Abstract
Background
Acute low‐back pain (LBP) is a common reason to consult a general practitioner. Debate continues on the comparative effectiveness of advice on bed rest and staying active as part of the primary care management.
Objectives
To determine the effects of advice to rest in bed or stay active for patients with acute low‐back pain or sciatica.
Search methods
We searched the Cochrane Back Review Group Trials Register, CENTRAL, MEDLINE, EMBASE, Sport, and SCISEARCH to May 2009, reference lists of relevant articles, and contacted authors of relevant articles.
Selection criteria
Randomised trials of the effectiveness of advice to stay active or rest in bed for patients with acute LBP or sciatica. The main outcomes were pain, functional status, recovery and return to work.
Data collection and analysis
Two authors independently selected trials, assessed the risk of bias and extracted data. The trials were combined qualitatively or statistically, depending on data availability and presentation.
Main results
We included ten RCTs with varying risk of bias. For patients with acute LBP, results from two trials (N = 401) suggest small improvements in pain relief (SMD 0.22 (95% CI: 0.02 to 0.41) and functional status (SMD 0.29 (95% CI: 0.09 to 0.49) in favour of advice to stay active. For patients with sciatica, there is moderate quality evidence of little or no difference in pain relief (SMD ‐0.03 (95% CI: ‐0.24 to 0.18)) or functional status (SMD 0.19 (95% CI: ‐0.02 to 0.41)), between advice to rest in bed or stay active.
Low quality evidence (3 RCTs, N = 931) suggests little or no difference between exercises, advice to rest in bed or stay active for patients with acute LBP. Low quality evidence (1 RCT, N = 250) suggests little or no difference between physiotherapy, advice to rest in bed or stay active for patients with sciatica. No trials that compared different ways of delivering advice.
Authors' conclusions
Moderate quality evidence shows that patients with acute LBP may experience small benefits in pain relief and functional improvement from advice to stay active compared to advice to rest in bed; patients with sciatica experience little or no difference between the two approaches. Low quality evidence suggests little or no difference between those who received advice to stay active, exercises or physiotherapy. Further research is very likely to have an important impact on the estimate of effect and is likely to change our confidence in it.
Keywords: Humans, Bed Rest, Exercise, Exercise Therapy, Low Back Pain, Low Back Pain/rehabilitation, Patient Education as Topic, Randomized Controlled Trials as Topic, Sciatica, Sciatica/rehabilitation
Plain language summary
Advice to rest in bed versus advice to stay active for acute low‐back pain and sciatica
Low‐back pain (LBP) is one of the most common conditions managed in primary care and a significant cause of absence from work and early retirement. Individuals, their families and society at large all carry part of the burden.
Many people get some relief from low‐back pain and sciatica (pain down the back and leg) by lying down. Until the late 1990s, advice to 'take it easy' and 'rest' was a common treatment approach. The exact nature of the 'rest' varied, but often it meant staying in bed full time, only taking trips to the bathroom and sometimes the kitchen.
However, as the millennium dawned, it became evident that extended periods of bed rest were potentially harmful because they may lead to the deterioration of muscles and body functions.
This review is the result of the combination of two previous reviews: Bed rest for acute low‐back pain and sciatica and Advice to stay active as a single treatment for low‐back pain and sciatica and an update of the literature search. No new randomised trials were identified.
This blended, updated review included 10 randomised trials (N = 1923). It evaluated the effects of advice to rest in bed or stay active on individuals with acute low‐back pain (pain lasting for less than 6 weeks) with or without sciatica.
Moderate quality evidence shows that patients with acute LBP may experience small improvements in pain relief and ability to perform everyday activities if they receive advice to stay active compared to advice to rest in bed. However, patients with sciatica experience little or no difference between the two approaches. Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality evidence suggests those patients with or without sciatica experienced little or no difference in pain relief or function, regardless of whether they received advice to stay active, exercises or physiotherapy. Further research is very likely to have an important impact on the estimate of effect and is likely to change our confidence in it.
Summary of findings
Summary of findings for the main comparison. bed rest compared to advice to stay active for acute low‐back pain.
| bed rest compared to advice to stay active for acute low‐back pain | ||||||
| Patient or population: patients with acute low‐back pain Settings: Intervention: bed rest Comparison: advice to stay active | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect (95% CI) | No of Participants (studies) | Quality of the evidence (GRADE) | Comments | |
| Assumed risk | Corresponding risk | |||||
| advice to stay active | bed rest | |||||
| Pain intensity, short‐term Follow‐up: 2 to 4 weeks | The mean Pain intensity, short‐term in the intervention groups was 0.02 standard deviations higher (0.16 lower to 0.2 higher) | 480 (3 studies) | ⊕⊕⊝⊝ low1,2 | SMD 0.02 (‐0.16 to 0.2) | ||
| Pain intensity, long‐term Follow‐up: median 12 weeks | The mean Pain intensity, long‐term in the intervention groups was 0.25 standard deviations higher (0.05 to 0.45 higher) | 393 (2 studies4) | ⊕⊕⊕⊝ moderate3 | SMD 0.25 (0.05 to 0.45) | ||
| Functional status, short‐term Follow‐up: 3 to 4 weeks | The mean Functional status, short‐term in the intervention groups was 0.29 standard deviations higher (0.09 to 0.49 higher) | 400 (2 studies4) | ⊕⊕⊕⊝ moderate3 | SMD 0.29 (0.09 to 0.49) | ||
| Functional status, long‐term Follow‐up: median 12 weeks | The mean Functional status, long‐term in the intervention groups was 0.24 standard deviations higher (0.04 to 0.44 higher) | 393 (2 studies4) | ⊕⊕⊕⊝ moderate3 | SMD 0.24 (0.04 to 0.44) | ||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; | ||||||
| GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate. | ||||||
1 One of the three studies is associated with high risk of bias (Wiesel (1985)). Malmivaara and Rozenberg were assessed as having low risk of bias, but we still consider that limitations in design (e.g. lack of blinding) may have introduced bias into the results. 2 Extensive inconsistencies between Wiesel on one hand, and Malmivaara / Rozenberg on the other hand. If the Wiesel trial is excluded ‐ and we think there are good reasons to do (i.e. limitations in study design and questionable generalisability) ‐ the heterogeneity disappears and the total effect estimate changes to SMD 0.22 (95% CI: 0.02 to 0.41). 3 It can not be ruled‐out that lack of blinding may have introduced bias into the results. 4 Only Malmivaara and Rozenberg were available for pooling as Wiesel did not report this outcome.
Summary of findings 2. bed rest compared to advice to stay active for sciatica.
| bed rest compared to advice to stay active for sciatica | ||||||
| Patient or population: patients with sciatica Settings: Intervention: bed rest Comparison: advice to stay active | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect (95% CI) | No of Participants (studies) | Quality of the evidence (GRADE) | Comments | |
| Assumed risk | Corresponding risk | |||||
| advice to stay active | bed rest | |||||
| Pain intensity, short‐term Follow‐up: 3 to 4 weeks | The mean Pain intensity, short‐term in the intervention groups was 0.03 standard deviations lower (0.24 lower to 0.18 higher) | 346 (2 studies) | ⊕⊕⊕⊝ moderate1 | SMD ‐0.03 (‐0.24 to 0.18) | ||
| Pain intensity, long‐term Follow‐up: median 10 weeks | The mean Pain intensity, long‐term in the intervention groups was 0.1 standard deviations higher (0.12 lower to 0.31 higher) | 330 (2 studies) | ⊕⊕⊕⊝ moderate1 | SMD 0.1 (‐0.12 to 0.31) | ||
| Functional status, short‐term Follow‐up: 3 to 4 weeks | The mean Functional status, short‐term in the intervention groups was 0.19 standard deviations higher (0.02 lower to 0.41 higher) | 346 (2 studies) | ⊕⊕⊕⊝ moderate1 | SMD 0.19 (‐0.02 to 0.41) | ||
| Functional status, long‐term Follow‐up: median 10 weeks | The mean Functional status, long‐term in the intervention groups was 0.12 standard deviations higher (0.1 lower to 0.33 higher) | 330 (2 studies) | ⊕⊕⊕⊝ moderate1 | SMD 0.12 (‐0.1 to 0.33) | ||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; | ||||||
| GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate. | ||||||
1 It can not be ruled‐out that lack of blinding may have introduced bias into the results.
Background
Low‐back pain (LBP) is one of the most common conditions managed in primary care, and a significant cause of absence from work and early retirement with a disability pension (Nachemson 2000). Thus, low‐back pain (LBP) represents a burden to many people as well as enormous costs for society.
Previously, the most commonly prescribed treatments for low‐back pain and sciatica were restricted activity, rest and bed rest (Waddell 1987). During the 1980s and 1990s, research revealed that prolonged inactivity, such as bed rest, is potentially harmful because it may lead to deterioration of many body functions (Bortz 1984; Allen 1999). By the end of the last century, Waddell and others introduced a more active approach to manage low‐back pain (Mayer 1988; Wheeler 1995; Waddell 1997).
From a clinical point of view it is interesting to review current evidence on the effect of advising patients with acute LBP or sciatica to stay in bed versus staying active. Moreover, it is interesting to determine whether other active treatments (e.g. exercises or manipulation) are more effective than simply giving advice to stay active or advice to stay in bed. Finally, it is interesting to observe whether the effect of advice to stay active is affected by the way the advice is delivered. The issues above were discussed in two previous separate Cochrane review articles (Hilde 2002; Hagen 2004), but since the questions are closely related, we decided to merge the two previous reviews. Hence, the aim of this review is to merge and update two previous Cochrane reviews: Bedrest for acute low‐back pain and sciatica (Hagen 2004) and Advice to stay active as a single treatment for low‐back pain and sciatica (Hilde 2002).
Objectives
The objective of this review was to determine the effects of advice to rest in bed or stay active for patients with acute low‐back pain or sciatica. The following comparisons were made:
Bed rest versus advice to stay active
Bed rest versus other treatments (besides staying active)
Shorter periods of bed rest versus longer periods of bed rest
Advice to stay active versus other treatments (besides bed rest)
Different ways to deliver advice to stay active (e.g. 'avoid bedrest' verus 'stay active')
Methods
Criteria for considering studies for this review
Types of studies
We included randomised controlled trials (RCT) where the effects of advice to rest in bed or advice to stay active were evaluated in the treatment of acute low‐back pain or sciatica.
Types of participants
We included trials with both male and female participants, between 16 and 80 years of age, who had acute (up to six weeks duration) LBP (the area bounded by the lowest palpable ribs superiorly and the gluteal folds inferiorly) or exacerbations of chronic pain lasting less than six weeks, in accordance with the recommendations of the Cochrane Back Review Group (Furlan 2009). In a previous version of this review we included two categories of patients, those with LBP without radiation below the knee and those with LBP with radiation of pain below the knee (a broad and unspecific definition of sciatica), but these categories have been refined to:
A) Acute simple LBP (i.e. LBP without neurological deficits)
B) Sciatica (i.e. LBP with verified neurological deficits)
C) Mixed LBP (i.e. LBP with and without verified neurological deficits)
Neurological deficits refer to those verified by clinical examination, not only those based on patients' subjective reporting. Trials including patients with inflammatory joint diseases, infections, neoplasms, metastases, osteoporosis or fractures were excluded.
Types of interventions
We included studies in which:
at least one group of subjects was advised to rest in bed (instructions to stay in bed for at least two days) and at least one group was not
at least one group of subjects was advised to stay active (instructions to stay as active as possible and continue normal daily activities) and at least one group was not
comparison groups were randomised to shorter or longer periods of bed rest
comparison group where randomised to receive the advice to stay active in different ways
Types of outcome measures
The main outcomes of interest were:
pain
back‐specific functional status
overall disability, such as length of sick leave, return to work
quality of life
adverse events
Outcomes were reported separately for short term (closest to four weeks) and long term effect (closest to one year).
