The Journal of Pain, Vol 15, No 1 (January), 2014: pp 2-15 Available online at www.jpain.org and www.sciencedirect.

com

Critical Review
Interventions Available Over the Counter and Advice for Acute Low Back Pain: Systematic Review and Meta-Analysis
Christina Abdel Shaheed,* Christopher G. Maher,y,z Kylie A. Williams,x and Andrew J. McLachlan*,jj
*Faculty of Pharmacy, University of Sydney, Sydney, Australia. y George Institute for Global Health, Sydney, Australia. z Sydney Medical School, University of Sydney, Sydney, Australia. x School of Pharmacy, University of Technology Sydney, Sydney, Australia. jj Centre for Education and Research on Ageing, Concord Hospital, Sydney, Australia.

Abstract: This systematic review evaluated evidence from randomized controlled trials investigating interventions available over the counter and advice that could be provided to people with acute low back pain. Searches were conducted on MEDLINE, Embase, Cochrane Database of Systematic Reviews, AMED, CENTRAL, and PsycINFO for eligible randomized controlled trials. The primary outcome measure was pain. Eligible controls included placebo, no treatment, or usual care. Two reviewers extracted data and rated study quality. A random effects model was used to pool trial effects with the overall strength of evidence described using the GRADE criteria. Thirteen randomized controlled trials (2,847 participants) evaluating advice, bed rest, simple analgesics (paracetamol, nonsteroidal anti-inammatory drugs), heat application, and a topical rubefacient were included. There was low-quality evidence that bed rest is ineffective and very-low-quality evidence that advice is ineffective in the short, intermediate, and long terms. There was very-low-quality evidence that nonsteroidal anti-inammatory drugs (ibuprofen and diclofenac when required dosing) provide an immediate analgesic effect (mean differences 10.9 [95% condence interval = 17.6 to 4.2] and 11.3 [95% condence interval = 17.8 to 4.9], respectively). There is very-low-quality evidence that heat wrap and a capsicum-based rubefacient provide an immediate analgesic effect (mean differences 13.5 [95% condence interval = 21.3 to 5.7] and 17.5, P < .001, respectively), but there was no information on longer-term outcomes. Perspective: There is limited evidence that nonsteroidal anti-inammatory drugs, heat wrap, and rubefacients provide immediate pain relief for acute back pain and that bed rest and advice are both ineffective. Future research is needed to provide evidence to support rational use of over-the-counter remedies and advice for people with acute low back pain. 2014 by the American Pain Society Key words: Low back pain, over the counter, advice, paracetamol, nonsteroidal anti-inammatory drugs.
No specic funding from any government or not-for-prot organization was sought for this research. C.G.M. and A.J.M. are investigators on the PACE trial (http://www.anzctr. org.au/ACTRN12609000966291.aspx), which is jointly funded by a project grant from the National Health and Medical Research Council and GlaxoSmithKline. C.A.S. is supported by an Australian Postgraduate Award. Supplementary data accompanying this article are available online at www.jpain.org/ and www.sciencedirect.com/. Address reprint requests to Andrew J. McLachlan, BPharm, PhD, Faculty of Pharmacy, University of Sydney, Pharmacy Building A15, Sydney, New South Wales 2006, Australia. E-mail: andrew.mclachlan@sydney. edu.au 1526-5900/$36.00 2014 by the American Pain Society http://dx.doi.org/10.1016/j.jpain.2013.09.016

ow back pain (LBP) is a common and costly condition worldwide,55 with up to 84% of adults experiencing an episode at some point in their lives.21 The Global Burden of Disease 2010 study31 has identied LBP as the leading contributor to disability. In Australia, it is estimated that more than $1 billion is spent annually on treatments for LBP, whereas a further $8 billion is spent on indirect costs.56 In the United States, this gure is estimated to be as high as $50 billion per year.11 Only around half of people with acute LBP consult a health professional,11,23,56,57 with use of remedies

Abdel Shaheed et al available over the counter (OTC) widespread. Findings from a 2008 U.S. National Health Interview Survey47 identied LBP as one of the most common conditions treated with OTC medicines. Many of these OTC interventions can be provided in a community pharmacy/ drug store and include nonprescription medicines and simple remedies such as heat or cold packs that can be provided to a person with LBP for self-administration. Despite the widespread use of OTC interventions and the provision of advice to those experiencing LBP, there are no systematic reviews summarizing the clinical effectiveness of these interventions for acute LBP. The aim of this review was to investigate the effectiveness of interventions that can be accessed OTC, without the need for a prescription, and advice that could be delivered in a primary contact setting.

The Journal of Pain

Data Extraction and Quality Assessment

Methodological quality of individual trials was assessed using the Physiotherapy Evidence Database (PEDro) scale.10,32,34 The PEDro scale is an 11-item scale (Table 1) that has been established as a valid and reliable method of rating methodological quality of individual RCTs.10,32,34 Each item is scored as either present (1) or absent (0). The PEDro summary score does not include the item related to external validity, thus giving a total score out of 10. Rating of trials is carried out by 2 independent raters, with disagreements resolved by an independent third rater. All PEDro raters undergo initial training, which involves practice with feedback, and raters do not begin rating trials until they pass an accuracy test. The level of disagreements for raters is monitored by the PEDro project manager, and if required additional training is provided for a rater. Finally, authors of trials are able to dispute ratings and present a case that a rating should change.42 Given the extensive quality control procedures used by the PEDro database, we adopted existing ratings for 12 of the 13 RCTs, with the remaining trial12 allocated a PEDro rating by 3 reviewers (C.A.S., C.G.M., and A.J.M.) using a joint consensus approach. Trials scoring <7/10 on the PEDro scale were dened as low quality; those scoring 7 or more were considered high quality. Two reviewers (C.A.S. and C.G.M.) extracted outcomes data from each individual study. Missing data were obtained by contacting authors or estimated using the methods endorsed by the Cochrane Collaboration.25 We adopted median scores for missing means and used standard deviation values from baseline (or the most similar eligible study) as a substitute for missing standard deviation values. An adapted version of the GRADE criteria3 endorsed by the Cochrane Back Review Group was used to evaluate the strength of recommendations and the overall quality

Methods
Data Sources and Searches
MEDLINE, Embase, Cochrane Database of Systematic Reviews, AMED, CENTRAL, and PsycINFO (inception to March 2013) were searched for randomized controlled trials (RCTs) evaluating OTC interventions and/ or advice for acute LBP (full search details in the Supplementary Appendix). We also screened studies and reference lists from systematic reviews in the area to identify potentially eligible RCTs. One reviewer (C.A.S.) screened titles and abstracts of retrieved studies. Two reviewers drawn from a pool of 3 reviewers (C.A.S., A.J.M., and C.G.M.) inspected the full manuscript of potentially eligible RCTs to determine eligibility, with disagreements resolved by consensus.

