Aktives individuelles Training mit Gruppentraining macht sich bezahlt
Abstract:
Study Design. Randomized controlled trial.
Objective. To compare two fast-access evidence-based interventions for the treatment of simple low back pain.
Subjects. People aged 18-65 with a new episode of simple back pain causing them to be off work or on modified work for less than 1 year.
Setting. A district general hospital using a consultation/treatment room and a rehabilitation gym.
Methods. One group received a back advice booklet and one session of advice and then followed the normal route of care as directed by their general practitioner. The other group received the back booklet and one session of advice and also a back program. The back program consisted of a full assessment, immediately followed by one individual treatment, and then exercise classes over 1 to 2 weeks.
Outcome Measures. (1) Rate of return to work, (2) pain scores and health status, and (3) cost effectiveness of interventions versus the financial cost to the individual or employer.
Results. On average, those receiving the individual treatment and group exercise took 7 days less off work. This represented a 35% reduction in the amount of time taken off work since study entry. The estimated cost saving of providing the extra service of a simple back program ranged between [pounds]250 (US$367, euro300) and [pounds]578 (US$850, euro694) for each patient.
Conclusion. The results indicate that the costs of this active back program are more than reimbursed as a consequence of earlier return to work.
(C) 2005 Lippincott Williams & Wilkins, Inc.
Fulltext: http://www.medscape.com/viewarticle/506674
Introduction
The effectiveness of return to work interventions which attempt to promote increased activities and early return to work is a crucial area in which there is limited scientific evidence.[1] The Clinical Standards Advisory Group for back pain (CSAG) guidelines suggest that the sooner an active exercise and rehabilitation approach is implemented, the better.[2]
The cost of setting up hospital-based fast-access programs for back pain has hindered their establishment. However, there has been increasing evidence that simple approaches can be effective for back pain, and this might indicate a direction for progress.
Once fast access to assessment for low back pain is provided, the question of what is the ideal treatment provides two polarized viewpoints. First, with the CSAG guidelines indicating that 90% of patients with low back pain spontaneously resolve in 6 weeks, there is an argument that medical intervention initially is unhelpful.[2] Indeed, Faas et al[3] felt that patients referred for exercise therapy or back school receive a lot of needless, undesirable and expensive attention. Research has shown that some medical intervention is less effective than the provision of information, for returning patients to work,[4] and another study showed that the provision of a back advice booklet (The Back Book) gave clinically important improvements in disability and fear-avoidance scores.[5] This educational booklet provides evidence-based information and advice consistent with clinical guidelines.
However, the second argument is that early intervention may help the 10% of patients who would otherwise progress to chronicity and subsequently take up a large proportion of health care costs. Early access to treatment for musculoskeletal pain showed a substantial benefit when compared with standard treatment in a Swedish study.[6] Individual physical treatments (comprising items of manipulation, steroid injection, and stretching) showed superior results for return to work when compared with standard activation physiotherapy.[7] In favor of group therapy rather than individual treatment is the randomized controlled trial of a Back to Fitness program for people with low back pain, which proved clinically and cost effective, with the intervention group taking fewer days off work.[8]
A trial was therefore proposed to measure any difference in outcome between the booklet and advice approach, one that added a single intervention of manipulation or injection and individual exercise, and then undertook group exercise. The single intervention, rather than repeated one-to-one treatments, before group exercise, was preferred to minimize the cost of that program.
Recruitment
Following approval of the study by the local ethics committee, recruitment of patients with a new episode of low back pain was through occupational health physicians and local general practitioners who gave information leaflets to patients who met the inclusion and exclusion criteria. The patients telephoned the project nurse at the district hospital if they wished to participate.
The inclusion criteria were as follows:
|
The exclusion criteria were as follows:
|
An appointment was offered, to those accepting the invitation to attend for the study, within 1 week. After an explanation of the study, written consent was requested.
Measurement and Outcomes
Those who agreed to participate in the study were allocated a study number. The date of stopping work or a change to light duties was recorded. Initial measurements were made, using the following questionnaires: (1) The SF-12 Health Survey, a multipurpose short-form (SF) generic measure of health status,[9] and (2) The Short Form McGill Pain Questionnaire, the most widely used method of assessing quality and intensity of pain.[10]
Follow-up assessments for both groups comprised a repeat of the SF-12 and The Short Form McGill Pain Questionnaire at 1 and 2 months by post (stamped addressed envelope supplied). A separate sheet asked for dates that they had returned to work or full duties. At 2 months the participants were additionally asked whether they had attended any other clinics or appointments since their entry into the trial. If patients had not returned their questionnaires after 2 weeks, they were telephoned once and encouraged to complete and return them.
