Clinical practice CPGs (CPGs) have been developed to summarize evidence related to the management of osteoarthritis (OA). CPGs facilitate uptake of evidence-based knowledge by consumers, health professionals, health administrators and policy makers. The objectives of the present review were: 1) to assess the quality of the CPGs on non-pharmacological management of OA; using a standardized and validated instrument - the Appraisal of Guidelines Research and Evaluation (AGREE II) tool - by three pairs of trained appraisers; and 2) to summarize the recommendations based on only high-quality existing CPGs. Scientific literature databases from 2001 to 2013 were systematically searched for the state of evidence, with 17 CPGs for OA being identified. Most CPGs effectively addressed only a minority of AGREE II domains. Scope and purpose was effectively addressed in 10 CPGs on the management of OA, stakeholder involvement in 12 CPGs, rigour of development in 10 CPGs, clarity/presentation in 17 CPGs, editorial independence in 2 CPGs, and applicability in none of the OA CPGs. The overall quality of the included CPGs, according to the 7-point AGREE II scoring system, is 4.8±0.41 for OA. Therapeutic exercises, patient education, transcutaneous electrical nerve stimulation, acupuncture, orthoses and insoles, heat and cryotherapy, patellar tapping, and weight control are commonly recommended for the non-pharmacological management of OA by the high-quality CPGs. The general clinical management recommendations tended to be similar among high-quality CPGs, although interventions addressed varied. Non-pharmacological management interventions were superficially addressed in more than half of the selected CPGs. For CPGs to be standardized uniform creators should use the AGREE II criteria when developing CPGs. Innovative and effective methods of CPG implementation to users are needed to ultimately enhance the quality of life of arthritic individuals.
Citation: Brosseau L, Rahman P, Toupin-April K, Poitras S, King J, et al. (2014) A Systematic Critical Appraisal for Non-Pharmacological Management of Osteoarthritis Using the Appraisal of Guidelines Research and Evaluation II Instrument. PLoS ONE 9(1): e82986. doi:10.1371/journal.pone.0082986
Editor: Adrian V. Hernandez, Universidad Peruana de Ciencias Aplicadas (UPC), Peru
Received: July 21, 2013; Accepted: October 29, 2013; Published: January 10, 2014
Copyright: © 2014 Brosseau 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.
Funding: This project was funded by the Canadian Institutes of Health Research's (CIHR) Institute of Musculoskeletal Health and Arthritis Community Outreach Grant: #ICS-123828. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Osteoarthritis (OA) is known as a degenerative disorder of the joint cartilage associated with hypertrophic bone changes  and it is recognized as the most common chronic joint disease in the world . It is expected that OA will be the fourth leading cause of disability by 2020 and the 6th leading cause of years lived with disability –. The annual absenteeism costs related to OA in North America are $10.3 billion . The management of OA in patients should be comprehensive and should target pain reduction, improvement and maintenance of joint function, a decrease in disability, and education of parents about disease and therapies . While people with severe and persistent OA symptoms may use pharmacological treatments such as nonsteroidal anti-inflammatory drugs (NSAIDs), cyclo-oxygenase-2 (COX 2) inhibitors, and undergo joint arthroplasty –, people with mild to moderate OA symptoms should consider conservative management by combining pharmacological and non-pharmacological interventions , . Non-pharmacological interventions are essential to the treatment and management of any chronic disease and they are as important as pharmacological interventions . According to Sakalauskiene (2010) , most non-pharmacological interventions 1) are low in cost; 2) incorporate self-management performed at home or in the community; and 3) have a substantial public health impact. Non-pharmacological interventions, such as therapeutic exercises and weight control, have been shown to be effective in reducing pain and improving function in OA and are usually safe –. However, their use is often suboptimal, which warrants further knowledge translation to clinicians and patients about their importance in improving health outcomes –. Numerous clinical practice CPGs (CPGs) exist in rheumatology, which are intended to facilitate knowledge translation to clinicians and evidence-based clinical decision making. In order to make optimal and accurate clinical decisions for their arthritic patients, health professionals should use high-quality CPGs. In previous systematic reviews –, CPGs that considered non-pharmacological and pharmacological interventions have been appraised. However, the CPGs which targeted only non-pharmacological interventions have never been assessed with the Appraisal of Guidelines Research and Evaluation II (AGREE II) tool –. Our paper focused on the quality assessment of non-pharmacological interventions, especially in terms of rigour of development. Non-pharmacological interventions include electrotherapy (e.g. transcutaneous electrical nerve stimulation), patient education, team approach (e.g. multidisciplinary team approach), therapeutic exercises (e.g. aquatics), weight management and other interventions (e.g. acupuncture assistive devices, etc.) –. This review will provide health care professionals a platform to compare the development of recommendations for non-pharmacological interventions from the AGREE II scoring. It will also help them accept and implement the recommended interventions in their health practice. The objective of this critical appraisal review is to 1) identify the CPGs focusing on non-pharmacological management of OA in all joints of the upper and lower extremities; 2) assess the quality of the CPGs using the updated AGREE II instrument (www.agreetrust.org); and 3) to document the non-pharmacological recommendations and identify the high quality CPGs.
