https://orcid.org/0000-0002-2013-7624
Correction
18 Jul 2024: Furlong B, Etchegary H, Aubrey-Bassler K, Swab M, Pike A, et al. (2024) Correction: Patient education materials for non-specific low back pain and sciatica: A systematic review and meta-analysis. PLOS ONE 19(7): e0307675. https://doi.org/10.1371/journal.pone.0307675 View correction
Figures
Abstract
Introduction
Guidelines recommend patient education materials (PEMs) for low back pain (LBP), but no systematic review has assessed PEMs on their own. We investigated the effectiveness of PEMs on process, clinical, and health system outcomes for LBP and sciatica.
Methods
Systematic searches were performed in MEDLINE, EMBASE, CINAHL, PsycINFO, SPORTDiscus, trial registries and grey literature through OpenGrey. We included randomized controlled trials of PEMs for LBP. Data extraction, risk of bias, and quality of evidence gradings were performed independently by two reviewers. Standardized mean differences or risk ratios and 95% confidence intervals were calculated, and effect sizes pooled using random-effects models. Analyses of acute/subacute LBP were performed separately from chronic LBP at immediate, short, medium, and long-term (6, 12, 24, and 52 weeks, respectively).
Results
27 studies were identified. Compared to usual care for chronic LBP, we found moderate to low-quality evidence that PEMs improved pain intensity at immediate (SMD = -0.16 [95% CI: -0.29, -0.03]), short (SMD = -0.44 [95% CI: -0.88, 0.00]), medium (SMD = -0.53 [95% CI: -1.01, -0.05]), and long-term (SMD = -0.21 [95% CI: -0.41, -0.01]), medium-term disability (SMD = -0.32 [95% CI: -0.61, -0.03]), quality of life at short (SMD = -0.17 [95% CI: -0.30, -0.04]) and medium-term (SMD = -0.23 [95% CI: -0.41, -0.04]) and very low-quality evidence that PEMs improved global improvement ratings at immediate (SMD = -0.40 [95% CI: -0.58, -0.21]), short (SMD = -0.42 [95% CI: -0.60, -0.24]), medium (SMD = -0.46 [95% CI: -0.65, -0.28]), and long-term (SMD = -0.43 [95% CI: -0.61, -0.24]). We found very low-quality evidence that PEMs improved pain self-efficacy at immediate (SMD = -0.21 [95% CI: -0.39, -0.03]), short (SMD = -0.25 [95% CI: -0.43, -0.06]), medium (SMD = -0.23 [95% CI: -0.41, -0.05]), and long-term (SMD = -0.32 [95% CI: -0.50, -0.13]), and reduced medium-term fear-avoidance beliefs (SMD = -0.24 [95% CI: -0.43, -0.06]) and long-term stress (SMD = -0.21 [95% CI: -0.39, -0.03]). Compared to usual care for acute LBP, we found high to moderate-quality evidence that PEMs improved short-term pain intensity (SMD = -0.24 [95% CI: -0.42, -0.06]) and immediate-term quality of life (SMD = -0.24 [95% CI: -0.42, -0.07]). We found low to very low-quality evidence that PEMs increased knowledge at immediate (SMD = -0.51 [95% CI: -0.72, -0.31]), short (SMD = -0.48 [95% CI: -0.90, -0.05]), and long-term (RR = 1.28 [95% CI: 1.10, 1.49]) and pain self-efficacy at short (SMD = -0.78 [95% CI: -0.98, -0.58]) and long-term (SMD = -0.32 [95% CI: -0.52, -0.12]). We found moderate to very low-quality evidence that PEMs reduced short-term days off work (SMD = -0.35 [95% CI: -0.63, -0.08]), long-term imaging referrals (RR = 0.60 [95% CI: 0.41, 0.89]), and long-term physician visits (SMD = -0.16 [95% CI: -0.26, -0.05]). Compared to other interventions (e.g., yoga, Pilates), PEMs had no effect or were less effective for acute/subacute and chronic LBP.
Conclusions
There was a high degree of variability across outcomes and time points, but providing PEMs appears favorable to usual care as we observed many small, positive patient and system impacts for acute/subacute and chronic LBP. PEMs were generally less effective than other interventions; however, no cost effectiveness analyses were performed to weigh the relative benefits of these interventions to the likely less costly PEMs.
Citation: Furlong B, Etchegary H, Aubrey-Bassler K, Swab M, Pike A, Hall A (2022) Patient education materials for non-specific low back pain and sciatica: A systematic review and meta-analysis. PLoS ONE 17(10): e0274527. https://doi.org/10.1371/journal.pone.0274527
Editor: Tariq Jamal Siddiqi, The University of Mississippi Medical Center, UNITED STATES
Received: June 4, 2022; Accepted: August 28, 2022; Published: October 12, 2022
Copyright: © 2022 Furlong et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data are provided in the article and the Supporting information files.
Funding: This study is supported by the Office of Research and Graduate Studies (Medicine) at Memorial University of Newfoundland and the Canadian Institute of Health Research, grant number 398 527. The funding bodies played no role in the design of the study; collection, analysis, and interpretation of data; or in writing the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Low back pain (LBP) accounts for more disability than any other musculoskeletal condition [1] and is among the five most common reasons why patients visit their family physicians [2]. It represents a substantial economic burden resulting from both direct (e.g., health care costs) and indirect costs [3] (e.g., productivity loss and compensation claims) [4, 5].
International, evidence-based guidelines for the treatment and management of LBP [6–10] recommend that for non-specific LBP (LBP that is not attributable to a recognizable, specific pathology) [11] investigations such as imaging are not required. Instead, they recommend that management should include reassurance, simple analgesics, self-care strategies, and advice and education. Patient education materials (PEMs) for LBP are intended to transfer accurate knowledge about diagnosis, prognosis, and ways to manage pain and aid recovery in order to correct false/unhelpful beliefs, reassure patients about prognosis, and manage their expectations of recovery. We hypothesized that by modifying beliefs and expectations, PEMs may reduce fear or concern related to pain, modify patients’ experience of pain and expectation for unnecessary tests or other referrals, and increase patients’ self-efficacy to engage in recommended strategies to manage pain which should facilitate recovery.
Indeed, Lim et al. [12] recently showed that people living with LBP want education–specifically, clear and consistent information about their LBP presented in language they can follow that includes self-management strategies and treatment options. Other systematic reviews have assessed patient education for LBP [13–25] as discussed in our protocol [26]. The most relevant review was published in 2008 [27], but variation in the education interventions of the 24 studies precluded meta-analysis limiting our understanding of the effectiveness of PEMs. Subsequent reviews have focused on clinical outcomes or broader interventions and therefore, none have fully assessed outcomes that would test our hypothesis.
This is the first systematic review and meta-analysis to investigate the effect of PEMs alone on a comprehensive set of outcomes for non-specific LBP and sciatica. The primary aim of this review is to provide up-to-date evidence on the effectiveness of these materials on immediate process outcomes such as knowledge, attitudes, and fear-avoidance beliefs; clinical outcomes such as pain and physical disability; and health system outcomes such as healthcare utilization and cost effectiveness in patients with acute and chronic non-specific LBP or sciatica.
Methods
We published our protocol for this systematic review and meta-analysis [26].
Search strategy
A professional librarian adapted the search strategy (S1 File) used by Engers et al. [27] which was later peer-reviewed following the Peer Review of Electronic Search Strategies (PRESS) guidelines [28]. They searched MEDLINE, EMBASE, CINAHL, PsycINFO, and SPORTDiscus from inception to March 24, 2022, as well as trial registries and grey literature using OpenGrey.
Study selection
Results from the electronic database search were de-duplicated in Endnote [29] and imported to Covidence systematic review software [30]. Google translate was used for all non-English articles and study authors were contacted for clarification if needed. Title and abstract and full-text review were conducted by two reviewers (BF, one of GD, AS, SG; see acknowledgements) using a screening form that included pre-specified inclusion and exclusion criteria (S2 File); conflicts were resolved by a third reviewer (AH). Reference lists of relevant studies were hand-searched, and authors of conference abstracts or ongoing trials were contacted to identify additional studies. If a paper related to a study identified in a conference abstract could not be found, it was excluded.