Search methods for identification of studies
Electronic searches
We identified relevant studies by updating the electronic search in:
The Cochrane Back Review Group's Specialized Register (May 2009)
The Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2009, issue 2)
MEDLINE (1998 ‐ May 2009)
EMBASE (1998 ‐ May 2009)
SPORT (1998 ‐ May 2009)
SCISEARCH (1998 ‐ May 2009)
No language restrictions were applied to the searches. The search strategy was developed by joining the search strategies from the previous two reviews and is described in detail in Appendix 3.
Searching other resources
We checked the references of included studies and searched the WHO International Clinical Trials Registry Platform (ICTRP) for any ongoing trials.
Data collection and analysis
Selection of studies
Two review authors (KTD and KGB) independently reviewed the titles and abstracts and obtained full copies of reports of trials that appeared to meet our inclusion criteria. Based on the full report, we selected the trials to be included in the review, using a form that was previously pilot‐tested. Any disagreements between the review authors were resolved by a discussion and consensus, with involvement of a third review author (KBH) when disagreements persisted.
Data extraction and management
Data were independently extracted by two review authors (KTD and GJ) using a pilot‐tested data extraction form. Disagreements were resolved by discussion. If the article did not contain data on characteristics of the study population or necessary data for statistical analyses, letters were sent to the investigators to request missing data.
Assessment of risk of bias in included studies
The risk of bias was assessed by using the twelve criteria recommended by the Cochrane Back Review Group (Furlan 2009; Appendix 1). All evaluations were performed independently by two review authors (KTD and KGB). To ensure congruency amongst the review team, we completed risk of bias tables for two trials that were not relevant for inclusion before we proceeded to articles that were relevant for inclusion in the present review. In brief, the twelve risk of bias criteria were rated "yes", “no” or "unclear", and text describing the judgments was entered into the risk of bias tables. We discussed our judgements and tried to reach consensus. A third review author (KBH) was involved in resolving any persisting disagreements. Overall assessments of risk of bias were based on a summary of these twelve criteria. Trials that fulfilled at least six criteria and did not have any fatal flaws were classified as 'low risk of bias' and the others as 'high risk of bias'. We used the risk of bias assessment in sensitivity analyses, either as stratified analyses (high versus low) or in meta‐regressions.
Data Analysis
For similar comparisons and continuous outcome measures, we calculated weighted mean differences (WMD), and 95% confidence intervals (95% CI). If similar but not identical instruments were used to measure pain or functional status, we calculated standardized mean differences (SMD) with 95% CI. Standardised mean differences are unit‐less measures that are not easy to interpret in a clinical context. As a rough guidance, however, changes in SMD scores less then 0.5 are usually interpreted as small changes, 0.5 to 0.8 as medium changes whereas changes greater than 0.8 are usually viewed as large effect (Cohen 1988; Furlan 2009). For dichotomous outcomes, we calculated overall relative risk and reported the results with 95% CI.
Comparisons were made separately according to the type of participant (i.e. acute LBP, sciatica or mixed LBP), the type of intervention and the type of control group. Pooling of the trials was attempted when more than one trial was available for a specific comparison and outcome. The meta‐analyses were performed by applying a fixed‐effect model in which the weight of each study was determined by the precision of it's effect estimates.
Regardless of whether there were sufficient data to use quantitative analyses to summarize the data, we assessed the overall quality of the evidence for each outcome. To accomplish this, we used an adopted GRADE approach, as recommended by the Cochrane Back Review Group (Furlan 2009).The quality of the evidence on a specific outcome was based on five main domains: limitations of the study design, inconsistency, indirectness (inability to generalize) and imprecision (insufficient or imprecise data) of results and publication bias across all studies that measured that particular outcome. The quality started at high when at least two RCTs with a low risk of bias provided results for the outcome, and reduced by a level for each of the domains not met.
High quality evidence = there were consistent findings among at least 75% of RCTs with no limitations of the study design, consistent, direct and precise data and no known or suspected publication biases. Further research is very unlikely to change the level of evidence or the estimate of effect.
Moderate quality evidence = one of the domains was not met. Further research is likely to have an important impact on confidence in the estimate of effect and may change the estimate.
Low quality evidence = two of the domains were not met. Further research is very likely to have an important impact on the estimate of effect and is likely to change our confidence in it.
Very low quality evidence = three of the domains were not met. There is great uncertainty about the estimate.
No evidence = no RCTs were identified that addressed this outcome
An assessment of clinical relevance of the results was independently assessed by two review authors (KBH, GJ). The five questions recommended in the 2009 Updated Method Guidelines were used to assess clinical importance (Furlan 2009; Appendix 2).
Reporting of results and conclusions
Reporting of results and conclusions drawn in this review were carried out in accordance with recommendations of the Cochrane Back Review Group (Furlan 2009).
In the text of the review, the reporting of the results included: quality of evidence, the number of trials (number of participants), results of quantitative analyses (effect size plus confidence interval), results of qualitative analysis (direction of the effect (more/less effective, no difference)), the intervention, the type of participants, the comparison treatment (specifically stated), the outcome measured and the timing (short‐term or long‐term) of the outcome measure.
Results
Description of studies
Comparison
Six of the studies (Wiesel 1980; Malmivaara 1995; Wilkinson 1995; Vroomen 1999; Hofstee 2002; Rozenberg 2002), compared advice to rest in bed with staying active, four studies (Gilbert 1985; Postacchini 1988; Malmivaara 1995; Hofstee 2002) compared advice to rest in bed with other treatments, two studies (Deyo 1986; Szpalski 1992) compared different lengths of bed rest, two studies (Malmivaara 1995; Hofstee 2002) compared advice to stay active with other treatments. No studies compared different ways of delivering advice to stay active.
Participants
The ten studies included 1923 patients with three types of LBP:
A) Six studies (Wiesel 1980; Gilbert 1985; Postacchini 1988; Szpalski 1992; Malmivaara 1995; Rozenberg 2002) included LBP with or without radiating pain, but excluded patients with neurological deficits, a positive Lasègue’s sign, or both (acute simple LBP). B) Two studies (Vroomen 1999; Hofstee 2002) included patients with verified neurological deficits (sciatica) only. C) Two studies (Deyo 1986; Wilkinson 1995) included patients with and without neurological deficits (mixed LBP).
Wiesel 1980 and Szpalski 1992 included men only, whereas the proportion of women in the other studies varied between 35% and 67%. Wiesel 1980 included combat trainees with an average age of 23 years, while Szpalski 1992 included students and self‐employed men with an average age of 35 years. The other studies had a similar age distribution, with mean ages between 35 and 46 years.
Outcomes
Outcomes were measured from immediately after treatment to one year after treatment was stopped. Pain intensity was reported by Wiesel 1980 (relative to the first day), Gilbert 1985 (3‐point scale), Szpalski 1992, Hofstee 2002 and Rozenberg 2002 (VAS‐scale), Malmivaara 1995 (11‐point scale) and Vroomen 1999 (McGill Questionnaire and VAS‐scale). Duration of pain was reported by Deyo 1986 and Malmivaara 1995. In two studies, pain was not reported separately; Postacchini 1988 used pain assessments in a combined score, while Wilkinson 1995 reported pain at baseline only. Functional status was reported by Deyo 1986 (physical dimension of Sickness Impact Profile), Malmivaara 1995 (Oswestry LBPQ) and Rozenberg 2002 (Eifel (French Version of Roland‐Morris DS)), Wilkinson 1995 and Vroomen 1999 (Oswestery LBPQ and Roland‐Morris DS) and Hofstee 2002 (20 ADL items). Recovery was reported by Gilbert 1985 (restrictions in activities of daily living), and Deyo 1986 (days of limited activity).
Results of the search
This review combined two previous reviews Hilde 2002, Hagen 2004 and includes an updated search for new and relevant literature. The new search led to the identification of 101 references, 93 of which were clearly irrelevant and eight of which were retrieved in full text for further screening. All eight articles were excluded, and thus, no new trials were included in this updated review.
Included studies
No new trials were included in this updated version of the review. Hence, ten trials with a total of 1923 patients were included in this updated version. Three of the studies were undertaken in North America and seven in Europe.
Excluded studies
Eight new trials were retrieved in full text following the updated literature search, but all were excluded. Six were excluded because the population was considered irrelevant (chronic LBP), one because interventions and comparisons were considered irrelevant (Hancock 2007), and one due to limitations in design (alternate allocation) (Godges 2008). Additionally, one trial that was included in a previous version of this review was excluded in this updated version due to limitations in study design (alternate allocation and thus not a RCT) Coomes 1961.
Risk of bias in included studies
Risk of bias tables were completed for each of the included studies (Figure 1). All studies suffered from inadequate blinding and unclear risk of selective outcome reporting, but on the other domains the risk of bias varied between the ten studies. Overall, seven studies were considered to have low risk of bias for the relevant outcomes whereas the remaining three studies were assessed to have high or unclear risk of bias ( Wiesel 1980; Postacchini 1988; Wilkinson 1995).
1.
Methodological quality summary: review authors' judgements about each methodological quality item for each included study.
Allocation
Eight trials (Gilbert 1985; Deyo 1986; Szpalski 1992; Malmivaara 1995; Wilkinson 1995; Vroomen 1999; Hofstee 2002; Rozenberg 2002) described that the sequence was generated from a random number table and that the randomisation procedure was concealed, whereas two trials failed to meet our criteria (Wiesel 1980; Postacchini 1988).
Blinding
Blinding of the participants is generally not feasible with these types of interventions, and because this review is based on subjective outcome parameters (assessed by the patient) the included studies also suffered from insufficient blinding of the outcome assessors. Some studies reported outcomes that were more objectively measured by assessors who were blinded to treatment allocation (e.g. results of physical examinations and clinical rating), but these outcomes were not relevant for inclusion in this review.
Incomplete outcome data
Seven studies reported and treated drop‐outs in a satisfactory way (Wiesel 1980; Gilbert 1985; Szpalski 1992; Malmivaara 1995; Vroomen 1999; Hofstee 2002; Rozenberg 2002). One study was rated 'no' (Wilkinson 1995), mainly because the drop‐out rate was high (20%) and because the authors did not report whether the participants who dropped out differed from the patients who completed the study.
Selective reporting
Study protocols were not available for any of the included studies. Without the protocol it is difficult to know whether the authors reported the outcomes they initially intended to, and the selective reporting domain is therefore rated as 'unclear' for all the studies.
Other potential sources of bias
The groups were considered to be similar at baseline in seven of the included trials (Gilbert 1985; Deyo 1986; Malmivaara 1995; Wilkinson 1995; Vroomen 1999; Hofstee 2002; Rozenberg 2002). Three studies (Wiesel 1980; Postacchini 1988; Szpalski 1992) were rated as 'unclear' because they lacked a detailed description of the participants at baseline. The use of co‐interventions seem to have been avoided in most studies, but in one study, the participants in the bed rest group reported more frequent use of self‐remedies and additional physiotherapy than the patients in the stay active group (Wilkinson 1995). Compliance was rated as acceptable in five studies and unclear in four studies. In Deyo 1986, patients allocated to the long bed rest group (advice to stay in bed for seven days) showed particularly low compliance, implying that the real difference between number of hours spend in bed is not very different between the short and the long bed rest groups.
Effects of interventions
1: Advice to rest in bed versus advice to stay active
1A: Acute LBP
Three trials were available for this comparison and included a total of 481 patients with acute low‐back pain (Malmivaara 1995; Rozenberg 2002; Wiesel 1980).