The systematic review of CPGs used the Cochrane Methodology (www.cochrane.org) to identify, select and analyze the data and the PRISMA statement to guide the reporting of the systematic review  (Appendix S1). Ethics approval was not required, as this work was based on a systematic literature review.
A systematic literature search was performed by an experienced librarian (JM) using relevant key words (Appendix S2) in specific databases: AMED, CINHAL Medline, Embase (Appendix S2). In addition, a hand search was also performed from two main sources: 1) existing guideline inventories and guideline homepages, such as PEDro (http://www.pedro.fhs.usyd.edu.au/index.html) , the National Guideline Clearinghouse (http://www.guideline.gov/) , Guideline International Network (http://www.g-i-n.net/) Turning Research into Practice (TRIP) (http://www.tripdatabase.com/)  2) the reference lists found in the selected/included CPGs.
In order to be inclusive of a wider range of non-pharmacological interventions, CPGs were included if they were current (2001–2013), consisted of a non-pharmacological component for adult patients (≥18 years of age) with osteoarthritis, and written in English. All CPGs considered included a grading system to evaluate the evidence and have been peer-reviewed prior to publication. Two pairs of independent evaluators assessed the eligibility of all potential CPGs and were considered if they fulfilled the inclusion criteria (Appendix S3) established prior to the literature search.
Data extraction procedures
The quality of each CPG was assessed using the AGREE II instrument (www.agreetrust.org). The AGREE II is a 23-item instrument which consists of six domains evaluating different aspects of CPG development: 1) scope and purpose: the objective of the guideline, the target population and health question; 2) stakeholder involvement: involvement of stakeholders in the guideline development process and patients views and preferences 3) rigour of development: the process to collect and synthesize evidence and the recommendation development process; 4) clarity and presentation: the language, structure and the presentation of the guideline; 5) applicability: looking at the barriers and facilitators for implementation, approach to improve uptake; and 6) editorial independence: identifying biases resulting from competing interests. The AGREE II uses a 7-point Likert scoring system ranging from 1 (strongly disagree), which corresponds to an item very poorly reported to 7 (strongly agree), which means an exceptional quality of reporting. A scoring from 2 to 6 was assigned if all the items in the domains were not considered; the scoring would increase depending on the consideration and fulfilment of each criterion (www.agreetrust.org).
Assessment of the methodological qualities of CPGs
Two pairs of evaluators with AGREE II experience independently assessed the selected CPGs with the AGREE II instrument and to prevent a potential information bias a third experienced evaluator was involved when a co-author of this review was a co-creator of an included CPG (Table 1). All three evaluators received training from the online tutorial and practice guideline (http://www.agreetrust.org/resource-centre/training/). For each CPG, the items from AGREE II instrument(http://www.agreetrust.org/wp-content/uploads/2013/06/AGREE_II_Users_Manual_and_23-item_Instrument_ENGLISH.pdf) was completed which included all relevant information and ratings on the 7-point scale for all six domains. The total scores were computed using the McMaster calculator (). The inter-rater agreement for AGREE II scores was also computed and a discussion about scoring occurred when the standard deviation (SD) between the two evaluators for each domain was greater than 1.50. The AGREE II consortium has not set a minimum or maximum range for domain score quality; therefore, it is difficult to differentiate between high and low quality CPGs. We used the criteria of previous guideline appraisals , , where domain scores greater than 60% are considered effectively addressed. The guideline is recommended and considered high quality based on the rigor of development and if three or more domains were effectively targeted (greater than 60%) .
Table 1. The Included CPGs.doi:10.1371/journal.pone.0082986.t001
Strength of recommendations for each non pharmacological interventions in the CPGs
In addition to assessing the methodological quality, the evaluators compiled all the non-pharmacological interventions assessed in at least one of the included CPGs (Table S1). The strength of the recommendations related to the interventions was translated into a 4-point hierarchical category system: 1) strongly recommended; 2) recommended; 3) weak evidence; and 4) insufficient evidence. Each of these categories was adjusted for equivalence and uniformity according to the individual scoring system of each of the seventeen CPGs included in this review. The grade “strongly recommended” corresponds to the highest grading and is usually based on at least one high-quality systematic review or randomized controlled trial (RCT) and represents a strong body of evidence. The grade “recommended” corresponds to a CPG grading usually based on at least one controlled clinical trial (CCT) or RCT of lower quality, and represents a good body of evidence that can be trusted to guide practice in most situations. “Weak evidence” is based on a strong clinical or expert opinion based on current practice, without the support of scientific evidence, and needs to be applied with caution. “Insufficient evidence” is assigned when there is a lack of scientific results or conflicting results not supported by clinical or expert opinion.
Inter-rater reliability study
The inter-rater reliability study for AGREE II was conducted to ensure the reliability of the CPGs quality scores between pairs of evaluators. Each individual evaluator completed the quality appraisal evaluation to determine the quality of each recommendation statement. The sub-total scores obtained by one evaluator for each domain were compared with those rated by the second evaluator. The statistical analysis involved intraclass correlation coefficients (ICCs) based on an ANOVA (random) procedure for repeated data . The analysis of the reliability study was performed with Statistical Package for Social Sciences (SPSS) version 20.
A systematic search provided a total of 1136 citations. Figure S1 provides a flow diagram of how the included CPGs were selected.
