J Manipulative Physiol Ther. 2014 (Jul); 37 (6): 343–362 ~ FULL TEXT
Alexander Tsertsvadze, MD, MSc, Christine Clar, PhD, Rachel Court, MA,
Aileen Clarke, MD, Hema Mistry, PhD, Paul Sutcliffe, DPhil
Senior Research Fellow,
Warwick Medical School,
University of Warwick, Coventry, UK.
OBJECTIVES: The purpose of this study was to systematically review trial-based economic evaluations of manual therapy relative to other alternative interventions used for the management of musculoskeletal conditions.
METHODS: A comprehensive literature search was undertaken in major medical, health-related, science and health economic electronic databases.
RESULTS: Twenty-five publications were included (11 trial-based economic evaluations). The studies compared cost-effectiveness and/or cost-utility of manual therapy interventions to other treatment alternatives in reducing pain (spinal, shoulder, ankle). Manual therapy techniques (eg, osteopathic spinal manipulation, physiotherapy manipulation and mobilization techniques, and chiropractic manipulation with or without other treatments) were more cost-effective than usual general practitioner (GP) care alone or with exercise, spinal stabilization, GP advice, advice to remain active, or brief pain management for improving low back and shoulder pain/disability. Chiropractic manipulation was found to be less costly and more effective than alternative treatment compared with either physiotherapy or GP care in improving neck pain.
CONCLUSIONS: Preliminary evidence from this review shows some economic advantage of manual therapy relative to other interventions used for the management of musculoskeletal conditions, indicating that some manual therapy techniques may be more cost-effective than usual GP care, spinal stabilization, GP advice, advice to remain active, or brief pain management for improving low back and shoulder pain/disability. However, at present, there is a paucity of evidence on the cost-effectiveness and/or cost-utility evaluations for manual therapy interventions. Further improvements in the methodological conduct and reporting quality of economic evaluations of manual therapy are warranted in order to facilitate adequate evidence-based decisions among policy makers, health care practitioners, and patients.
From the FULL TEXT Article
Manual therapy is a skilled nonsurgical conservative management using the practitioner's hands and/or fingers on the patient's body for the purpose of assessing, diagnosing, and treating a variety of symptoms and conditions. [1, 2] Manual therapy is used within the traditional medical (eg, physiotherapy, orthopedics, and sports medicine) and complementary and alternative medicine context (eg, chiropractic and osteopathy) and consists of different techniques (eg, manipulation, mobilization, static stretching, and muscle energy techniques). The definition and purpose of manual therapy vary across health care professionals.
The use of manipulation and mobilization has been recommended in clinical practice guidelines in the United States, Great Britain, Canada, and the Netherlands. [3-9] Although past research evidence on the clinical effectiveness [10-19] and safety [20-27] of manual therapy relative to other interventions is abundant, the evidence on cost-effectiveness is insufficient and inconclusive. [28-36] Moreover, to our best knowledge, a systematic review of full economic evaluations of recent evidence (ie, cost-effectiveness [CEA] and/or cost-utility analysis [CUA]) alongside randomized controlled trials (RCTs) of manual therapy has not been conducted.
In light of limited health care resources, policy makers, health care providers, and researchers need to make informed decisions in prioritizing and allocating resources to the provision of health care interventions that are both effective and cost saving. Ideally, the decision-making process should be based on high-quality evidence summarizing incremental costs and effects of a health care intervention of interest compared with alternative interventions.
The aim of this review was to systematically identify, appraise, and evaluate the evidence on trial-based economic evaluations (cost-effectiveness and/or cost-utility) of manual therapy relative to other alternative interventions used for the management of musculoskeletal conditions.
This review identified limited evidence indicating that manual therapy techniques (eg, osteopathic spinal manipulation, physiotherapy consisting of manipulation and mobilization techniques, and chiropractic manipulation), in addition to other treatments or alone, are more cost-effective than usual GP care (alone or with exercise), spinal stabilization, GP advice, advice to remain active, or brief pain management for improving low back pain and/or disability. Similarly, one study  demonstrated that spinal manipulation in addition to GP care was more cost-effective than GP care alone in reducing shoulder pain and related disability. The extra costs needed for 1-unit improvement in low back or shoulder pain/disability score or 1 QALY gained were lower than the WTP thresholds reported across the studies.
The cost-effectiveness of manual therapy for improving neck pain, disability, and QALYs gained in comparison with other treatments was not consistent across the studies. For example, one trial [63, 64] demonstrated the domination of chiropractic manipulation over physiotherapy or GP care in improving neck pain and QALYs gained. In 2 other trials, either alternative intervention (behavioral graded activity) was more cost-effective than manual therapy  or the probability for manual therapy being more cost-effective compared with advice plus exercise was too low. 