Pain intensity
Short‐termed improvements in pain intensity were reported in all three studies, but no significant differences in pain improvement were observed between advice to rest in bed and advice to stay active when the results were combined in a fixed effect model (SMD 0.02 (95% CI: ‐0.16 to 0.20)). The quality of this evidence is downgraded to low because of considerable heterogeneity (I² = 92%) and contradicting results between Wiesel 1980 on the one hand, and Malmivaara 1995 and Rozenberg 2002 on the other. However, there are good reasons to exclude Wiesel 1980 from the analysis. First, Wiesel 1980 is assessed as having a high risk of bias, whereas Malmivaara 1995 and Rozenberg 2002 are considered to have a low risk of bias. Second, Wiesel 1980 only includes young combat trainees who are hospitalised, and thus, the applicability to the general population is questionable. Exclusion of Wiesel 1980 from the analysis results in a statistically significant difference in favour of advice to stay active (SMD 0.22 (95% CI: 0.02 to 0.41)), a difference that is considered to be of limited clinical importance. The latter result is consistent with pain measured at 12‐week follow‐up (SMD 0.25 (95% CI: 0.05 to 0.45)), an outcome that is only reported in Malmivaara 1995 and Rozenberg 2002.
Functional status
Wiesel 1980 did not report changes in functional status, whereas pooling of Malmivaara 1995 and Rozenberg 2002 resulted in statistically significant differences in favour of staying active at four‐week follow‐up (SMD 0.29 (95% CI: 0.09 to 0.49)) and at 12‐week follow‐up (SMD 0.25 (95% CI: 0.02 to 0.48)). Therefore, there is moderate quality evidence that advice to stay active is more effective than advice to stay in bed for improving functional status.
Length of sick leave
Length of sick leave was reported in all three trials. Rozenberg 2002 analysed the frequency and duration of sick leave for a subgroup of patients, employees with complete 12‐week follow‐up, and found that a higher proportion of the bed rest group had initial sick leave than did the normal activity group (86% versus 52%, P < 0.001). Malmivaara 1995 measured the difference in number of sick days during the first three weeks (WMD 3.4 days (95% CI: 1.64 to 5.16)) and at 12‐week follow up (WMD 4.5 days (95% CI: 1.37 to 7.63)), suggesting significantly fewer sick days among the patients who were randomised to advice to stay active. In contrast, Wiesel 1980 reported that combat trainees who were randomised to stay active needed on average 4.5 more days to resume full active duty than participants in the bed rest group (WMD ‐4.47 days (95% CI: ‐5.31 to ‐3.64)). Pooling of Wiesel and Malmiware in a fixed effect model result did not show any statistical difference in the number of sick days between those who received advice to stay active and those who received advice to rest in bed (SMD ‐0.15 (‐0.48 to 0.19)). However, the enormous heterogeneity (I² = 99 %) suggests that the two trials are too different to justify pooling.
1B: Sciatica
Two trials with a total of 348 patients were included in this comparison (Vroomen 1999; Hofstee 2002) and both trials were assessed to have a low risk of bias.
Pain intensity
The overall results did not show differences in pain intensity for patients randomised to either the bed rest or the stay active group at either short‐term (SMD ‐0.03 (95% CI: ‐0.24 to 0.18)) or 12‐week follow‐up (SMD 0.10 (95% CI: ‐0.12 to 0.31)). These results are based on two consistent trials with low risk of bias.In summary, we conclude that there is moderate quality evidence that advice to rest in bed does not lead to better pain improvement than advice to stay active.
Functional status
Improvement in functional status was measured in both Vroomen 1999 and Hofstee 2002, and the results of the two trials were consistent. Pooling did not reveal statistically significant differences between advice to stay active and advice to rest in bed at short‐term (SMD 0.19 (95% CI: ‐0.02 to 0.41)) or at long‐term follow‐up (SMD 0.12 (95% CI: ‐0.10 to 0.33)). We find it reasonable to conclude that there is moderate quality evidence that advice to rest in bed has little or no better effect on functional status than advice to stay active.
Length of sick leave
Hofstee 2002 did not report the length of sick leave. The results from Vroomen 1999 suggest no difference in the length of sick leave between bed rest and stay active groups (median values were 65 days for bed rest and 66 days for watchful waiting), but we consider the quality of the total evidence to be of low quality for this outcome. With respect to other important outcomes, Vroomen 1999 reported no differences in satisfaction with care or overall improvement, while Hofstee 2002 found no difference in treatment failure or the need for surgery.
1C: Mixed LBP
One small trial with high risk of bias was available for this comparison (Wilkinson 1995). Changes in pain intensity were not reported in this study. No differences were found for length of sick leave (WMD ‐1.0 day (95% CI: ‐8.73 to 6.73)) or for changes in functional status, but for the latter outcome the study groups were unbalanced in favour of the stay active group at baseline. Existing evidence therefore suggests that advice to rest in bed and advice to stay active are equally effective when studying mixed LBP populations (i.e. patients with acute LBP or sciatica), but the evidence is very low quality, which makes it difficult to draw clear conclusions. However, the relative lack of studies that include a mixed population needs to be seen in context with the quality of evidence for respective subpopulations; that is either acute LBP alone (comparison 1A) or sciatica alone (comparison 1B).
2: Advice to rest in bed versus other treatments
2A: Acute LBP
Three trials (N = 931) were available for this comparison Gilbert 1985; Postacchini 1988; Malmivaara 1995. One of the trials was assessed as having a high risk of bias (Postacchini 1988), and found no difference in improvement on a combined pain, disability and physical exam score between bed rest and manipulation, drug therapy, physiotherapy, back school or placebo. The other trials were assessed as having low risk of bias, and compared advice to rest in bed with exercises for patients with acute LBP (Gilbert 1985; Malmivaara 1995). Gilbert 1985 did not find significant differences in pain or restrictions in activities of daily living at six‐week,12‐week, or one‐year follow‐up, but the results were reported in a format that did not allow pooling. Malmivaara 1995 found no difference in pain, functional status or sick leave at three and 12‐week follow‐up. A challenge is that the actual exercises used in Malmivaara 1995 are poorly defined, and the directness of the evidence is therefore unclear. Thus, we consider that there is low quality of evidence suggesting little or no difference in pain, functional status or sick leave between advice to rest in bed and exercise. For other comparisons (e.g. advice to rest in bed versus manipulation) there is limited or no available evidence.
2B: Sciatica
Hofstee 2002 (N = 167) compared advice to rest in bed with physiotherapy, without further specification of the treatment that was given. No differences were found in pain intensity at four and at 12‐week follow‐up. For functional status, there was a statistically significant difference in favour of physiotherapy at four‐week follow‐up (WMD 6.9 points (on a 0 to 100 scale) (95% CI: 1.09 to 12.74)), and insignificant difference between bed rest and physiotherapy groups at 12‐week follow‐up (WMD 5.4 (95% CI: ‐2.10 to 12.90)). Even though Hofstee 2002 was associated with low risk of bias, we choose to downgrade the quality of the current evidence because the actual components of the physiotherapy were not described in sufficient detail. Hence, we find low quality evidence of little or no difference in pain intensity and functional status between patients who were advised to rest in bed and patients who received physiotherapy.
2C: Mixed LBP
No relevant studies were identified for this population.
3: Short bed rest versus longer bed rest
3A: Acute LBP
One trial with high risk of bias was available for this comparison and reported statistically insignificant differences in pain intensity between patients who were randomised to three and seven days of bed rest respectively (Szpalski 1992). Consequently, there is very low quality evidence of little or no difference in pain intensity between groups prescribed short or long bed rest.
3B: Sciatica
No relevant studies were identified for this comparison.
3C: Mixed LBP
One trial was available for this comparison, and our assessment suggested that the study was associated with low risk of bias for the relevant outcome measurements (Deyo 1986). The trial reported insignificant differences in pain intensity, functional status or time to recovery between patients who were randomised to two versus seven days of bed rest . Additionally, subgroup analysis of patients who were employed (i.e. about one third) showed that two days of bed rest resulted in fewer days off work (mean: 3.1 days) than for those assigned to seven days of bed rest (mean 5.6 days), a difference that was statistically significant (P = 0.01). In conclusion, there is low quality evidence showing that advice for shorter or longer bed rest has little or no impact on pain intensity, time to recovery, or functional status.
4: Staying active versus other treatments (besides bed rest)
4A: Acute LBP
One trial with low risk of bias compared advice to stay active with exercise and found no difference in pain at three (WMD 0.0 points (on a 11 point scale) (95% CI: ‐0.91 to 0.91)) or 12‐week (WMD 0.5 points (95% CI: ‐0.28 to 1.28)) follow‐up (Malmivaara 1995). In contrast, statistically significant differences in functional status in favour of advice to stay active could be observed at four‐week follow‐up (WMD 8.6 points (on a 0 to 100 scale) (95% CI: 3.32 to 13.88)), but the difference did not persist until the 12‐week follow‐up (WMD 3.4 points (95% CI: ‐1.12 to 7.92)). Length of sick leave was not statistically different between the stay active and the exercise group at four week (WMD 1.6 days 95% CI: ‐0.25 to 3.45)) or at 12‐week follow‐up (WMD 2.5 days (95% CI: ‐0.58 to 5.58)). A challenge is that the actual exercises used in Malmivaara 1995 are poorly defined, and the directness of the results are therefore unclear. Overall, we found low quality evidence showing that exercises add no clinically relevant benefit for patients with acute LBP when compared to advice to stay active. We did not identify trials that could shed new light on other possibly relevant comparisons (e.g. effect of advice to rest in bed versus manipulation).
4B: Sciatica
Hofstee 2002 compared the effectiveness of advice to stay active versus physiotherapy. No statistical differences were found between the two treatment strategies when evaluating pain intensity or functional status at four‐ or 12‐week follow‐up. Hofstee 2002 was assessed as low risk of bias, but the generalisability of the results is unclear because the actual components of the physiotherapy are not described in sufficient detail. Thus, when improvement in pain and functional status are considered as outcome measures, we conclude that there is low quality evidence showing that physiotherapy adds no further benefit to advice to stay active for patients with sciatica. We did not identify trials that could shed new light on other possibly relevant comparisons (e.g. effect of advice to rest in bed versus manipulation).
4C: Mixed LBP
No trials were available for this comparison in this population.
5: Different ways to deliver advice to stay active
No trials were available for this comparison.
Discussion
Summary of main results
Ten trials met the inclusion criteria of this review, and included a total of 1923 patients. Three trials compared advice to rest in bed with advice to stay active for acute LBP, and the overall results suggest that advice to stay active adds no benefit to main outcomes (Wiesel 1980; Malmivaara 1995; Rozenberg 2002). However, after excluding the one trial with high risk of bias and questionable generalisability (Wiesel 1980), the pooled results suggest small benefits for advice to stay active in improving pain and functional status.
Two trials compared advice to rest in bed with advice to stay active for patients with sciatica ( Vroomen 1999; Hofstee 2002). No significant differences in pain intensity were found, but for functional status, there was a borderline significant difference in favour of staying active at three to four weeks follow‐up (SMD 0.19, P = 0.07). The latter result suggests that advice to rest in bed is not a favourable option for patients with sciatica.
When comparing the effect of advice to stay active with other treatment strategies, the quality of available evidence is generally low. Malmivaara 1995 did not find statistically significant differences in pain intensity and length of sick leave between advice to stay active and exercise for patients with acute simple LBP, but observed a significant difference in favour of advice to stay active for functional status measured at four weeks. Hofstee 2002 reported no difference in pain intensity and functional status between advice to stay active and physiotherapy for patients with sciatica.
Overall completeness and applicability of evidence
The present review only considers patients with acute low‐back pain (duration of pain < 6 weeks). With one exception, the included trials seem to include representative patients, but the Wiesel trial only included male combat trainees. The participants in this study were, on average, 23 years old, and the participants in the bed rest group were hospitalised during the study period. The directness of the evidence and the applicability to a more general population is therefore highly questionable. Consequently, we have chosen to present our results in too different ways: with and without result of the Wiesel trial.
