Michaleff ZA, Lin CW, Maher CG, van Tulder MW.
The George Institute for Global Health,
The University of Sydney,
Missenden Road, Sydney, NSW 2050, Australia.
BACKGROUND: Spinal manipulative therapy (SMT) is frequently used by health professionals to manage spinal pain. With many treatments having comparable outcomes to SMT, determining the cost-effectiveness of these treatments has been identified as a high research priority.
OBJECTIVE: To investigate the cost-effectiveness of SMT compared to other treatment options for people with spinal pain of any duration.
METHODS: We searched eight clinical and economic databases and the reference lists of relevant systematic reviews. Full economic evaluations conducted alongside randomised controlled trials with at least one SMT arm were eligible for inclusion. Two authors independently screened search results, extracted data and assessed risk of bias using the CHEC-list.
RESULTS: Six cost-effectiveness and cost-utility analysis were included. All included studies had a low risk of bias scoring =16/19 on the CHEC-List. SMT was found to be a cost-effective treatment to manage neck and back pain when used alone or in combination with other techniques compared to GP care, exercise and physiotherapy.
CONCLUSIONS: This review supports the use of SMT in clinical practice as a cost-effective treatment when used alone or in combination with other treatment approaches. However, as this conclusion is primarily based on single studies more high quality research is needed to identify whether these findings are applicable in other settings.
From the Full-Text Article:
Spinal pain, including neck pain and back pain, is a common condition in modern society (Woolf and Pfleger, 2003; Côté et al., 2003 ). It presents major social and economic burdens due to the high levels of chronicity and resultant long term disability which are associated with high costs in health care and losses of productivity (e.g. sick leave) (Woolf and Pfleger, 2003). While existing practice guidelines inform the individual, clinicians and policy makers on the effectiveness of a range of interventions, few provide information on the cost-effectiveness of treatments. It is arguable that cost-effectiveness of treatment is an equally important consideration as effectiveness, as all health administrators need to make decisions about how they allocate scarce health resources.
Economic evaluations are frequently conducted alongside randomised controlled trials of treatment effectiveness and involve the identification, measurement, valuation and then comparison of the costs and consequences (benefits) of two or more alternatives (Drummond et al., 2005). Economic evaluations are most useful when the treatments under question have been evaluated in terms of efficacy (can the treatment work in those who comply with the recommendations), effectiveness (is the treatment acceptable and does the treatment work in those who the treatment is offered) and availability (is the treatment accessible to all who would benefit from it). The result of an economic evaluation supplements the evidence base on treatment effectiveness by providing information on the efficiency or ‘‘value for money’’ of treatment alternatives (Drummond et al., 2005). This information can be used to inform consumers, insurers, governments and policy makers where the health budget should be spent.
Spinal manipulative therapy (SMT), including both manipulation (a high velocity thrust technique) and mobilisation (low velocity technique), is frequently used by a number of health professions, including physiotherapists, chiropractors and osteopaths, to manage people with neck pain and back pain (Gross et al., 2010; Assendelft et al., 2004). The effectiveness of SMT to treat spinal pain has been summarised in recent Cochrane Reviews (Gross et al., 2010; Assendelft et al., 2004; Rubinstein et al., 2011). Overall the evidence suggests that SMT provides greater improvements for pain and function than a placebo or no treatment but similar improvements to many competing treatments such as general practitioner management, medication and exercise. With many treatments for spinal pain having comparable outcomes to SMT, determining the cost effectiveness of these treatment alternatives has been identified as a high priority (Rubinstein et al., 2011). The purpose of this systematic review is to investigate the costeffectiveness of SMT compared to other treatment options for people with spinal pain of any duration.
Cost-effectiveness of SMT
4.1. SMT compared to GP care
One study investigated the cost-effectiveness of SMT versus general practitioner (GP) care (advice, education and drug prescription). In patients with neck pain, Korthals-de Bos et al. (2003) demonstrated SMT to be dominant over GP care, from a societal perspective, in terms of recovery and quality of life, as SMT was associated with lower total costs and higher rates of recovery. Interestingly, no difference was shown in the cost-effectiveness of SMT versus GP care for pain intensity and functional disability (Korthals-de Bos et al., 2003) which are the outcomes typically selected to judge effectiveness of SMT (Table 3). No studies were found to investigate the cost-effectiveness of SMT versus GP care for LBP.
4.2. SMT compared to exercise
Two studies adopting a societal perspective reported on SMT versus exercise in patients with neck pain (Bosmans et al., in press; Korthals-de Bos et al., 2003). Both studies found SMT to be a costeffective treatment option compared to an exercise program in terms of pain, recovery and QALY gains (Table 4). The cost-effectiveness plane by Korthals-de Bos et al. (2003) showed 98% of the cost-effect pairs for pain located in the South East quadrant suggesting that SMT is dominant compared with exercise. No studies were found to investigate the cost-effectiveness of SMT versus exercise in patients with LBP.
4.3. SMT plus GP care compared to other
Two studies reported on the cost-effectiveness of SMT plus GP care compared to GP care alone in LBP (UK BEAM Trial Team, 2004) or neck and LBP ( Williams et al., 2004); in addition one of the studies also compared SMT plus GP to GP care plus exercise (UK BEAM Trial Team, 2004). From a UK health care perspective both trials found SMT plus GP care to be a cost-effective treatment compared to GP care alone as both ICERs fell below the NICE threshold (GBP 20,000–GBP 30,000 per QALY gained) despite the different pain regions being under investigation (Williams et al.: ICER £3560 per QALY gained in 1999/2000 GBP; UK BEAM Trial Team: ICER = £4800 per QALY gained in 2000/2001 GBP). SMT plus GP care was also shown to be a cost-effective treatment when compared to GP care plus exercise with an ICER of £2300 per QALY gained in 2000/2001 GBP (Table 5).
