Arthritis Care & Research

Vol. 75, No. 9, September 2023, pp 1897–1907

DOI 10.1002/acr.25101
© 2023 The Authors. Arthritis Care & Research published by Wiley Periodicals LLC on behalf of American College of Rheumatology.
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits
use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or
adaptations are made.

Core Recommendations for Osteoarthritis Care: A

Systematic Review of Clinical Practice Guidelines
Brooke Conley,1 Samantha Bunzli,2 Jonathan Bullen,3 Penny O’Brien,1 Jennifer Persaud,4
Tilini Gunatillake, Michelle M. Dowsey, Peter F. M. Choong,1 and Ivan Lin5
1 1

Objective. To evaluate the quality of clinical practice guidelines (CPGs) for interventions in management of osteo-
arthritis (OA) and to provide a synthesis of high-quality CPG recommendations.
Methods. Five databases (OvidSP Medline, Cochrane, Cumulative Index to Nursing and Allied Health Literature
[CINAHL], Embase, and the Physiotherapy Evidence Database [PEDro]) and 4 online guideline repositories were
searched. CPGs for the management of OA were included if they were 1) written in English and published from January
2015 to February 2022, focused on adults age ≥18 years, and met the criteria of a CPG as defined by the Institute of
Medicine; and 2) were rated as high quality on the Appraisal of Guidelines for Research and Evaluation II (AGREE II)
instrument. CPGs for OA were excluded if they were available via institutional access only, only addressed recommen-
dations for the system/organization of care and did not include interventional management recommendations, and/or
included other arthritic conditions.
Results. Of 20 eligible CPGs, 11 were appraised as high quality and included in the synthesis. Of interest were the
hip, knee, hand, and glenohumeral joints and/or polyarticular OA. Consistent recommendations were that care should
be patient centered and include exercise, education, and weight loss (where appropriate). Nonsteroidal antiinflamma-
tory drugs and surgical interventions were recommended for disabling OA that had not improved with nonsurgical care.
Hand orthoses should be recommended for patients with hand OA.
Conclusion. This synthesis of high-quality CPGs for OA management offers health care providers with clear,
simple guidance of recommended OA care to improve patient outcomes.

INTRODUCTION limitations and reduced quality of life (5,6). OA is associated with

substantial direct health care costs due to health care visits, diag-
Osteoarthritis (OA) is a degenerative joint disease that can nostic procedures, medications and surgery, and indirect costs
affect any joint, but it most commonly occurs in the hip, knee, related to lost workplace productivity (4,6).
and hand (1,2). Symptoms often include joint pain, stiffness, and Clinical practice guidelines (CPGs) are a set of health care
reduced range of movement (2). OA affects 303 million people recommendations developed by reviewing scientific literature
worldwide, with prevalence expected to increase with aging and consensus from an expert panel (7). The aim of CPGs is to
populations and rising obesity rates globally (3,4). OA is a leading guide health care decision-making, thereby reducing practice var-
cause of pain and disability among adults worldwide and inflicts a iability and improving patient outcomes (8,9). Several global CPGs
significant burden on the individuals affected, including activity have been published in recent years for the management of OA

Supported by the Australian Commonwealth Government (Arthritis Bullen, Bus.Info.Sys., GradDipPsychStudies, PhD: Curtin University, Perth,
Australia grant GA76063). Professor Dowsey’s work was supported by the Western Australia, Australia; 4Jennifer Persaud, BSc (Hons) Physio, MSc Sports
Australian National Health and Medical Research Council (Career Develop- and Ex Med, GradCert Social Impact: Arthritis and Osteoporosis Western
ment Fellowship grant APP1122526) and the University of Melbourne (Dame Australia and Sir Charles Gairdner Hospital, Perth, Western Australia,
Kate Campbell Fellowship). Professor Choong’s work was supported by the Australia; 5Ivan Lin, BSc(Physio), M. Manip. Ther., PhD, FACP: University of
Australian National Health and Medical Research Council (Practitioner Fellow- Western Australia and Geraldton Regional Aboriginal Medical Service, Gerald-
ship grant APP1154203). ton, Western Australia, Australia.
1
Brooke Conley, BExSci, MPhysio, Penny O’Brien, BHealthSci (Hons), PhD, Author disclosures are available at https://onlinelibrary.wiley.com/action/
Tilini Gunatillake, BSc (Hons), PhD, Michelle M. Dowsey, BHealthSci, MEpi, downloadSupplement?doi=10.1002%2Facr.25101&file=acr25101-sup-0001-
PhD, Peter F. M. Choong, MBBS, MD, FRACS, FAOrthA, FAAHMS, MAICD: The Disclosureform.pdf.
University of Melbourne, Melbourne, Victoria, Australia; 2Samantha Bunzli, Address correspondence via email to Brooke Conley, BExSci, MPhysio, at
BPhty (Hons), PhD: The University of Melbourne, Melbourne, Victoria, [email protected].
Australia, Griffith University, Nathan, Queensland, Australia, and Royal Bris- Submitted for publication October 25, 2022; accepted in revised form
bane and Women’s Hospital, Brisbane, Queensland, Australia; 3Jonathan February 7, 2023.