Study Selection
Studies were restricted to English language RCTs evaluating OTC remedies for acute (pain duration <12 weeks) nonspecic LBP. We restricted interventions to those that could be self-administered by a person with LBP and were readily accessible from a community pharmacy/ drug store without prescription (Supplementary Appendix Tables 1 and 2). Eligible controls included placebo, no treatment, or usual care. Eligible interventions included OTC medicines (eg, paracetamol or nonsteroidal anti-inammatory drugs [NSAIDs]), complementary/herbal remedies (eg, comfrey), and topical applications (eg, heat or cold packs) (see Supplementary Appendix). We also included advice that could be provided to a person with LBP in an OTC setting. We excluded interventions that required a prescription (such as single-ingredient opioid analgesics, analgesic adjuvants, or muscle relaxants) and physical interventions such as spinal manipulation, acupuncture, and laser therapy. Trials were included if they reported endpoints such as pain, disability, global perceived recovery, sickness leave, or adverse events outcomes. We considered pain as the primary outcome.

Table 1.

PEDro Ratings for Eligible Trials

1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 3 1 1 1 1 1 1 1 1 1 1 4 1 1 1 1 1 5 1 6 1 1 1 1 1 1 1 7 1 1 1 1 1 1 1 1 1 1 8 1 1 1 1 1 1 1 1 9 1 1 1 1 1 1 1 1 1 1 1 1 10 TOTAL 1 1 1 1 1 1 1 1 1 1 1 9 4 6 6 9 6 6 6 7 6 7 5 5

STUDY Dreiser 200312 Milgrom 199336 Nadler 200338 Nadler 200339 Pengel 200743 Ginsberg 198719 Jellema 200528 Indahl 199527 Storheim 200351 Gilbert 198518 Malmivaara 199535 Wilkinson 199558 Rozenberg 200249

NOTE. 1 = randomization; 2 = concealed allocation; 3 = baseline comparability; 4 = subject blinding; 5 = therapist/physician blinding; 6 = assessor blinding; 7 = adequate follow-up (>85%); 8 = intention-to-treat analysis; 9 = betweengroup statistical comparisons; and 10 = point measures and measures of variability.

The Journal of Pain

Over-the-Counter Interventions and Advice for Acute Low Back Pain This approach to transforming data has been described46 and used45,46 in recent systematic reviews evaluating these outcome measures. We present results in mean differences rather than standardized mean differences, as the benchmarks for clinically important difference in pain and disability are expressed in points on a 0 to 100 pain scale, not proportions of a standard deviation.14,20,41 The proposed thresholds for clinically important changes in pain and disability range from 10 to 30 points on this scale of 0 to 100.14,20,41 We therefore considered a mean difference of at least 10 points on this 0 to 100 scale as clinically signicant. Outcomes were grouped into 4 time categories: immediate term (#2 weeks after randomization), short term (>2 weeks, <3 months), intermediate term ($3 months, <12 months), and long term ($12 months). Where multiple time points were available for a single category, the time closest to 1 week was chosen for immediate term, 8 weeks for the short term, 6 months for the intermediate term, and 12 months for the long term. Results from dichotomous outcomes have been expressed as risk ratios (RR). Descriptive statistics were used to report on adverse events. Trials deemed clinically homogenous were grouped according to intervention, outcomes, and time points. Meta-analysis was carried out using Review Manager (RevMan) version 5.1 (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark). Pooled effects were calculated only when inconsistencies among trials were low (I2 statistics < 40%) using a random effects model. When trials were not sufciently homogenous, pooling of data via meta-analysis was not performed and individual effect sizes were reported. For calculation of effect size, see Supplementary Appendix Tables 3 to 5.

of the evidence for a given intervention. This method is described in detail elsewhere,45,46 but briey the quality of evidence was downgraded a level for each of the 4 factors considered in this review: poor study design (25% or more of trials, weighted by sample size, have a low PEDro score <7/10), inconsistency of results (25% or more of the trials, weighted by sample size, have results that are not in the same direction), imprecision (sample size <300 for each outcome), and indirectness (where the trial context, eg, setting, patients, intervention, comparison, or outcomes, is not exactly the same as the review question). We did not assess for publication bias using funnel plots because too few studies were included in the meta-analysis. Overall quality of evidence was dened as  High qualityall domains satised, no suspected reporting bias, future research unlikely to change our condence in the estimated effect  Moderate quality1 domain not met, future research likely to have a signicant impact on our condence in the estimated effect and is likely to change the effect  Low quality2 domains not met, future research likely to have a signicant impact on our condence in the estimated effect and is likely to change the effect  Very low quality3 or more domains not met, uncertain about the estimated effect. For each outcome, the quality of evidence was downgraded 1 level for indirectness as the interventions were not delivered specically within the OTC setting. Single RCTs with sample sizes <300 provided very-lowquality evidence as these were also deemed both inconsistent and imprecise. Note: Results from single RCTs provided inconsistent data as there needed to be more than 1 trial to compare results with in order to determine consistency.

Results
The database searches identied 4,336 studies. After screening titles and abstracts from the database search and a hand search of references, 22 potentially relevant RCTs were inspected for full review, with 13 found to be eligible. A summary of the search results is provided in Fig 1. Eligible studies investigated paracetamol/acetaminophen,36 NSAIDs,12,36 a capsicum-based topical rubefacient,19 local heat application,38,39 bed rest vs normal activity,18,35,49,58 and advice delivered by a health care provider as a stand-alone intervention,43 in addition to written educational material,28 or involving referral to an allied health care professional (HCP) such as a physiotherapist27,51 (excluding physical interventions). The risk of bias (PEDro rating) of each trial is summarized in Table 1. The overall quality of evidence for OTC interventions, bed rest, or advice in acute LBP is of low to very low quality. Most studies on OTC interventions (medicines, local heat application) were single studies with small sample sizes (ie, <300) and/or had deciencies in study design, that is, low PEDro rating. Common limitations with the advice trials centered on poor study design and inconsistency in ndings (25%