Based on feedback on previous versions of this review, we also present separate analyses for patients with acute LBP without neurological deficits and patients with verified neurological deficits (sciatica). For both diagnoses, it is a widespread understanding that patients should be advised to stay active; therefore, the main results of this review support current clinical practice. None of the included studies reported significant side effects of bed rest, but adverse physiological effects of bed rest have previously been reported Convertino 1997. These results may be relevant to consider in clinical practice, and further demonstrate that advice to rest in bed should be avoided for patients with acute low‐back pain.
Few trials compared the effect of advice to stay active versus other treatments and the total quality of evidence is low for this comparison. Therefore, It is uncertain whether patients with acute low‐back pain will experience benefits from other treatment strategies (e.g. physiotherapy, exercises or manipulation) compared to advice to stay active on its own. To some extent, this uncertainty is reflected in the current clinical practice where some patients are just advised to stay active whereas others receive physiotherapy, manipulation therapy, etc.
Due to a low number of relevant trials, we did not complete meta‐regressions in this review. Moreover, due to heterogeneous outcome reporting in the included primary studies, we used standardized mean difference effect size in most of our meta‐analysis. Standardised mean differences are unit‐less measures that are not easy to interpret in a clinical context. As a rough guidance, however, changes in SMD scores less then 0.5 are usually interpreted as small changes, 0.5 and 0.8 as medium changes whereas changes greater than 0.8 are usually viewed as large effects (Cohen 1988; Furlan 2009).
Quality of the evidence
We completed risk of bias tables for all the included studies, as suggested by the Editorial Board of the Back Review Group (Furlan 2009). Risk of bias tables provide standardized and explicit assessments of potential methodological sources of error for all included primary studies, revealing general methodological weaknesses as well as more study specific methodological pitfalls.
In general, the potential risk of selective outcome reporting was scored “unclear” in all included studies. This reflects the fact that we did not have access to the individual study protocols, and without study protocols it is not possible to assess whether the individual studies reported the outcomes they had initially intended to report. Moreover, in general, the included studies did not implement blinding of providers, patients or outcome assessors (outcomes are usually assessed by the patients themselves). It can be argued that blinding is not feasible in this kind of study and that lack of blinding should not be considered a weakness, but the absence of blinding can be associated with bias even though blinding is not feasible.
Some studies were assessed as having a particularly high risk of bias. The study by Wiesel 1980 reported significantly better outcomes for bed rest and was clearly different from all other results. The study suffered from poor methodological reporting (i.e. “unclear” scores in several risk of bias domains), and since the applicability to the general population is also questionable (it included young, male combat trainees), the results of this study were downgraded. The trial by Wilkinson 1995 compared advice to rest in bed with advice to stay active for patients with and without neurological deficits. The results of this trial were downgraded due to the heterogeneous population and high risk of bias in general.
Furthermore, we used the GRADE approach to evaluate the quality of evidence, as recommended by Cochrane Handbook (Higgins 2009 ) and by the Editorial Board of the Cochrane Back Review Group (Furlan 2009). For bed rest compared to staying active for acute low‐back pain (LBP) and for sciatica, we included two studies with approximately 400 participants, and concluded that there is high quality evidence that bed rest is not more effective than staying active. However, for most comparisons and outcomes, the quality of the evidence was downgraded due to few available studies, risk of bias and imprecise effect estimates. GRADE assessments are partly based on subjective judgements and are not definite, nevertheless GRADE does provide a transparent and consistent classification of the quality of evidence for relevant comparisons and outcomes.
Potential biases in the review process
New risk of bias tables and the use of GRADE are implemented in this update and will hopefully contribute to increased transparency in the review process. From the risk of bias tables, it can be seen that some domains are scored “unclear”, implying that the primary publications do not supply enough details to assess this point. The number of domains assessed as “unclear” should ideally be reduced by obtaining supplementary information from the primary authors, but for the course of simplicity we have chosen to complete our risk of bias tables based on information that is printed in the primary papers and information brought in from authors during preparation of earlier versions of this review. Consequently, no new attempts to obtain supplementary information from the authors of primary studies were made during the preparation of this updated version.
The updated search covers the period up to April 2009, but did not result in inclusion of any new studies. Even though our search strategy was comprehensive and not language restricted, there is always a risk that relevant citations may have been lost in the review process.
Authors' conclusions
Implications for practice.
| Advice to stay active is probably slightly more effective than bed rest for people with acute LBP, whereas for patients with sciatica, little or no difference is seen between advice to rest in bed and advice to stay active. A couple of studies suggest little or no difference between the effect of advice to stay active and other treatment strategies (e.g. physiotherapy, exercises or manipulation therapy), but the quality of evidence is too low to draw clear conclusions. Further research is very likely to have an important impact on the estimate of effect and is likely to change our confidence in it. |
Implications for research.
| The updated version of this review suggests that advice to rest in bed is not effective in the treatment of acute LBP. Studies comparing advice to stay active versus other treatment and different forms of advice to stay active are needed. |
Acknowledgements
We would like to thank M. Winnem and G. Hilde for their contribution to the two previous reviews. Moreover, we acknowledge Rachel Couban, the Trial Search Co‐ordinator from the Cochrane Back Review Group for developing the search strategies and performing the updated literature search.
Appendices
Appendix 1. Criteria for a judgment of 'yes' for the sources of risk of bias
1. Was the method of randomisation adequate?
A random (unpredictable) assignment sequence. Examples of adequate methods are coin toss (for studies with two groups), rolling a dice (for studies with two or more groups), drawing of balls of different colours, drawing of ballots with the study group labels from a dark bag, computer‐generated random sequence, pre‐ordered sealed envelops, sequentially‐ordered vials, telephone call to a central office, and pre‐ordered list of treatment assignments
Examples of inadequate methods are: alternation, birth date, social insurance/security number, date in which they are invited to participate in the study, and hospital registration number
2. Was the treatment allocation concealed?
Assignment generated by an independent person not responsible for determining the eligibility of the patients. This person has no information about the persons included in the trial and has no influence on the assignment sequence or on the decision about eligibility of the patient.
Was knowledge of the allocated interventions adequately prevented during the study?
3. Was the patient blinded to the intervention?
This item should be scored “yes” if the index and control groups are indistinguishable for the patients or if the success of blinding was tested among the patients and it was successful.
4. Was the care provider blinded to the intervention?
This item should be scored “yes” if the index and control groups are indistinguishable for the care providers or if the success of blinding was tested among the care providers and it was successful
5. Was the outcome assessor blinded to the intervention?
Adequacy of blinding should be assessed for the primary outcomes. This item should be scored “yes” if the success of blinding was tested among the outcome assessors and it was successful or:
for patient‐reported outcomes in which the patient is the outcome assessor (e.g., pain, disability): the blinding procedure is adequate for outcome assessors if participant blinding is scored “yes”
for outcome criteria assessed during scheduled visit and that supposes a contact between participants and outcome assessors (e.g., clinical examination): the blinding procedure is adequate if patients are blinded, and the treatment or adverse effects of the treatment cannot be noticed during clinical examination
for outcome criteria that do not suppose a contact with participants (e.g., radiography, magnetic resonance imaging): the blinding procedure is adequate if the treatment or adverse effects of the treatment cannot be noticed when assessing the main outcome
for outcome criteria that are clinical or therapeutic events that will be determined by the interaction between patients and care providers (e.g., co‐interventions, hospitalisation length, treatment failure), in which the care provider is the outcome assessor: the blinding procedure is adequate for outcome assessors if item for caregiver is scored “yes”
for outcome criteria that are assessed from data of the medical forms: the blinding procedure is adequate if the treatment or adverse effects of the treatment cannot be noticed on the extracted data
Were incomplete outcome data adequately addressed?
6. Was the drop‐out rate described and acceptable?
The number of participants who were included in the study but did not complete the observation period or were not included in the analysis must be described and reasons given. If the percentage of withdrawals and drop‐outs does not exceed 20% for short‐term follow‐up and 30% for long‐term follow‐up and does not lead to substantial bias a 'yes' is scored. (N.B. these percentages are arbitrary, not supported by literature).
7. Were all randomised participants analysed in the group to which they were allocated?
All randomised patients are reported/analysed in the group they were allocated to by randomisation for the most important moments of effect measurement (minus missing values) irrespective of non‐compliance and co‐interventions.
8. Are reports of the study free of suggestion of selective outcome reporting?
In order to receive a ‘yes’, the review author determines if all the results from all pre‐specified outcomes have been adequately reported in the published report of the trial. This information is either obtained by comparing the protocol and the report, or in the absence of the protocol, assessing that the published report includes enough information to make this judgment.
Other sources of potential bias:
9. Were the groups similar at baseline regarding the most important prognostic indicators?
In order to receive a “yes”, groups have to be similar at baseline regarding demographic factors, duration and severity of complaints, percentage of patients with neurological symptoms, and value of main outcome measure(s).
10. Were co‐interventions avoided or similar?
This item should be scored “yes” if there were no co‐interventions or they were similar between the index and control groups.
11. Was the compliance acceptable in all groups?
The reviewer determines if the compliance with the interventions is acceptable, based on the reported intensity, duration, number and frequency of sessions for both the index intervention and control intervention(s). For example, physiotherapy treatment is usually administered over several sessions; therefore it is necessary to assess how many sessions each patient attended. For single‐session interventions (for ex: surgery), this item is irrelevant.
12. Was the timing of the outcome assessment similar in all groups?
Timing of outcome assessment should be identical for all intervention groups and for all important outcome assessments.
Appendix 2. Questions to determine if results are clinically relevant
Based on the data provided, can you determine if the results will be clinically relevant?
Are the patients described in detail so that you can decide whether they are comparable to those that you see in your practice?
Are the interventions and treatment settings described well enough so that you can provide the same for your patients?
Were all clinically relevant outcomes measured and reported?
Is the size of the effect clinically important?
Are the likely treatment benefits worth the potential harms?