1897
1898 CONLEY ET AL

(e.g., guideline*.mp. or Practice Guideline/ or Guideline/) and OA

SIGNIFICANCE & INNOVATIONS were used (see Supplementary Appendix A, available on the Arthritis
• Eleven clinical practice guidelines for osteoarthritis Care & Research website at http://onlinelibrary.wiley.com/doi/10.
(OA) were appraised as high quality. 1002/acr.25101). Other arthritis conditions were included in the
• Consistent recommendations were that care should search terms as this systematic review is part of a wider body of work
be patient centered and include exercise, educa-
to inform arthritis management. Eligibility criteria are presented in
tion, and weight loss (where appropriate). Nonste-
roidal antiinflammatory drugs and surgical Table 1. The search included CPGs published between January
interventions are recommended for disabling OA 2015 and December 2020 and was updated to include CPGs pub-
that had not improved with nonsurgical care. Hand lished between December 2020 and February 14, 2022. This cutoff
orthoses should be recommended for patients with date was selected as CPGs >5 years old may be out of date (7).
hand OA.
• To implement recommendations in practice, future
priorities include identifying core skill sets and com- Protocol changes. In the original protocol, CPGs addres-
petencies among health care workers, developing sing 1 treatment modality (e.g., medication prescribing) were
training/education resources, and creating a frame- excluded. To improve comprehensiveness, the scope was
work to improve quality of OA care.
expanded during the study selection phase to include all OA man-
agement options. Due to time that has elapsed since the original
search, the original timeline (January 2015 and December 2020)
(10–29). These typically include nonpharmacologic (e.g., exercise and
was extended to include CPGs published up until February 2022.
education), pharmacologic (e.g., acetaminophen and nonsteroidal
antiinflammatory drugs [NSAIDs]), and surgical options such as total
joint replacement (10–29). However, uptake of recommendations Study selection. After importing search results into End-
from CPGs into practice is variable, especially for first-line, nonphar- Note (Clarivate), duplicates were removed and titles/abstracts were
macologic treatments such as exercise (30). Gaps between evidence uploaded into Covidence systematic review software (Veritas
Health Innovation; available at www.covidence.org). Titles and
and practice may in part be due to a lack of clarity about what is being
abstracts were screened by 2 independent reviewers (BC and TG
recommended and conflicting recommendations across different
or IL); any disagreements were resolved through consensus dis-
CPGs, a situation exacerbated when CPGs for OA are not developed
cussion with a third reviewer. Following this, full texts were
rigorously and their recommendations are less trustworthy (30–36).
uploaded into Covidence and screened through the same process.
As CPGs are costly to develop, this is an inefficient use of resources
and further adds to the confusion for clinicians (37). To encourage
the uptake of evidence and delivery of appropriate OA care, clinicians Data appraisal (quality assessment of guidelines).
require clear, consistent management recommendations (31). The Appraisal of Guidelines for Research and Evaluation II (AGREE
The purpose of this systematic review was to evaluate the II) instrument was used to assess CPG quality (38). This is an interna-
quality of the CPGs for the management of OA and to provide a tionally validated tool that has been widely applied in CPG research
synthesis of high-quality CPG recommendations. By synthesizing (38,39). It consists of 23 items grouped into 6 domains: 1) scope
recommendations across high-quality CPGs, the aim was to offer and purpose; 2) stakeholder involvement; 3) rigor of development;
health care providers with clear, simple guidance of recom- 4) clarity of presentation; 5) applicability; and 6) editorial indepen-
mended OA care to improve patient outcomes. dence. Seven reviewers (BC, SB, JB, PO, JP, TG, and IL) were