Data Synthesis and Analysis

Pain and disability measures were converted to a common 0 to 100 scale (0 = no pain or disability to 100 = worst possible pain or disability). The pain intensity measures used in the meta-analysis were visual analog scale scores (range, 0100) and numerical rating scale (NRS) scores (range, 010). The NRS was converted to the same 0 to 100 scale as in the visual analog scale because these 2 pain measures have been shown to be highly correlated and, when transformed, can be used interchangeably.5,26 The disability measures used to calculate pooled effects were the Oswestry Disability Index scores (range, 0100), RolandMorris Disability Questionnaire (RMDQ) scores (range, 024), and the Eifel Index scores (range, 024), which is the validated French-language equivalent of the RMDQ.8 The RMDQ/ Eifel index scores were converted to the same 0 to 100 scale as in the Oswestry Disability Index. The Oswestry Disability Index and RMDQ are highly correlated and share similar psychometric properties, and they are the most commonly employed measures of disability in the back pain literature.48

Abdel Shaheed et al

The Journal of Pain

Figure 1. Summary of search. or more of the trials, weighted by sample size, have results that are not in the same direction). The characteristics of all eligible studies have been summarized in Table 2. sample size >300 but downgraded for inconsistency, indirectness, and poor study design). The effects of advice on disability are similar to those for pain, with pooled results showing no clinically significant effect for the short28,43 and long terms28,43: MDs 1.4 (95% CI = 5.8 to 3.0) and .2 (95% CI = 2.9 to 2.6), respectively. Similarly, the GRADE classication for these effects at each time point is very low quality (combined sample size >300 but downgraded for inconsistency, indirectness, and poor study design). Three RCTs18,35,49 investigating the effect of bed rest on pain were identied. Pooling showed no effect of bed rest in the immediate, short, and intermediate terms: MDs 3.4 (95% CI = .7 to 7.5), 3.9 (95% CI = .3 to 7.5), and 4.9 (95% CI = .5 to 9.2), respectively, with results favoring normal activity. The GRADE classication for these effects is low quality (total sample size >300; however, downgraded for indirectness and poor study design). The effects of bed rest on disability match those for pain, with pooled estimates from 3 studies35,49,58 similarly showing no signicant effect of bed rest in the immediate, short,

Pain and Disability

The effects of OTC interventions and advice on pain and disability are shown in Figs 2 and 3, respectively. Clinically signicant effects on both pain and disability were identied for NSAIDs alone, whereas heat wrap treatment showed clinically important effects for pain only. None of the trials evaluating advice or bed rest reported statistically and clinically important effects at any time point. Pooled results from 2 trials28,43 on advice following a biopsychosocial model of patient care showed no signicant effect on pain in the intermediate and long terms: mean differences (MDs) .8 (95% condence interval [CI] = 4.7 to 3.2) and .2 (95% CI = 4.2 to 3.9), respectively. The GRADE classication for these effects is very low quality for each time point (combined

Table 2.
STUDY

Characteristics of Included Studies

PATIENT POPULATION 219 patients (119 F, 100 M) mean age 36.1 with acute (less than 3 months) nonspecic LBP. Pain intensity had to be more than 1 on a 6-point scale 76 patients (49 F, 27 M) with acute nonspecic LBP 369 patients (187 F, 182 M) mean age 40.8 with untreated acute LBP; onset within 2 days. Pain >50 mm on a 100-mm VAS SETTING 2 community-based research centers INTERVENTION COMPARISON LENGTH OF TREATMENT Heat wrap worn for 3 consecutive days FOLLOW UP Day 5 ELIGIBLE OUTCOME MEASURE Disability: RMDQ 024

6
The Journal of Pain

Nadler 200338

Oral placebo 2 tablets 3 Heat wrap times daily spaced 6 hours (Thermacare Heat apart. n = 88 wrap; Procter & Gamble, Cincinnati, OH); heats to 40 C within 30 minutes of exposure to air; maintains this temperature continuously for 8 hours. n = 92

Nadler 200339

Five clinical testing research sites

Dreiser 200312

54 clinics of general practitioners in France

Milgrom 199336 395 male infantry recruits mean age 18 with acute overexertion back pain

Israeli infantry recruits

As above. Heat wrap Oral placebo 2 tablets applied 1520 minutes before taken 1520 minutes sleep and worn for 8 hours for before bed each night for 3 3 consecutive nights. n = 31 consecutive nights. n = 32 Diclofenac 12.5 mg (rapidPlacebo initial dose 2 tablets, acting formulation), initial 7 days exible dose 46 dose 2 tablets, 7 days tablets/day. n = 116 exible dose 46 tablets a day. n = 119 Ibuprofen 200 mg initial dose Placebo initial dose 2 tablets, 2 tablets, 7 days exible dose 7 days exible dose 46 to 46 tablets a day. n = 118 tablets/day. n = 116 Ibuprofen 800 mg 3 times No drug treatment. n = 22 a day for 7 days. n = 24 Paracetamol 1000 mg 3 times No treatment. n = 22 a day for 7 days. n = 24

Heat wrap worn for 3 consecutive nights 7 days

Day 5

Pain affect (0100) RMDQ

Variable Day 8

As above

As above

Pain (100-mm VAS) mean changed score (SD) Side effects RMDQ 024 As above

Over-the-Counter Interventions and Advice for Acute Low Back Pain

7 days 7 days

Variable time End of basic training (14 weeks) Immediate

Cure/no cure Cure/no cure

Ginsberg 198719 40 patients with acute Clinical settingaccess to mechanical LBP. radiologic examination and Each patient was given 45 laboratory tests paracetamol 250-mg tablets. No other analgesic, antiinammatory drug, or physical treatment was allowed during the 12-week period

Rado-Salil ointment (containing Placebo (containing only the 14 days 17.64 mg ethylsalicylate excipient with 3 times the 26.47 mg methylsalicylate, amount of lavendula and 8.82 mg glycosalicylate etc in bergamot essences) matched the form of a 40-g stick for appearances. n = 20 applied as needed. n = 20

Pain evaluation on a 10-cm linear scale

Table 2.
STUDY

Continued
PATIENT POPULATION SETTING INTERVENTION COMPARISON LENGTH OF TREATMENT 4 days FOLLOW UP ELIGIBLE OUTCOME MEASURE

Abdel Shaheed et al

Gilbert 198518

252 patients (124 F, 128 M) Family practice Bed rest: patients instructed to No instruction or treatment. mean age 40.6 with acute LBP stay in bed for at least 4 days n = 48 with or without radiation and given written instructions down the leg and free from depicting appropriate LBP for more than 30 days positions for bed rest. n = 47 before the current episode Storheim 200351 93 patients (48 F, 45 M) mean Initial clinical examination was Cognitive group n = 34 (ITT) Control group received age 40.8 sick listed for 8 followed by an appointment Baseline intervention: usual care (no referral). 12 weeks due to subacute LBP at an outpatient clinic standard clinical examination n = 29 (ITT) where back was examined, imaging results explained, and patients encouraged to continue with normal activity Cognitive intervention took place at the outpatient clinic and was a collaborative effort between a specialist in physical medicine and a physical therapist. It involved a discussion of pain mechanisms, completion of a questionnaire, functional examination, instruction on movement methods, reassurance, and advice on how to manage new attacks. Two consultations offered, each lasting 3060 minutes 7 university hospitals and Advice sessions followed that Sham advice and Pengel 200743 259 patients (124 F, 135 M) mean age 49.9 primary care clinics in of Indahl et al,27 where the sham exercise health provider with subacute LBP Australia and New Zealand n = 68 (physiotherapist) encouraged (excluding recent a graded return to normal surgery) activities, explained the benign nature of the condition, addressed any unhelpful beliefs about back pain, cautioned against being overly careful, eg, avoiding light activity as this could delay recovery. n = 63

10 days, 6 weeks, Pain 0100-point and 12 weeks scale

Two consultations Reported at were offered 18 weeks (duration of each 3060 minutes).