Appendix 3. Electronic search strategies
MEDLINE 1 clinical trial.pt. 2 randomized.ab,ti. 3 placebo.ab,ti. 4 dt.fs. 5 randomly.ab,ti. 6 trial.ab,ti. 7 groups.ab,ti. 8 or/1‐7 9 Animals/ 10 Humans/ 11 9 not (9 and 10) 12 8 not 11 13 dorsalgia.ab,ti. 14 exp Back Pain/ 15 exp Low Back Pain/ 16 backache.ab,ti. 17 (lumbar adj pain).ab,ti. 18 coccyx.ab,ti. 19 coccydynia.ab,ti. 20 exp Sciatica/ 21 sciatica.ab,ti. 22 spondylosis.ab,ti. 23 lumbago.ab,ti. 24 or/13‐23 25 exp Spine/ 26 discitis.ab,ti. 27 exp Spinal Diseases/ 28 (disc adj degeneration).ab,ti. 29 (disc adj prolapse).ab,ti. 30 (disc adj herniation).ab,ti. 31 spinal fusion.sh. 32 spinal neoplasms.sh. 33 (facet adj joints).ab,ti. 34 intervertebral disk.sh. 35 postlaminectomy.ab,ti. 36 arachnoiditis.ab,ti. 37 (failed adj back).ab,ti. 38 or/25‐37 39 exp activities of daily living/ 40 activities of daily living.tw. 41 day to day activ$.ti,ab. 42 daily activit$.ti,ab. 43 ordinary activit$.ti,ab. 44 normal activit$.ti,ab. 45 stay$ active.ti,ab. 46 exp Bed Rest/ 47 bed rest.mp. 48 or/39‐47 49 24 or 38 50 12 and 49 51 48 and 50
EMBASE 1 Clinical Article/ 2 clinical study/ 3 clinical trial/ 4 controlled study/ 5 randomized controlled trial/ 6 major clinical study/ 7 double blind procedure/ 8 multicenter study/ 9 single blind procedure/ 10 phase 3 clinical trial/ 11 phase 4 clinical trial/ 12 crossover procedure/ 13 placebo/ 14 or/1‐13 15 allocat$.mp. 16 assign$.mp. 17 blind$.mp. 18 (clinic$ adj25 (study or trial)).mp. 19 compar$.mp. 20 control$.mp. 21 cross?over.mp. 22 factorial$.mp. 23 follow?up.mp. 24 placebo$.mp. 25 prospectiv$.mp. 26 random$.mp. 27 ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).mp. 28 trial.mp. 29 (versus or vs).mp. 30 or/15‐29 31 14 or 30 32 human/ 33 nonhuman/ 34 animal/ 35 animal experiment/ 36 33 or 34 or 35 37 32 not 36 38 31 not 36 39 31 and 37 40 38 not 39 41 dorsalgia.mp. 42 exp back pain/ 43 backache.mp. 44 (lumbar adj pain).mp. 45 coccyx.mp. 46 coccydynia.mp. 47 sciatica.mp. 48 sciatica/ 49 spondylosis.mp. 50 lumbago.mp. 51 or/41‐50 52 exp spine/ 53 discitis.mp. 54 exp spinal diseases/ 55 (disc adj degeneration).mp. 56 (disc adj prolapse).mp. 57 (disc adj herniation).mp. 58 spinal fusion.mp. 59 spinal neoplasms.mp. 60 (facet adj joints).mp. 61 intervertebral disk.mp. 62 postlaminectomy.mp. 63 arachnoiditis.mp. 64 (failed adj back).mp. 65 or/52‐64 66 exp activities of daily living/ 67 activities of daily living.mp. 68 day to day activ$.mp. 69 ordinary activit$.mp. 70 normal activit$.mp. 71 stay$ active.mp. 72 exp Bed Rest/ 73 bed rest.mp. 74 or/66‐73 75 51 or 65 76 40 and 75 77 74 and 76
CINAHL 1 Randomized Controlled Trials/ 2 Animals/ not Human/ 3 1 not 2 4 clinical trial.pt. 5 exp Clinical Trials/ 6 (clin$ adj25 trial$).tw. 7 ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).tw. 8 Placebos/ 9 placebo$.tw. 10 random$.tw. 11 Research Design/ 12 (latin adj square).tw. 13 or/4‐12 14 13 not 12 15 14 not 3 16 Comparative Study/ 17 exp Evaluation Studies/ 18 Follow‐Up Studies/ 19 Prospective Studies/ 20 (control$ or prospective$ or Volunteer$).tw. 21 or/16‐20 22 21 not 2 23 22 not (3 or 15) 24 dorsalgia.ti,ab. 25 exp Back Pain/ 26 backache.ti,ab. 27 (lumbar adj pain).ti,ab. 28 coccyx.ti,ab. 29 coccydynia.ti,ab. 30 sciatica.ti,ab. 31 sciatica/ 32 spondylosis.ti,ab. 33 lumbago.ti,ab. 34 or/24‐33 35 exp Spine/ 36 discitis.ti,ab. 37 exp Spinal Diseases/ 38 (disc adj degeneration).ti,ab. 39 (disc adj prolapse).ti,ab. 40 (disc adj herniation).ti,ab. 41 spinal fusion.ti,ab. 42 (facet adj joints).ti,ab. 43 intervertebral disk.ti,ab. 44 postlaminectomy.ti,ab. 45 achnoiditis.ti,ab. 46 (failed adj back).ti,ab. 47 or/35‐46 48 exp activities of daily living/ 49 activities of daily living.tw. 50 day to day activ$.ti,ab. 51 daily activit$.ti,ab. 52 ordinary activit$.ti,ab. 53 normal activit$.ti,ab. 54 stay$ active.ti,ab. 55 or/48‐54 56 exp Bed Rest/ 57 bed rest.mp. 58 or/48‐57 59 34 or 47 60 23 and 59 61 58 and 60
CENTRAL 1 MeSH descriptor Back explode all trees #2 MeSH descriptor Buttocks, this term only #3 MeSH descriptor Leg, this term only #4 MeSH descriptor Back Pain explode tree 1 #5 MeSH descriptor Back Injuries explode all trees #6 MeSH descriptor Low Back Pain, this term only #7 MeSH descriptor Sciatica, this term only #8 (low next back next pain) #9 (lbp) #10 (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9) #11 MeSH descriptor Activities of Daily Living explode all trees #12 activities of daily living in Clinical Trials #13 day to day activ* in Clinical Trials #14 daily activit* in Clinical Trials #15 ordinary activit* in Clinical Trials #16 normal activit* in Clinical Trials #17 stay* active in Clinical Trials #18 bed rest in Clinical Trials #19 MeSH descriptor Bed Rest explode all trees #20 (#11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19) #21 (#20 AND #10) #22 (#21), from 2003 to 2008
SportDiscus S25 S24 Limiters ‐ Year Published from: 2003‐2008 S24 S23 and S16 S23 S22 or S21 or S20 or S19 or S18 or S17 S22 bed rest S21 DE "REST" S20 staying active S19 activity S18 DE "PHYSICAL education & training" S17 DE "ACTIVITIES of daily living" S16 S15 and S10 S15 S14 or S13 or S12 or S11 S14 DE "LUMBAR vertebrae" or DE "LUMBOSACRAL region" S13 DE "SCIATICA" S12 low back pain S11 DE "BACKACHE" S10 S9 or S8 or S7 or S6 or S5 or S4 or S3 or S2 or S1 S9 single blind S8 random allocation S7 SU randomized controlled trial S6 SU clinical trials S5 clinical trials S4 placebo S3 controlled clinical trial S2 double blind S1 randomi?ed controlled trial
Data and analyses
Comparison 1. Bed rest versus advice to stay active.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 1 Pain Intensity, 3‐4 weeks | 5 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 1.1 Acute LBP | 3 | 480 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.02 [‐0.16, 0.20] |
| 1.2 Sciatica | 2 | 346 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.03 [‐0.24, 0.18] |
| 2 Pain Intensity, 12 weeks | 4 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 2.1 Acute LBP | 2 | 393 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.25 [0.05, 0.45] |
| 2.2 Sciatica | 2 | 330 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.10 [‐0.12, 0.31] |
| 3 Functional status, 3‐4 weeks | 5 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 3.1 Acute LBP | 2 | 400 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.29 [0.09, 0.49] |
| 3.2 Sciatica | 2 | 346 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.19 [‐0.02, 0.41] |
| 3.3 Mixed LBP | 1 | 34 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.20 [‐0.49, 0.88] |
| 4 Functional status, 12 weeks | 4 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 4.1 Acute LBP | 2 | 393 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.24 [0.04, 0.44] |
| 4.2 Sciatica | 2 | 330 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.12 [‐0.10, 0.33] |
| 5 Length of sick leave, 3‐4 weeks | 3 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 5.1 Acute LBP | 2 | 203 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.15 [‐0.48, 0.19] |
| 5.2 Mixed LBP | 1 | 22 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.10 [‐0.94, 0.74] |
| 6 Length of sick leave, 12 weeks | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 6.1 Acute LBP | 1 | 123 | Mean Difference (IV, Fixed, 95% CI) | 4.50 [1.37, 7.63] |
1.1. Analysis.
Comparison 1 Bed rest versus advice to stay active, Outcome 1 Pain Intensity, 3‐4 weeks.
1.2. Analysis.
Comparison 1 Bed rest versus advice to stay active, Outcome 2 Pain Intensity, 12 weeks.
1.3. Analysis.
Comparison 1 Bed rest versus advice to stay active, Outcome 3 Functional status, 3‐4 weeks.
1.4. Analysis.
Comparison 1 Bed rest versus advice to stay active, Outcome 4 Functional status, 12 weeks.
1.5. Analysis.
Comparison 1 Bed rest versus advice to stay active, Outcome 5 Length of sick leave, 3‐4 weeks.
1.6. Analysis.
Comparison 1 Bed rest versus advice to stay active, Outcome 6 Length of sick leave, 12 weeks.
Comparison 2. Bed rest versus other treatments.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 1 Pain Intensity, 10 days ‐ 3 weeks | 2 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 1.1 Acute LBP | 1 | 104 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.23 [‐0.16, 0.63] |
| 1.2 Sciatica | 1 | 162 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.10 [‐0.21, 0.41] |
| 2 Pain Intensity, 12 weeks | 2 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 2.1 Acute LBP | 1 | 99 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.14 [‐0.26, 0.54] |
| 2.2 Sciatica | 1 | 159 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.12 [‐0.19, 0.43] |
| 3 Functional status, 3 weeks | 2 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 3.1 Acute LBP | 1 | 104 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.17 [‐0.57, 0.22] |
| 3.2 Sciatica | 1 | 162 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.36 [0.05, 0.67] |
| 4 Functional status, 12 weeks | 2 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 4.1 Acute LBP | 1 | 104 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.08 [‐0.31, 0.48] |
| 4.2 Sciatica | 1 | 159 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.22 [‐0.09, 0.54] |
| 5 Length of sick leave, 3 weeks | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 5.1 Acute LBP | 1 | 110 | Mean Difference (IV, Fixed, 95% CI) | 1.80 [‐0.37, 3.97] |
| 6 Length of sick leave, 12 weeks | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 6.1 Acute LBP | 1 | 110 | Mean Difference (IV, Fixed, 95% CI) | 2.00 [‐1.85, 5.85] |
2.1. Analysis.
Comparison 2 Bed rest versus other treatments, Outcome 1 Pain Intensity, 10 days ‐ 3 weeks.
2.2. Analysis.
Comparison 2 Bed rest versus other treatments, Outcome 2 Pain Intensity, 12 weeks.
2.3. Analysis.
Comparison 2 Bed rest versus other treatments, Outcome 3 Functional status, 3 weeks.
2.4. Analysis.
Comparison 2 Bed rest versus other treatments, Outcome 4 Functional status, 12 weeks.
2.5. Analysis.
Comparison 2 Bed rest versus other treatments, Outcome 5 Length of sick leave, 3 weeks.
2.6. Analysis.
Comparison 2 Bed rest versus other treatments, Outcome 6 Length of sick leave, 12 weeks.
Comparison 3. Short bed rest versus long bed rest.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 1 Pain intensity | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 1.1 Acute LBP | 1 | 47 | Mean Difference (IV, Fixed, 95% CI) | 0.20 [‐0.15, 0.55] |
| 2 Proportion of patients recovered | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 2.1 Mixed LBP | 1 | 189 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.08 [0.60, 1.96] |
3.1. Analysis.
Comparison 3 Short bed rest versus long bed rest, Outcome 1 Pain intensity.
3.2. Analysis.
Comparison 3 Short bed rest versus long bed rest, Outcome 2 Proportion of patients recovered.
Comparison 4. Advice to stay active versus other treatment.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 1 Pain Intensity, 3‐4 weeks | 2 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 1.1 Acute LBP | 1 | 103 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.0 [‐0.39, 0.39] |
| 1.2 Sciatica | 1 | 163 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.02 [‐0.33, 0.29] |
| 2 Pain Intensity, 12 weeks | 2 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 2.1 Acute LBP | 1 | 102 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.26 [‐0.14, 0.66] |
| 2.2 Sciatica | 1 | 156 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.02 [‐0.30, 0.33] |
| 3 Functional status, 3‐4 weeks | 2 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 3.1 Acute LBP | 1 | 103 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.68 [0.28, 1.09] |
| 3.2 Sciatica | 1 | 163 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.04 [‐0.35, 0.27] |
| 4 Functional status, 12 weeks | 2 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 4.1 Acute LBP | 1 | 103 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.32 [‐0.08, 0.71] |
| 4.2 Sciatica | 1 | 156 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.06 [‐0.37, 0.26] |
| 5 Length of sick leave, 3‐4 weeks | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 5.1 Acute LBP | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | 1.60 [‐0.25, 3.45] |
| 6 Length of sick leave, 12 weeks | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 6.1 Acute LBP | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | 2.5 [‐0.58, 5.58] |
4.1. Analysis.
Comparison 4 Advice to stay active versus other treatment, Outcome 1 Pain Intensity, 3‐4 weeks.