Table 1. Clinical practice guidelines (CPGs) selection criteria

MATERIALS AND METHODS Inclusion criteria
Published between January 2015 and February 14, 2022
Search strategy and eligibility criteria. This systematic For the interventional management of osteoarthritis
review was registered on the Open Science Framework (DOI 10. For adults (individuals age ≥18 years)
Published in the English language or has a complete English
17605/OSF.IO/UB3Y7) and followed the Preferred Reporting Items
language version available
for Systematic Reviews (PRISMA) guidelines. A database search of Is a CPG, as defined by inclusion of a systematic review of the
Medline, Cochrane, CINAHL, Embase, and the Physiotherapy Evi- literature, and developed by an expert multidisciplinary panel (33)
Represents an original body of work, i.e., not solely an adaptation
dence Database (PEDro), and a further search of 4 online guideline
or systematic review of existing guidelines
repositories (Guidelines International Network, National Health and Exclusion criteria
Medical Research Council, Agency for Health Care Research and Does not include interventional management recommendations
Quality, and the National Institute for Health and Care Excellence) Includes other arthritic conditions
Only addresses recommendations for the system/organization of
was conducted to identify all relevant CPGs. The search strategy care
was developed in consultation with a reference librarian. Medical Unavailable via institutional access, i.e., requires additional
subject headings and key words associated with CPGs payment
OA CARE CORE RECOMMENDATIONS 1899

provided with the AGREE II user manual and undertook the online spreadsheet. Extracted data comprised CPG characteristics
AGREE II practice exercise to participate in CPG quality appraisal (e.g., title, country of publication), methodology, and guideline
(38,40). In accordance with the AGREE II manual, each item was topic target users (see Supplementary Table 2, available on the
rated independently by 2 reviewers using a 7-point Likert scale rang- Arthritis Care & Research website at http://onlinelibrary.wiley.
ing from 1 (strongly disagree) to 7 (strongly agree) (38). To calculate com/doi/10.1002/acr.25101). From each CPG, extracted rec-
scores for each domain, the following formula was used: obtained ommendations were ranked as either “should do,” “could do,”
score – minimum possible score / maximum possible score – mini- “do not do,” or “uncertain” (see Supplementary Table 3, avail-
mum possible score (38). There is no uniform criterion for overall able on the Arthritis Care & Research website at http://
quality; the AGREE II developers recommend that research teams onlinelibrary.wiley.com/doi/10.1002/acr.25101). Recommen-
define their own criteria based on their own study context (38). For dation ratings were consistent with language used in the CPGs
the purposes of this review, and consistent with previous reviews in and definitions from a previous musculoskeletal systematic
musculoskeletal pain management, the authors defined a quality cut- review of CPGs (36) (Table 2). Language among the CPGs var-
off score of ≥60% of the maximum possible score in 3 domains ied, although recommendations were ranked according to the
deemed the most important for validity: stakeholder involvement same criteria, either the Grading of Recommendations Assess-
(domain 2); rigor of development (domain 3); and editorial indepen- ment, Development and Evaluation (GRADE) method
dence (domain 6) (36,37,41,42). CPGs that did not meet this defini- (10,12,13,20,22,26,29), Oxford Centre for Evidence-Based
tion were excluded. Medicine standards (13,24), or PEDro scores (17). Extracted
data and recommendation rankings were checked by 2 authors
Interrater agreement. The domain percentages and (SB and IL), and any discrepancies were resolved by consensus
overall quality rating (%) were independently calculated for each discussion between the 3 reviewers while consulting the original
reviewer. We defined acceptable interrater agreement as excel- citation.
lent with intraclass coefficient values of ≥80 and domain percent-
ages and an overall quality rating of ≤20% difference between
reviewers (43,44). Where variation of ≥20% between scores Narrative summary. A narrative summary was drafted by
existed, a consensus discussion took place with a third reviewer the first author (BC) and then reviewed and refined by 2 authors
engaged when necessary to agree on a rating (see Supplemen- (SB and IL). The summary detailed how many CPGs reported on
tary Table 1, available on the Arthritis Care & Research website an intervention, what the recommendations involved, and how
at http://onlinelibrary.wiley.com/doi/10.1002/acr.25101). consistent/inconsistent recommendations were across CPGs
regarding OA interventions (see Supplementary Table 4, available
Data extraction. Data extraction was performed by the on the Arthritis Care & Research website at http://onlinelibrary.
first author (BC) using a purpose-designed Excel (Microsoft) wiley.com/doi/10.1002/acr.25101).