Pain (VAS) Disability (RMDQ)

6 weeks

6 weeks, 3 months, 12 months

Pain (010 linear scale) Disability (RMDQ) The Journal of Pain

Table 2.
STUDY Rozenberg 200249

Continued
PATIENT POPULATION SETTING INTERVENTION COMPARISON LENGTH OF TREATMENT 4 days, total trial period 3 months FOLLOW UP ELIGIBLE OUTCOME MEASURE

8
The Journal of Pain

Wilkinson 199558

Malmivaara 199535

Indahl 199527

Jellema 200528

278 patients (147 F, 131 M) Multicenter study involving Bed rest for at least 16 of Advised to continue with mean age 43.6 with acute LBP ambulatory patients recruited 24 hours for the rst 4 days. normal daily activity (to with pain >40 mm on VAS by rheumatologists and the extent that pain Patients instructed to remain general practitioners in bed except to eat and for allowed). n = 140 personal care. n = 138 33 patients with acute LBP of Recruitment of participants by 48 hours of strict bed Advised to stay active less than 7 days duration. 9 general practitioners from rest. n = 15 and avoid daytime Mean age 35.2 and 41.2 (bed practices in West Midlands rest. n = 18 rest and control group respectively) 186 patients (124 F, 62 M) Occupational health care center Bed rest for 2 days with only Continuation of normal mean age 40.3 with acute LBP in Helsinki, Finland essential walking activities as tolerated. (eg, to go to bathroom) n = 61 allowed. n = 62 975 patients (378 F, 597 M) County of Oetsfold Examination whereby Control group did not receive mean age 42.5 with subacute Spine Clinic participants received advice a formal examination but LBP on sickness leave from largely following principles of instead received standard work for more than 8 weeks a biopsychosocial model. The care. n = 512 condition was explained to the patient, light activity was encouraged, emotional aspects of pain discussed, and fears about the condition addressed. Information was reinforced at 3 months and 1 year during follow-up appointments. Initial consultation  2.5 hours. n = 463 314 patients with LBP of less 60 general practitioners in 41 General practitioner Usual care. than 12 weeks duration general practices. consultation with emphasis n = 163 on psychosocial prognostic factors (at least 1 consultation of 20 minutes) plus educational booklet. n = 142

1 week, 1 month, Pain intensity 3 months (VAS) Disability (Eifel index)

48 hours

1 week, 4 weeks Disability (Oswestry Index)

2 days

3 and 12 weeks

Pain scale (010)

Over period 200 days and of consultation 400 days

Number (n) not on sickness leave

Over-the-Counter Interventions and Advice for Acute Low Back Pain

Over period of consultation

6 weeks, 13 weeks, and 1 year

Pain (VAS) Disability (RMDQ)

Abbreviations: F, female; M, male; VAS, visual analog scale; SD, standard deviation; ITT, intention to treat.

Abdel Shaheed et al

The Journal of Pain

Figure 2. Effects of OTC remedies on pain. Pain is measured on a 0 (no pain) to 100 (maximum) scale. Effects are mean differences in
outcome with 95% CI. For studies 12 and 51, the effects are computed from change scores from baseline. NSAID 1 = ibuprofen 200 mg p.r.n.; NSAID 2 = diclofenac 12.5 mg p.r.n.

and intermediate terms (MDs 4.6 [95% CI = 1.3 to 8.0], 4.7 [95% CI = 1.9 to 7.4], and 3.2 [95% CI = .8 to 5.7], respectively). The overall quality of evidence is considered low quality (downgraded for indirectness and poor study design). A single RCT of heat wrap therapy39 reported statistically signicant and clinically important pain relief in the immediate term (MD 13.5 [95% CI = 21.3 to 5.7]); however, no outcomes were available beyond day 5. The GRADE classication for this effect is very low quality (single study downgraded for imprecision, indirectness, and poor study design). Pooled results from 2 low-quality RCTs on heat wrap treatment38,39 showed a statistically signicant but not clinically signicant effect on disability (MD 8.90 [95% CI = 13.7 to 4.0]) (Fig 3). The individual studies provide imprecise estimates of treatment effect (sample size <300), and the GRADE classication is very low quality for this intervention (downgraded for imprecision, indirectness, and poor study design). Evidence around NSAIDs was limited, with only 1 study12 reporting on eligible pain and disability measures. This study12 used a 3-arm study design (ibuprofen 200 mg p.r.n. [when required] vs diclofenac 12.5 mg p.r.n. vs placebo p.r.n.), with treatment effects for each intervention described in Figs 2 and 3. There were not enough clinical

data (pain and disability) on NSAIDs to allow pooling of effects, as such individual effect sizes have been reported. There were clinically signicant effects on pain, with ibuprofen 200 mg and diclofenac 12.5 mg p.r.n. dosing MDs 10.9 (95% CI = 17.6 to 4.2) and 11.3 (95% CI = 17.8 to 4.9), respectively (Fig 2), but there are no outcomes available beyond day 8 of treatment. Ibuprofen 200 mg and diclofenac 12.5 mg p.r.n. dosing were also found to have clinically signicant effects on reducing disability in the immediate term, MDs 10.0 (95% CI = 15.6 to 4.4) and 12.1 (17.9 to 6.2), respectively (Fig 3), but similarly there are no outcomes beyond day 8 of treatment. The GRADE classication for these effects on both pain and disability is very low quality (single study, downgraded for inconsistency, imprecision, and indirectness). One RCT19 investigating a capsicum-based topical rubefacient showed a statistically signicant improvement in pain compared to placebo at day 3 (mean score 19.0 vs 1.5, respectively) (MD 17.5, P < .001) and provides very-low-quality evidence for the effectiveness of this formulation in acute LBP (single study downgraded for imprecision, indirectness, and poor study design).