4.2. Analysis.
Comparison 4 Advice to stay active versus other treatment, Outcome 2 Pain Intensity, 12 weeks.
4.3. Analysis.
Comparison 4 Advice to stay active versus other treatment, Outcome 3 Functional status, 3‐4 weeks.
4.4. Analysis.
Comparison 4 Advice to stay active versus other treatment, Outcome 4 Functional status, 12 weeks.
4.5. Analysis.
Comparison 4 Advice to stay active versus other treatment, Outcome 5 Length of sick leave, 3‐4 weeks.
4.6. Analysis.
Comparison 4 Advice to stay active versus other treatment, Outcome 6 Length of sick leave, 12 weeks.
Characteristics of studies
Characteristics of included studies [ordered by study ID]
Deyo 1986.
| Methods | Design: Randomised (parallel group) controlled trial | |
| Participants |
|
|
| Interventions | Patient randomly assigned to one of two intervention protocols:
|
|
| Outcomes | Evaluation of treatment outcome was obtained three weeks and three months after the initial visit:
Physical examinations and the clinicians' ratings were carried out, and the patient self‐rating was elicited by a specially trained nurse practitioner or a physician who was blinded to study‐group assignment |
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Low risk | Sequence generated from a random number table. |
| Allocation concealment? | Low risk | The subjects were assigned to the intervention groups by sealed envelopes handed directly to the patient. |
| Blinding? All outcomes ‐ patients | High risk | patients aware of advice received |
| Blinding? All outcomes ‐ providers | High risk | providers aware of advice given |
| Blinding? All outcomes ‐ assessors | High risk |
Only patient reported outcomes are used in the review. |
| Incomplete outcome data addressed? All outcomes ‐ drop‐outs? | Unclear risk |
No reason for missing data was given, but those who were lost to follow‐up differed from the remaining subjects by having a significantly shorter mean duration of pain prior to entry, having substantially lower SIP scores, and being more likely to be employed. |
| Incomplete outcome data addressed? All outcomes ‐ ITT analysis | Low risk | Drop‐outs excluded from the analysis, but all others were analysed in the group to which they were allocated. |
| Free of selective reporting? | Unclear risk | No protocol available, and thus unclear whether the outcomes were defined a priori. |
| Similarity at baseline characteristics? | Low risk | Baseline demographic and clinical characteristics seemed to be similar. |
| Co‐interventions avoided or similar? | Low risk | Quote: "Uniform recommendations for non‐pharmacologic treatments were given to all subjects, whereas a physician (blinded to patient allocation) prescribed drug therapy individually." |
| Compliance acceptable? | High risk | Quote: "Modal duration of bed rest in the two groups were two and seven days, respectively, whereas the means were 2.3 days (group I) and 3.9 days (group 2). Compliance with the seven‐day recommendation was especially limited: 73 of 99 subjects (74 %) reported fewer than seven days of actual bed rest." |
| Timing outcome assessments similar? | Low risk | Assessment 3 weeks and 3 months following randomisation in both groups |
Gilbert 1985.
| Methods | Design: Randomised (parallel group) controlled trial | |
| Participants |
|
|
| Interventions | Patient randomised to one of four treatment strategies:
|
|
| Outcomes | Patients were evaluated at different time‐points after treatment randomisation:
Additionally, medical records were reviewed to validate the accuracy of the patient reports. |
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Low risk | The sequence generation is not described in detail, but it was performed at a centralized patient registry and randomisation service |
| Allocation concealment? | Low risk | Quote: "The subjects were assigned to one of four groups by a physician who telephoned a central randomisation service. Patients were stratified by whether the physician prescribed minor or major medication." |
| Blinding? All outcomes ‐ patients | High risk | patients were aware of the intervention received |
| Blinding? All outcomes ‐ providers | High risk | providers were aware of the intervention delivered |
| Blinding? All outcomes ‐ assessors | High risk |
Only patient reported outcomes are used in the review. |
| Incomplete outcome data addressed? All outcomes ‐ drop‐outs? | Low risk | 270 participants were randomised, whereas 252 eligible patients began treatment (eight were considered to not meet the trial's predefined inclusion and exclusion criteria when reviewing baseline data, and ten patients refused to accept the treatment that was randomly assigned to them). 87% returned to their physician for follow‐up evaluation, 89% completed at least one 10‐day patient diary, 96% completed 6‐ and 12‐week telephone follow up, and 90% completed the long term (1 year) follow‐up questionnaire. |
| Incomplete outcome data addressed? All outcomes ‐ ITT analysis | High risk | Quote: "Ten patients refused to accept the treatment that was randomly assigned to them and were excluded from the study, nine of these patient were allocated to the bed rest or the bed rest + exercise group." |
| Free of selective reporting? | Unclear risk | No protocol available, and thus unclear whether the outcomes were defined a priori. |
| Similarity at baseline characteristics? | Low risk | Quote: "Patients who were randomised to educational programme were better at baseline on three measures, but these baseline differences were accounted for in the analysis." |
| Co‐interventions avoided or similar? | Low risk | Patients were stratified by whether the physician prescribed minor (i.e. muscle relaxants and/or analgesics containing fewer than eight aspirins per day) or major medication (i.e. any non‐steroidal anti‐inflammatory drug or analgesics containing eight or more aspirins per day). |
| Compliance acceptable? | Low risk | Patient diaries showed that patients who were randomised to bed rest spent on average three days longer in bed than non‐bed rest patients. No objective measure of compliance with bed rest could be obtained. Quote: "All but two of the patients who were randomised to physiotherapy and education saw the physiotherapist at least once, and none of the patients in the non‐physiotherapy groups received the physiotherapy and education programme." |
| Timing outcome assessments similar? | Low risk | Assessment 10 days, 6 and 12 weeks and 1 year following randomisation in both groups |
Hofstee 2002.
| Methods | Design: Randomised (parallel group) controlled trial | |
| Participants |
|
|
| Interventions | Patients randomised to one of three interventions:
|
|
| Outcomes | Evaluation at one, two and six months after the baseline visit. Patient reported outcomes:
Outcomes assessed by investigator:
|
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Low risk | Lists supplied by a statistician. |
| Allocation concealment? | Low risk | Concealment approach not reported, but judged to "YES" based on correspondence with the authors. |
| Blinding? All outcomes ‐ patients | High risk | patients aware of interventions received |
| Blinding? All outcomes ‐ providers | High risk | providers aware of interventions delivered |
| Blinding? All outcomes ‐ assessors | High risk |
Only patient reported outcomes are used in the review. |
| Incomplete outcome data addressed? All outcomes ‐ drop‐outs? | Low risk | At the last follow‐up (6 months) the drop‐out rate ranged from 7% in the bed rest group to 13% in the physiotherapy group. Quote: "Reasons for dropping out of the study in the first 2 months of follow‐up were relocating (four patients), choosing another hospital (two patients), and unknown (six patients). Another 13 patients missed the 6‐month follow‐up visit, one because he relocated and 12 for unknown reasons." |
| Incomplete outcome data addressed? All outcomes ‐ ITT analysis | Low risk | Quote: "Comparison of the treatment groups was undertaken on the basis of the intention‐to‐treat principle. Data obtained in patients who dropped out, those who were noncompliant, and those who underwent surgery were included in the statistical analyses." |
| Free of selective reporting? | Unclear risk | No protocol available, and thus unclear whether the outcomes were defined a priori. |
| Similarity at baseline characteristics? | Low risk | Baseline clinical and demographic characteristics are thoroughly summarized and seem comparable. |
| Co‐interventions avoided or similar? | Low risk | Quote: "All patients were allowed to use analgesic medication and to call investigators for help if they had problems or questions." |
| Compliance acceptable? | Unclear risk | In the bed rest group two patients decided to consult a chiropractor; in the physiotherapy group one patient opted for bed rest; and in the control group two patients consulted a physiotherapist and one chose bed rest during the first 2 months. Compliance among bed rest patients (i.e. the actual number of hours in bed) was not measured. |
| Timing outcome assessments similar? | Low risk | Assessement 1,2 and 6 month following randomisation in both groups |
Malmivaara 1995.
| Methods | Design: Randomised (parallel group) controlled trial | |
| Participants |
|
|
| Interventions | Patients randomised to one of three interventions:
|
|
| Outcomes | Patients visited a physiotherapist for evaluation 3 and 12 weeks after start of intervention: Patient reported outcomes:
Outcomes assessed by physiotherapist blinded to treatment allocation:
|
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Low risk | Random number tables. |
| Allocation concealment? | Low risk | Quote: "The patients were examined by occupational health physicians at the centers, at which time the criteria for inclusion in the study were checked. The physicians opened the sealed envelopes and gave the treatment instructions to each patient at the end of the initial visit." |
| Blinding? All outcomes ‐ patients | High risk | patients aware of interventions received |
| Blinding? All outcomes ‐ providers | High risk | providers aware of interventions delivered |
| Blinding? All outcomes ‐ assessors | High risk |
Only patient reported outcomes are used in the review. |
| Incomplete outcome data addressed? All outcomes ‐ drop‐outs? | Low risk | 202 patients were randomised, but 16 patients were excluded because base‐line questionnaires were not obtained or were filled in too late or because the physicians’ initial determination that they fulfilled the criteria for inclusion in the study proved to be incorrect. The decision to withdraw these patients from the study was made without knowledge of their treatment assignments.
Reasons for dropping‐out are not stated, but "baseline characteristics of the patients who did not return for follow‐up did not differ markedly from the characteristics of those who returned." |
| Incomplete outcome data addressed? All outcomes ‐ ITT analysis | Low risk | Analyses are based on the patients for whom follow‐up data are available (Drop‐outs excluded from the analysis). |
| Free of selective reporting? | Unclear risk | No protocol available, and thus unclear whether the outcomes were defined a priori. |
| Similarity at baseline characteristics? | Low risk | The bed rest group included a larger fraction of patient with pain radiating below the knee (31%) compared to exercise (19%) and control groups (13%), and the control group consisted of more patients with heavy physical work (33%) compared to the bed rest group (15%). Otherwise, demographic and clinical characteristics in the three intervention groups were similar. |
| Co‐interventions avoided or similar? | Low risk | Quote: "Antiinflammatory drugs or analgesics were prescribed for 93%, 91%, and 93% of the patients in the bed‐rest, exercise, and control groups, respectively. In the bed‐rest group, one patient underwent back surgery because of a disk prolapse." |
| Compliance acceptable? | Low risk | Quote: "In the follow‐up questionnaires, all patients were asked on how many days they had taken bed rest during the day, and for how many hours on average. The patients were also asked on how many days they had done back exercises and how often they did them per day on average. If they received any health care services apart from those prescribed in the protocol, these were recorded." Compliance with bed rest at three‐week follow‐up: patients in the bed rest group had, on average, 22 hours of day‐time rest, compared with two hours in the control group. |
| Timing outcome assessments similar? | Low risk | Assessment 3 and 12 weeks following randomisation in all groups |
Postacchini 1988.