Data extraction
Two reviewers (BF, one of AS, SG; see acknowledgements) independently extracted data for all studies using standardized data extraction forms in Microsoft Excel, and conflicts were resolved by a third reviewer (AH). Data items included study information (authors, year of publication, country of data collection, LBP type and duration, sample size, outcome measures, study design, intervention group description, comparison group description), intervention details using the 12 variables in the TIDieR checklist [31] and outcome information (measurement tools, measurement scales, scoring methods and interpretation, means, and standard deviations).
Risk of bias assessment
Risk of bias was assessed using the PEDro scale [32]. A study was at high risk of bias if 0–3 criteria on the scale were satisfied, moderate if 4–6 criteria were satisfied, and low if 7–10 criteria were satisfied. However, if randomization was not appropriate (e.g., quasi-randomization) or there was less than 85% follow-up, the study was considered to be at high risk of bias. PEDro scores were extracted from the PEDro database if available; otherwise, two reviewers (BF, AH) independently assessed risk of bias for each study. Conflicts were discussed and, if necessary, reviewed with a third author (AP) to reach consensus.
Data synthesis
We included the following contrasts:
- PEMs alone vs. no intervention
- PEMs alone vs. another intervention
- PEMs + another intervention vs. the same intervention without PEMs
Analyses were conducted separately for acute/sub-acute (pain<12 weeks) and chronic (pain≥ 12 weeks) populations for all outcomes at immediate, short, medium, and long-term (defined as the closest follow-up time point to 6, 12, 24 and 52 weeks, respectively). For immediate-term follow-up only, if a study measured more than once during our defined timeframe (e.g., at both 2 weeks and 6 weeks), we chose the closest follow-up measure after the intervention was provided to get a more accurate depiction of the intervention’s “immediate” effect. For other time points, if a study measured more than once within our specified timeframe, we chose the time point closest to 12, 24, or 52 weeks.
Effectiveness analysis.
Point estimates of effect size and 95% confidence intervals were used to estimate the treatment effect. Review Manager (RevMan) 5.4.1 (The Cochrane Collaboration) was used for the analysis [33]. Since different measurement tools were used for each outcome, we used the standardized mean difference for all analyses of continuous outcomes. Risk ratios were used for dichotomous outcomes. Where outcome data from multiple studies was pooled but the measurement scales pointed in different directions (e.g., one scale increased with disease severity while the others did not), we multiplied the point estimates by –1 to reverse the direction as described in the Cochrane handbook [34]. Where data for the same outcome were reported continuously and dichotomously between studies, we transformed dichotomous data into the SMD where possible using the methods described in the Cochrane handbook [35]. to allow for pooling of treatment effects. Otherwise, SMD and RR were reported separately. A random-effects model was used for each contrast since variation between each intervention was likely. We pooled the results if the participants, interventions, and outcomes were sufficiently homogenous, allowing for a small degree of clinical heterogeneity in the types of PEMs (e.g., content or delivery of the intervention) and populations assessed (e.g., duration of low back pain). If I2 > 75%, which represents potential for considerable statistical heterogeneity [36], we investigated both the level of clinical heterogeneity as well as the magnitude and direction of the differences in effect sizes across studies to determine if it remained reasonable to pool the results.
Certainty of the evidence.
To assess the level of certainty of the evidence, a summary of findings table was developed for each outcome using the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) approach [37]. GRADE was assessed independently by two reviewers (BF, AH); our process for downgrading each of the five domains can be found in our published protocol [26] and in S2 File. Conflicts were discussed and, if necessary, reviewed with a third author (AP) to reach consensus.
Sensitivity and subgroup analyses.
Our primary analyses included all studies, but we excluded studies judged to be at high risk of bias due to concerns about the randomization process in a sensitivity analysis to determine if these studies influenced the results.
Missing data.
In cases where only the between group mean difference was provided in a study and we could not obtain the individual group summary data from the study’s authors, we used the generic inverse variance method to pool this data with that of the other studies [38]. A more complete explanation of missing data treatment is described in our protocol [26].
Protocol deviations
We made minor deviations (further described in S2 File) to our published protocol [26]. Of note, due to small number of studies with physician-provided PEMs, we expanded our criteria to include studies where a member of the study’s research team was responsible for providing the PEMs.
Results
Description of included trials (Table 1)
Of the 6435 unique records identified, 537 full texts were reviewed, and 27 included in the review (Fig 1). Most trials were conducted in the United States [39–48], followed by three in the United Kingdom [49–51], two each in Spain [52, 53], Sweden [54, 55], and Thailand [56, 57], and one each in Australia [58], Croatia [59], Finland [60], Germany [61], Iran [62], the Netherlands [63], and New Zealand [64]. One trial was conducted in both Denmark and Norway [65]. There were 21 RCTs [39–49, 51–56, 58, 59, 61, 65] and six cluster RCTs [50, 57, 60, 62–64], and participants were recruited largely through primary care [41, 42, 45–53, 59–61, 63–65]. Twelve trials included participants with acute LBP [39, 41, 49–51, 54, 55, 57, 60, 61, 63, 64] and 15 with chronic LBP [40, 42–48, 52, 53, 56, 58, 59, 62, 65]. PEMs interventions were compared to usual care in 14 studies [39, 48–51, 53, 55, 58, 60–65] and other interventions in 13 studies including Pilates [52], Yoga [45–47, 59], exercise [57], stretching [40], proprioceptive neuromuscular facilitation [56], massage [42], walking [44], chiropractic manipulation [41], and cognitive behavioral therapy [43, 54].
Description of the interventions using the TIDieR checklist (Table 2)
PEMs were provided by physicians [48–51, 59, 61, 63, 64] or researchers [39–47, 52–58, 62, 65] via a hard copy booklet, leaflet or pamphlet [39–42, 44–52, 54, 56, 57, 59–61, 63, 64] with several newer studies using digital formats [43, 53, 55, 58, 62, 65]. PEMs content was similar across studies and included anatomy, causes of LBP, posture and movement, proper lifting techniques, exercises, how to manage flare-ups, pain management, importance of staying active, self-management strategies, and treatment options. Six studies intended to and/or measured delivery of the PEMs to the patient by audio-recording GP consultations [64], asking participants if they read the materials [42, 46, 54, 63] or recording participant activity in a mobile application [65].
Risk of bias (Table 3)
10 studies had high risk of bias [39, 40, 49, 51, 56, 58–62], eight had moderate risk of bias [43, 44, 50, 53, 55, 57, 63, 64], and nine had low risk of bias [41, 42, 45–48, 52, 54, 65]. The most common source of bias was lack of blinding. Due to the nature of the intervention, none of the 27 included studies satisfied the criteria for blinding of subjects or providers and only nine of 27 studies reported blinding of outcome assessors. Nine of 10 high risk of bias studies [39, 40, 49, 56, 58–62] were the result of insufficient follow-up. Only one of six cluster RCTs [50, 57, 60, 62–64] adequately reported adjusting for clustering [60].
Effectiveness of patient education materials for acute/subacute LBP
Patient education materials alone vs. no intervention or usual care.
Nine trials [39, 49–51, 55, 60, 61, 63, 64] compared the effect of PEMs to usual care on LBP-related outcomes for acute/subacute LBP patients. In the usual care arm, patients could carry on with any LBP care as they normally would outside of the study. In one study [61], the usual care group also received a booklet with information unrelated to LBP as a control intervention. The most commonly measured outcome was disability (n = 8), followed by measures of pain intensity (n = 5), pain self-efficacy (n = 4), knowledge (n = 4), quality of life (n = 4), fear-avoidance beliefs (n = 3), catastrophizing (n = 3), anxiety (n = 3), days off work (n = 3), and physician visits (n = 3). Single studies measured global improvement, cost, imaging, and referrals. No studies measured function, general beliefs, attitudes, coping, stress, or depression. A summary of findings for eight key outcomes are presented in Table 4 (a summary of all other outcomes and forest plots for all analyses are presented in S3 and S4 Files, respectively).
Pain intensity (n = 5). We found high-quality evidence that PEMs were significantly more effective for reducing pain intensity compared to usual care at short-term (4 RCTs, n = 1101; SMD = -0.24; 95% CI: -0.42, -0.06; p = 0.01; I2 = 55%). We found high-quality evidence that PEMs had no effect on pain intensity compared to usual care at immediate (3 RCTs, n = 910; SMD = -0.13; 95% CI: -0.27, 0.01; p = 0.07; I2 = 14%) and medium-term (2 RCTs, n = 515; SMD = -0.03 95% CI: -0.20, 0.15; p = 0.77; I2 = 0%), and moderate-quality evidence of no effect at long-term (3 RCTs, n = 892; SMD = -0.11; 95% CI: -0.24, 0.02; p = 0.11; I2 = 0%).