Characteristics of studies
A total of 17 CPGs – were included for appraisal according to selection criteria and all of the included CPGs provided recommendation for the management of hand, hip or knee (Table 1). Out of the 17 CPGs, 9 CPGs –, –,  were considered non-pharmacological CPGs exclusively and 8 CPGs –, –,  included both pharmacological and non-pharmacological interventions (Table 2).
Table 2. CPGs that considered pharmacological and non-pharmacological interventions.doi:10.1371/journal.pone.0082986.t002
The AGREE II quality scores of the studies
The quality scores for each domain of the AGREE II instrument for the 17 included CPGs – are found in table 3 and the individual scores for each item can be found in appendix S4. The methodological quality ranged from 45%–80% for the various CPGs, domain 1 being of highest quality and domain 5 being of lowest quality (Table 3). D1: scope and purpose had a mean score of 70%±20% for the 17 – included CPGs and both the Philadelphia Panel – and the series of Ottawa Panel CPGs – scored the highest at 97%. D2: stakeholder involvement obtained a lower mean score of 61%±21%. The National Institute of for Health and Clinical Excellence for the Care and Management of Osteoarthritis (NICE)  had the highest score at 89% and the series of EULAR – CPGs scored the lowest for this domain with 28%. Domain 3: rigor of development had a mean score of 63%±11%. American Academy of Orthopaedic Surgeons  scoring the highest for this domain at 80% and out of the 17 CPGs, AAOS  underwent the most extensive development process. Domain 4: clarity of presentation scored a mean of 86%±9.1% with Peter et al CPG scoring the highest at 97%. Domain 5: applicability had a poor mean score of 14%±16%, for this domain three CPGs , ,  obtained 0%. Domain 6: editorial independence also had a poor mean score at 31%±19.1%.
Table 3. Quality Scores using AGREE II Instruments for included CPGs on OA.doi:10.1371/journal.pone.0082986.t003
Overall, ten –, –,  out of the seventeen CPGs were deemed high quality (>60%) based on rigour of development and effectively targeting 4–5 out of the 6 domains, one of the CPGs  was able to effectively target five out of the six domains and the remaining CPGs , –,  effectively targeted four domains out of the six. Clarity of presentation was the domain that was most effectively addressed by all of the CPGs and applicability was the most poorly addressed domain. None of the CPGs effectively addressed domain 5 and CPG scores ranged from 0%–56%. Only a minority of the domains were effectively targeted by most of the CPGs (domain 1, 2, 3 & 4). None of the included CPGs effectively targeted all six of the domains.
Interestingly, out of the nine CPGs that solely looked at non-pharmacological interventions (Table 2), only the Philadelphia Panel – and the series of Ottawa Panel CPGs – are considered to be a high quality guideline (>60%). Roddy et al , EULAR  & Peter et al  only effectively targeted two out of the six domains and Kjeken et al  effectively targeted only one out of the six domains.
For CPGs that targeted pharmacological interventions and non-pharmacological interventions (Table 2), five –, –,  out of the eight CPGs were deemed high quality and effectively addressed four to five domains. All five CPGs scored high for two domains in particular: Scope and purpose and clarity of presentation. Consequently, all five CPGs –, ,  were recommended for practice. The series of EULAR – CPGs were the only CPGs which targeted both pharmacological and non-pharmacological interventions and not graded as a high quality CPG with a low score of 4/7 for overall quality of the CPG. Both appraisers recommended this guideline for practice, but with modifications.
When asked to rate the overall quality of the CPGs, ten –, , , –, – out of the seventeen were recommended for practice by both appraisers and seven CPGs –, , , , were recommended but with modifications (Appendix S4).
The strength of the recommendations
The strength of the recommendations for each intervention was provided by the developers of the included CPGs – (Table S1). The recommendations formulated by these CPGs were categorized by alphabetical order: 1) Electrotherapy; 2) Other Interventions; 3) Patient Education; 5) Team Approaches; 6) Therapeutic Exercises; 7) Weight Management (Table S1). Generally speaking, the strength of the recommendation varied largely amongst the 17 CPGs. Table S1 summarizes the strength of all the recommendations of the included CPGs according to interventions.
Transcutaneous electric nerve stimulation (TENS) were recommended by the majority of the included CPGs. Ten , –, – out of the seventeen CPGs recommended or strongly recommended TENS. Only one CPG  found insufficient evidence to recommend TENS.
Acupuncture was recommended or strongly recommended by 9 CPGs , –, – and 2 CPGs ,  found insufficient evidence to recommend it for the management of osteoarthritis. Foot orthoses and insoles were recommended or strongly recommended by ten CPGs –, –. Again for heat/cryotherapy a total of twelve –, , –, – out of seventeen mentioned this intervention in their CPGs and ten –, , – recommended it and two – found insufficient evidence. Patellar taping was recommended in six CPGs –, ; However one  CPG found insufficient evidence to recommend this intervention.
Only one guideline  mentioned multidisciplinary approach and physiotherapy as an intervention. However, the guideline found weak evidence to recommend it for the management of OA.
For therapeutic exercises the two interventions recommended by the majority of the CPGs were aerobic exercises, behavioral approach combined with therapeutic exercises and strengthening exercises. Aerobic exercises were recommended by thirteen –, –,  out of the seventeen CPGs, Behavioral approach combined with therapeutic exercises was recommended by eight , , –,  out of seventeen and strengthening exercises were recommended by nine , , , –.