Interventions
Randomization was achieved using a computer program (Sampsize v.2.0, 1997). A sealed envelope containing the randomized group number was given to the patient to open.
Group 1 patients received The Back Book,[11] a 23 page booklet designed to change beliefs and behavior in relation to back pain, and verbal advice, and they continued their normal route of care as directed by their general practitioners. The advice given to patients reinforced that of The Back Book. Additionally, simple, practical advice on how to modify physical activities specific to the individual's work situation was discussed.[2]
Group 2 received The Back Book and advice regarding work modification as for group 1. The patients were assessed by a musculoskeletal physician or a senior physiotherapist and received one treatment. Treatment varied depending on the assessment findings. Examples of treatments offered were manipulation (where joint dysfunction and restriction were identified), joint and soft tissue mobilization, steroid injection including caudal epidural injection (where segmental paravertebral or facet joint tenderness was marked, or where the sciatic stretch test was positive), and specific exercises (according to the movement restrictions found). The specific exercises were for lumbar flexion and/or extension (McKenzie technique), carried out for an average of 10 repetitions, three or four times daily. Subsequently, patients attended group exercise sessions for 1 hour, three times a week in a gym. The group exercises comprised a circuit of nine stations, providing aerobic exercise and a focus on proprioception, spinal stability, strengthening of quadriceps, glutei, abdominals, and other spine stabilizing muscles based on the routine described by Moffett and Frost.[12,13] Motivation was encouraged by self-record of increasing repetition and difficulty. Patients were scheduled to attend for 2 weeks. If the patient felt able to go back to work sooner, they did not need to attend all sessions.
Analysis
Rate of Return to Work. Kaplan-Meier survival curve analysis was used to assess whether there was a significant difference between the groups with regard to who returned to work the quickest.
Days Off Work. The average number of days taken off work after study entry was calculated by substituting 62 days for any patient who had not returned to work within the follow-up time of 2 months. The two groups were compared.
Light Duties. Information was also collected regarding those who were on light duties but not off work at the onset of the study, but these cases were not included in the analysis for cost-effectiveness of return to work.
The Cost of Being Off Work. The potential cost to an individual or their employer of being off work was assessed for both groups. This was calculated using the average number of days taken off work since study entry, multiplied by the average daily wage. The two groups were compared.
Cost of Service Provision. For each group an analysis of expenditure and cost of service provision was made.
Pain and Health Status. The Short Form McGill Pain Questionnaire and the SF-12 Health Survey were completed three times: at study entry, 1 month, and 2 months. The scores were compared between the two groups. Analysis of covariance (ANCOVA), taking baseline values as a covariate, was used to evaluate any statistical significance of observed differences between the control and treatment groups[14]: the statistical software used was Minitab v.12 (3081 Enterprise Drive, State College, PA 16801-3008).
Other Health Care Interventions. Other therapies, clinics or departments attended within the 2 month follow-up period were recorded. The groups were compared
Baseline
The clinical and demographic characteristics of the patients recruited are shown in Table 1 .
Study Population
One hundred and nineteen patients initially met the research criteria (Figure 1). Eight were excluded before randomization: 2 people did not attend, 4 no longer met the inclusion/exclusion criteria, and 2 said they would not be able to attend the exercise classes if selected for that group. One hundred and eleven patients remained for the study. Of the 56 patients in group 1, 10 were lost to follow-up leaving 46 patients, and nine of those were on light duties and therefore not off work; therefore, 37 were left for analysis. Group 2 lost five to follow-up leaving 50 patients, of whom seven were not off work, but on light duties, leaving the data of 43 patients for analysis. The median number of days duration of back pain before entry to the study was 20, the mode being 14 days; the study group was therefore almost entirely acute or subacute.
 |
Figure 1. (click image to zoom)
Flow chart showing patients' progress through the study. |
Rate of Return to Work
Kaplan-Meier curves (Figure 2) illustrate the pattern of return to work rates for each of the two study groups. A log-rank test, used to compare the two curves, reveals a statistically significant difference at the 5% level (P = 0.034). The median number of days to return to work was 20 for group 1, as compared with 13 for group 2.
 |
Figure 2. (click image to zoom)
Return to work experience of each of the two study groups. |
Days Off Work
As the follow-up period for the trial was 2 months, 62 days off work was given for each patient who did not return to work within this period. The average number of days off work could then be calculated. The results (Figure 2) show that on average, group 2 patients returned to work 7 days earlier.
