Skip Navigation

Rheumatology 2007 46(9):1495-1501; doi:10.1093/rheumatology/kem183
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Manca, A.
Right arrow Articles by Eaton, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Manca, A.
Right arrow Articles by Eaton, S.
Related Collections
Right arrow Psychology: Measurement and Management of Pain
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Randomized trial of two physiotherapy interventions for primary care back and neck pain patients: cost–effectiveness analysis

A. Manca, J. C. Dumville1, D. J. Torgerson1, J. A. Klaber Moffett2, M. P. Mooney2, D. A. Jackson2 and S. Eaton2

Centre for Health Economics, University of York, York YO10 5DD, 1Department of Health Sciences, University of York, York YO10 5DD and 2Institute of Rehabilitation, University of Hull, Hull HU3 2PG, UK.

Correspondence to: Dr Andrea Manca, Centre for Health Economics, Alcuin A Block, University of York, York YO10 5DD, UK. E-mail: am126{at}york.ac.uk


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Acknowledgements
 References
 
Objectives. To assess the cost-effectiveness of a brief physiotherapy pain management approach using cognitive-behavioural principles (Solution-Finding Approach) when compared with a commonly used traditional method of physical therapy (McKenzie Approach).

Methods. Economic evaluation conducted alongside a randomized trial. The study related incremental differences in costs and benefits associated with the Solution Finding and McKenzie approaches over 12 months. Costs were measured in UK pounds sterling. Benefit was measured as health-related quality of life using the EQ-5D, which was used to estimate patient-specific quality adjusted life years (QALYs).

Results. The McKenzie treatment required, on average, one extra physiotherapist visit (4.15 vs 3.10). Over a 12-month period, Solution Finding was associated with a lower per patient cost of £–24.4 (95% CI £–49.6 to £0.789). The mean difference in QALYs between the two groups was –0.020 (95% CI –0.057 to 0.017); favouring those receiving McKenzie. Relating incremental mean costs and QALYs gave an incremental cost effectiveness ratio of £1220 (–24.4/–0.020) suggesting the McKenzie treatment is cost effective.

Conclusions. Results suggest that the additional cost associated with the McKenzie treatment when compared with the Solution Finding Approach may be worth paying, given the additional benefit the approach seems more likely to provide. Further research is needed to assess the extent to which the difference in physiotherapy visits between the two strategies is generalizable to other treatment settings.

KEY WORDS: Neck and back pain, Primary care, Cost, Benefits, Physiotherapy


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Acknowledgements
 References
 
Musculoskeletal back and neck pain are both common. A recent UK survey estimated the (1-month period) prevalence of spinal pain to be 29% (95 CI 27–31%) [1]. The lifetime prevalence of these conditions is also high—it is estimated that ~70–85% of the population will experience some spinal pain during their lifetime, although often pain will be transient [2–4].

Such musculoskeletal conditions result in pain, loss of function and reduced quality of life. They are one of the most common reasons for primary care physician consultation in the UK [5] and the US [6]. In 1998, the cost of lower back pain alone to the UK National Health Service (NHS) was estimated at £1 billion [7], with over £200 million being spent on physiotherapy. Spinal pain-related morbidity also results in work absenteeism and loss of productivity, which incurs additional costs to society that run into billions [4, 8–10].

The treatment of non-specific musculoskeletal back and neck conditions in primary care is variable [11]. However, guidelines state that in the first instance patients should be encouraged to remain active, with the prescription of anti-inflammatory drug and/or analgesia where required [12, 13]. If, after a period of weeks, this approach is unsuccessful, a physiotherapy referral may be considered. In a US survey of primary care, 38% of patients presenting with non-serious and non-specific spinal problems were referred to a physical therapist [14]. In The Netherlands, 22% of patients also presenting with non-specific lower back pain were referred to a physiotherapist after their first consultation and 50% after a follow-up consultation [15].

Physiotherapy treatments aimed at alleviating the physical causes of back and neck pain include: advice, exercise programmes, massage, mobilization and manipulation [16], with varying amounts of research evidence regarding the clinical effectiveness [17–21] and cost effectiveness [22, 23] of these treatments. Conclusive recommendations are complicated by the range of different approaches available.

Additionally, a new type of intervention for treating back and neck pain has recently been developed, triggered by growing awareness that psychosocial factors play an important role in musculoskeletal complaints [24–28]. These behavioural interventions have different compositions depending on the specific theory underpinning the approach. A systematic review of such interventions in the treatment of back pain reported that cognitive–behavioural treatments that aim to modify the way patients view their pain, may be clinically effective when compared with waiting list controls and as effective as exercise therapy [29]. There is now a growing interest in the effectiveness of these interventions when delivered by physiotherapists in the place of more traditional physical therapies [30, 31].