Control of weight was recommended or strongly recommended by thirteen , , –, –,  of the seventeen CPGs and only four CPGs – strongly recommended diet alone, diet combined with physical activity as well as physical activity as a sole intervention.
Inter-rater reliability study
The AGREE II scores exhibits an overall very good inter-rater reliability with an ICCs values ranging from 0.86 (good reliability) to 0.95 (high reliability) depending on the domain assessed (Table 4). These results indicate the AGREE II quality scores of the included CPGs obtained between pairs of evaluators are reliable (Table 4).
This review identified a total of 17 CPGs on the non-pharmacological management of OA in all joints of the upper and lower extremities. Among the 17 CPGs considered, nine of these solely focus on non-pharmacological interventions and the remaining eight comprise a combination of pharmacological and non-pharmacological interventions. According to the AGREE II instrument, ten CPGs –, –,  were recognized as good quality CPGs with high scores for rigor of development and because they effectively targeted four to five domains. Therapeutic exercises, patient education, Transcutaneous Electrical Nerve Stimulation, acupuncture, orthoses and insoles, heat and cryotherapy, patellar tapping and weight control are commonly recommended for the non-pharmacological management of OA by the high-quality CPGs. It was noted that common recommendations were found by the majority of the CPGs; however, the strength of the recommendations varied between the CPGs.
In the past, OA CPGs have been evaluated with the AGREE I instrument –, , . To our knowledge, the AGREE II instrument has been used to appraise pharmacological and non-pharmacological CPGs for the management of OA; however, no quality scoring was presented in the literature . Therefore, the results of this review can only be compared partially with previous literature for CPGs on non-pharmacological and pharmacological interventions.
Assessment of the quality for the included CPGs
Domain 1: Scope and purpose.
All the included CPGs obtained a high (>60%) AGREE II quality score for domain 1 (scope and purpose), except for Peter et al. , Kjeken et al.  & OARSI  CPGs and all three CPGs were not able to effectively target this domain at 56%. The score of 56% was due to failure to provide detailed descriptions of the health question covered by the CPGs (item 2) and the target population (item 3). For a guideline to obtain a high score for item 2, the Population Intervention Comparison Outcome PICO model should be used when describing the health question(s) covered , –. The series of Ottawa – and the Philadelphia panels  CPGs were the only ones that described the population using the PICO model . Finally, although the CPGs did provide the overall objective, sufficient details about the objective were not provided to the readers . These results are consistent with previous reviews conducted with the AGREE I instrument on OA CPGs –, , where the scope and purpose was effectively addressed by a majority of the OA CPGs
Domain 2: Stakeholder Involvement.
All the CPGs, except for EULAR –, Roddy et al.  & Kjeken et al. , described the stakeholder involvement. The Kjeken et al.  CPG was targeted to occupational therapists, but the CPG lacked credentials on the professionals who were involved in the development process. In addition, both EULAR – and Roddy et al.  were graded low for items 5 & 6, as they did not clearly define the target users of the CPG. Although the majority of the CPGs provided information about the development groups, most did not provide details on whether patient preferences and views were considered during the recommendation development phase , –. The series of Ottawa Panel –, NICE , RACGP  and EULAR  CPGs were the only ones which considered patient views and preferences during the development phase of the CPGs. Both the series of Ottawa Panel Guideline – and EULAR  CPGs chose patients with OA to be part of the panel. The NICE  guideline consisted of a patient with OA, a consumer expert, researcher and organization representatives on the guideline development group and the RACGP  had patient representatives in the RACGP  working group , , –, . Previous reviews by Poitras et al. , Misso et al  and Penchaz et al.  with the AGREE I also had a significant low quality score for this domain as a result of the majority of the CPGs not reporting seeking patient views and preferences.
Domain 3: Rigour of development.
This domain obtained a relatively low scoring for the majority of the CPGs, but more specifically for Kjeken et al. ; EULAR – and Roddy et al. . For all of the CPGs considered, systematic literature searches were conducted and the evidence found was rated using a grading system , –. Some of the CPGs provided search strategies in the document (i.e. appendix and reference) – and a few provided online links to the literature , , however the links were not functioning. The grading systems differed for each CPG. Peter et al.  used the EMBRO system; EULAR –, Kjeken et al.  and Roddy et al.  used an evidence hierarchy; ACR  used the GRADE system (www.cochrane.org); RACGP  used the SIGN appraisal tools and NHMRC assessment; NICE  used the NICE method; AAOS  used AMSTAR and evidence hierarchy; OARSI  used Oxman and Guyatt and the Jadad scale, and the series of Ottawa Panel CPGs – used the Jaded scale. Kjeken et al. , EULAR – and Roddy et al.  all received a score lower than 60% ranging from (45%–57%). These CPGs had insufficient information on how the body of evidence was evaluated for bias (item 9), the strengths and limitations of the body of evidence (item 9), how the development process influenced the recommendations (item 10), and a procedure for updating the guideline (item 14). Since a majority of the CPGs failed to provide detailed information on item 14 for whether it would be updated and/or the updating procedures, this resulted in an overall lower score in domain 3. This review's results are consistent with previous reviews on appraisals with AGREE I –, . A majority of the CPGs provided recommendations with evidence, highlighted key recommendations, and considered health benefits and harm when formulating the recommendation. However, the mean quality score for rigor of development from AGREE I for the two previous reviews was found to be 47% (13) and 43% (14) significantly lower than the mean of 63%±11% based on the AGREE II instrument. A rationale for the difference in scoring can be attributed to using a rating scale from 1–7 as opposed to 1–4 in AGREE I instrument.