Disability (n = 8). We found high-quality evidence that PEMs had no effect on disability compared to usual care at immediate (6 RCTs, n = 1220; SMD = -0.05; 95% CI: -0.17, 0.06; p = 0.35; I2 = 0%), short (6 RCTs, n = 1272; SMD = -0.06; 95% CI: -0.18, 0.05; p = 0.30; I2 = 7%), and medium-term (3 RCTs, n = 563; SMD = 0.09; 95% CI: -0.08, 0.27; p = 0.31; I2 = 6%) and moderate-quality evidence of no effect at long-term (4 RCTs, n = 938; SMD = -0.09; 95% CI: -0.27, 0.08; p = 0.28; I2 = 37%).
Quality of life (n = 4). We found moderate-quality evidence that PEMs are significantly more effective than usual care for improving quality of life at immediate-term (2 RCTs, n = 524; SMD = -0.24; 95% CI: -0.42, -0.07; p = 0.006; I2 = 0%). We found high-quality evidence that PEMs had no effect on quality of life compared to usual care at short-term (3 RCTs, n = 804; SMD = -0.20; 95% CI: -0.43, 0.03; p = 0.09; I2 = 58%). We found very low-quality evidence of no effect at medium-term (1 RCT, n = 286; SMD = 0.00; 95% CI: -0.23, 0.23; p = 1.00) and moderate-quality evidence of no effect at long-term (2 RCTs, n = 470; SMD = 0.01; 95% CI: -0.17, 0.19; p = 0.94; I2 = 0%).
Global improvement (n = 1). We found very low-quality evidence that PEMs had no effect compared to usual care on global improvement at immediate (1 RCT, n = 305; RR = 1.07; 95% CI: 0.80, 1.43; p = 0.64), short (1 RCT, n = 305; RR = 1.03; 95% CI: 0.75, 1.42; p = 0.85), medium (1 RCT, n = 299; RR = 1.05; 95% CI: 0.75, 1.47; p = 0.76), and long-term (1 RCT, n = 288; RR = 1.15; 95% CI: 0.81, 1.65; p = 0.43), where RR > 1 favors usual care.
Knowledge (n = 5). We found low-quality evidence that PEMs are significantly more effective than usual care for improving knowledge in the immediate (4 RCTs, n = 699; SMD = -0.51; 95% CI: -0.72, -0.31; p < 0.00001; I2 = 47%) and short-term (2 RCTs, n = 502; SMD = -0.48; 95% CI: -0.90, -0.05; p = 0.03; I2 = 71%). We found very low-quality evidence that PEMs are significantly more effective than usual care for improving long-term knowledge (1 RCT, n = 777; RR = 1.28; 95% CI: 1.10, 1.49; p = 0.001).
Pain self-efficacy (n = 4). We found moderate quality evidence that PEMs had no effect on pain self-efficacy compared to usual care at immediate-term (3 RCTs, n = 650; SMD = -0.28; 95% CI: -0.63, 0.07; p = 0.12; I2 = 73%). We found very low-quality evidence that PEMs are significantly more effective than usual care for improving self-efficacy at short (1 RCT, n = 398; SMD = -0.78; 95% CI: -0.98, -0.58; p < 0.00001) and long-term (1 RCT, n = 421; SMD = -0.32; 95% CI: -0.52, -0.12; p = 0.002).
Fear-avoidance beliefs (n = 3). We found high quality evidence that PEMs had no effect on fear-avoidance beliefs compared to usual care at immediate-term (3 RCTs, n = 611; SMD = -0.14; 95% CI: -0.36, 0.09; p = 0.23; I2 = 44%), and very low-quality evidence of no effect at short (1 RCT, n = 114; SMD = 0.00; 95% CI: -0.38, 0.38; p = 1.00) and long-term (1 RCT, n = 150; SMD = 0.10; 95% CI: -0.15, 0.35; p = 0.43).
Catastrophizing (n = 3). We found high quality evidence that PEMs had no effect on catastrophizing compared to usual care at immediate-term (3 RCTs, n = 879; SMD = -0.01; 95% CI: -0.22, 0.20; p = 0.92; I2 = 60%), and very low-quality evidence of no effect at short (1 RCT, n = 398; SMD = -0.12; 95% CI: -0.31, 0.07; p = 0.22) and long-term (1 RCT, n = 248; SMD = 0.07; 95% CI: -0.18, 0.32; p = 0.58).
Anxiety (n = 3). We found moderate-quality evidence that PEMs had no effect on anxiety compared to usual care at immediate-term (2 RCTs, n = 485; SMD = -0.01; 95% CI: -0.45, 0.43; p = 0.98; I2 = 83%) and low-quality evidence of no effect at long-term (2 RCTs, n = 673; SMD = -0.13; 95% CI: -0.52, 0.26; p = 0.53; I2 = 85%).
Days off work (n = 3). We found low-quality evidence that PEMs were significantly more effective for reducing days off work compared to usual care at short-term (2 RCTs, n = 612; SMD = -0.35; 95% CI: -0.63, -0.08; p = 0.01; I2 = 22%). We found very low-quality evidence that PEMs had no effect on days off work compared to usual care at immediate (1 RCT, n = 248; RR = 0.83; 95% CI: 0.49, 1.42; p = 0.50) and medium-term (1 RCT, n = 244; RR = 0.33; 95% CI: 0.10, 1.16; p = 0.08) and moderate-quality evidence of no effect at long-term (3 RCTs, n = 1535; SMD = -0.10; 95% CI: -0.32, 0.12; p = 0.37; I2 = 62%). Sensitivity analysis for long-term follow-up revealed no difference when removing one study [51] due to concerns about their randomization method (SMD = -0.23; 95% CI: -0.46, 0.00; p = 0.05; I2 = 11%).
Imaging (n = 1). We found very low-quality evidence that PEMs are significantly more effective for reducing imaging for LBP compared to usual care at long-term (1 RCT, n = 364; RR = 0.60; 95% CI: 0.41, 0.89; p = 0.01). We found very low-quality evidence that PEMs had no effect on imaging compared to usual care at short-term (1 RCT, n = 364; RR = 0.64; 95% CI: 0.38, 1.09; p = 0.10).
Physician visits (n = 3). We found moderate-quality evidence that PEMs are significantly more effective for reducing physician visits compared to usual care at long-term (3 RCTs, n = 1721; SMD = -0.16; 95% CI: -0.26, -0.05; p = 0.003; I2 = 0%). We found very low-quality evidence of no effect at short-term (1 RCT, n = 364; SMD = -0.07; 95% CI: -0.27, 0.13; p = 0.49). Sensitivity analysis for long-term follow-up revealed no difference when removing one study [51] due to concerns about their randomization method (SMD = -0.16; 95% CI: -0.31, -0.02; p = 0.03; I2 = 0%).
Referrals (n = 1). We found very low-quality evidence that PEMs are significantly more effective than usual care for reducing specialist referrals at long-term (1 RCT; n = 936; RR = 0.85; 95% CI: 0.58, 1.23; p = 0.38).
Cost (n = 1). We found very low-quality evidence that PEMs had no effect on cost compared to usual care at medium-term (1 RCT, n = 226; SMD = -0.11; 95% CI: -0.37, 0.16; p = 0.43).
Patient education materials alone vs. other interventions.
Three trials [41, 54, 57] compared the effect of PEMs to other interventions on LBP-related outcomes for acute/subacute LBP patients. The comparator interventions were cognitive behavioural therapy [54], chiropractic manipulation [41], and an exercise program [57]. The studies included measures of pain intensity (n = 3), disability (n = 3), and days off work (n = 2), and one study measured fear-avoidance beliefs, catastrophizing, anxiety, depression, and physician visits. No studies measured quality of life, global improvement, function, knowledge, self-efficacy, attitudes, general beliefs, coping, stress, imaging, referrals, or cost. A summary of findings for eight key outcomes are presented in Table 5 (a summary of all other outcomes and forest plots for all analyses are presented in S3 and S4 Files, respectively).