Recovery
Two RCTs28,36 reported on recovery. Studies reporting on recovery measured perceived recovery as reported

10

The Journal of Pain

Over-the-Counter Interventions and Advice for Acute Low Back Pain

Figure 3. Effects of OTC remedies on disability. Disability is measured on a 0 (no disability) to 100 (maximum) scale. Effects are mean
differences in outcome with 95% CI. For studies 12 and 51, the effects are computed from change scores from baseline. NSAID 1 = ibuprofen 200 mg when required (p.r.n.); NSAID 2 = diclofenac 12.5 mg when required (p.r.n.).

by the patients themselves. In 1 trial,28 this was reported on a Likert-type scale ranging from slightly improved to very much worse, with data on no recovery used as a primary outcome measure. In the other trial36 recovery was simply expressed as cure vs no cure. (Note that this review presents data on no recovery/no cure.) One study36 on NSAIDs found no signicant difference between unrecovered participants treated with paracetamol 1,000 mg 3 times daily for 7 days and the control group treated with placebo, RR 2.52 (95% CI = 0.96 to 6.77). The same study reported no signicant difference between unrecovered participants treated with ibuprofen 800 mg 3 times daily for 7 days compared with the control group, RR 1.38 (95% CI = .67 to 2.81). The GRADE classication for these effects is very low quality (single study downgraded for imprecision, indirectness, and poor study design). One RCT28 investigating advice accompanied by an educational booklet found no signicant difference between unrecovered participants in the intervention and control (usual care) groups in the short, intermediate, and long terms: RRs 1.03 (95% CI = .75 to 1.42), 1.05 (95% CI = .75 to 1.47), and 1.15 (95% CI = .81 to 1.65), respectively (Supplementary Appendix Table 5). The

GRADE classication is very low quality for each time point (single study, downgraded for inconsistency, indirectness, and poor study design).

Sickness Leave
Two studies investigating advice delivered by a health professional27,28 following a biopsychosocial model to patient care and 3 studies on bed rest35,49,58 reported on sickness leave as a study outcome. Studies reporting on sickness leave were included if they reported the number (n) of people on sickness leave (as opposed to the number of sickness days) at any time point specied in this review (immediate, short, intermediate, long terms). Pooling of data shows a signicant effect of advice in the intermediate term, RR .5 (95% CI = .43 to .58) (Fig 4), with the GRADE classication for this effect considered low quality (downgraded for indirectness and poor study design). Pooled results from 2 studies on bed rest35,49 showed a statistically signicant negative effect of bed rest in the immediate term, RR 1.8 (95% CI = 1.5 to 2.2) with results favoring normal activity. The GRADE classication for

Abdel Shaheed et al

The Journal of Pain

11

Figure 4. Effects of advice on sickness leave. this effect is very low quality (downgraded for imprecision, inconsistency, and poor study design). However, pooled estimates from 2 trials35,58 showed no statistically signicant effect of bed rest in the short term, RR 1.3 (.9, 1.9) (Fig 5), with the GRADE classication for this effect considered low quality (downgraded for imprecision and indirectness). with the diclofenac (16/124) and ibuprofen treatment groups (14/122). Common side effects for the NSAIDs were gastrointestinal: abdominal discomfort, diarrhea, nausea, and vomiting. There was 1 case of rectum hemorrhage in the ibuprofen group. Somnolence was reported by a total of 3 participants in the diclofenac group and other nervous system reactions reported by a further 4 participants treated with NSAIDs (3 diclofenac, 1 ibuprofen). Sixteen participants were withdrawn from the study because of an adverse event (8 placebo, 4 diclofenac, 4 ibuprofen; specic event not described).

Adverse Effects
Data on adverse effects of interventions were identied for heat application, a topical capsicum-based rubefacient, and NSAIDs. One study on heat wrap therapy38 reported 1 case of local skin discoloration by day 5 of the study; however, it resolved without treatment. The study intervention was well tolerated among 94/95 participants. The study of a capsicum-based rubefacient19 reported 1 case of skin irritation on day 7 of treatment and 3 more cases between days 10 and 12. These effects resolved spontaneously. One study on NSAIDs12 reported that side effects were more common in the placebo group (19/122) compared

Discussion
There is no convincing evidence of effectiveness for any intervention available OTC or advice in the management of acute LBP. At best there is very-low-quality evidence that heat wrap, NSAIDs, and rubefacients provide an immediate analgesic effect but no information about longer-term effects on pain. There is low- to very-low-quality evidence that bed rest and advice are ineffective. This review has highlighted an area of

Figure 5. Effects of bed rest on sickness leave.

12

The Journal of Pain

Over-the-Counter Interventions and Advice for Acute Low Back Pain in the intermediate term. Results from one of these studies27 involving referral to an allied HCP and reinforcement of key messages at follow-up visits showed signicant effects in the intermediate and long terms. This represents preliminary evidence of the potential benets of multidisciplinary management while highlighting a need for ongoing review of individuals with nonspecic LBP. There are efforts underway4 to evaluate the effectiveness and cost-effectiveness of a biopsychosocial multidisciplinary intervention in LBP. Results from such research are important to inform the evidence around the benets of these interventions, which may support their inclusion as part of a government-reimbursed service. With regard to pharmacologic management, current guidelines recommend paracetamol (4 g/day in divided doses) as rst-line treatment ahead of NSAIDs.40,53 This is supported by evidence that identied no benet with the addition of diclofenac 50 mg twice daily to simple rst-line care (advice and paracetamol).22 This review identied a paucity of evidence around simple interventions (NSAIDs and paracetamol) as stand-alone interventions, with very-low-quality evidence that rapid-acting diclofenac 12.5 mg (the potassium formulation as opposed to the sodium formulation) and ibuprofen 200 mg taken on a p.r.n. regimen has clinically signicant effects on pain and disability. However, a signicant limitation with this study12 was that assessment of outcome measures was available for the immediate term only. A better analysis would have been to follow up patients over a longer period to ascertain the short-, intermediate-, and long-term clinical benets of these interventions. There was also no statistical evaluation on the side-effect prole of these low-dose NSAIDs compared with placebo; however, no marked dissimilarities were noted, with the primary complaint in each study arm being gastrointestinal complications (eg, abdominal discomfort, nausea). Gastrointestinal complications are the most common side effects reported with NSAIDs,17,37,50 although ibuprofen and diclofenac taken at low doses appear to have the least risk of gastrointestinal complications compared with any other NSAID.17,50 There was an absence of pain data with the high dose of ibuprofen36 investigated in this review (800 mg 3 times daily) to allow a comparison with the simpler p.r.n. regimen of ibuprofen12 evaluated. There are suggestions that starting with a higher dose of a shortacting NSAID, for example, ibuprofen, then lowering the dose once analgesic efcacy is reached may be favorable.17 However, given that ibuprofen has a ceiling analgesic effect at doses of 400 mg29 (the initial dose administered to participants in the trial by Dreiser et al12 [2 200 mg tablets]), there seems little rationale to administering doses higher than this, particularly with there being no evidence of a dose-response relationship between 400 and 800 mg in terms of clinical efcacy.29 Based on the analgesic and side effects dose-response curves for NSAIDs, higher doses are associated with an increased risk of side effects but may produce very little additional pain relief.29,37,50