Status Change
Although not part of the planned analysis, we assessed the percentage of patients either returning to work or changing from light duties to full duties at 1 month, 6 weeks, and 2 months (Figure 3). Four weeks after recruitment, the percentage of patients who had achieved a change in their work status was 50% in group 1 compared with 72% in group 2. At 6 weeks the figures rose to 61% in group 1 compared with 80% in group 2, and at 8 weeks these figures rose again to 65 and 84%, respectively.
 |
Figure 3. (click image to zoom)
Percentage of patients either returning to work or changing from adapted to full duties. |
The Cost to Individual or Employer of Being Off Work
Table 2 shows the approximate cost of being off work using the average days off work taken and the average daily wage.
The Cost of Service Provision
Table 3 shows the weekly cost of providing the services for group 1 and group 2. The costs comprised the National Health Service (NHS) professional salaries, administration, and material cost and overheads at 45%. The cost per patient would depend on how many patients were seen. For example, in group 2, if 10 patients were seen (half capacity), the cost per patient in group 2 would be £85.79 (US$126, €103). If the service for group 2 worked at full capacity (20 patients), the cost per patient would be £42.90 (US$63, €51).
Financial Implications
With the varied estimates of cost of being off work, and the relative costs of treatment according to numbers treated in each group of patients, this calculation gives the possibility of wide variations. The lowest cost of being off work for 7 days using the lowest figure for the average wage is £336.77 (US$495, €404). If the back program worked at half capacity, the cost per patient would be £85.79 (US$126, €103). Therefore, the cost saving of group 2 patients may be as low as £250 (US$367, €300). If the highest average daily wage is used combined with the back program working at full capacity, the saving may be as much as £578 (US$850, €694) per patient.
Pain Level and Severity
Table 4 and Figures 4-7 show the scores for the four elements of the Short Form McGill Pain Questionnaire at baseline, 1 month, and 2 months for each group. The gradients indicate a faster reduction in pain scores of both groups within the first month compared to the second month. The ANCOVA adjusted means at 1 month exhibit a statistically significant benefit for the group 2 subjects, relative to those in group 1, in respect of pain score, present pain intensity, and sensory scores: all P < 0.05. At 2 months the relative benefit is statistically significant (P < 0.05) for present pain intensity only. No significant benefit was demonstrated with respect to mean emotional scores over the course of the study.
 |
Figure 4. (click image to zoom)
One of the four elements of the Short Form McGill Pain Questionnaire, showing mean score variations with time. VAS, Visual Analogue Scale. |
 |
Figure 5. (click image to zoom)
One of the four elements of the Short Form McGill Pain Questionnaire, showing mean score variations with time. PPI, present pain intensity. |
 |
Figure 6. (click image to zoom)
One of the four elements of the Short Form McGill Pain Questionnaire, showing mean score variations with time. |
 |
Figure 7. (click image to zoom)
One of the four elements of the Short Form McGill Pain Questionnaire, showing mean score variations with time. |
General Health/Function
Table 5 shows the Physical and Mental Component Summary Scores for the SF-12 Health Survey. The scores in both groups improve over time, with group 2 showing a slightly larger improvement.
Other Health Care Interventions
Figure 8 shows other health care interventions received by each group. Group 1 patients indicated the use of 28 other interventions compared with 7 in group 2. Unfortunately, this use of other departments, treatments, or investigations cannot be used for cost analysis purposes, as full details of sessions are unknown. However, it is noted that there is a large difference between the groups, which would have had a cost implication in favor of group 2.
 |
Figure 8. (click image to zoom)
Other interventions received during the two months study period |
Discussion
The study demonstrated that an intervention for acute and subacute low back pain that included one session of individualized physical treatment or injection and a simple back program combined with a package of information and advice (group 2) resulted in a speedier return to work than intervention that provided information, advice, and the normal route of care as directed by the general practitioner (group 1).
Group 2 patients returned to work on average 7 days earlier than group 1, an improvement of 35%. The reduction in time off work in favor of group 2 was confirmed to be statistically significant using the Kaplan-Meier survival curve analysis.
The CSAG guidelines state that 90% of patients suffering simple low back pain recover within 6 weeks.[2] Our study patients took longer, but the inclusion criteria for our study required them to be off work or on restricted duties. Stanley et al[18] found that of their patients taking up a new prompt access physiotherapy service in primary care, only 8% had taken any time off work. Of that 8%, only 18% took more than a week off work. At 6 weeks the percentage of our patients who had returned to work was only 61% in group 1 and only 80% in group 2.