Pain intensity (n = 3). We found very low-quality evidence that PEMs are more effective for reducing pain intensity compared to other interventions at medium-term (1 RCT, n = 31; SMD = -0.89; 95% CI: -1.66, -0.11; p = 0.02). We found very low-quality evidence that PEMs are less effective than other interventions at immediate-term (1 RCT, n = 178; SMD = 0.51; 95% CI: 0.20, 0.83; p = 0.001), low-quality evidence that PEMs have no effect on pain intensity when compared to other interventions at short-term (2 RCTs, n = 212; SMD = 0.07; 95% CI: -0.81, 0.95; p = 0.88; I2 = 79%), and very low-quality evidence of no effect at long-term (1 RCT, n = 155; SMD = 0.04; 95% CI: -0.28, 0.36; p = 0.81).
Disability (n = 3). We found very low-quality evidence that PEMs had no effect on disability compared to other interventions at immediate (1 RCT, n = 178; SMD = 0.27; 95% CI: -0.04, 0.58; p = 0.09) and medium-term (1 RCT, n = 31; SMD = -0.15; 95% CI: -0.88, 0.58; p = 0.69), moderate-quality evidence of no effect at short-term (2 RCTs, n = 212; SMD = 0.23; 95% CI: -0.06, 0.51; p = 0.12; I2 = 0%), and low-quality evidence of no effect at long-term (2 RCTs, n = 343; SMD = 0.20; 95% CI: -0.04, 0.43; p = 0.10; I2 = 0%).
Fear-avoidance beliefs (n = 1). We found very low-quality evidence that PEMs had no effect on fear-avoidance beliefs compared to other interventions at long-term (1 RCT, n = 155; SMD = 0.17; 95% CI: -0.16, 0.49; p = 0.31).
Catastrophizing (n = 1). We found very low-quality evidence that PEMs had no effect on catastrophizing compared to other interventions at long-term (1 RCT, n = 155; SMD = -0.06; 95% CI: -0.38, 0.27; p = 0.73).
Anxiety (n = 1). We found very low-quality evidence that PEMs had no effect on anxiety compared to other interventions at long-term (1 RCT, n = 155; SMD = -0.05; 95% CI: -0.37, 0.27; p = 0.74).
Depression (n = 1). We found very low-quality evidence that PEMs had no effect on depression compared to other interventions at long-term (1 RCT, n = 155; SMD = 0.00; 95% CI: -0.32, 0.32; p = 1.00).
Days off work (n = 2). We found low-quality evidence that PEMs are significantly less effective than other interventions for reducing days off work at long-term (2 RCTs, n = 343; SMD = 0.36; 95% CI: 0.09, 0.63; p = 0.01; I2 = 0%).
Physician visits (n = 1). We found very low-quality evidence that PEMs were less effective than other interventions on reducing physician visits (1 RCT, n = 155; SMD = 0.53; 95% CI: 0.20, 0.85; p = 0.002) at long-term.
Effectiveness of patient education materials for chronic LBP
Patient education materials alone vs. no intervention or usual care.
Five trials [48, 53, 58, 62, 65] compared the effect of PEMs to usual care on LBP-related outcomes for chronic LBP patients. A protocol for usual care was not described in four of these studies; rather, patients could continue any LBP care as they normally would outside of the study. In one study [58], the comparator group was unguided internet use where participants were asked to seek out information about LBP on their own; we considered this similar to usual care. Outcomes measured included pain intensity (n = 5), disability (n = 5), quality of life (n = 4), fear-avoidance beliefs (n = 2), and one study measured global improvement, self-efficacy, stress, and depression. No studies measured function, knowledge, attitudes, general beliefs, catastrophizing, coping, anxiety, days off work, imaging, physician visits, referrals, or cost. A summary of findings for eight key outcomes are presented in Table 6 (a summary of all other outcomes and forest plots for all analyses are presented in S3 and S4 Files, respectively).
Pain intensity (n = 5). We found moderate-quality evidence that PEMs were significantly more effective for reducing pain intensity compared to usual care at immediate (4 RCTs, n = 890; SMD = -0.16; 95% CI: -0.29, -0.03; p = 0.02; I2 = 0%) and long-term (2 RCTs, n = 757; SMD = -0.21; 95% CI: -0.41, -0.01; p = 0.04; I2 = 47%), and low-quality evidence of the same observation at short (4 RCTs, n = 925; SMD = -0.44; 95% CI: -0.88, 0.00; p = 0.05; I2 = 89%) and medium-term (4 RCTs, n = 907; SMD = -0.53; 95% CI: -1.01, -0.05; p = 0.03; I2 = 90%).
Disability (n = 5). We found moderate-quality evidence that PEMs are significantly more effective for reducing disability compared to usual care at medium-term (4 RCTs, n = 939; SMD = -0.32; 95% CI: -0.61, -0.03; p = 0.03; I2 = 74%). We found moderate-quality evidence of no effect at immediate (4 RCTs, n = 919; SMD = -0.12; 95% CI: -0.31, 0.07; p = 0.23; I2 = 38%), short (4 RCTs, n = 964; SMD = -0.23; 95% CI: -0.48, 0.03; p = 0.08; I2 = 68%), and long-term (2 RCT, n = 770; SMD = -0.12; 95% CI: -0.27, 0.02; p = 0.09; I2 = 0%).
Quality of life (n = 4). We found moderate-quality evidence that PEMs are significantly more effective for increasing quality of life compared to usual care at short (4 RCTs, n = 934; SMD = -0.15; 95% CI: -0.28, -0.03; p = 0.02; I2 = 0%) and medium-term (4 RCT, n = 902; SMD = -0.23; 95% CI: -0.41, -0.04; p = 0.02; I2 = 39%). We found moderate-quality evidence of no effect at immediate (3 RCT, n = 839; SMD = -0.04; 95% CI: -0.18, 0.09; p = 0.55; I2 = 0%) and long-term (2 RCT, n = 748; SMD = -0.13; 95% CI: -0.28, 0.01; p = 0.07; I2 = 0%).
Global improvement (n = 1). We found very low-quality evidence that PEMs were significantly more effective at increasing global improvement ratings compared to usual care at immediate (1 RCT, n = 461; SMD = -0.40; 95% CI: -0.58, -0.21; p < 0.0001), short (1 RCT, n = 461; SMD = -0.42; 95% CI: -0.60, -0.24; p < 0.00001), medium (1 RCT, n = 461; SMD = -0.46; 95% CI: -0.65, -0.28; p < 0.00001), and long-term (1 RCT, n = 461; SMD = -0.43; 95% CI: -0.61, -0.24; p < 0.00001).
Self-efficacy (n = 1). We found very low-quality evidence that PEMs were significantly more effective at increasing self-efficacy compared to usual care at immediate (1 RCT, n = 461; SMD = -0.21; 95% CI: -0.39, -0.03; p = 0.02), short (1 RCT, n = 461; SMD = -0.25; 95% CI: -0.43, -0.06; p = 0.009), medium (1 RCT, n = 461; SMD = -0.23; 95% CI: -0.41, -0.05; p = 0.01), and long-term (1 RCT, n = 461; SMD = -0.32; 95% CI: -0.50, -0.13; p = 0.0007).
Fear-avoidance beliefs (n = 2). We found very low-quality evidence that PEMs were significantly more effective for reducing fear-avoidance beliefs compared to usual care at medium-term (1 RCT, n = 461; SMD = -0.24; 95% CI: -0.43, -0.06; p = 0.01). We found high-quality evidence that PEMs had no effect on fear-avoidance beliefs compared to usual care at immediate-term (2 RCTs, n = 505; SMD = -0.15; 95% CI: -0.33, 0.02; p = 0.09; I2 = 0%), and very low-quality evidence of no effect at short (1 RCT, n = 461; SMD = -0.09; 95% CI: -0.27, 0.09; p = 0.33) and long-term (1 RCT, n = 461; SMD = -0.16; 95% CI: -0.34, 0.02; p = 0.08).
Stress (n = 1). We found very low-quality evidence that PEMs were significantly more effective at decreasing stress compared to usual care at long-term (1 RCT, n = 461; SMD = -0.21; 95% CI: -0.39, -0.03; p = 0.02). We found very low-quality evidence that PEMs had no effect on stress compared to usual care at immediate (1 RCT, n = 461; SMD = -0.13; 95% CI: -0.32, 0.05; p = 0.15), short (1 RCT, n = 461; SMD = -0.13; 95% CI: -0.31, 0.06; p = 0.18), and medium-term (1 RCT, n = 461; SMD = -0.15; 95% CI: -0.33, 0.03; p = 0.11).