Domain 4 (clarity of presentation).
All the CPGs effectively addressed this domain and most received a relatively high score (75%–97%). The NICE  and the RACGP  CPGs consisted of clinical algorithms, which may facilitate clinical decision making , . Physiotherapy (PT) and occupational therapy (OT) are rehabilitation professions which provide a large spectrum of non-pharmacological interventions to patients with OA. Consequently more precision about PT and OT treatment modalities needs to be provided, especially regarding specific therapeutic interventions; CPGs should not refer to the professions of PT and OT as interventions. However, the AGREE II scores are consistent with previous reviews –, , domain 4: clarity and presentation was the most effectively addressed out of all the domains.
Domain 5 (applicability).
This domain obtained the lowest AGREE II scores for all the CPGs on OA, as most CPGs only describe a pilot study without providing results, or do not mention an implementation strategy besides dissemination or publication. Compared to all of the other CPGs, the NICE  guideline scored the highest as it provided an implementation section which consisted of algorithms and online resources for stakeholders . None of the CPGs effectively addressed all of the items in this domain. Reviews conducted by Poitras et al. , Misso et al. , and Penchaz et al.  also found a clear weakness for this domain, as a majority of the CPGs failed to provide a strategy of dissemination and implementation. In addition, this domain also had the lowest scoring for all three reviews –, .
Domain 6 (Editorial independence).
None of the CPGs effectively addressed all of the items for this domain. While many of the CPGs mentioned the funding body, very few were able to provide information on competing interests of the development group except AAOS  and OARSI . While many of the CPGs mentioned the funding body, very few failed to provide information on competing interests of the development group. Similar to the results, the reviews conducted by Poitras et al. , Misso et al. , and Penchaz et al.  found this domain to be poorly addressed.
Strength of recommendations
A common theme found when assessing the quality of the CPGs was that the majority of interventions, with the strongest evidence, looked at both pharmacological and non-pharmacological interventions –, . However, minority interventions or infrequently used interventions were considered as recommendations in CPGs that only looked at non-pharmacological interventions. The series of Ottawa Panel CPGs – was an exception where both the common and infrequent interventions were addressed.
Implications of score
There is existing high-quality CPGs for the non-pharmacological management of mild to moderate OA. However, the dissemination as well as the implementation of non-pharmacological CPGs in rheumatology is a challenging issue. This is also reflected in the CPG development, and observed with the lower ratings (0%–56%) obtained for applicability of AGREE II (domain 5) for the 11 CPGs included in this review (Table 3). This observation was made over the years regarding the low application rate of non-pharmacological interventions among health professionals in rheumatology . Indeed, family doctors and rheumatologists – prescribed less than 50% (30%–45%) physical interventions, not necessarily recommended, such as aerobic and range of motion exercises, orthoses, and assistive devices for ambulation, acupuncture, and OT or energy conservation. Surprisingly, very low medical referral rates are observed for patient education (29%), as well as for arthritis self-management programs (3%) , although these interventions are recommended by most CPGs (Table S1). Furthermore, individuals with arthritis adopted non-pharmacological interventions more frequently by themselves than prescribed by health professionals .
The AGREE II instrument is intended to be an improved version of the original AGREE I instrument –. The AGREE II instrument was available in 2009 with validation studies conducted in 2010. –. Despite this update, more recent CPGs published after this date –, ,  did not benefit from the improved instrument in their overall quality assessment. This is found for items in domain 5 (applicability), where more detail is required for tools, facilitators, and barriers for application; and domain 6 (editorial independence), where CPGs must provide details on the description and methods by which potential competing interests were identified.
The AGREE II instrument focuses on the methodological process for guideline development in domain 3. Although this is important to consider, the scorings received in domain 3 do not ensure that the recommendations are valid because the process to develop the guideline does not equate with its quality . For example, the AGREE II assesses in domain 3 if the literature for the CPG was systematically searched, but fails to consider if the recommendations of each intervention were analyzed using a quantitative approach. Domain 3 also failed to consider if the primary comparative controlled trials underlying the CPGs have high-quality standards. The studies included in the CPGs should be assessed by instruments such as Jadad  and Pedro scale ,  for CCSs and Amstar  for systematic reviews.
The grading system and the potential subjectivity could lead to question interpretation and could be considered as a weakness of the AGREE II. There is a user manual for AGREE II which the appraisers can use to evaluate the CPGs; however, when only some of the criteria for an item are met the appraiser must use their best judgment to score the item.
The weight of each item per domain is also problematic. This was especially seen with the scoring for domain 3, where all six items for the domain were regarded equally and as a result the majority of the CPGs received a lower score for this domain. Thus, even if the CPG received higher scores and effectively addressed the other items (i.e. the systematic methods used, criteria for selecting the evidence, method for formulating the recommendations etc.), the overall quality percentage was deemed low because they failed to address item 14 (a procedure for updating the guideline). This was especially seen with the series of Ottawa Panel CPGs –. This guideline received high scores for the five out of the six items in domain three, but only scored 65% on the overall quality of rigor of development.