The evidence regarding cost-effectiveness of manual therapy compared with physiotherapy for reducing pain and disability related to ankle fractures, as reported in one study,  has been insufficient and inconclusive because of small sample size and uncertainty around the cost-effectiveness measure.
It is difficult to draw definitive conclusions regarding the comparative cost-effectiveness of manual therapy techniques in patients with spinal pain due to the paucity, clinical heterogeneity (eg, different techniques, wide variety of comparators), and study-related shortcomings (eg, small sample, short follow-up, high uncertainty in the estimates of ICERs) of the identified evidence. For example, the use of different manual therapy techniques (eg, manipulation, mobilization, and chiropractic care) in combination with other interventions (eg, physiotherapy, exercise, and GP care) leads to differential effectiveness profiles, thereby limiting the comparability of results across studies. The nonspecific or contextual effects (eg, intervention fidelity, placebo effect, practitioner's experience) due to the complexity of interventions and lack of patient blinding may have biased the study results for subjective outcome measures such as pain, disability, and QOL. Because none of the studies used a sham/control arm, it is difficult to tease out the specific effects of treatment from patients' differential expectation (or practitioner's experience/skill set) across the study treatment arms. 
All the included studies were trial-based economic evaluations. None of the studies used economic modeling to extrapolate beyond the trial data to look at the longer-term cost-effectiveness of the different interventions. Studies reporting cost-effectiveness acceptability curves (CEACs) used bootstrapping, none of the studies used simple one-way or multiway sensitivity analyses to check for uncertainty in any of the key cost factors, which may be driving the ICER.
Limitations and Strengths
The findings of this review are not directly comparable with those of other systematic reviews, [28-33, 71-81] given the differences in scope, research question, study inclusion/exclusion criteria, types of economic evaluation, and interventions. The findings of these reviews were either inconclusive because of the paucity and heterogeneity of the evidence for manual therapy [28-33] or showed some cost-effectiveness of manual therapy over alternative treatments (eg, usual care and exercise). [71, 75, 76, 78, 79, 81]
The applicability of findings of the included studies, despite them being pragmatic, may be limited to only countries with similar health care system and considerations of utility (eg, calculations based on the same QOL instrument). The applicability may also be limited by the differences in components of manual therapy interventions and short follow-ups of the studies.
The strengths of the current review include the reviewer's use of systematic and independent strategies to minimize the ROB in searching, identifying, selecting, extracting, and appraising the primary studies. The search strategy was applied to multiple electronic databases and other sources such as references of relevant primary studies and systematic reviews. Also, this review summarized the evidence from studies that evaluated costs and effectiveness simultaneously through cost-effectiveness and/or CUAs by providing ICERs. As a limitation, this review included only RCT-based cost-effectiveness evaluations.
This paper provides a platform for further research into the cost-effectiveness of manual therapy for the management of musculoskeletal conditions. The findings underscore the paucity of good-quality published evidence on this issue. This is based on the small number of identified RCTs focus of which is rather limited (ie, nonspecific spinal pain). The insufficient evidence on cost-effectiveness may be explained by difficulties in obtaining cost data, lack of expertise in economic outcomes, and/or perceived societal discomfort with assigning monetary units to human health.  Raising awareness among the chiropractic community about the importance of undertaking more high quality economic evaluations is needed.
Because several studies did not use QALYs as an outcome measure, this presents difficulty for decision makers if they wish to compare value for money across musculoskeletal conditions with other health conditions such as cancer and cardiovascular disease, in line with the cost-effectiveness thresholds set by NICE. Consideration of the competing demand/supply side issues of manual therapy and how these issues may vary across countries is needed. Furthermore, it is not clear whether the affordability of manual therapy in countries where the provision of such services fall outside publicly funded arrangements is likely to influence utilization; this raises questions about the generalizability of the current reported findings.
We recommend that future studies report unit cost calculation with costs broken down by each service to allow the judgment as to whether all relevant costs for a given perspective were considered and how the total costs were calculated. If ethically justifiable, future trials need to include sham or no treatment arm to allow the assessment and separation of nonspecific effects (eg, patient's expectation) from treatment effects. More exploration is warranted about which characteristics of manual therapy (eg, mode/frequency of administration or choice of spinal regions) are important for clinically relevant and patient-centered outcomes. Finally, greater consideration is needed to improve reporting quality of primary studies evaluating manual therapy.
Preliminary evidence from this review shows some economic advantage of manual therapy relative to other interventions used for the management of musculoskeletal conditions. However, at present, there is a paucity of evidence on the cost-effectiveness and/or cost-utility evaluations for manual therapy interventions. Further improvements in the methodological conduct and reporting quality of economic evaluations of manual therapy are warranted in order to facilitate adequate evidence-based decisions among policy makers, health care practitioners, and patients.