practice where there is a paucity of research to inform clinical decision making. Current LBP management guidelines40,53 recommend that individuals avoid bed rest and make a gradual return to normal activity. Findings from this review have shown no favorable effect of bed rest on pain or disability, with results favoring normal activity, concordant with previous analyses that also suggest that bed rest can delay recovery from LBP.9 However, encouragement of physical activity and avoidance of bed rest in LBP are not to be considered in isolation. Guidelines40,53 have emphasized the importance for HCPs to reassure individuals of a favorable prognosis, recognizing that catastrophizing and fear avoidance among patients represent potential barriers to recovery.44 Although this is a component of rst-line care echoed in most advice trials,27,43 this review found only low-quality evidence that advice delivered by HCPs has some benet on sickness leave. The majority of advice trials delivered an intervention principally aligned with a biopsychosocial model of care.54 With this model, there is the opportunity to address any concerns or negative beliefs the individual has about his or her condition. Individuals are provided reassurance of a favorable outcome and encouraged to stay active. With all the advice trials, the intervention was compared with usual care; however, differences such as the accompaniment of written educational material may have affected the overall results. Written information may serve to reinforce verbal advice,52 although the clinical benets with this combination remain largely unclear in acute LBP. Another explanation for the lack of effect of advice may be related to variable times spent with the patient. In the pooled analysis of the advice trials shown in Figs 2 and 3,28,43,51 the length of single consultations varied from 20 minutes28 to 30 to 60 minutes,51 and also follow-up of participants varied between these studies. Furthermore, the interventions were delivered by different HCPs, physiotherapists43 in one and primary care physicians in another trial.28 Evidence suggests that patients with acute LBP are more likely to benet from comprehensive consultations,15 although this is not loosely characterized by the length of time spent with the patient but includes appropriate information gathering and quality of advice delivered. Furthermore, the training that the HCPs received may have affected the delivery of the intervention. Most of these studies made no reference to the nature or length of training received by the HCPs. Previously it had been suggested that short-term (<10 hours) training sessions may not allow HCPs to deliver adequate care,30 although more recent ndings suggest that training periods of <10 hours may be equally effective as a more intensive training program.13 There may, however, have been interhealth professional variability so that even if the training was uniformly delivered, personal delivery of the intervention by the HCP affected the overall results. Pooled estimates from 2 studies27,28 showed a statistically signicant effect of advice on sickness leave

Abdel Shaheed et al In the absence of evidence of clinical benets with higher doses of NSAIDs in acute LBP, a trial of simpler regimens is encouraged initially to avoid possible doserelated complications. Among the single interventions investigated, the greatest effect observed was on pain reduction with a topically applied rubefacient (MD 17.5 P < .001). Most other interventions produced differences too small to be clinically signicant or, in the case of NSAIDs and heat application, changes in pain and/or disability were only slightly above the minimum clinically signicant threshold of 10 points on a 0- to 100-point scale. We adopted the most lax criterion for clinically signicant changes in pain and disability6,7,14,16,20,41 suggested in the literature, and given these ndings, it is compelling to postulate that a combination of pharmacologic and nonpharmacologic interventions, for example, simple analgesic or rubefacient plus heat application, may have a synergistic effect and subsequently clinically superior outcomes; however, future research is needed to support this. It is noteworthy that acute LBP has a favorable prognosis, with up to 90% recovering within the rst month of pain onset.1 This provides a plausible explanation for the lack of effect of interventions not compared with placebo. However, the theory that the small treatment effects of interventions in trials of acute LBP could largely be attributed to the favorable natural history of this condition has been challenged in the placebocontrolled setting. Findings from a systematic review33 evaluating mean baseline and follow-up pain scores (on a 0100 point scale) from 18 placebo-controlled trials of interventions for acute LBP found that mean differences of up to 40 points can be demonstrated between placebo and treatment groups. Of the 4 placebo-controlled trials included in the analysis of this review, the smallest MD observed was 10.9 points (dicofenac 12.5 mg vs placebo). Although low, this is a clinically signicant result and represents a differ-

The Journal of Pain

13

ence of 15% from baseline, concordant with the ndings from the aforementioned review.33 This review has identied an absence of quality evidence in an important area, particularly given the enormity of OTC products marketed for LBP management and the confusion surrounding effectiveness that often arises. Our ndings provide reliable information to safeguard against elaborate or poorly substantiated claims. The strengths of this review include an extensive and highly sensitive search strategy, including OTC interventions universally available that are either indicated or marketed for use in LBP, making the ndings globally relevant. Furthermore, we excluded interventions that are no longer available; therefore, ndings have not been skewed by outdated data and are immediately relevant to current practice. The PEDro scale was used to rate the quality of individual trials as it has been found to be a reliable and valid rating tool,10,32,34 whereas limitations have been reported for the Cochrane risk of bias scale.2,24 Limitations of this study include possible publication bias, as only published studies included in peerreviewed journals were included. Furthermore, only English-language studies were included in the analysis.

Conclusion
There is limited evidence that NSAIDs, heat wrap, and rubefacients provide immediate pain relief for acute LBP and that bed rest and advice are both ineffective. There is a need for future, high-quality research to inform the rational use of OTC remedies and advice in acute LBP.

Supplementary Data
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.jpain.2013.09.016.