The 7-point scale used in the AGREE II instrument is based on the idea that if all the elements of a particular item are fully addressed, then it is given a score of seven for that particular item. Conversely, if none of the elements were present then it was given a score of 1. An initial score of 1 (absence of information) is considered a systematic error, because AGREE II does not consider “not applicable” as a response according to its scoring system.
There is a potential publication bias, as only CPGs published in English were chosen according to the selection criteria. In addition, the recommendations based on the CPGs are difficult to establish because there is a contradiction between the strength of the recommendation. CPGs often referencing the same non-pharmacological studies will grade the strength differently, ranging from insufficient evidence to strongly recommended. A rationale for this inconsistency could be attributed to the panel's difficulty in dealing with conflicting data from primary methods. A technique to address this is to engage in a quantitative method such as the Cochrane Collaboration methodology to resolve conflicting results between two RCTs.
From the total seventeen CPGs included, based on the AGREE II scoring, we found ten good quality CPGs –, –,  where the rigor of development was >60%. There are good-quality CPGs available for health professionals. There was consensus for some of the recommendations, such as therapeutic exercises, patient education, Transcutaneous Electrical Nerve Stimulation, acupuncture, orthoses and insoles, heat and cryotherapy, patellar tapping and weight control for the management of OA.
Recommendations for the Management of Osteoarthritis. ACR: American College of Rheumatology; EULAR: The European League against rheumatism; NICE: National Institute for health and Clinical Excellence; OARSI: Osteoarthritis Research Society International; PGrip: People Getting a Grip on Arthritis RACGP: Royal Australian College of General Practitioners TENS: Transcutaneous Electric Nerve Stimulation.
Prisma flow diagram of included CPGs.
Prisma 2009 Checklist.
Literature Search Strategy.
The Eligibility Criteria.
The authors of this paper would like to thank Ms. Christine Smith and Ms. Sabrina Cavallo for providing valuable comments, ideas and assistance in the development of the manuscript and for their hard work and dedication in editing the final draft of the manuscript. The authors would also like to acknowledge and thank the support of Ms. Jennifer Burt, President, the Arthritis Health Professions Association (AHPA) and Ms. Carol Miller, Director, Practice and Research, the Canadian Physiotherapy Association.
Conceived and designed the experiments: LB PR KTA SP JK LC GP LL GD JM. Analyzed the data: LB PR KTA. Wrote the paper: LB PR KTA SP JK LC GP LL GD JM.
- 1. Lawrence RC, Helmick CG, Arnett FC, Deyo RA, Felson DT, et al. (1998) Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States. Arthritis Care Res 41: 778–799.
- 2. Fransen M, Bridgett L, March L, Hoy D, Penserga E, et al. (2011) The epidemiology of osteoarthritis in Asia. Int J Rheum Dis 14: 113–121.
- 3. Woolf AD, Pfleger B (2003) Burden of major musculoskeletal conditions. Bulletin of the World Health Organization 81: 646–656.
- 4. Ay S, Tur B, Kucukdeveci A (2008) Evaluation of Disability in patients with degenerative and inflammatory arthritis. Int J Rehabil Res 31: 159–163.
- 5. Kotlarz H (2010) Osteoarthritis and absenteeism costs: Evidence from US national survey data. J Occup Environ Med 52: 263.
- 6. Sakalauskiene G, Jauniskiene D (2010) Osteoarthritis: Etiology, Epidemiology, Impact on the Individual and Society and the Main Principles of Management. Medicina (Kaunas) 46: 790–797.
- 7. Bennell KL, Hunter DJ, Hinman RS (2012) Management of osteoarthritis of the knee. BMJ doi: 10.1136/bmj.e4934.
- 8. Richmond JC (2008) Surgery for Osteoarthritis of the Knee. Rheum Dis Clin North Am 39: 203–211.
- 9. Lützner J, Kasten P, Günther KP, Kirschner S (2009) Surgical options for patients with osteoarthritis of the knee. Nat Rev Rheumatol 5: 309–316.
- 10. Iwamoto J, Sato Y, Takeda T, Matsumoto H (2011) Effectiveness of exercise for osteoarthritis of the knee: A review of the literature. World J Orthop 2: 37–42.
- 11. Cottrell JE, Jonas M, Bergsten U, Blaas E, de la Torre Aboki J, et al. (2013) The nurse's role in addressing unmet treatment and management needs of patients with rheumatoid arthritis: Delphi-based recommendations. Int J Nurs Knowl 24: 66–76 doi: ;10.1111/j.2047-3095.2013.01231.x; 10.1111/j.2047-3095.2013.01231.x.
- 12. Porcheret M, Jordan K, Croft P (2007) Primary Care Rhumatology Society (2007) Treatment of knee pain in older adults in primary care: Development of an evidence-based model of care. Rheumatology (Oxford) 46: 638–648 doi: 10.1093/rheumatology/kel340.
- 13. Misso ML, Pitt VJ, Jones KM, Barnes HM, Piterman L, et al. (2008) Quality and consistency of clinical practice guidelines for diagnosis and management of osteoarthritis of the hip and knee: a descriptive overview of published guidelines. MJA 189: 394–399.