Depression (n = 1). We found very low-quality evidence that PEMs had no effect on depression compared to usual care at immediate (1 RCT, n = 461; SMD = -0.18; 95% CI: -0.36, 0.01; p = 0.06), short (1 RCT, n = 461; SMD = -0.09; 95% CI: -0.27, 0.09; p = 0.35), medium (1 RCT, n = 461; SMD = -0.11; 95% CI: -0.29, 0.07; p = 0.24), and long-term (1 RCT, n = 461; SMD = -0.15; 95% CI: -0.33, 0.03; p = 0.10).
Patient education materials alone vs. other interventions.
Ten trials [40, 42–47, 52, 56, 59] compared the effect of PEMs to other interventions (Table 2) on LBP-related outcomes for chronic LBP patients. The most commonly measured outcome was pain intensity (n = 10), followed by disability (n = 9), quality of life (n = 5), global improvement (n = 3), anxiety (n = 2), and depression (n = 2). Single studies measured function, pain self-efficacy, fear-avoidance beliefs, catastrophizing, coping, stress, days off work. No studies measured knowledge, attitudes, general beliefs, imaging, physician visits, referrals, or cost. A summary of findings for eight key outcomes are presented in Table 7 (a summary of all other outcomes and forest plots for all analyses are presented in S3 and S4 Files, respectively).
Pain intensity (n = 10). We found high-quality evidence that PEMs are less effective than other interventions for decreasing pain intensity at immediate (8 RCT, n = 732; SMD = 0.30; 95% CI: 0.03, 0.56; p = 0.03; I2 = 63%) and short-term (7 RCT, n = 815; SMD = 0.54; 95% CI: 0.20, 0.88; p = 0.002; I2 = 80%). We found moderate and very-low quality evidence that PEMs had no effect on pain intensity compared to other interventions at medium (4 RCT, n = 450; SMD = 0.22; 95% CI: -0.25, 0.69; p = 0.35; I2 = 81%) and long-term (1 RCT, n = 168; SMD = 0.18; 95% CI: -0.12, 0.48; p = 0.24), respectively.
Disability (n = 9). We found high-quality evidence that PEMs are less effective than other interventions for decreasing disability at immediate (7 RCTs, n = 714; SMD = 0.47; 95% CI: 0.12, 0.83; p = 0.009; I2 = 79%) and short-term (8 RCT, n = 881; SMD = 0.64; 95% CI: 0.25, 1.02; p = 0.001; I2 = 85%). We found high and very-low quality evidence that PEMs had no effect on disability compared to other interventions at medium (4 RCT, n = 450; SMD = 0.29; 95% CI: -0.09, 0.67; p = 0.13; I2 = 72%) and long-term (1 RCT, n = 168; SMD = -0.07; 95% CI: -0.37, 0.23; p = 0.65), respectively.
Quality of life (n = 5). We found low-quality evidence that PEMs were less effective than other interventions for improving quality of life at immediate-term (2 RCTs, n = 62; SMD = 1.25; 95% CI: 0.14, 2.36; p = 0.03; I2 = 73%). Two studies (2 RCTs; n = 221) could not be pooled in the analysis but both found no difference of effect. We found low-quality evidence that PEMs had no effect on quality of life compared to other interventions at short-term (2 RCTs, n = 228; SMD = 1.01; 95% CI: -0.99, 3.01; p = 0.32; I2 = 96%). Two studies (2 RCTs; n = 221) could not be pooled, but one found there to be no difference of effect (n = 66), and the other found PEMs to be significantly less effective than other interventions (n = 168). Finally, we found very low-quality evidence that PEMs had no effect on quality of life compared to other interventions medium (1 RCT; n = 63) and long-term (1 RCT; n = 159).
Global improvement (n = 3). We found moderate-quality evidence that PEMs are less effective than other interventions on global improvement ratings at immediate (2 RCTs, n = 327; SMD = 0.53; 95% CI: 0.21, 0.84; p = 0.001; I2 = 22%) and medium-term (2 RCTs, n = 327; SMD = 0.55; 95% CI: 0.19, 0.91; p = 0.003; I2 = 44%), and high-quality evidence of the same observation at short-term (3 RCTs, n = 509; SMD = 0.60; 95% CI: 0.16, 1.04; p = 0.008; I2 = 75%).
Function (n = 1). We found very low-quality evidence that PEMs are significantly less effective than other interventions for improving performance-based function measures on the 6-Minute Walk test (1 RCT, n = 19; SMD = 1.34; 95% CI: 0.32, 2.36; p = 0.01) and Sit-to-Stand test (1 RCT, n = 17; SMD = 1.26; 95% CI: 0.18, 2.34; p = 0.02) at immediate-term. We found very low-quality evidence that PEMs had no effect compared to other interventions on the Sit-and-Reach test (1 RCT, n = 19; SMD = 0.95; 95% CI: -0.02, 1.91; p = 0.05) at immediate-term.
Pain self-efficacy (n = 1). We found very low-quality evidence that PEMs had no effect on pain self-efficacy compared to other interventions at immediate (1 RCT, n = 199; SMD = 0.05; 95% CI: -0.23, 0.33; p = 0.74), short (1 RCT, n = 199; SMD = 0.06; 95% CI: -0.22, 0.34; p = 0.67), and medium-term (1 RCT, n = 199; SMD = 0.04; 95% CI: -0.24, 0.32; p = 0.77).
Fear-avoidance (n = 1). We found very low-quality evidence that PEMs had no effect on fear-avoidance beliefs compared to other interventions at immediate (1 RCT, n = 199; SMD = 0.13; 95% CI: -0.15, 0.41; p = 0.35), short (1 RCT, n = 199; SMD = 0.08; 95% CI: -0.20, 0.36; p = 0.57), and medium-term (1 RCT, n = 199; SMD = 0.00; 95% CI: -0.28, 0.28; p = 1.00).
Catastrophizing (n = 1). We found very low-quality evidence that PEMs are significantly less effective than other interventions for reducing catastrophizing thoughts at immediate (1 RCT, n = 199; SMD = 0.50; 95% CI: 0.21, 0.78; p = 0.0006), short (1 RCT, n = 199; SMD = 0.42; 95% CI: 0.14, 0.70; p = 0.003), and medium-term (1 RCT, n = 199; SMD = 0.44; 95% CI: 0.15, 0.72; p = 0.002).
Coping (n = 1). We found very low-quality evidence that PEMs had no effect on coping compared to other interventions at immediate (1 RCT, n = 199; SMD = 0.13; 95% CI: -0.14, 0.41; p = 0.34), short (1 RCT, n = 199; SMD = 0.22; 95% CI: -0.05, 0.50; p = 0.12), and medium-term (1 RCT, n = 199; SMD = 0.17; 95% CI: -0.10, 0.45; p = 0.22).
Anxiety (n = 2). We found very low-quality evidence that PEMs had no effect on anxiety compared to other interventions at immediate (1 RCT, n = 199; SMD = 0.07; 95% CI: -0.20, 0.35; p = 0.60), and medium-term (1 RCT, n = 199; SMD = 0.13; 95% CI: -0.15, 0.40; p = 0.38), and low-quality evidence of no difference in effect at short-term (1 RCT, n = 199; SMD = 0.65; 95% CI: -0.58, 1.87; p = 0.30; I2 = 88%).
Stress (n = 1). We found very low-quality evidence that PEMs had no effect on stress compared to other interventions immediate (1 RCT, n = 199; SMD = 0.17; 95% CI: -0.10, 0.45; p = 0.22) and medium-term (1 RCT, n = 199; SMD = 0.26; 95% CI: -0.02, 0.54; p = 0.07). We found very low-quality evidence that PEMs are significantly less effective than other interventions for decreasing stress at short-term (1 RCT, n = 199; SMD = 0.31; 95% CI: 0.03, 0.59; p = 0.03).
Depression (n = 2). We found very low-quality evidence that PEMs had no effect on depression compared to other interventions at immediate (1 RCT, n = 199; SMD = 0.03; 95% CI: -0.25, 0.31; p = 0.84) and medium-term (1 RCT, n = 199; SMD = 0.18; 95% CI: -0.10, 0.46; p = 0.21), and low-quality evidence of no effect at short-term (1 RCT, n = 199; SMD = 0.79; 95% CI: -0.56, 2.14; p = 0.25; I2 = 90%).