As with all medical decisions regarding the implementation of alternative treatment opinions, the cost effectiveness of physiotherapy treatments must be assessed. Whilst clinical data is vital for informing policy and practice; elucidating whether a treatment offers good value for money in terms of cost vs benefit must also be considered. A previous study has found that a brief physiotherapy-delivered intervention, based on cognitive–behavioural principles, may be cost effective when compared with a standard physiotherapy approach in the treatment of neck pain [32]. Physiotherapy-delivered exercise classes using a cognitive–behavioural approach have also been reported to reduce health costs and work absenteeism over 12 months, when compared with continued treatment in primary care [33].

This study reports the cost–effectiveness analysis of data collected alongside a recently completed randomized controlled trial [34]. The trial compared two physiotherapists delivered interventions for musculoskeletal back and neck pain, which aimed to promote return to normal activities. The first of these, the Solution Finding Approach, was a brief physiotherapy intervention based on cognitive–behavioural principles. The approach takes a patient-centred view and, in this context, aims to help patients identify reasons for their pain and to provide solutions and long-term management strategies. The second approach was the more traditional biomechanical approach used by physiotherapists, the McKenzie approach, which involves classification of the patient's spinal condition and the prescription of specific therapeutic exercises. The McKenzie approach is commonly used by physiotherapists in the UK and has been assessed in numerous studies [35].


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Acknowledgements
 References
 
Study design
Full details of the trial design are provided elsewhere [34]. Briefly, participants were recruited from general practitioner referrals to seven different NHS physiotherapy departments in West and East Yorkshire, UK. People were screened for potential inclusion into the trial if they were aged 18 or over and had suffered from back or neck pain that was considered to be non-systemic in origin for more than 2 weeks. Individuals with a score of less than 4 on the Roland Morris Disability Questionnaire [36] or less than 10 (9 if they were non-car drivers) on the Neck Pain Questionnaire [37] were excluded. Patients were also excluded if they had received any physical therapy within the previous 3 months or were currently planning to use private physical therapy alongside NHS physiotherapy; had possible serious pathology; were pregnant or had been referred by a hospital consultant. Ethical approval for the study was obtained from the relevant Research and Ethics Committees. All participants gave informed consent. The trial has been assigned the International Standardised Randomised Controlled Trial Number ISRCTN48919562 [controlled-trials.com] . When recruited to the trial, participants were randomized into one of the following interventions.

Interventions
McKenzie approach
The system has been clearly documented [38] and is an approach commonly used by physiotherapists. The physiotherapist conducted a biomechanical assessment using repeated movements of the spine and, based on these findings, prescribed specific exercises for the patients to work on repeatedly themselves. The approach relies on active compliance with the exercises and advice provided. Follow up sessions to monitor progress were offered based on the physiotherapists’ clinical discretion.

Solution finding approach
Based on cognitive–behavioural principles, this approach involves physiotherapists guiding patients in identifying any problems related to their pain, developing solutions and setting goals. Patients allocated to this approach received an initial assessment, which included an interview, a brief physical examination, explanation about the condition, reassurance and goal setting. One or two follow-up sessions were offered to guide progress.

Trial physiotherapists were trained to deliver either intervention. All the physiotherapists delivering the McKenzie approach were experienced in this method and had undertaken McKenzie Institute training (courses A–D). The physiotherapists also had a day's training with two senior physiotherapists, recognized by the McKenzie Institute, to familiarize them with the trial protocol and procedures. Since the Solution Finding Approach was a newly developed intervention, physiotherapists were provided with a day and a half of training that included evidence-based therapy and practical application of enhanced communication skills and cognitive-behavioural principles. More details of this are provided elsewhere [34]. A small observational study enabled a sample of the consultations to be scrutinized in depth [39]. This provided evidence that for the most part, although physiotherapists were asked to deliver both types of intervention, delivery of these was close to that described in the trial protocol.

After randomization to physiotherapy treatment approach, trial participants were also randomized to receive an educational booklet (the Back Book [39] or the Neck Book [40]). This book was inexpensive (less than £1.30 per copy [23]) and was delivered to (half of) both groups. It would not have caused incremental differences between trial arms and was therefore not considered in this economic analysis.

Resource use measurement
Contact with physiotherapist
The number of physiotherapy visits made by each participant was recorded by the treating physiotherapist over the duration of the trial. The physiotherapist also recorded the number of telephone calls they received from participants for consultation. To ensure that the length of the initial and subsequent physiotherapist's appointment for the McKenzie and Solution Finding approaches were accurately estimated, a survey of trial physiotherapists was conducted.