| Methods | Design: Randomised (parallel group) controlled trial | |
| Participants |
|
|
| Interventions | Patients randomised to one of six interventions:
|
|
| Outcomes | Outcome assessment at three weeks, two months and six months. A combined score ranging from 5 (poor clinical status) to 32 (excellent clinical status) was computed from a set of two subjective (patient reported) parameters and objective parameters, to each of which a score of 1 to 4 was assigned: Patient reported data:
Objective parameters:
|
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Unclear risk | Randomisation procedure not reported |
| Allocation concealment? | Unclear risk | Concealment approach not reported |
| Blinding? All outcomes ‐ patients | High risk | patients aware of interventions received |
| Blinding? All outcomes ‐ providers | High risk | providers aware of interventions delivered |
| Blinding? All outcomes ‐ assessors | High risk | A combined score of patient reported (not blinded) and objectively measured parameters (blinding of assessors not stated) |
| Incomplete outcome data addressed? All outcomes ‐ drop‐outs? | Unclear risk | 459 patients entered the study, 23 were lost to follow up and 38 interrupted or changed assigned treatment, resulting in a 13% drop‐out. Not stated whether withdrawals were evenly distributed in all groups. |
| Incomplete outcome data addressed? All outcomes ‐ ITT analysis | High risk | Patients who interrupted or changed assigned treatment (N = 38) were excluded from the analysis, as was also the patients who were lost to follow‐up (N = 23). |
| Free of selective reporting? | Unclear risk | Not stated whether the composition of the combined outcome score was defined a priori, or whether all important outcomes are reported. |
| Similarity at baseline characteristics? | Unclear risk | Distribution of sex, mean age and mean durations of symptoms are reported for each subgroup of patients (1A, 1B, 1C, 2A and 2B), but similar data for bed rest versus placebo group are not supplied. |
| Co‐interventions avoided or similar? | Unclear risk | This topic is not addressed |
| Compliance acceptable? | Unclear risk | Compliance with bed rest not reported |
| Timing outcome assessments similar? | Low risk | Assessment 2 weeks, 2 and 6 months following randomisation in both groups |
Rozenberg 2002.
| Methods | Design: Randomised (parallel group) controlled trial | |
| Participants |
|
|
| Interventions | Patients randomised to one of two treatments:
|
|
| Outcomes | Outcome was assessed by the patient treating physician on day 6 or 7, after 1 month and after 3 months:
|
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Low risk | Computer‐generated random numbers |
| Allocation concealment? | Low risk | Group assignment by calling a central office, stratified according to the patient's type of work |
| Blinding? All outcomes ‐ patients | High risk | patients aware of interventions received |
| Blinding? All outcomes ‐ providers | High risk | providers aware of interventions delivered |
| Blinding? All outcomes ‐ assessors | High risk | Neither patients nor the treating physicians were blinded to treatment allocation |
| Incomplete outcome data addressed? All outcomes ‐ drop‐outs? | Low risk | 281 patients were included in the two treatment groups. Excluded from ITT‐analysis: 3 withdrawals (allocation by the practitioner without calling the randomisation centre) and one loss to follow‐up. |
| Incomplete outcome data addressed? All outcomes ‐ ITT analysis | Low risk | Quote: "A per treatment analysis would have introduced a bias in favour of the treatment hypothesis being tested because it would have excluded the patients who had not complied with the bed rest period (essentially those doing best in the rest group and worst in the normal activity group). Accordingly, the principal analysis was an intention to treat analysis." |
| Free of selective reporting? | Unclear risk | No protocol available, and thus unclear whether the outcomes were defined a priori. |
| Similarity at baseline characteristics? | Low risk | Demographic and clinical characteristics for the two groups seem comparable |
| Co‐interventions avoided or similar? | Low risk | Quote: "Whatever allocation group, all subjects received the same medication during the first 4 days: 1g paracetamol three times per day and 8 mg of muscle relaxant, Thiocolchicoside, two times per day. Non‐steroidal antiinflammatory treatment was authorized during the study in the case of real need. Any analgesic or muscle relaxant treatment other than the study treatment was forbidden during the first week. Local or systemic corticoids and local treatments such as massage, physical therapy, vertebral manipulations, or braces were forbidden throughout the 3 months of the trial." |
| Compliance acceptable? | Low risk | Quote: "The prescribed treatment was followed by 95 patients (72%) in the bed rest group (mean duration of bed rest at least 16 hours daily for 4 days), as compared with 122 (89.7%) of those in the normal activity group (mean duration of bed rest 12 or fewer hours daily for 4 days) (P = 0.001). The mean number of actual days of bed rest was 3.0 (1.5) in the bed rest group and 0.1 (0.4) in the normal activity group. The mean number of hours of bed rest during the 4 days was 17.9 in the bed rest group and 9.9 in the activity group (night rest included). The groups did not differ in compliance with the drug treatment. Overall, 80% complied, with duration of treatment ranging from 3 to 5 days." |
| Timing outcome assessments similar? | Low risk | Assessment 1 and 3 months following randomisation in both groups |
Szpalski 1992.
| Methods | Design: Randomised (parallel group) controlled trial | |
| Participants |
|
|
| Interventions | Patient randomised to one of two bed rest strategies:
|
|
| Outcomes | Assessment at day one (both groups) and at day five (group 1) or day nine (group 2):
|
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Low risk | Randomisation table |
| Allocation concealment? | Low risk | Judged "YES" following correspondence with authors |
| Blinding? All outcomes ‐ patients | High risk | patients aware of interventions received |
| Blinding? All outcomes ‐ providers | High risk | providers aware of interventions delivered |
| Blinding? All outcomes ‐ assessors | High risk |
Only patient reported outcomes are used in the review. |
| Incomplete outcome data addressed? All outcomes ‐ drop‐outs? | Low risk | Quote: "3 drop‐outs (12%) in group 1 and 1 drop‐out (4%) in group 2." Reasons for drop‐outs are not stated. |
| Incomplete outcome data addressed? All outcomes ‐ ITT analysis | Low risk | Analysis conducted on the 47 patient who completed the study, whereas drop‐outs are excluded. |
| Free of selective reporting? | Unclear risk | No protocol available, and thus unclear whether the outcomes were defined a priori. |
| Similarity at baseline characteristics? | Unclear risk | No significant differences in the measured baseline patient characteristics, but a limited number of parameters was reported. For example, improvements in the Isostation B200 measurements are only measured in percent, baseline measurements for the two groups are not reported. |
| Co‐interventions avoided or similar? | Low risk | Quote: "Paracetamol at a dose of 1.5 G/day was prescribed for both groups to be taken for 2 days. Subjects in both groups were instructed not to perform heavy lifting tasks for 2 weeks after beginning treatment." |
| Compliance acceptable? | Unclear risk | Compliance with the bed rest recommendations (number of resting hours) not reported. |
| Timing outcome assessments similar? | High risk | Assessment at day one (both groups) and at day five (group 1) or day nine (group 2). As time in itself may be associated with improvements the results may be biased in favour of the outcomes being assessed at the later time‐point. |
Vroomen 1999.
| Methods | Design: Randomised (parallel group) controlled trial | |
| Participants |
|
|
| Interventions | Patients were randomised to one of two treatment strategies:
A pain medication algorithm including paracetamol and codeine or naproxen was applied in both groups. After being examined at two weeks, the patients were referred to either a general practitioner or a specialist for usual care, which ranged from advice to resume normal activities to surgery, but in all cases they were advised against bed rest. |
|
| Outcomes | Evaluation at 2 (the end of treatment), 3, and 12 weeks: Patient reported/rated outcomes:
Investigators (blinded) assessment:
All completed questionnaires were evaluated by a blinded investigator |
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Low risk | Computer‐generated random‐number table |
| Allocation concealment? | Low risk | Subjects were allocated to the intervention groups by a nurse |
| Blinding? All outcomes ‐ patients | High risk | patients were not blinded to the interventions received |
| Blinding? All outcomes ‐ providers | High risk | Quote: "Patient’s care giver was unaware of the treatment‐group assignments except in the cases of a few patients whose group assignment and outcome were thought to be helpful to the care giver in planning further management." However, the patient had to decide whether to comply with the advice, and as the patients themselves were not blinded to treatment allocation providers can not be said to be blinded. |
| Blinding? All outcomes ‐ assessors | High risk |
Only patient reported outcomes are used in the review. |
| Incomplete outcome data addressed? All outcomes ‐ drop‐outs? | Low risk | Quote: "Of the 183 patients who entered the study, 1 in each group did not return after two weeks; both reported improvement. After 12 weeks, further seven patients (9%) in each group were lost to follow‐up. Their base‐line characteristics and success rates after two weeks were similar to those of the 167 remaining patients." |
| Incomplete outcome data addressed? All outcomes ‐ ITT analysis | Low risk | Quote: "The results were analysed according to ITT principle." (All analysis are based on the 183 patients.) |
| Free of selective reporting? | Unclear risk | No protocol available, and thus unclear whether the outcomes were defined a priori. |
| Similarity at baseline characteristics? | Low risk | Baseline characteristics are summarized in separate table, and are seemingly similar. The degree of "pain in the leg" as measured on VAS is statistically higher in the control group (68 versus 62), but is probably of minor importance. |
| Co‐interventions avoided or similar? | Low risk | Quote: "The patients were allowed to take acetaminophen (1000 mg three times a day) for pain, supplemented by codeine (10 to 40 mg six times a day) or naproxen (500 mg three times a day) when necessary. Temazepam (10 mg once daily) was prescribed for insomnia. Patients were asked to record any other treatments they used for radicular symptoms, although these were discouraged." Quote: "The patients in the bed‐rest group, on average, took temazepam once during the two weeks of treatment, as compared with three times for the patients in the control group, else the use of co‐interventions was similar." |
| Compliance acceptable? | Low risk | Quote: "According to the patients’ diaries, those assigned to the bed‐rest group were in bed for a mean (±SD) of 21±4 hours daily, as compared with 10±4 hours for the patients in the control group. The most common reasons for not complying with the treatment instructions on a particular day were “resolution of complaints” (5% in the bed‐rest group and 4% in the control group), “incompatibility with work” (20% and 10%), and “it didn’t work” (12% and 9%). The number of hours of bed rest had no relation to the likelihood of improvement in either group." |
| Timing outcome assessments similar? | Low risk | Assessment 2,3 and 12 weeks following randomisation in both groups |
Wiesel 1980.
| Methods | Design: Randomised (parallel group) controlled trial | |
| Participants |
|
|
| Interventions | The study was divided into three major parts. One part investigated the role of antiinflammatory medications: patients were admitted to hospital for bedrest and randomised to three different medication regimens. The second part of the study investigated the effect of analgesic medication, and patients were randomly treated with bed rest plus either acetaminophen, codeine or oxycodone. The part of the study that is relevant for the current systematic review investigates the effect of bed rest versus staying active:
|
|
| Outcomes | Patient reported outcomes:
|
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Unclear risk | Not reported |
| Allocation concealment? | Unclear risk | Not reported |
| Blinding? All outcomes ‐ patients | High risk | patients were not blinded to the interventions received |
| Blinding? All outcomes ‐ providers | High risk | providers were not blinded to the interventions provided |
| Blinding? All outcomes ‐ assessors | High risk | outcomes were self‐reported level of symptoms (subjective) and return to full duty (objective) |
| Incomplete outcome data addressed? All outcomes ‐ drop‐outs? | Low risk | All included patients seemed to complete the study |
| Incomplete outcome data addressed? All outcomes ‐ ITT analysis | Low risk | All included patients seemed to complete the study |
| Free of selective reporting? | Unclear risk | No protocol available, and thus unclear whether the outcomes were defined a priori. |
| Similarity at baseline characteristics? | Unclear risk | Baseline characteristics not reported |
| Co‐interventions avoided or similar? | Low risk | All patients received one acetaminophen tablet twice daily, other co‐interventions avoided. |
| Compliance acceptable? | Low risk | Drill sergeants made sure that the control group were kept on their feet, and medical corpsmen overlooked the bed rest group |
| Timing outcome assessments similar? | Low risk | Assessment 14 days following randomisation in both groups |
Wilkinson 1995.