Days off work (n = 1). We found very low-quality evidence that PEMs had no effect on days off work compared to other interventions at short-term (1 RCT, n = 168). No summary data for this outcome was provided in the study so no point estimate can be provided.
Discussion
We found 27 trials that evaluated the effectiveness of PEMs for acute or chronic LBP. Most were at moderate to high risk of bias (most commonly due to insufficient follow-up). We hypothesized that knowledge provided by PEMs would modify beliefs, expectations, and pain self-efficacy, and these changes would positively influence patients’ experience or perception of pain, expectations for unnecessary tests or other referrals, and adherence to advice to facilitate recovery compared to those who did not receive PEMs. Compared to usual care for acute LBP, PEMs appear to have at least some positive impacts both for patients and health systems, such as improved short-term pain intensity and immediate-term quality of life. Though the evidence was fairly low quality, knowledge appears to increase with the provision of PEMs across all measured time periods, as well as pain self-efficacy in the short to long-term. For health systems, the evidence was again fairly low quality, but PEMs reduced the short-term number of days off work and long-term physician visits and imaging. Compared to usual care for chronic LBP, PEMs were associated with improved pain intensity, global improvement ratings, and pain self-efficacy across all time periods, and quality of life from short to medium-term with variable levels of very low to moderate evidence. At medium-term, PEMs decreased disability but showed no impact at any other time measurement. The effect of PEMs on fear-avoidance beliefs and stress was more variable: fear-avoidance beliefs decreased in the medium-term, while stress decreased in the long term, with no other measurable impact in the other time periods. PEMs had no impact on depression.
Compared to other interventions, PEMs appear to have limited effectiveness in acute LBP. Though there were only one to two studies in all analyses and the quality of evidence was low to very low, PEMs were less effective in reducing immediate-term pain intensity and the number of long-term days off work and physician visits, with no effect on fear-avoidance beliefs, anxiety, depression, and disability. PEMs showed only a small impact on reducing pain intensity in the medium-term, but not short or long-term. Compared to other interventions in chronic LBP, PEMs had no effect or were less effective for every outcome measured.
Comparison with existing literature
Though we are the first to assess PEMs alone, our results are supported by and expand on previous literature investigating the effectiveness of patient education for LBP. We found many under-assessed outcomes in the LBP patient education literature, including knowledge. Nevertheless, we did find improvements in knowledge across all measured time points, and we provide the first evidence of effect on this outcome for LBP. Looking to the wider literature, we find similar results for PEMs on knowledge for other conditions like diabetes [66] and cancer [67]. Imaging was another under-assessed outcome measured by only one study in our review. However, we found LBP PEMs can reduce imaging rates, which is consistent with studies where PEMs are used as part of larger multi-component interventions to reduce imaging [68–70].
Our findings differ from those of Traeger et al., [24] who found that individual patient education (with or without PEMs) improved reassurance for acute/subacute LBP. Despite including many of the same studies, we did not find any measures of reassurance. Looking more closely at their methods, we see they combined several proxy outcomes (e.g., anxiety, fear-avoidance, and catastrophizing) as their measure of reassurance. We included these outcomes but analysed them as separate constructs and while many favoured PEMs, they were mostly not statistically significant. This highlights the importance of using validated measures of outcomes.
Our results also expanded on those of Engers et al., [27] who found no studies comparing individual patient education to usual care for chronic LBP. We updated this literature with five recent studies and found PEMs were effective on several clinical and process outcomes. Compared to usual care for acute LBP, they found patient education was significantly more effective in some studies but not others. We had similar findings in this comparison, but since we pooled the results in meta-analyses, we were able to find a trend towards a benefit of PEMs over usual care for most clinical outcomes at most time points. Compared to other interventions, we had similar findings that PEMs had no effect or were less effective for chronic LBP.
Implications for practice
Our review showed that offering PEMs to patients is preferable to usual care for both acute and chronic LBP. Given that PEMs are relatively inexpensive to produce, easy to provide, and unlikely to cause harm, clinicians may find them an effective adjunct to care. Unfortunately, we could not obtain copies of many of the PEMs that were the focus of the papers in our review despite reaching out to all authors. Additional work will be required to effectively translate these materials into practice and realize their potential.
Implications for research
Overall, we were disappointed to find that many of the studies included in our review used unvalidated and modified outcome measures (especially for process outcomes) despite the existence of validated measurement tools. This clouds our understanding of the effectiveness of interventions and we recommend that researchers use unmodified, validated tools to measure all outcomes. In addition, many key outcomes were rarely measured (e.g., quality of life, knowledge, pain self-efficacy) or not measured at all (e.g., attitudes, general beliefs). To standardize reporting in clinical trials, we recommend that researchers more frequently assess quality of life as it is a core clinical outcome for LBP alongside pain and disability [71], and suggest developing a similar set of core domains for important process outcomes related to LBP (e.g., fear-avoidance beliefs, catastrophizing, coping, pain self-efficacy) since measures of these outcomes varied substantially across LBP trials. Researchers should work with LBP patients to choose a core set of prioritized, patient-reported outcomes. Finally, PEMs literature lacks adequate reporting on material development as well as measures of intervention adherence and other outcomes related to intervention fidelity, making it difficult to fully understand their effectiveness. We recommend that researchers assess and report these outcomes to determine if the interventions are being provided and received as planned by following intervention reporting guidelines such as the TIDieR checklist [31].
Future research
PEMs compared to usual care for chronic LBP appear to have more success than those for acute LBP, perhaps because the majority were comprehensive digital interventions (as opposed to the physical booklets most often used for acute LBP) with one or often more of the following: (i) co-development with patients, (ii) text- and video-based information, (iii) instant, tailored feedback based on automated questions, (iv) interactive or gamification components including quizzes and rewards, (v) reminders to use the material and follow recommendations, and (vi) could be accessed anywhere at any time. We recommend future studies compare these newer PEMs to other guideline-recommended interventions (e.g., exercise therapies, massage, CBT) since most studies we found in this comparison used standard physical booklets. Furthermore, most of these studies treated the PEMs group as a control or usual care group, which may have introduced bias to the comparison and hindered our ability to interpret the results.
Strengths and limitations
The primary strengths of this review were our adherence to best practices for conducting systematic reviews. We followed all guidance provided in the Cochrane [72] and GRADE [37] handbooks, conducted a sensitive search strategy that adhered to the PRESS guidelines [28], and followed the TIDieR recommendations [31] for reporting of intervention details, which allowed for a more thorough assessment of PEMs. Additionally, we included a comprehensive list of outcomes that are important to all stakeholders, including patients, policymakers, researchers, and clinicians, and compared PEMs to other interventions that are commonly used in practice to provide relative effectiveness. We also sought and obtained additional data from authors who did not report the data within their study. Limitations to this review include the use of unvalidated and modified outcome measures and the conversion of dichotomized data to SMDs where it was necessary to pool the results. Both decisions could have influenced the resulting effect sizes and increased the degree of variability across outcome measures and time periods.
Conclusion
Due to the degree of variability in the impact of PEMs on all outcomes and across all time periods (likely a result of the heterogeneity of measures and definitions across studies), it is difficult to succinctly and concisely state conclusions for all outcomes. However, it certainly appears that providing PEMs is better than doing nothing (i.e., usual care) as we observed small positive patient and system impacts for both acute and chronic LBP. Given their low cost and relative ease of provision, PEMs appear preferable to usual care, although the quality of evidence is fairly low for this conclusion. Compared to other interventions, PEMs had no effect or were less effective for almost every outcome measured; however, cost effectiveness was not assessed in any of these studies, and it is likely that PEMs were substantially less costly than all other studied interventions. Additionally, in recent years more comprehensive digital PEMs have been developed, and we recommend these are compared to other interventions before making conclusions about their relative usefulness.
Supporting information
S2 File. Inclusion, exclusion, and GRADE criteria, protocol deviations.
https://doi.org/10.1371/journal.pone.0274527.s002
(DOCX)
S3 File. Summary of findings (all outcomes and comparisons).
https://doi.org/10.1371/journal.pone.0274527.s003
(DOCX)
Acknowledgments
We would like to thank Georgia Darmonkow, Anika Shama, and Senem Gözel for their contributions to study screening and data extraction.