- 14. Poitras S, Avouac J, Rossignol M, Avouac B, Cedraschi C, et al. (2007) A critical appraisal of guidelines for the management of knee osteoarthritis using Appraisal of Guidelines Research and Evaluation criteria. Arthritis Res Ther 9: 1–12.
- 15. Toupin KA, Benkhalti MJ, Paterson G, Ueffing E, Rader T et al.. (2013) Clinical practice guidelines for osteoarthritis. In: Martel-Pelletier J, Pelletier JP, editors. Addressing unmet needs in osteoarthritis. London: Future Medicine Ltd. pp. 122–138.
- 16. Brand CA, Ackerman IN, Bohensky MA, Bennell KL (2013) Chronic Disease Management – A Review of Current Performance Across Quality of Care Domains and Opportunities for Improving Osteoarthritis Care. Rheum Dis Clin N Am 39: 123–143.
- 17. Brandt KD (2004) Non-surgical treatment of osteoarthritis: A half century of “advances. Ann Rheum Di 63: 117–122.
- 18. Rannou F (2010) Non-pharmacological approaches for the treatment of osteoarthritis. Best Pract Res Clin Rheumatol 24: 93.
- 19. Moher D, Liberati A, Tetzlaff J, Altman DG (2009) PRISMA Group (2009) Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. J Clin Epidemiol 62: 1006–1012 doi: ;10.1016/j.jclinepi.2009.06.005; 10.1016/j.jclinepi.2009.06.005.
- 20. Physiotherapy Evidence Database (PEDro). Available: http://www.pedro.fhs.usyd.edu.au/index.html. Accessed 2013 May 23.
- 21. National Guideline Clearinghouse. Available: http://www.guideline.gov/. Accessed 2013 May 23
- 22. Guidelines International Network (GIN). Available: http://www.g-i-n.net/. Accessed 2013 May 23
- 23. Trip Database. Available: http://www.tripdatabase.com/. Accessed 2013 October 11
- 24. Yan J, Min J, Zhou B (2012) Diagnosis of pheochromocytoma: A clinical practice guideline appraisal using AGREE II instrument. J Eval Clin Pract doi: ;10.1111/j.1365-2753.2012.01873.x; 10.1111/j.1365-2753.2012.01873.x.
- 25. Fleiss JL (1986) The design and analysis of clinical experiments. Wiley-Intersciences, New York. pp. 1–128.
- 26. American Academy of Orthopaedic Surgeons (2008) Treatment of osteoarthritis of the knee (non-arthroplasty). American Academy of Orthopaedic Surgeons 263.
- 27. Hochberg MC, Altman RD, April KT, Benkhalti M, Guyatt G, et al. (2012) American College of Rheumatology 2012 recommendations for the use of nonpharmacologic and pharmacologic therapies in osteoarthritis of the hand, hip, and knee. Arthritis Care Res 4: 465–467.
- 28. European League Against Rheumatism (2003) EULAR Recommendations 2003: an evidence based approach to the management of knee osteoarthritis: Report of a Task Force of the Standing Committee for International Clinical Studies Including Therapeutic Trials (ESCISIT). Ann Rheum Dis 62: 1145–1155.
- 29. European League Against Rheumatism (2004) EULAR evidence based recommendations for the management of hip osteoarthritis: report of a task force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT). Ann Rheum Dis 64: 669–681.
- 30. European League Against Rheumatism (2007) EULAR evidence based recommendations for the management of hand osteoarthritis – report of a task force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT). Ann Rheum Dis 66: 377–388.
- 31. Fernandes L, Hagen KB, Bijlsma JW, Andreassen O, Christensen P, et al. (2013) EULAR recommendations for the non-pharmacological core management of hip and knee osteoarthritis. Ann Rheum Dis 72: 1125–35 doi: 10.1136/annrheumdis-2012-202745.
- 32. Kjeken I (2011) Occupational therapy-based and evidence-supported recommendations for assessment and exercises in hand osteoarthritis. Scand J Occup Ther 18: 265–281 doi: 10.3109/11038128.2010.514942.
- 33. Conaghan PG, Dickson J, Grant RL (2008) Care and management of osteoarthritis in adults: Summary of NICE guidance. doi: 10.1136/bmj.39490.608009.AD
- 34. Zhang W, Moskowitz RW, Nuki G, Abramson S, Altman RD, et al. (2008) OARSI recommendations for the management of hip and knee osteoarthritis, Part II: OARSI evidence-based, expert consensus guidelines. Osteoarthritis Cartilage 16 (2) 137–62.
- 35. Brosseau L, Wells G, Tugwell P, Egan M, Dubouloz CJ, et al. (2005) Ottawa Panel evidence-based clinical practice guidelines for therapeutic exercises and manual therapy in the treatment of osteoarthritis. Phys Ther 85: 907–971.
- 36. Brosseau L, Egan M, Wells GA, Tugwell P, Dubouloz CJ, et al. (2010) Ottawa Panel evidence-based clinical practice guidelines for patient education programs in the treatment of osteoarthritis. Health Educ J 70: 318–358.