References
1. Andersson GB: Epidemiological features of chronic lowback pain. Lancet 354:581-585, 1999 2. Armijo-Olivo S, Stiles CR, Hagen NA, Biondo PD, Cummings GG: Assessment of study quality for systematic reviews: A comparison of the Cochrane Collaboration Risk of Bias Tool and the Effective Public Health Practice Project Quality Assessment Tool: Methodological research. J Eval Clin Pract 18:12-18, 2012 3. Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, Guyatt GH, Harbour RT, Haugh MC, Henry D, Hill S, Jaeschke R, Leng G, Liberati A, Magrini N, Mason J, Middleton P, Mrukowicz J, OConnell D, Oxman AD, Phillips B, Schunemann HJ, Edejer T, Varonen H, Vist GE, Williams JW Jr, Zaza S, GRADE Working Group: Grading quality of evidence and strength of recommendations. BMJ 328:1490, 2004 4. Berenguera A, Pujol-Ribera E, Rodriguez-Blanco T, Violan C, Casajuana M, de Kort N, Trapero-Bertran M: Study protocol of cost-effectiveness and cost-utility of a bio-

psychosocial multidisciplinary intervention in the evolution of non-specic sub-acute low back pain in the working population: Cluster randomised trial. BMC Musculoskelet Disord 12:194, 2011 5. Bijur PE, Latimer CT, Gallagher EJ: Validation of a verbally administered numerical rating scale of acute pain for use in the emergency department. Acad Emerg Med 10:390-392, 2003 6. Chou R, Huffman LH: Medications for acute and chronic low back pain: A review of the evidence for an American Pain Society/American College of Physicians clinical practice guideline. Ann Intern Med 147:505-514, 2007 7. Chou R, Huffman LH: Nonpharmacologic therapies for acute and chronic low back pain: A review of the evidence for an American Pain Society/American College of Physicians clinical practice guideline. Ann Intern Med 147:492-504, 2007 8. Coste J, Le Parc JM, Berge E, Delecoeuillerie G, Paolaggi JB: French validation of a disability rating scale for the evaluation of low back pain (EIFEL questionnaire). Rev Rhum Ed Fr 60:335-341, 1993

14

The Journal of Pain

Over-the-Counter Interventions and Advice for Acute Low Back Pain

24. Hartling L, Ospina M, Liang Y, Dryden DM, Hooton N, Krebs Seida J, Klassen TP: Risk of bias versus quality assessment of randomised controlled trials: Cross sectional study. BMJ 339:b4012, 2009 25. Higgins JPT, Green S: Cochrane Handbook for Systematic Reviews of Interventions, version 5.0.2. Cochrane Collaboration, 2009 26. Hjermstad MJ, Fayers PM, Haugen DF, Caraceni A, Hanks GW, Loge JH, Fainsinger R, Aass N, Kaasa S, European Palliative Care Research Collaborative (EPCRC): Studies comparing Numerical Rating Scales, Verbal Rating Scales, and Visual Analogue Scales for assessment of pain intensity in adults: A systematic literature review. J Pain Symptom Manage 41:1073-1093, 2011 27. Indahl A, Velund L, Reikeraas O: Good prognosis for low back pain when left untampered: A randomised clinical trial. Spine 20:473-477, 1995 28. Jellema P, Van Der Windt DAWM, Van Der Horst HE, Twisk JWR, Stalman WAB, Bouter LM: Should treatment of (sub)acute low back pain be aimed at psychosocial prognostic factors? Cluster randomised clinical trial in general practice. BMJ 331:84-87, 2005 29. Laska EM, Sunshine A, Marrero I, Olson N, Siegel C, McCormick N: The correlation between blood levels of ibuprofen and clinical analgesic response. Clin Pharmacol Ther 40:1-7, 1986 30. Lewin SA, Skea ZC, Entwistle V, Zwarenstein M, Dick J: Interventions for providers to promote a patient-centred approach in clinical consultations. Cochrane Database Syst Rev;CD003267, 2001 31. Lim S, Vos T, Flaxman A, Danaei G, Shibuya K, AdairRohani H, the Global Burden of Disease Study Group: A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 19902010: A systematic analysis for the Global Burden of Disease Study 2010. Lancet 300:2224-2260, 2012 32. Macedo LG, Elkins MR, Maher CG, Moseley AM, Herbert RD, Sherrington C: There was evidence of convergent and construct validity of Physiotherapy Evidence Database quality scale for physiotherapy trials. J Clin Epidemiol 63:920-925, 2010 33. Machado LAC, Kamper SJ, Herbert RD, Maher CG, McAuley JH: Analgesic effects of treatments for nonspecic low back pain: A meta-analysis of placebocontrolled randomized trials. Rheumatology (Oxford) 48: 520-527, 2009 34. Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M: Reliability of the PEDro scale for rating quality of randomized controlled trials. Phys Ther 83:713-721, 2003 35. Malmivaara A, Hakkinen U, Aro T, Heinrichs ML, Koskenniemi L, Kuosma E, Lappi S, Paloheimo R, Servo C, Vaaranen V, Hernberg S: The treatment of acute low back pain-bed rest, exercises, or ordinary activity? N Engl J Med 332:351-355, 1995 36. Milgrom C, Finestone A, Lev B, Wiener M, Floman Y: Overexertional lumbar and thoracic back pain among recruits: A prospective study of risk factors and treatment regimens. J Spinal Disord 6:187-193, 1993 37. Munir MA, Enany N, Zhang J-M: Nonopioid analgesics. Anesthesiol Clin 25:761-774, 2007 38. Nadler SF, Steiner DJ, Erasala GN, Hengehold DA, Abeln SB, Weingand KW: Continuous low-level heatwrap