References
- 1. Hoy D, March L, Brooks P, Blyth F, Woolf A, Bain C, et al. The global burden of low back pain: estimates from the Global Burden of Disease 2010 study. Ann Rheum Dis. 2014;73: 968–974. pmid:24665116
- 2. Sauver JLS, Warner DO, Yawn BP, Jacobson DJ, McGree ME, Pankratz JJ, et al. Why patients visit their doctors: assessing the most prevalent conditions in a defined American population. Mayo Clinic Proceedings. Elsevier; 2013. pp. 56–67.
- 3. Kent PM, Keating JL. The epidemiology of low back pain in primary care. Chiropr Osteopat. 2005;13: 13. pmid:16045795
- 4. Katz JN. Lumbar disc disorders and low-back pain: socioeconomic factors and consequences. JBJS. 2006;88: 21–24. pmid:16595438
- 5.
Lidgren L. The bone and joint decade 2000–2010. SciELO Public Health; 2003.
- 6.
Group TOPLBPW. Evidence-Informed Primary Care Management of Low Back Pain. Edmont Can Optim Pract. 2015.
- 7. de Campos TF. Low back pain and sciatica in over 16s: assessment and management NICE Guideline [NG59]. J Physiother. 2017;63: 120.
- 8.
Innovation NA for C. Management of people with acute low back pain: model of care. Chatswood. 2016.
- 9. Airaksinen O, Brox JI, Cedraschi C, Hildebrandt J, Klaber-Moffett J, Kovacs F, et al. European guidelines for the management of chronic nonspecific low back pain. Eur Spine J. 2006;15: s192.
- 10. Qaseem A, Wilt TJ, McLean RM, Forciea MA. Noninvasive treatments for acute, subacute, and chronic low back pain: a clinical practice guideline from the American College of Physicians. Ann Intern Med. 2017;166: 514–530. pmid:28192789
- 11. Machado LAC, Kamper SJ, Herbert RD, Maher CG, McAuley JH. Analgesic effects of treatments for non-specific low back pain: a meta-analysis of placebo-controlled randomized trials. Rheumatology. 2009;48: 520–527. pmid:19109315
- 12. Lim YZ, Chou L, Au RT, Seneviwickrama KMD, Cicuttini FM, Briggs AM, et al. People with low back pain want clear, consistent and personalised information on prognosis, treatment options and self-management strategies: a systematic review. J Physiother. 2019;65: 124–135. pmid:31227280
- 13. Ainpradub K, Sitthipornvorakul E, Janwantanakul P, van der Beek AJ. Effect of education on non-specific neck and low back pain: a meta-analysis of randomized controlled trials. Man Ther. 2016;22: 31–41. pmid:26585295
- 14. Barbari V, Storari L, Ciuro A, Testa M. Effectiveness of communicative and educative strategies in chronic low back pain patients: a systematic review. Patient Educ Couns. 2020;103: 908–929. pmid:31839351
- 15. Brox JI, Storheim K, Grotle M, Tveito TH, Indahl A, Eriksen HR. Systematic review of back schools, brief education, and fear-avoidance training for chronic low back pain. Spine J. 2008;8: 948–958. pmid:18024224
- 16. Clarke CL, Ryan CG, Martin DJ. Pain neurophysiology education for the management of individuals with chronic low back pain: A systematic review and meta-analysis. Man Ther. 2011;16: 544–549.
- 17. Du S, Hu L, Dong J, Xu G, Chen X, Jin S, et al. Self-management program for chronic low back pain: a systematic review and meta-analysis. Patient Educ Couns. 2017;100: 37–49. pmid:27554077
- 18. Du S, Yuan C, Xiao X, Chu J, Qiu Y, Qian H. Self-management programs for chronic musculoskeletal pain conditions: a systematic review and meta-analysis. Patient Educ Couns. 2011;85: e299–e310. pmid:21458196
- 19. Nicholl BI, Sandal LF, Stochkendahl MJ, McCallum M, Suresh N, Vasseljen O, et al. Digital support interventions for the self-management of low back pain: a systematic review. J Med Internet Res. 2017;19: e179. pmid:28550009
- 20. Oliveira VC, Ferreira PH, Maher CG, Pinto RZ, Refshauge KM, Ferreira ML. Effectiveness of self-management of low back pain: Systematic review with meta-analysis. Arthritis Care Res. 2012;64: 1739–1748. pmid:22623349
- 21. Parreira P, Heymans MW, van Tulder MW, Esmail R, Koes BW, Poquet N, et al. Back Schools for chronic non-specific low back pain. Cochrane Database Syst Rev. 2017;2017. pmid:28770974
- 22. Straube S, Harden M, Schröder H, Arendacka B, Fan X, Moore RA, et al. Back schools for the treatment of chronic low back pain: possibility of benefit but no convincing evidence after 47 years of research—systematic review and meta-analysis. Pain. 2016;157: 2160. pmid:27257858
- 23. Tegner H, Frederiksen P, Esbensen BA, Juhl C. Neurophysiological pain education for patients with chronic low back pain. Clin J Pain. 2018;34: 778–786.
- 24. Traeger AC, Huebscher M, Henschke N, Moseley GL, Lee H, McAuley JH. Effect of primary care–based education on reassurance in patients with acute low back pain: systematic review and meta-analysis. JAMA Intern Med. 2015;175: 733–743. pmid:25799308
- 25. Zahari Z, Ishak A, Justine M. The effectiveness of patient education in improving pain, disability and quality of life among older people with low back pain: A systematic review. J Back Musculoskelet Rehabil. 2019; 1–10.
- 26. Furlong B, Aubrey-Bassler K, Etchegary H, Pike A, Darmonkow G, Swab M, et al. Patient education materials for non-specific low back pain and sciatica: a protocol for a systematic review and meta-analysis. BMJ Open. 2020;10: e039530. pmid:32878763
- 27. Engers AJ, Jellema P, Wensing M, van der Windt DA, Grol R, van Tulder MW. Individual patient education for low back pain. Cochrane Database Syst Rev. 2008. pmid:18254037
- 28. McGowan J, Sampson M, Salzwedel DM, Cogo E, Foerster V, Lefebvre C. PRESS peer review of electronic search strategies: 2015 guideline statement. J Clin Epidemiol. 2016;75: 40–46. pmid:27005575
- 29.
EndNote. Clarivate Analytics PLC; 2020. https://endnote.com/
- 30.
Innovation VH. Covidence systematic review software. Veritas Health Innovation Melbourne, VIC; 2016.
- 31. Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Moher D, et al. Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide. Bmj. 2014;348: g1687. pmid:24609605
- 32. Armijo-Olivo S, da Costa BR, Cummings GG, Ha C, Fuentes J, Saltaji H, et al. PEDro or Cochrane to assess the quality of clinical trials? A meta-epidemiological study. PloS One. 2015;10. pmid:26161653
- 33.
Review Manager (RevMan) [Computer program]. The Cochrane Collaboration; 2020.
- 34. Higgins JP, Li T, Deeks JJ. Choosing effect measures and computing estimates of effect. Cochrane Handb Syst Rev Interv. 2019; 143–176.
- 35. Deeks J, Higgins J, Altman D. Chapter 9: Analysing Data and Undertaking Meta-analyses: Cochrane Handbook for Systematic Reviews of Interventions Version 5.1. 0 [updated March 2011]. Cochrane Handb Syst Rev Interv Version. 2011;5.
- 36. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. Bmj. 2003;327: 557–560. pmid:12958120
- 37. Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, et al. GRADE guidelines: 1. Introduction—GRADE evidence profiles and summary of findings tables. J Clin Epidemiol. 2011;64: 383–394. pmid:21195583
- 38.
Higgins JPT, Green S. Extracting study results and converting to the desired format. Cochrane Handb Syst Rev Interv Lond UK Cochrane Collab. 2009.