Additional resource use
At 6 and 12 months follow-up, each trial participant was asked to complete a postal questionnaire and case record form (CRF). The CRF recorded each participant's use of additional NHS resources by asking them to recall the number of visits to general practice, outpatient clinic or hospital specialists over the interim 6-month period. The CRF also asked participants to supply details of any private expenditure they had incurred due to their back or neck pain. This included appointments with any private health professionals or therapists, as well as the cost of any equipment purchased. Finally, participants were asked to record the number of days of work they had missed due to their back or neck pain if they were in paid employment. Participants not in paid employment were asked to record how many days of normal activities they had missed. In all cases where blanks were left on the CRF in response to resource use questions, we assumed a value of zero.

Unit costs
The cost–effectiveness analysis was carried out from the perspective of the UK Health Provider, the NHS [41]. Healthcare and private resources utilized by the trial participants and recorded in the CRF were valued using 2003–04 costs, in pounds sterling: adjusting of costs to this price year was undertaken where required. The cost of the physiotherapy treatment sessions in each trial arm was calculated using UK average physiotherapist costs [42], which were applied to average session times provided by physiotherapists in the survey circulated. The unit cost of further NHS and private health professionals were taken from a recently published trial in this area [32], and up-rated as required [42]. The cost of a private podiatrist visit was identified through a survey of local practitioners (n = 10). The costs of items purchased by the participants were taken directly from the CRF.

Table 1 shows the unit costs of healthcare resources. Given the time horizon of 12 months costs were not discounted. Private expenditure incurred by participants was not included in the base case analysis; following National Institute for Health and Clinical Excellence (NICE) guidelines for technology assessment this analysis was reported separately [41].


View this table:
[in this window]
[in a new window]

  		TABLE 1. Unit costs employed in healthcare resource use valuation

 
Health outcomes
Each trial participant was asked to complete the Euro-Qol-5D questionnaire (EQ-5D) at baseline, 6 weeks, 6 months and 12 months. The EQ-5D questionnaire is a widely recognized and validated generic measure of health related quality of life [43]. It contains five questions, each covering a specific dimension: mobility, self-care, ability to undertake usual activity, pain/discomfort. For each of the five questions there are three possible responses: no problems, moderate problems and severe problems. Based on their combined answers to the EQ-5D questionnaire, participants were classified as being in one of 243 possible health states (plus unconscious or dead). Each of these health states has an associated utility score, derived from interviews with a large sample of the UK population. These scores are on a scale of 0 (dead) to 1 (excellent health). Individual's EQ-5D values were used to calculate their specific Quality Adjusted Life Years (QALYs) at 12 months using the area under curve method [44, 45]. The use of QALYs is widely recognized and is the measure of health benefit used by NICE [41], and has the advantage of reflecting both health-related quality of life and mortality into a single index. Given the time horizon of 12 months, QALYs were not discounted.

Cost–effectiveness analysis
Cost–effectiveness analysis was carried out by relating incremental costs at the end of the study period (i.e. mean costs for Solution Finding arm minus mean costs for McKenzie arm) to incremental QALYs (i.e. mean QALYs for Solution Finding arm minus mean QALYs for McKenzie arm). In relating incremental costs and QALYs one of the following scenarios could occur:

  1. The Solution Finding Approach is both more effective (that is, generates larger QALYs), or at least as effective as, the McKenzie approach and is less costly;
  2. The Solution Finding Approach is less effective than the McKenzie approach and more (or at least as) costly.
  3. The Solution Finding Approach is both less (more) effective and less (more) costly compared with the McKenzie approach.

In a cost–effectiveness analysis, if the results indicate either scenarios (a) or (b) above, then one approach is clearly more cost effective, that is, it dominates the other. However, if the results indicate scenario (c) then a ‘decision rule’ is required to assess which is the most cost effective treatment. The decision rule requires the calculation of the incremental cost–effectiveness ratio (ICER)—the ratio of the mean incremental costs and the mean incremental QALYs between the two trial arms. The ICER represents the additional cost that decision makers are (on average) expected to pay to achieve an additional QALY in this population. A treatment strategy is considered to be cost-effective if the decision maker's willingness to pay for an additional QALY is at least equal to (or greater than) the ICER.

To aid decision-making, the results from the cost-effectiveness analysis are combined with the ‘decision rule’ discussed earlier, and graphically presented through a cost–effectiveness acceptability curve (CEAC) [46, 47]. The CEAC shows the probability that Solution Finding is more cost-effective than McKenzie for a range of different amounts the decision-maker may be willing to pay for an additional QALY.

The analysis was carried out in STATA [48]. Mean incremental costs and QALYs between trial arms were calculated using bivariate normal regression, adjusting for baseline EQ-5D score [45], whilst maintaining the correlation between costs and QALYs. For the base case analysis missing EQ-5D data were imputed using a multiple imputation approach [49, 50], assuming that data were missing at random [51, 52]. Using this technique, missing values are imputed from a distribution based on the relationship between chosen study predictors and the variable of interest. Imputation is carried out from this distribution multiple times, leading to the creation of several data sets. The results from each data set are then pooled together into a single analysis [52].