- 37. Brosseau L, Wells GA, Tugwell P, Egan M, Dubouloz CJ, et al. (2011) Ottawa Panel evidence-based clinical practice guidelines for the management of osteoarthritis in adults who are obese or overweight. Phys Ther 91 (6) 843–6.
- 38. Loew L, Brosseau L, Wells GA, Tugwell P, Kenny GP, et al. (2012) Ottawa panel evidence-based clinical practice guidelines for aerobic walking programs in the management of osteoarthritis. Archives of Physical Medicine and Rehabilitation 93 (7) 1269–1285.
- 39. People Getting a Grip on Arthritis Program (PGRIP). Available: www.arthritis.ca/peoplegettingagrip. Accessed 2013 May 1.
- 40. Albright J, Allman R, Bonfiglio R, Conill A, Dobkin B, et al. (2001) Philadelphia Panel evidence-based clinical practice guidelines on selected rehabilitation interventions: overview and methodology. Phys Ther 81: 1629–40.
- 41. Albright J, Allman R, Bonfiglio R, Conill A, Dobkin B, et al. (2001) Philadelphia Panel evidence-based clinical practice guidelines on selected rehabilitation interventions for knee pain. Phys Ther 81: 1675–700.
- 42. Peter WF, Jansen MJ, Hurkmans EJ, Bloo H, Dekker J, et al. (2011) Physiotherapy in hip and knee osteoarthritis: Development of a practice guideline concerning initial assessment, treatment and evaluation. Acta Reumatologica Portuguesa 36: 268–281.
- 43. Roddy E, Zhang W, Doherty M, Arden NK, Barlow J, et al. (2005) Evidence-based recommendations for the role of exercise in the management of osteoarthritis of the hip or knee–the MOVE consensus. Rheumatology (Oxford) 44: 67–73 doi: 10.1093/rheumatology/keh399.
- 44. Brand C, Buchbinder R, Wluka A, Ruth D, McKenzie S, et al.. (2009) Guideline for the Non-Surgical Management of Hip and Knee Osteoarthritis, RACGP. Melbourne: The Royal Australian College of General Practitioners. 68 p.
- 45. Pencharz JN, Grigoriadis E, Jansz GF, Bombardier C (2002) A critical appraisal of clinical practice guidelines for the treatment of lower-limb osteoarthritis. Arthritis Res 4: 36–44.
- 46. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS (1996) Evidence based medicine: what it is and what it isn't. Clin Orthop Relat Res 455: 3–5.
- 47. Appraisal of Guidelines Research and Evaluation website. Available: www.agreetrust.org. Accessed 2013 May 1.
- 48. DeHaan MN, Guzman J, Theodore Bayley M, Bell MJ (2007) Knee osteoarthritis clinical practice guidelines—how are we doing? J Rheumatol 34: 2099–2105.
- 49. Li LC, Maetzel A, Pencharz JN, Maguire L, Bombardier C, et al. (2004) Use of mainstream nonpharmacologic treatment by patients with arthritis. 51: 203–209 doi: 10.1002/art.20244.
- 50. Glazier RH, Badley EM, Wright JG, Coyte PC, Williams JI, et al. (2003) Patient and provider factors related to comprehensive arthritis care in a community setting in Ontario, Canada. J Rheumatol 30: 1846–1850.
- 51. Jordan KM, Sawyer S, Coakley P, Smith HE, Cooper C, et al. (2004) The use of conventional and complementary treatments for knee osteoarthritis in the community. Rheumatology (Oxford, England) 43 (3) 381–384 doi: 10.1093/rheumatology/keh045.
- 52. Denoeud L, Mazieres B, Payen-Champenois C, Ravaud P (2005) First line treatment of knee osteoarthritis in outpatients in France: Adherence to the EULAR 2000 recommendations and factors influencing adherence. Ann Rheum Dis 64: 70–74 doi: 10.1136/ard.2003.015263.
- 53. Bekkering GE, Hendriks HJ, van Tulder MW, Knol DL, Hoeijenbos M, et al. (2005) Effect on the process of care of an active strategy to implement clinical guidelines on physiotherapy for low back pain: a cluster randomised controlled trial. Qual Saf Health Care 14: 107–112.
- 54. Brouwers MC, Kho ME, Browman GP, Burgers JS, Cluzeau F, et al. (2010) AGREE Next Steps Consortium. Development of the AGREE II:part 1: performance, usefulness and areas for improvement. CMAJ 182: 1045–1052.
- 55. Brouwers MC, Kho ME, Browman GP, Burgers JS, Cluzeau F, et al. (2010) AGREE Next Steps Consortium. Development of the AGREE II: part 2: assessment of validity of items and tools to support application. CMAJ 182: 472–478.
- 56. Matthys J, De Meyere M, van Driel ML, De Sutter A (2007) Differences among international pharyngitis guidelines: not just academic. Ann Fam Med 5: 436–443.
- 57. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJM, et al. (1996) Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 17: 1–12.
- 58. Olivo SA, Macedo LG, Gadotti IC, Fuentes J, Stanton T, et al. (2008) Scales to assess the quality of randomized controlled trials: a systematic review. Phys Ther 88: 156–175.
- 59. Shea BJ, Grimshaw JM, Wells GA, Boers M, Andersson N, et al. (2007) Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Medical Research Methodology 7: 1–10.