- 39. Lorig KR, Laurent DD, Deyo RA, Marnell ME, Minor MA, Ritter PL. Can a Back Pain E-mail Discussion Group improve health status and lower health care costs?: A randomized study. Arch Intern Med. 2002;162: 792–796. pmid:11926853
- 40. Brodsky M, Hansen A, Bjerke W. Randomized pilot trial for a community-based group stretching exercise program for chronic low back pain. Glob Adv Health Med. 2019;8: 2164956119846055. pmid:31069163
- 41. Cherkin DC, Deyo RA, Battié M, Street J, Barlow W. A comparison of physical therapy, chiropractic manipulation, and provision of an educational booklet for the treatment of patients with low back pain. N Engl J Med. 1998;339: 1021–1029. pmid:9761803
- 42. Cherkin DC, Eisenberg D, Sherman KJ, Barlow W, Kaptchuk TJ, Street J, et al. Randomized trial comparing traditional Chinese medical acupuncture, therapeutic massage, and self-care education for chronic low back pain. Arch Intern Med. 2001;161: 1081–1088. pmid:11322842
- 43. Chiauzzi E, Pujol LA, Wood M, Bond K, Black R, Yiu E, et al. painACTION-back pain: a self-management website for people with chronic back pain. Pain Med. 2010;11: 1044–1058. pmid:20545873
- 44. Ferrell BA, Josephson KR, Pollan AM, Loy S, Ferrell BR. A randomized trial of walking versus physical methods for chronic pain management. Aging Clin Exp Res. 1997;9: 99–105. pmid:9177592
- 45. Saper RB, Lemaster C, Delitto A, Sherman KJ, Herman PM, Sadikova E, et al. Yoga, physical therapy, or education for chronic low back pain: a randomized noninferiority trial. Ann Intern Med. 2017;167: 85–94. pmid:28631003
- 46. Sherman KJ, Cherkin DC, Erro J, Miglioretti DL, Deyo RA. Comparing yoga, exercise, and a self-care book for chronic low back pain: a randomized, controlled trial. Ann Intern Med. 2005;143: 849–856. pmid:16365466
- 47. Sherman KJ, Cherkin DC, Wellman RD, Cook AJ, Hawkes RJ, Delaney K, et al. A randomized trial comparing yoga, stretching, and a self-care book for chronic low back pain. Arch Intern Med. 2011;171: 2019–2026. pmid:22025101
- 48. Weiner DK, Gentili A, Rossi M, Coffey-Vega K, Rodriguez KL, Hruska KL, et al. Aging back clinics—A geriatric syndrome approach to treating chronic low back pain in older adults: Results of a preliminary randomized controlled trial. Pain Med. 2020;21: 274–290. pmid:31503275
- 49. Little P, Roberts L, Blowers H, Garwood J, Cantrell T, Langridge J, et al. Should we give detailed advice and information booklets to patients with back pain?: a randomized controlled factorial trial of a self-management booklet and doctor advice to take exercise for back pain. Spine. 2001;26: 2065–2072. pmid:11698879
- 50. Roberts L, Little P, Chapman J, Cantrell T, Pickering R, Langridge J. The back home trial: general practitioner-supported leaflets may change back pain behavior. Spine. 2002;27: 1821–1828. pmid:12221342
- 51. Roland M, Dixon M. Randomized controlled trial of an educational booklet for patients presenting with back pain in general practice. J R Coll Gen Pract. 1989;39: 244–246. pmid:2556518
- 52. Valenza MC, Rodríguez-Torres J, Cabrera-Martos I, Díaz-Pelegrina A, Aguilar-Ferrándiz ME, Castellote-Caballero Y. Results of a Pilates exercise program in patients with chronic non-specific low back pain: a randomized controlled trial. Clin Rehabil. 2017;31: 753–760. pmid:27260764
- 53.
Pascual FV. The influence of a web-based biopsychosocial pain education intervention on pain, disability, and pain cognition in patients with chronic low back pain in primary care: a mixed methods approach. Universitat de Lleida. 2019.
- 54. Linton SJ, Andersson T. Can chronic disability be prevented?: A randomized trial of a cognitive-behavior intervention and two forms of information for patients with spinal pain. Spine. 2000;25: 2825–2831. pmid:11064530
- 55. Irvine AB, Russell H, Manocchia M, Mino DE, Glassen TC, Morgan R, et al. Mobile-Web app to self-manage low back pain: randomized controlled trial. J Med Internet Res. 2015;17: e3130. pmid:25565416
- 56. Areeudomwong P, Wongrat W, Neammesri N, Thongsakul T. A randomized controlled trial on the long-term effects of proprioceptive neuromuscular facilitation training, on pain-related outcomes and back muscle activity, in patients with chronic low back pain. Musculoskeletal Care. 2017;15: 218–229. pmid:27791345
- 57. Sihawong R, Waongenngarm P, Janwantanakul P. Efficacy of risk factor education on pain intensity and disability in office workers with nonspecific neck or low back pain: A pilot cluster randomized clinical trial. J Back Musculoskelet Rehabil. 2021;34: 251–259. pmid:33185585
- 58. Hodges PW, Hall L, Setchell J, French S, Kasza J, Bennell K, et al. Effect of a Consumer-Focused Website for Low Back Pain on Health Literacy, Treatment Choices, and Clinical Outcomes: Randomized Controlled Trial. J Med Internet Res. 2021;23: e27860. pmid:34128822
- 59. Kuvačić G, Fratini P, Padulo J, Antonio DI, De Giorgio A. Effectiveness of yoga and educational intervention on disability, anxiety, depression, and pain in people with CLBP: A randomized controlled trial. Complement Ther Clin Pract. 2018;31: 262–267. pmid:29705466
- 60. Simula AS, Jenkins HJ, Hancock MJ, Malmivaara A, Booth N, Karppinen J. Patient education booklet to support evidence-based low back pain care in primary care–a cluster randomized controlled trial. BMC Fam Pract. 2021;22: 1–15.
- 61. Bücker B, Butzlaff M, Isfort J, Koneczny N, Vollmar HC, Lange S, et al. Effect of written patient information on knowledge and function of patients with acute uncomplicated back pain (PIK Study). Gesundheitswesen Bundesverb Arzte Offentlichen Gesundheitsdienstes Ger. 2010;72: e78–88.
- 62. Kazemi S-S, Tavafian S-S, Hiller CE, Hidarnia A, Montazeri A. The effectiveness of social media and in-person interventions for low back pain conditions in nursing personnel (SMILE). Nurs Open. 2021;8: 1220–1231. pmid:33905171
- 63. Jellema P, Van Der Windt DA, Van Der Horst HE, Twisk JW, Stalman WA, Bouter LM. Should treatment of (sub) acute low back pain be aimed at psychosocial prognostic factors? Cluster randomised clinical trial in general practice. Bmj. 2005;331: 84. pmid:15967762
- 64. Darlow B, Stanley J, Dean S, Abbott JH, Garrett S, Mathieson F, et al. The Fear Reduction Exercised Early (FREE) approach to low back pain: study protocol for a randomised controlled trial. Trials. 2017;18: 1–14.
- 65. Sandal LF, Bach K, Øverås CK, Svendsen MJ, Dalager T, Jensen JSD, et al. Effectiveness of app-delivered, tailored self-management support for adults with lower Back pain–related disability: a selfBACK randomized clinical trial. JAMA Intern Med. 2021;181: 1288–1296. pmid:34338710
- 66. Loveman E, Cave C, Green C, Royle P, Dunn N, Waugh N. The clinical and cost-effectiveness of patient education models for diabetes: a systematic review and economic evaluation. Health Technol Assess Winch Engl. 2003;7: iii, 1–190. pmid:13678547
- 67. Oldenmenger WH, Geerling JI, Mostovaya I, Vissers KC, de Graeff A, Reyners AK, et al. A systematic review of the effectiveness of patient-based educational interventions to improve cancer-related pain. Cancer Treat Rev. 2018;63: 96–103. pmid:29272781
- 68. Kullgren JT, Krupka E, Schachter A, Linden A, Miller J, Acharya Y, et al. Precommitting to choose wisely about low-value services: a stepped wedge cluster randomised trial. BMJ Qual Saf. 2018;27: 355–364. pmid:29066616
- 69. Schectman JM, Schroth WS, Verme D, Voss JD. Randomized controlled trial of education and feedback for implementation of guidelines for acute low back pain. J Gen Intern Med. 2003;18: 773–780. pmid:14521638
- 70. Min A, Chan VW, Aristizabal R, Peramaki ER, Agulnik DB, Strydom N, et al. Clinical decision support decreases volume of imaging for low back pain in an urban emergency department. J Am Coll Radiol. 2017;14: 889–899. pmid:28483544
- 71. Chiarotto A, Deyo RA, Terwee CB, Boers M, Buchbinder R, Corbin TP, et al. Core outcome domains for clinical trials in non-specific low back pain. Eur Spine J. 2015;24: 1127–1142. pmid:25841358
- 72.
Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al. Cochrane handbook for systematic reviews of interventions. John Wiley & Sons; 2019.