Sensitivity analyses
To further assess the impact of this missing data on the study results three additional analyses were carried out [53]. Firstly, complete case analysis was conducted. Secondly, where data were missing in each dimension of the EQ-5D, the mode value (calculated for each treatment arm) was imputed. Finally an extreme sensitivity analysis replaced missing EQ-5D data in the cost effective arm with the worst possible value (i.e. 3), the lowest EQ-5D dimension score, whilst missing EQ-5D data in the non-cost effective arm was replaced with 1 (i.e. no problems).


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Acknowledgements
 References
 
The clinical findings of this study suggested that both the McKenzie and Solution Finding approaches lead to improvements in patient outcomes over time, with no significant difference between the two treatments [34]. As indicated in the clinical report the findings of this cost effectiveness are vital in informing the decision-making process regarding the use of these treatments.

In total 315 participants took part in this trial, 161 participants were randomized to the McKenzie arm, and 154 to the Solution Finding arm. Four participants were excluded from the cost–effectiveness analysis (one participant from the Solution Finding arm and three from McKenzie) since these participants did not contribute any cost or outcome data during follow-up.

Resource use
Details of resource use associated with the two trial treatments are reported in Table 2. The McKenzie treatment required, on average, one extra subsequent visit to the physiotherapist (4.15 vs 3.10). Very few participants accessed their physiotherapist by phone contact during the trial. Uptake of further NHS services outside the trial treatments was limited and there was little difference between the two arms.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Main healthcare resource use (total n = 311)

 
Table 3 shows the costs associated with the resource use during the trial. From an NHS perspective the incremental mean cost between the two arms was £–24.4 (95% CI £–49.6 to £0.789), with the Solution Finding arm incurring a slightly lower cost to the NHS. This cost difference was attributed to the reduced mean physiotherapist visit in this group. The cost of private expenditure, excluding days usual activity lost, was slightly higher in the Solution Finding group, thus when NHS and private expenditure were combined the incremental mean cost was reduced to £–8.30 (95% CI £–72.8 to £56.3). However, a larger proportion of participants in the McKenzie trial arm missed at least 1 day of usual activity compared with the Solution Finding arm (26 vs 19%), thus when the cost of number of days of usual activity lost was also included in the analysis, the cost of the McKenzie group further increased as did the incremental mean cost in favour of the Solution Finding Approach £–117 (95% CI £–530 to £295).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Costs associated with trial treatments (2003–04 prices in £ Sterling)

 
Health outcomes
Table 4 details the EQ-5D scores at baseline, 6 weeks, 6 months and 12 months. On average, patients in both treatment arms showed continued improvement at each follow-up period up to 12 months. For the base case analysis, over the 12 month period the mean adjusted difference in QALYs between the two groups was –0.020 (95% CI –0.057 to 0.017); with those trial participants receiving McKenzie having increased utility when compared with those receiving the Solution Finding Approach.


View this table:
[in this window]
[in a new window]

  		TABLE 4. EQ-5D and QALY estimates

 
Cost–effectiveness analysis
The results presented so far indicate that the Solution Finding Approach is slightly cheaper than the McKenzie approach but confers marginally lower benefit, hence resulting in scenario (c) as described in the Methods section. Relating the trial incremental mean costs and incremental mean QALYs gives an ICER of £1220 (–24.4/–0.020). The policy maker needs to decide whether she or he is willing to invest additional health care resources funding the McKenzie approach and pay an average of £1220 per patient, in order to acquire an additional unit of health outcome (e.g. QALY) in this patient population. Given that in the UK NICE is commonly quoted as regarding willingness to pay between £20 000 and £30 000 per additional QALY, these results suggest that the additional cost associated with the McKenzie treatment is worth paying, given the additional benefit it provides. Figure 1 confirms that as soon as a provider is willing to pay £1000 or more per additional QALY, the Solution Finding Approach has very little probability of being cost effective. Using the combined NHS and private cost, excluding days lost to usual activity, reduces the ICER further to £415 (–8.30/–0.020) in favour of McKenzie. When the cost of lost days are also included the ICER increases to £5850 (–117/–0.020), thus the McKenzie approach remains cost effective.


Figure 1
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 1. Cost–effectiveness acceptability curve of Solution Finding vs McKenzie— base case analysis.

 
Sensitivity analyses
The mean difference in QALYs for the complete case analysis was –0.023 (95% CI –0.066 to 0.019), leading to an ICER or £1061 and a CEAC similar to Fig. 1 (Fig. 2). When QALYs were calculated from EQ-5D scores where missing data were replaced by treatment arm-specific dimension's modes, the improvement in utility at each time point remained—however, the EQ-5D means at each follow-up period were consistently higher when compared with the base case analysis and the complete case analysis (Table 4). The incremental mean QALYs over 12 months was larger compared with the base case and complete case analysis at –0.034 (95% CI –0.064 to –0.004), giving an ICER of £718—further reducing the likelihood of Solution Finding being cost effective (Fig. 3).