4.4. SMT plus other treatment compared other
Three studies investigated the cost-effectiveness of a combined treatment approach, which involved SMT plus advice (delivered by a physiotherapist or GP) and exercise (Table 6). Two studies, one from a societal perspective (Niemisto et al., 2005) and the other from a health sector perspective (UK BEAM Trial Team, 2004), compared the combined treatment approach to GP care alone for LBP (UK BEAM Trial Team, 2004; Niemisto et al., 2005). From a societal perspective, Niemisto et al. (2005) found no difference between groups in terms of quality of life however the data suggest that the combined treatment incurred lower annual costs compared to GP care alone. With respect to pain and disability outcomes the data suggested the combined treatment was dominant over GP care alone however this was not supported by the conclusions made by the authors. From a healthcare perspective the UK BEAM Trial found the combined treatment to be cost-effective over GP care alone with a low ICER (£3800 per QALY gained in 2000/2001 GBP) (UK BEAM Trial Team, 2004).
For people with neck pain the most cost-effective treatment was dependant on the perspective, societal or health sector, and of the threshold for willingness to pay. Advice and exercise was generally more cost-effective in terms of changes to neck disability scores from both a societal and healthcare perspective. In terms of QALY gained the combined approach was more cost-effective from a societal perspective however from a healthcare perspective there was more uncertainty as to the most cost-effective treatment with SMT appearing slightly more advantageous (Lewis et al., 2007).
Six economic evaluations were included in this systematic review which evaluated the cost-effectiveness of SMT compared to other treatment options for people with neck and back pain. The studies which evaluated treatments for back pain were primarily UK studies conducted from a health sector perspective, while the studies of neck pain were Dutch and Finnish studies conducted from a societal perspective. Regardless of the perspective employed or the region of pain, SMT appears to be a cost-effective treatment when used alone or in combination with GP care or advice and exercise compared to GP care alone, exercise or any combination of these. However, as a result of only six studies being eligible for inclusion the majority of these conclusions are based on the findings of single studies.
The findings of this review have important clinical, research and policy implications. Clinically, SMT is a treatment technique frequently used by a number of health professionals to manage neck and back pain. Based on the available literature, this review supports the use of SMT in clinical practice as a cost-effective treatment when used alone or in combination with other treatment approaches. In some studies SMT was also shown to be less costly per unit gained when compared to education, exercise and GP care. These results are applicable to clinicians, who make recommendations about treatment options, as well to health consumer who wish to make informed decision about available health care options.
From a research perspective this review highlights the need for more high quality economic evaluations to be conducted alongside randomised controlled trials of treatment effectiveness. While this review summarised the available literature, due to a limited number of studies with heterogeneous populations, perspectives, settings and analyses it was not possible to pool the results of included studies. This resulted in a number of conclusions as to the cost-effectiveness of SMT for neck or back pain to be based on one study alone. Systematic reviews of randomized controlled trials have shown that results of one study are often not reliable and precise. There is no reason to believe that this would be different for economic evaluations, especially because sample sizes of economic evaluations are often (too) small, and that economic consequences differ across different health settings. Of the comparisons supported by two studies it is worth noting the agreement between the studies which increases the robustness of the conclusions which can be made. For example, a comparable ICER was reported by Williams et al. (2004) and (UK BEAM Trial Team, 2004) when SMT plus GP care was compared to GP care alone despite the different pain regions and price years reported.
A number of tools are available to assess the risk of bias of economic evaluations (e.g. CHEC-List Evers et al., 2005, BMJ Check-list (Drummond and Jefferson, 1996) and The Quality of Health Economic Studies instrument (Chiou et al., 2003). In our study we used the CHEC-List as it was generated through the consensus of international experts in a Delphi method (Evers et al., 2005). In general, the economic evaluations included in this study were of low risk of bias with all studies scoring P16 out of 19 on the CHEC-List. The key item identified for improvement in future studies is the inclusion of a sensitivity analysis (Item 15). While not an item in the CHEC-List, other methods of investigating the level of uncertainty in cost-effectiveness estimates include the cost-effectiveness acceptability curves and cost-effectiveness plane. These methods were presented in four of the six included studies (Bosmans et al., in press; Lewis et al., 2007; Niemisto et al., 2005; Williams et al., 2004) and five of the six included studies (Bosmans et al., in press; Lewis et al., 2007; Niemisto et al., 2005; Korthals-de Bos et al., 2003; Williams et al., 2004), respectively. Measures of uncertainty such as these provide a systematic way of dealing with the level of uncertainty around the results and should be an essential component in all economic evaluations to assess the robustness of the conclusions made (van der Roer et al., 2005; Briggs et al., 1994).
Economic evaluations are an essential consideration to inform and support health care and funding decision made by insurers, governments and policy developers. In treatments for spinal pain, as in the case of SMT, effectiveness studies are often able to demonstrate treatment effects when compared to no treatment, but fail to demonstrate which active treatment is more effective (Gross et al., 2010; Rubinstein et al., 2011). The findings of this review show that while the treatment effectiveness of SMT is comparable to other treatments, SMT is a cost-effective treatment option. Furthermore, this review found the most effective treatment or the least costly treatment to not always be the most cost-effective treatment. This demonstrates the valuable information that can be provided by economic evaluations beyond results of effectiveness alone and supports the need for more economic evaluation to be conducted.
This systematic review found SMT to be a cost-effective treatment to manage spinal pain when used alone or in combination with GP care or advice and exercise compared to GP care alone, exercise or any combination of these. The findings were primarily based on single studies conducted in the UK and the Netherlands. More high quality studies can support whether the findings of this review are applicable in other settings.