| Methods | Design: Randomised (parallel group) controlled trial | |
| Participants |
|
|
| Interventions | Patient randomised to one of two treatment strategies:
|
|
| Outcomes | Outcomes assessed 7 and 28 days after start of treatment:
|
|
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Adequate sequence generation? | Low risk | Sequence generation is not described in detail, but the fact that randomisation was achieved by sealed envelopes suggest that the sequence was generated prior to (and independent of) patient enrolment. |
| Allocation concealment? | Low risk | Sealed envelope |
| Blinding? All outcomes ‐ patients | High risk | patients were not blinded to the interventions received |
| Blinding? All outcomes ‐ providers | High risk | providers were not blinded to the interventions provided |
| Blinding? All outcomes ‐ assessors | High risk |
Only patient‐reported outcomes are used in the review. |
| Incomplete outcome data addressed? All outcomes ‐ drop‐outs? | High risk | 20 were allocated to the 48 hours' bed rest group and 22 to the control group. Thirty three subjects (79%), 15 from the bed rest group and 18 from the control group, returned for assessment on day seven and 34 subjects (81%), 14 from the bed rest group and 20 from the control group, completed the day 28 questionnaire. Four subjects failed to attend the day seven assessment and failed to complete the day 28 questionnaire. It is not reported whether drop‐outs differed from the rest. |
| Incomplete outcome data addressed? All outcomes ‐ ITT analysis | Low risk | Analysis are based on the patients who attended the physical examinations and completed the questionnaires. |
| Free of selective reporting? | Unclear risk | No protocol available, and thus unclear whether the outcomes were defined a priori. |
| Similarity at baseline characteristics? | Low risk | At baseline, the disability scores were higher in the bed rest group than in the control group, but else demographic and prognostic details of the included subjects seem to be similar. |
| Co‐interventions avoided or similar? | High risk | Quote: "All subjects received ibuprofen or, if this was contraindicated, co‐proxamol for analgesia. Subjects did not receive physiotherapy during the trial, and other treatments, including self‐remedies and physical therapies (apart from local application of heat), were discouraged. Self‐remedies and physical therapies reported by the 42 subjects were: local application of heat (10 subjects in the bed rest group and five in the control group); exercises (seven and four subjects, respectively); massage (five and three subjects, respectively); rubefacients (four and four subjects, respectively); chiropractic (one and one subject, respectively); and physiotherapy (two and no subjects, respectively). There were more recordings of treatments by subjects in the bed rest group than in the control group (29 and 17, respectively), but the difference was not significant." |
| Compliance acceptable? | Unclear risk | Compliance was assessed by using treatment diaries, completed by patients, that included hours of bed rest: Quote: "Thirty two of the 42 treatment diaries (76%) were returned. During the first 48 hours, subjects who were prescribed 48 hours' bed rest spent twice as many hours resting compared with controls, 12.6 daytime hours (SD 6.0 hours) and 6.1 daytime hours (SD 2.6 hours), respectively." |
| Timing outcome assessments similar? | Low risk | Assessment 7 and 28 days following randomisation in both groups |
Characteristics of excluded studies [ordered by study ID]
| Study | Reason for exclusion |
|---|---|
| Coomes 1961 | Alternate allocation, not RCT |
| Godges 2008 | Alternate allocation, not RCT |
| Hancock 2007 | Intervention= not staying active versus other treatments (besides bed rest); not comparison between different forms of advice to stay active (e.g. avoiding bed rest, staying active) |
| Paatelma 2008 | Participant with acute to chronic LBP, the first or recurrent episode. |
| Pal 1986 | Intervention = not bed rest versus staying active; not bed rest versus other treatment; not short bed rest versus long bed rest. |
| Rupert 1988 | Co‐intervention (drugs) is not used similarly across comparison groups |
Differences between protocol and review
Meta‐regression not performed due to few available studies.
Contributions of authors
The review authors contributed as follows: KTD, KBH and GJ planned and developed the protocol. KTD and KGB identified and assessed the trials and extracted data. KTD, KBH, GJ and KGB drafted and finalized the review.
Sources of support
Internal sources
Norwegian Knowledge Centre for the Health Services and National Resource Centre for Rehabilitation in Rheumatology, Norway.
External sources
No sources of support supplied
Declarations of interest
None
New
References
References to studies included in this review
Deyo 1986 {published data only}
- Deyo RA, Diehl AK, Rosenthal M. How many days of bed rest for acute low back pain? A randomized clinical trial. N Engl J Med 1986;315(17):1064‐70. [DOI] [PubMed] [Google Scholar]
Gilbert 1985 {published data only}
- Gilbert JR, Taylor DW, Hildebrand A, Evans C. Clinical trial of common treatments for low back pain in family practice. Br Med J Clin Res Ed 1985;291(6498):791‐4. [DOI] [PMC free article] [PubMed] [Google Scholar]
Hofstee 2002 {published and unpublished data}
- Hofstee DJ, Gutenbeek JMM, Hoogland PH, Houwelingen HC, Kloet A, Lötters F, et al. Westeinde sciatica trial: randomized controlled study of bed rest and physiotherapy for acute sciatica. J Neurosurg 2002;96(1 Suppl):45‐9. [DOI] [PubMed] [Google Scholar]
Malmivaara 1995 {published and unpublished data}
- Malmivaara A, Hakkinen U, Aro T, Heinrichs ML, Koskenniemi L, Kuosma E, et al. The treatment of acute low back pain ‐ bed rest, exercises, or ordinary activity?. N Engl J Med 1995;332(6):351‐5. [DOI] [PubMed] [Google Scholar]
Postacchini 1988 {published data only}
- Postacchini F, Facchini M, Palieri P. Efficacy of various forms of conservative treatment in low back pain: a comparative study. Neuro‐orthopedics 1988;6(1):28‐35. [Google Scholar]
Rozenberg 2002 {published data only}
- Rozenberg S, Delval C, Rezvani Y, Olovieri‐Apicella N, Kuntz JL, Legrand E, et al. Bed rest or normal activity for patients with acute low back pain: a randomized controlled trial. Spine 2002;27(14):1487‐93. [DOI] [PubMed] [Google Scholar]
Szpalski 1992 {published data only}
- Szpalski M, Hayez JP. How many days of bed rest for acute low back pain? Objective assessment of trunk function. Eur Spine J 1992;1(1):29‐31. [DOI] [PubMed] [Google Scholar]
Vroomen 1999 {published data only}
- Vroomen PCAJ, Marc CTFM, Wilmink JT, Kester ADM, Knottnerus JA. Lack of effectiveness of bed rest for sciatica. N Engl J Med 1999;340(6):418‐23. [DOI] [PubMed] [Google Scholar]
Wiesel 1980 {published data only}
- Wiesel SW, Cuckler JM, Deluca F, Jones F, ZeideMS, Rothman RH. Acute low‐back pain ‐ An objective analysis of conservative therapy. Spine 1980;5(4):324‐30. [DOI] [PubMed] [Google Scholar]
Wilkinson 1995 {published data only}
- Wilkinson MJ. Does 48 hours’ bed rest influence the outcome of acute low back pain?. Br J Gen Pract 1995;45(398):481‐4. [PMC free article] [PubMed] [Google Scholar]
References to studies excluded from this review
Coomes 1961 {published data only}
- Coomes NE. A comparison between epidural anaesthesia and bedrest in sciatica. BMJ 1961;1(5218):20‐4. [DOI] [PMC free article] [PubMed] [Google Scholar]
Godges 2008 {published data only}
- Joseph J Godges, Marie A Anger, Grenith Zimmerman, Anthony Delitto. Effects of Education on Return‐to‐Work Status for People With Fear‐Avoidance Beliefs and Acute LowBack Pain. Physical Therapy 2008;88(2):231‐239. [DOI] [PubMed] [Google Scholar]
Hancock 2007 {published data only}
- Mark J Hancock, Chris G Maher, Jane Latimer, Andrew J McLachlan, Chris W Cooper, Richard O Day, Megan F Spindler, James H McAuley. Assessment of diclofenac or spinal manipulative therapy, orboth, in addition to recommended fi rst‐line treatment foracute low back pain: a randomised controlled trial. Lancet 2007;370:1638‐43. [DOI] [PubMed] [Google Scholar]
Paatelma 2008 {published data only}
- Paatelma M, Kilpikoski S, Simonen R, Heinonen A, Alen M, Videman T. Orthopedic manual therapy, Mckenzie method or advice onely for low back pain in working adults: A randomizedcontrolled trial with one year follow‐up. J Rehabil Med 2008;40:858‐863. [DOI] [PubMed] [Google Scholar]
Pal 1986 {published data only}
- Pal B, Mangion P, Hossain MA, Diffey BL. A controlled trial of continuous lumbar traction in the treatment of back pain and sciatica. Br J Rheumatol 1986;25:1181‐3. [DOI] [PubMed] [Google Scholar]
Rupert 1988 {published data only}
- Rupert RL, Wagnon R, Thompson P, Ezzeldin MT. Chiropractic adjustments: results of a controlled trial in Egypt. Intl Rev Chiro 1985:58‐60. [Google Scholar]
Additional references
Allen 1999
- Allen C, Glasziou P, Mar C. Bed rest: a potentially harmful treatment needing more careful evaluation. Lancet 1999;354(9):1229‐33. [DOI] [PubMed] [Google Scholar]
Bortz 1984
- Bortz WM. The disuse syndrome. West J Med Nov. 1984;141(5):691‐4. [PMC free article] [PubMed] [Google Scholar]
Cohen 1988
- Cohen J. Statistical power analysis for the behavioral sciences. 1st Edition. New York, San Francisco, London: Academic press, 1988. [Google Scholar]
Convertino 1997
- Convertino VA, Bloomfield SA, Greenleaf JE. An overview of the issues: physiological effects of bed rest and restricted activity. Med Sci Sports Exerc 1997;29:187‐90. [DOI] [PubMed] [Google Scholar]
Furlan 2009
- Furlan AD, Pennick V, Bombardier C, Tulder M, Editorial Board, Cochrane Back Review Group. 2009 updated method guidelines for systematic reviews in the Cochrane Back Review Group. Spine 2009;34(18):1929‐41. [DOI] [PubMed] [Google Scholar]
Hagen 2004
- Hagen KB, Hilde G, Jamtvedt G, Winnem M. Bed rest for acute low‐back pain and sciatica. Cochrane Database of Systematic Reviews 2004, Issue 4. [DOI: 10.1002/14651858.CD001254.pub2] [DOI] [PubMed] [Google Scholar]
Higgins 2009
- Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.2 [updated September 2009]. The Cochrane Collaboration, 2009. Available from www.cochrane‐handbook.org.
Hilde 2002
- Hilde G, Hagen KB, Jamtvedt G, Winnem M. Advice to stay active as a single treatment for low‐back and sciatica. Cochrane Database of Systematic Reviews 2006, Issue 2. [DOI: 10.1002/14651858.CD003632.pub2] [DOI] [Google Scholar]
Mayer 1988
- Mayer TG, Gatchel RJ. Functional restoration for spinal disorders. The sports medicine approach. Philadelphia: Lea & Febiger, 1988. [Google Scholar]
Nachemson 2000
- Nachemson A, Waddell G, Nordlund A. Prevalence of pain in the neck and lower back [Förekomst av smärta i nacken och ländryggen]. In: Nachemson A, Johnson E editor(s). Prevalence neck and low back pain. in Neck and Back Pain. Stockholm: The Swedish Council on Technology Assessment in Health Care, 2000. [Google Scholar]
Waddell 1987
- Waddell G. 1987 Volvo award in clinical sciences. A new clinical model for the treatment of low‐back pain. Spine 1987;12:632‐44. [DOI] [PubMed] [Google Scholar]
Waddell 1997
- Waddell G, Feder G, Moshie L. Systematic reviews of bed rest and advice to stay active for acute low back pain. British Journal of General Practice 1997;47:647‐52. [PMC free article] [PubMed] [Google Scholar]
Wheeler 1995
- Wheeler AH, Hanley EN. Nonoperative treatment for low back pain. Rest to restoration. Spine 1995;20(3):375‐8. [DOI] [PubMed] [Google Scholar]