neck pain as well as baseline variables associated with unfavorable trajectories were identified. Study II was an analysis aiming to determine the effect of deep tissue massage, strengthening and stretching exercises and a combined therapy including both components, using advice to stay active as a reference group. Two primary outcomes: pain intensity and pain-related disability, and two secondary outcomes: self-perceived recovery and sickness absence, were measured at 7, 12, 26 and 52 weeks. In Study III, participants were asked to report and describe adverse events debuting after the sessions of therapy. That information was contrasted against the proportion of participants in each group achieving perceived recovery at seven weeks, in order to calculate measures of harm in relation to benefits. In Study IV, costs resulting from neck pain were estimated, including those directly and indirectly related to the interventions given in the STONE trial. The costs associated with gains in health-related quality of life due to the given interventions were calculated.

Results: In Study I, six different trajectories were identified, and a quarter of participants had unfavorable courses of neck pain characterized by high pain intensity, either constant or fluctuating. High pain intensity at baseline, being a woman and having depressive symptoms at baseline were among the factors associated with such unfavorable courses. In Study II, compared to advice, massage alone or in combination with exercise resulted in less minimal clinically important improvement (MCII) in pain intensity in the short term, and exercise alone resulted in less MCII in pain intensity in the mid-term. Massage and/or exercise resulted in similar MCII in pain intensity compared to advice in the long term. Moreover, no differences were observed between treatment arms for MCII in pain-related disability or sickness absence after one year. On the other hand, compared to advice, all the other therapies resulted in better self-perceived recovery. In Study III, it was found that around a third of participants reported adverse events that were classified as highly bothersome. The most common adverse events were tiredness, muscle soreness, increased pain and stiffness. None of the adverse events were serious. No clear differences between treatment arms were observed in terms of harms in relation to benefits. In Study IV, massage alone or combined with exercise were found to be more costly and resulting in less gains of quality of life than advice. Exercise, on the other hand, was found to be cost-effective compared to advice to stay active.

Discussion and conclusions: Non-specific neck pain is a subjective, individual and complex experience. Therefore, evaluations of interventions should consider the interplay of various biological and psychosocial factors. Compared to advice, massage and exercise therapy are associated with modest effects in terms of minimal clinically important improvement in pain intensity and no effects in minimal clinically important improvement in pain-related disability. However, improvements in other dimensions of pain – that were probably captured by the outcome “perceived recovery” – result from the mentioned interventions. Furthermore, the therapies are safe, and exercise seems to be cost-effective compared to advice. The STONE trial used a rigorous procedure to ensure a proper randomization and allocation concealment. Despite blinding participants not being possible, well-defined criteria to assess the outcomes were followed. In addition, significant efforts were made to provide the therapies according to pre-established protocols and to achieve high response rates. Appropriate methods for the analysis of the data were followed. All these elements combined ensure the internal validity of the trial. The STONE trial is a predominantly pragmatic trial, while aspects such as intensive measurement and the use of a single center for the provision of the therapies correspond more to an explanatory trial, a good balance between rigorousness and pragmatism was achieved. This balance allows the results from this trial to be generalized to populations with subacute and persistent non-specific neck pain.