Figure 2
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 2. Cost–effectiveness acceptability curve of Solution Finding vs McKenzie—complete case analysis.

 

Figure 3
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 3. Cost–effectiveness acceptability curve of Solution Finding vs McKenzie—imputation of mode EQ-5D dimension value.

 
To further investigate the impact of the missing EQ-5D data on the base case findings the results were challenged with an extreme sensitivity analysis (results not reported in the table). All missing EQ-5D dimension data in the cost effective approach were assumed to have the worst utility possible, while all missing EQ-5D dimension data in the Solution Finding arm, the best. In this situation the ICER is £–369 (–24.4/0.066), this results is described by scenario (a) in the Methods section, Solution Finding is both cheaper and more effective and thus dominates. However, this situation is extreme and thus unlikely to occur.


   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Acknowledgements
 References
 
This study reports the economic evaluation of a physiotherapist-delivered brief cognitive–behavioural intervention compared with the McKenzie physiotherapy approach in treating back and neck pain. The mean per patient cost of the McKenzie approach was £129, with the Solution Finding Approach offering approximately a £24 saving over the 12-month follow-up period. This saving represented one less visit to a physiotherapist in the Solution Finding Group. Whilst the McKenzie approach was, on average, the more costly approach, its additional patient benefit meant that it was more likely to be the cost-effective treatment, as shown in Fig. 1. The McKenzie approach remained the cost-effective treatment in sensitivity analyses, unless the data were challenged with an extreme situation in terms of missing data. The McKenzie approach also remained the cost-effective option, when private and societal costs were included in the analysis. This finding highlights the importance of relating costs to benefits in an economic analysis rather than focusing on cost alone. When allocating health care resources from a limited budget, as in the NHS, decision makers aim to maximize public health benefit given available resources.

The clinical study that accompanies this cost-effectiveness analysis [34] suggested that both the McKenzie and Solution Finding Approach led to clinical improvements, measured using standard assessment instruments (Roland Morris Disability Questionnaire and Neck Pain Scores), with no significant difference in scores between the two trial arms. However, the McKenzie approach did increase patient satisfaction when compared with the Solution Finding Approach [34]. This cost effectiveness analysis further justifies use of the McKenzie approach in terms of value for money.

When considering the generalizability of these findings, the costs used were NHS costs and thus represent physiotherapists working in the UK public sector. The McKenzie treatment was delivered by trained and experienced staff across a number of centres and is taken to represent normal practice. However, the Solution Finding Approach must be recognized as just one type of brief intervention. It might have been more effective if the physiotherapists had been trained to be more confident and competent in their delivery. As noted previously, the Solution Finding Approach was a new intervention with a fundamentally different emphasis from the traditional physiotherapy approach, which relies on clinical reasoning based mainly on pathology and biomechanics. The Solution Finding Approach required a change in mind set for the therapist and competence in the use of newly learned skills. The ability to apply cognitive–behavioural skills would be important and could be pivotal for its effectiveness. The participating physiotherapists were provided with a day and a half's training and additional supporting written material. They were also asked to practice the techniques on five patients prior to participating in the study. However, this may have been insufficient for optimal competence. In terms of cost differences, in this study a further reduction in costs (i.e. a reduction in the number of visits in the Solution Finding arm, or increase in the McKenzie arm) would, alone, have been unlikely to make the Solution Finding Approach cost-effective. An increase in patient benefit in the Solution Finding arm when compared with the McKenzie arm would also be required.

A previous cost–effectiveness analysis of a UK randomized controlled trial has compared the costs and benefits of a brief physiotherapy-delivered intervention, based on cognitive–behavioural principles, with usual physiotherapy care in the treatment of neck pain over a 12-month period [32]. The study found an incremental mean difference in costs of –£68 (95% CI: –103 to –35), also in favour of the brief intervention (£152 in the usual care arm vs £84 in the brief intervention arm). The incremental difference in mean QALYs was –0.001 (95% CI: –0.030 to 0.028) in favour of usual care. Such a small difference in benefits meant that, in this case, results were driven by costs. Thus, the brief intervention could be the most cost-effective choice. However, there was much uncertainly around the decision and, at a willingness-to-pay figure of £30 000, the probability that the brief intervention was cost-effective was only approximately 50%. The study results were modified by an interaction between a patient's preference for a treatment and the treatment received after randomization. That is, the brief intervention seemed cost-effective for those patients who had a preference for it prior to randomization, and who were then randomly allocation to this trial arm.