9. Dahm KT, Brurberg KG, Jamtvedt G, Hagen KB: Advice to rest in bed versus advice to stay active for acute low-back pain and sciatica. Cochrane Database Syst Rev (6): CD007612, 2010 10. De Morton NA: The PEDro scale is a valid measure of the methodological quality of clinical trials: A demographic study. Aust J Physiother 55:129-133, 2009 11. Deyo RA, Weinstein JN: Low back pain. N Engl J Med 344: 363-370, 2001 12. Dreiser RL, Marty M, Ionescu E, Gold M, Liu JH: Relief of acute low back pain with diclofenac-K 12.5 mg tablets: A exible dose, ibuprofen 200 mg and placebo-controlled clinical trial. Int J Clin Pharmacol Ther 41:375-385, 2003 13. Dwamena F, Holmes-Rovner M, Gaulden CM, Jorgenson S, Sadigh G, Sikorskii A, Lewin S, Smith RC, Coffey J, Olomu A: Interventions for providers to promote a patient-centred approach in clinical consultations. Cochrane Database Syst Rev 12:CD003267, 2012 14. Dworkin RH, Turk DC, Wyrwich KW, Beaton D, Cleeland CS, Farrar JT, Haythornthwaite JA, Jensen MP, Kerns RD, Ader DN, Brandenburg N, Burke LB, Cella D, Chandler J, Cowan P, Dimitrova R, Dionne R, Hertz S, Jadad AR, Katz NP, Kehlet H, Kramer LD, Manning DC, McCormick C, McDermott MP, McQuay HJ, Patel S, Porter L, Quessy S, Rappaport BA, Rauschkolb C, Revicki DA, Rothman M, Schmader KE, Stacey BR, Stauffer JW, von Stein T, White RE, Witter J, Zavisic S: Interpreting the clinical importance of treatment outcomes in chronic pain clinical trials: IMMPACT recommendations. J Pain 9:105-121, 2008 15. Engers A, Jellema P, Wensing M, van der Windt DAWM, Grol R, van Tulder MW: Individual patient education for low back pain. Cochrane Database Syst Rev;CD004057, 2008 16. Farrar JT, Portenoy RK, Berlin JA, Kinman JL, Strom BL: Dening the clinically important difference in pain outcome measures. Pain 88:287-294, 2000 17. Garcia Rodriguez LA: Variability in risk of gastrointestinal complications with different nonsteroidal antiinammatory drugs. Am J Med 104:30S34S; discussion 41S42S, 1998 18. Gilbert JR, Taylor DW, Hildebrand A, Evans C: Clinical trial of common treatments for low back pain in family practice. BMJ (Clin Res Ed) 291:791-794, 1985 19. Ginsberg F, Famaey JP: A double-blind study of topical massage with Rado-Salil ointment in mechanical low-back pain. J Int Med Res 15:148-153, 1987 20. Giraudeau B, Rozenberg S, Valat JP: Assessment of the clinically relevant change in pain for patients with sciatica [Letter]. Ann Rheum Dis 63:1180-1181, 2004 21. Goubert L, Crombez G, De Bourdeaudhuij I: Low back pain, disability and back pain myths in a community sample: Prevalence and interrelationships. Eur J Pain 8:385-394, 2004 22. Hancock MJ, Maher CG, Latimer J, McLachlan AJ, Cooper CW, Day RO, Spindler MF, McAuley JH: Assessment of diclofenac or spinal manipulative therapy, or both, in addition to recommended rst-line treatment for acute low back pain: A randomised controlled trial. Lancet 370: 1638-1643, 2007 23. Hart LG, Deyo RA, Cherkin DC: Physician ofce visits for low back pain. Frequency, clinical evaluation, and treatment patterns from a U.S. national survey. Spine 20:11-19, 1995

Abdel Shaheed et al
therapy for treating acute nonspecic low back pain. Arch Phys Med Rehabil 84:329-334, 2003 39. Nadler SF, Steiner DJ, Petty SR, Erasala GN, Hengehold DA, Weingand KW: Overnight use of continuous low-level heatwrap therapy for relief of low back pain. Arch Phys Med Rehabil 84:335-342, 2003 40. NHMRC: Evidence-based management of acute musculoskeletal pain: A guide for clinicians. In: NHMRC, editor. Canberra, 2004 41. Ostelo RW, Deyo RA, Stratford P, Waddell G, Croft P, Von Korff M, Bouter LM, de Vet HC: Interpreting change scores for pain and functional status in low back pain: Towards international consensus regarding minimal important change. Spine (Phila Pa 1976) 33:90-94, 2008 42. PEDro: Physiotherapy evidence database. How are trials rated? Available at: http://www.pedro.org.au/english/faq/ #question_ve. Accessed July 2013 43. Pengel LH, Refshauge KM, Maher CG, Nicholas MK, Herbert RD, McNair P: Physiotherapist-directed exercise, advice, or both for subacute low back pain: A randomized trial. Ann Intern Med 146:787-796, 2007 44. Pincus T, Burton AK, Vogel S, Field AP: A systematic review of psychological factors as predictors of chronicity/ disability in prospective cohorts of low back pain. Spine 27:E109-E120, 2002 45. Pinto RZ, Maher CG, Ferreira ML, Ferreira PH, Hancock M, Oliveira VC, McLachlan AJ, Koes B: Drugs for relief of pain in patients with sciatica: Systematic review and meta-analysis. BMJ 344:e497, 2012 46. Pinto RZ, Maher CG, Ferreira ML, Hancock M, Oliveira VC, McLachlan AJ, Koes B, Ferreira PH: Epidural corticosteroid injections in the management of sciatica: A systematic review and meta-analysis. Ann Intern Med 157: 865-877, 2012 47. Report of the Consumer Health Care Products Associations Clinical/Medical Committee: White paper on the benets of OTC medicines in the United States. Pharmacy Today October 2010, pp 68-79 48. Roland M, Fairbank J: The Roland-Morris Disability Questionnaire and the Oswestry Disability Questionnaire. Spine (Phila Pa 1976) 25:3115-3124, 2000

The Journal of Pain

15

49. Rozenberg S, Delval C, Rezvani Y, Olivieri-Apicella N, Kuntz J-L, Legrand E, Valat JP, Blotman F, Meadeb J, Rolland D, Hary S, Duplan B, Feldmann JL, Bourgeois P: Bed rest or normal activity for patients with acute low back pain: A randomized controlled trial. Spine 27: 1487-1493, 2002 50. Sostres C, Gargallo CJ, Arroyo MT, Lanas A: Adverse effects of non-steroidal anti-inammatory drugs (NSAIDs, aspirin and coxibs) on upper gastrointestinal tract. Baillieres Best Pract Res Clin Gastroenterol 24:121-132, 2010 51. Storheim K, Brox JI, Holm I, Koller AK, Bo K: Intensive group training versus cognitive intervention in sub-acute low back pain: Short-term results of a singleblind randomized controlled trial. J Rehabil Med 35: 132-140, 2003 52. Symptomatic treatment for acute low back pain. NPS MedicineWise. Available at: http://www.nps.org.au/ conditions-and-topics/conditions/nervous-system-problems/ pain/for-individuals/pain-conditions/low-back-pain/backpain-acute-low/symptomatic-management-pad. Accessed February 2013 53. Van Tulder M, Becker A, Bekkering T, Breen A, Del Real MTG, Hutchinson A, Koes B, Laerum E, Malmivaara A: Chapter 3: European guidelines for the management of acute nonspecic low back pain in primary care. Eur Spine J 15:S169-S191, 2006 54. Waddell G: 1987 Volvo award in clinical sciences. A new clinical model for the treatment of low-back pain. Spine 12: 632-644, 1987 55. Walker BF: The prevalence of low back pain: A systematic review of the literature from 1966 to 1998. J Spinal Disord 13:205-217, 2000 56. Walker BF, Muller R, Grant WD: Low back pain in Australian adults: The economic burden. Asia-Pacic J Public Health 15:79-87, 2003 57. Wilk V, Palmer HD, Stosic RG, McLachlan AJ: Evidence and practice in the self-management of low back pain: Findings from an Australian internet-based survey. Clin J Pain 26: 533-540, 2010 58. Wilkinson MJ: Does 48 hours bed rest inuence the outcome of acute low back pain? Br J Gen Pract 45: 481-484, 1995