The secondary outcome measures of pain and health status were not significantly different between the two groups at entry to the study. Both groups showed a more rapid reduction in pain scores within the first month, which leveled out in the second month. The visual analog pain scores were significantly better for group 2, but interestingly in the questionnaires at 2 months, the emotional and present pain intensity pain scores deteriorated slightly. This did not occur in group 1. As other scores and the return to work status were better in group 2, the deterioration was unlikely to be through relapse or reoccurrence. Input from health care professionals can medicalize a problem, creating the expectation that the problem can be fixed with the patient taking little responsibility for their own health.[19] Thus when professional input is withdrawn, deterioration occurs, as patients are not empowered to help themselves. The interventional back program, however, was deliberately simple and designed to promote self-confidence. Patients followed a standard structure of one assessment, one treatment, and then group exercise sessions. The aim was to achieve early pain relief and to avoid repeated individual sessions but to move on to a fitness program with its peer group benefits. And whereas individual sessions were based on the treatment options used in Blomberg's study of manipulative treatment and steroid injections,[7] this was followed by demedicalized group exercise routines.
The objective of demonstrating that the investment in a simple back program can be cost effective in terms of the financial consequences for the individual or employer was achieved. Moffett et al,[8] when evaluating the economics of a randomized controlled trial of an exercise program (for patients with back pain duration 4-24 weeks), used £45 (US$66, €54) per day as the daily cost of being off work. After a 12-month follow-up period, the cost saving for their intervention group was suggested to be up to £442 (US$650, €530) per patient.
Our active program, as received by group 2, showed a positive cost-effective benefit of between £250 and £578 for each patient (US$367-850, €300-694).
These figures must be interpreted with caution. It is recognized that days off work do not necessarily result in production losses to industry and that cost savings based on days off work alone are likely to overestimate the economic burden.[20]
There is an argument that these figures underestimate the cost savings, however. In work situations such as the NHS, sick nurses are replaced with agency or bank staff at a higher than original cost. These research trial figures also do not take into account the reduced use of other health services by group 2 in comparison with group 1. This cost saving could not be accurately assessed because the number and cost of additional treatments was unknown. The amount of medication taken by each group was also unknown.
The trial was pragmatic. Patients were excluded from analysis only if they did not provide any information on work status or questionnaires. Patient compliance could not be assessed in group 1, and therefore patients in group 2 were not excluded if they missed scheduled exercise classes. Consequently, the trial represented a closer to life situation rather than an artificial experimental one.
The trial was also pragmatic in that the active program used the combined inputs of physician/physiotherapist expert assessment, manipulation or injection, and group exercise. In studies that have evaluated cost-effectiveness, we are aware of one similar study by Niemisto et al[21] who used physician assessment, manipulation, and trunk stability exercises, but the study was for chronic, not acute or subacute, cases.
Some weaknesses in the trial design were identified. It would have been preferable to have a no treatment control group. Ethically this was not possible. Our ethical committee required a comprehensive patient information leaflet. Patients were aware of the content of each treatment protocol, which could have effected their actions and responses. We found that group 1 patients were more likely to drop out by not returning any follow-up information.
Our follow-up period was short, and it is well known that the reoccurrence rate of low back pain is high. A systematic review quotes a pooled cumulative risk of recurrence within 3 months of 26% and within 12 months between 66 and 84%.[22] However, rereferral to the same program could be advantageous in returning patients to work quickly by reinforcing the original messages. The interventional back program could also identify those patients needing additional help and intervention.
The overall simplicity of this back program lends itself to replication in a variety of locations, including hospitals, primary care and local leisure center, or occupational settings. The musculoskeletal physician had osteopathic training, and the physiotherapist had some manipulative training. The clinical nurse specialist had a special interest in ergonomics. However, the health professionals involved did not receive any extra training to implement the back program, and therefore the skills required could be obtained in other geographical areas at no added cost.
Conclusion
The study demonstrated that an intervention including information, advice, and a simple back program that offered manipulation, steroid injection, and group exercise therapy resulted in a speedier return to work than an intervention that provided information, advice, and the normal route of care as directed by the general practitioner.
Our original objective of reducing time off work by 5% was surpassed to approximately 35%. The active intervention group returned to work on average 7 days earlier.
The objective of demonstrating that the investment in a simple back program can be cost effective in terms of the financial consequences for the individual or employer was also achieved. This back program was calculated to provide an approximate cost saving of between £250 and £578 for every patient (US$367-850, €300-694).
Key Points
|