A second recent study [23] reported that the use of a (non-behavioural) brief physiotherapy advice intervention of one session was more cost effective than usual physiotherapy in the treatment of back pain over a 12-month period, but only when private expenditure was considered. From a health provider's perspective it was unclear which treatment was cost-effective, although the traditional approach had a 73% probability of being cost-effective if willingness to pay was £30 000/QALY. Interestingly, this study reported a mean cost difference of £20 favouring the brief intervention and a mean QALY difference of 0.02 in favour of traditional physiotherapy.

The findings reported here suggest there is evidence that McKenzie physiotherapy may be a cost-effective approach for the treatment of back and neck pain. It is important to note that traditional statistical inferential decision rules are unhelpful in cost–effectiveness analysis. Healthcare cost and QALYs are often secondary (if not tertiary) outcomes in randomized clinical trials which mean that trials are almost always underpowered to detect statistically significant differences in cost and effects. This trial is not an exception. Policy decisions regarding treatment for the patient group considered in this trial will rest on the best current evidence, to which this study contributes. In this sense, this study suggests that Solution Finding has a low probability of being cost-effective (~20%) when the decision maker is willing to pay £1000 or more per additional QALY.

Finally, in placing these findings in context, the aim of any cost–effectiveness analysis is to aid decision-making. In the case of physiotherapy treatment for back and neck pain, where there are a number of approaches and a large body of evaluative trials reporting a range of comparisons, it is difficult to draw firm or detailed conclusions from a single study. For example, there is evidence that spinal manipulation is a cost-effective treatment for back pain, when compared with exercise therapy and primary care treatment [22]. Yet, the relative cost–effectiveness of manipulation versus other approaches, such as behavioural interventions, is unclear. In such situations the application of more comprehensive evidence synthesis approaches to inform policy making [54, 55] allow the evidence regarding a number of treatment options to be compared simultaneously. Such analysis could be utilized further to help draw existing clinical and cost data together to aid practice and policy decision-making regarding physiotherapy treatments for the spine in general. In addition, further work is required to understand the importance of patient preference in this area.

Formula


   Acknowledgements
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Acknowledgements
References
 
The authors gratefully acknowledge the participation in this trial of all patients and staff in all the collaborating centres. We also thank the two reviewers for their helpful suggestions. This study was funded by the Arthritis Research Campaign. Andrea Manca is recipient of a Wellcome Trust funded post-doctoral Training Fellowship in Health Services Research (grant number GR071304MA). The views and opinions expressed therein are the authors’ and do not necessarily reflect those of the funding institutions.

The authors have declared no conflicts of interest.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Acknowledgements
 References
 

  1. Webb R, Brammah T, Lunt M, Urwin M, Alison T, Symmons D. Prevalence and predictors of intense, chronic and disabling neck and back pain in the UK general population. Spine (2003) 28:1195–202.[CrossRef][Web of Science][Medline]
  2. Andersson G. Epidemiological features of chronic low-back pain. Lancet (1999) 354:581–5.[CrossRef][Web of Science][Medline]
  3. Cote P, Cassidy JD, Carroll L. The Saskatchewan health and back pain survey. The prevalence of neck pain and related disability in Saskatchewan adults. Spine (1998) 23:1689–98.[CrossRef][Web of Science][Medline]
  4. Boswell MV, Shah RV, Everett CR, et al. Interventional techniques in the management of chronic spinal pain: evidence-based practice guidelines. Pain Physician (2005) 8:1–47.[Medline]
  5. McCormick A, Fleming D, Charlton J. Morbidity statistics from general practice. Fourth national study 1991–92. In: Office of Popfpulation Censuses and Surveys (1995) London: HMSO.
  6. Carey TS, Evans A, Hadler N, Kalsbeek W, McLaughlin C, Fryer J. Care-seeking among individuals with chronic low back pain. Spine (1995) 20:312–7.[Web of Science][Medline]
  7. Maniadakis N, Gray A. The economic burden of back pain in the UK. Pain (2000) 84:95–103.[CrossRef][Web of Science][Medline]
  8. Clinical Standards Advisory Group. Back pain: report of a Clinical Standards Advisory Group on back pain (1994) London: HMSO.
  9. Rizzo JA, Abbott TA 3rd, Berger ML. The labor productivity effects of chronic backache in the United States. Med Care (1998) 36:1471–88.[CrossRef][Web of Science][Medline]
  10. Borghouts JA, Koes BW, Vondeling H, Bouter LM. Cost-of-illness of neck pain in The Netherlands in 1996. Pain (1999) 80:629–36.[CrossRef][Web of Science][Medline]
  11. Frankel BS, Moffett JK, Keen S, Jackson D. Guidelines for low back pain: changes in GP management. Fam Pract (1999) 16:216–22.[Abstract/Free Full Text]
  12. Waddell G, McIntosh A, Lewis M, Hutchinson A. Low back pain evidence review (1996) London: Royal College of General Practitioners.
  13. Cost B13 Working Group. European guidelines for the management of low back pain: Research Commission Research Directorate General. Accessed March 2007. http://www.backpaineurope.org.
  14. Freburger JK, Carey TS, Holmes GM. Management of back and neck pain: who seeks care from physical therapists? Phys Ther (2005) 85:872–86.[Abstract/Free Full Text]
  15. Schers H, Braspenning J, Drijver R, Wensing M, Grol R. Low back pain in general practice: reported management and reasons for not adhering to the guidelines in The Netherlands. Brit J Gen Pract (2000) 50:640–4.
  16. van Tulder M, Koes B, Malmivaara A. Outcome of non-invasive treatment modalities on back pain: an evidence-based review. Eur Spine J (2006) 15:S64–81.[CrossRef][Web of Science][Medline]
  17. Gross AR, Hoving JL, Haines TA, et al. Cervical overview group. Manipulation and mobilisation for mechanical neck disorders. In: Cochrane Database of Systematic Reviews 2004. (Issue 1). Art. No.: CD004249. DOI: 10.1002/14651858.CD004249.pub2.
  18. Hayden JA, van Tulder MW, Malmivaara A, Koes BW. Exercise therapy for treatment of non-specific low back pain. In: Cochrane Database of Systematic Reviews 2005. (Issue 3). Art. No.: CD000335. DOI: 10.1002/14651858.CD000335.pub2.
  19. Assendelft WJJ, Morton SC, Yu Emily I, Suttorp MJ, Shekelle PG. Spinal manipulative therapy for low-back pain. In: Cochrane Database of Systematic Reviews 2004. (Issue 1). Art. No.: CD000447. DOI: 10.1002/14651858.CD000447.pub2.
  20. Haraldsson BG, Gross AR, Myers CD, et al. Massage for mechanical neck disorders. In: Cochrane Database of Systematic Reviews 2006. (Issue 3). Art. No.: CD004871. DOI: 10.1002/14651858.CD004871.pub3.
  21. UK BEAM Trial Team. United Kingdom back pain exercise and manipulation (UK BEAM) randomised trial: effectiveness of physical treatments for back pain in primary care. Br Med J (2004) 329:1377.[Abstract/Free Full Text]
  22. UK BEAM Trial Team. United Kingdom back pain exercise and manipulation (UK BEAM) randomised trial: cost effectiveness of physical treatments for back pain in primary care. Br Med J (2004) 329:1381.[Abstract/Free Full Text]
  23. Rivero-Arias O, Gray A, Frost H, Lamb SE, Stewart-Brown S. Cost-utility analysis of physiotherapy treatment compared with physiotherapy advice in low back pain. Spine (2006) 31:1381–7.[CrossRef][Web of Science][Medline]
  24. Gatchel RJ, Polatin PB, Mayer TG. The dominant role of psychosocial risk factors in the development of chronic low back pain disability. Spine (1995) 20:2702–9.[Web of Science][Medline]
  25. Gatchel RJ. Comorbidity of chronic pain and mental health disorders: the biopsychosocial perspective. Am Psychol (2004) 59:795–805.[CrossRef][Medline]
  26. Vlaeyen JW, Linton SJ. Fear-avoidance and its consequences in chronic musculoskeletal pain: a state of the art. Pain (2000) 85:317–32.[CrossRef][Web of Science][Medline]
  27. Linton SJ. A review of psychological risk factors in back and neck pain. Spine (2000) 25:1148–56.[CrossRef][Web of Science][Medline]
  28. Pincus T, Vlaeyen JWS, Kendall NAS, Von Korff MR, Kalauokalani DA, Reis S. Cognitive-behavioural therapy and psychosocial factors in low back pain: directions for the future. Spine (2002) 27:E133–8.[CrossRef][Medline]
  29. Ostelo RWJG, van Tulder MW, Vlaeyen JWS, Linton SJ, Morley SJ, Assendelft WJJ. Behavioural treatment for chronic low-back pain. In: Cochrane Database of Systematic Reviews 2005. (Issue 1). Art. No.: CD002014. DOI: 10.1002/14651858.CD002014.pub2.
  30. Klaber Moffett JA, Jackson DA, Richmond SJ, et al. Randomised trial of a brief physiotherapy intervention compared with usual physiotherapy for neck pain patients: outcomes and patients’ preference. Br Med J (2005) 330:53–4.[Free Full Text]
  31. Hay EM, Mullis R, Lewis M, et al. Comparison of physical treatments versus a brief pain-management programme for back pain in primary care: a randomised clinical trial in physiotherapy practice. Lancet (2005) 365:2024–30.[CrossRef][Web of Science][Medline]
  32. Manca A, Epstein D, Torgerson DJ, et al. Randomized trial of a brief physiotherapy intervention compared with usual physiotherapy for neck pain patients: cost-effectiveness analysis. Int J Technol Assess Health Care (2006) 22:67–75.[Web of Science][Medline]
  33. Klaber Moffett JA, Torgerson DJ, Bell-Syer S, et al. Randomised controlled trial of exercise for low back pain: clinical outcomes, costs, and preferences. Br Med J (1999) 319:279–83.[Abstract/Free Full Text]
  34. Klaber Moffett JA, Jackson DA, et al. Randomized trial of two physiotherapy interventions for primary care neck and back pain patients: ‘McKenzie’ vs brief physiotherapy pain management. Rheumatol (2006) 45:1514–21.[Abstract/Free Full Text]
  35. Aina A, May S, Clare H. The centralization phenomenon of spinal symptoms — a systematic review. Man Ther (2004) 9:134–43.[CrossRef][Web of Science][Medline]
  36. Roland M, Morris R. A study of the natural history of back pain. Part I: development of a reliable and sensitive measure of disability in low-back pain. Spine (1983) 8:141–4.[Web of Science][Medline]
  37. Leake A, Cooper J, Dyer S, Williams K, Turner-Stokes L, Frank A. The Northwick Park Neck Pain Questionnaire, devised to measure neck pain and disability. Br J Rheumatol (1994) 33:469–74.[Abstract/Free Full Text]
  38. McKenzie R MS. The lumbar spine—mechanical diagnosis and therapy (2003) 2nd edn. Waikanae, New Zealand: Spinal Publications Ltd.
  39. Roland M, Waddell G, Klaber Moffett J, Burton K, Main C. The back book (2002) London: The Stationery Office. 1–21.
  40. Waddell G, Klaber Moffett J, Burton K. The neck book (2004) London: The Stationery Office. 1–27.
  41. NICE. Guide to the methods of technology appraisal. (2004) April. London: National Institute for Clinical Excellence. www.nice.org.uk.
  42. Netten A, Dennett J, Knight J. Unit costs of health and social care. In: PSSRU (2004) Canterbury: University of Kent.
  43. Kind P. The EuroQol instrument: an index of health-related quality of life. In: Quality of life and pharmacoeconomics in clinical trials—Spiker B, ed. (1996) Philadelphia: Lippincott-Raven.
  44. Manca A, Hawkins N, Sculpher MJ. Estimating mean QALYs in a trial-based cost-effectiveness analysis: the importance of controlling for baseline utility. Health Econ (2005) 14:487–96.[CrossRef][Web of Science][Medline]
  45. Matthew JNS, Altman D, Campbell MJ. Analysis of serial measurements in medical research. Br Med J (1990) 300:230–5.[Abstract/Free Full Text]
  46. Fenwick E, Claxton K, Sculpher MJ. Representing uncertainty: the role of cost-effectiveness acceptability curves. Health Econ (2001) 10:779–87.[CrossRef][Web of Science][Medline]
  47. Van Hout BA, Al MJ, Gordon GS, Rutten FFH. Costs, effects and C/E ratios alongside a clinical trial. Health Econ (1994) 3:309–19.[Web of Science][Medline]
  48. StataCorp. Stata statistical software Version 9.0. (2005) College Station: TX Stata Corporation.
  49. Royston P. Multiple imputation of missing values. Stata J (2004) 4:227–41.
  50. Royston P. Multiple imputation of missing values: update. Stata J (2005) 5:188–201.
  51. Rubin D. Inference and missing data. Biomertrika (1976) 63:581–92.[CrossRef]
  52. Rubin D. Multiple imputation for non-response in surveys (1987) New York: John Wiley Sons.
  53. Manca A, Palmer S. Handling missing data in patient-level cost-effectiveness analysis alongside randomised clinical trials. Appl Health Econ Health Policy (2005) 4:65–75.[CrossRef][Medline]
  54. Ades AE, Sculpher M, Sutton A, et al. Bayesian methods for evidence synthesis in cost-effectiveness analysis. Pharmacoeconomics (2006) 24:1–19.[CrossRef][Medline]
  55. Sutton AJ, Abrams KR. Bayesian methods in meta-analysis and evidence synthesis. Stat Methods Med Res (2001) 10:277–303.[Abstract/Free Full Text]
  56. ONS. New earnings survey: average gross weakly earnings by area (2002) London: Office of National Statistics.
Submitted 28 March 2007; revised version accepted 7 June 2007.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Manca, A.
Right arrow Articles by Eaton, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Manca, A.
Right arrow Articles by Eaton, S.
Related Collections
Right arrow Psychology: Measurement and Management of Pain
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?