Table 2. Recommendation classification, definition, and examples of terminology for each classification*
Recommendation classification Definition (37) Examples of terminology from CPGs
Should do “Should do” recommendations were those that the authors “Should do” (18) and “strongly
determined should be applied in all circumstances unless recommend” (11)
there is a rationale not to. These were based on strong
evidence, for example, multiple high-quality studies reporting
clinically relevant positive effects, benefits that outweigh risks,
or when in the opinion of CPG development group members
the benefits were unequivocal.
Could do “Could do” recommendations were those that the authors “Could be used” (18), “may be
determined could be applied depending on the beneficial” (9,18), “can” and
circumstances of individual patients. They were usually based “consider” (26)
on consistent evidence from multiple lesser-quality studies or
1 high-quality study and where benefits outweigh harms.
Do not do “Do not do” recommendations were those for which the “Are not recommended” (18), “do not
authors determined that there was strong evidence of no recommend” (11), and “do not offer”
benefit and/or harms outweighing benefits. (26)
Uncertain “Uncertain” recommendations were those for which the “Cannot recommend for or against” (9)
authors determined that there was no recommendation for and “unable to recommend either for
or against a practice because of incomplete or inconsistent or against” (11)
research findings. Not all CPGs provided uncertain
recommendations.
* CPG = clinical practice guideline.
1900 CONLEY ET AL

RESULTS those responsible for commissioning care (see Supplementary

Table 2, available at http://onlinelibrary.wiley.com/doi/10.1002/
Characteristics of included CPGs. Twenty CPGs met
acr.25101).
the eligibility criteria, and 11 were CPGs appraised as high quality
and included (11–13,17–19,21,24,25,27,29) (Figure 1). Most
CPGs were developed by medical societies (n = 9, 82%), while Quality of CPGs. The AGREE II quality assessment scores
the remaining were developed by an expert panel (n = 2, 18%). for each CPG are provided in Supplementary Table 1, available
Five CPGs were published by medical societies or expert panels at http://onlinelibrary.wiley.com/doi/10.1002/acr.25101. The
in the US (10,11,19,25,29), 4 in Europe (13,21,24,27), 1 in quality assessment results of each guideline included in the sys-
Canada (17), and 1 in Australia (12). Of interest were the hip tematic review are provided in Table 3. Those that were excluded
(n = 6), knee (n = 7), hand (n = 4), and glenohumeral (n = 1) joints based on not achieving a high-quality cutoff score are presented
and/or polyarticular OA (n = 1). Target users included health pro- in Supplementary Table 1. The quality of the included CPGs was
fessionals, decision/policy makers, patients, their families, the assessed across the following 6 domains: scope and purpose
pharmaceutical industry, health insurance companies, and (range 75–100%); stakeholder involvement (range 58–89%); rigor

Identification of studies via databases and registers Identification of studies via other methods

Records identified from:

Databases (n = 2,505) Records removed before screening:
Identification

Medline (n = 633) Duplicate records removed Records identified from:

CINAHL (n = 863) (n = 813) Websites (n = 9)
Embase (n = 982) Records marked as ineligible by Organizations (n = 5)
PEDro (n = 27) automation tools (n = 0) Citation searching (n = 4)
Records removed for other etc.
reasons (n = 0)

Reports sought for Reports not

Records screened Records excluded
retrieval retrieved
(n = 1,692) (n = 1,571)
(n = 9) (n = 0)
Screening

Reports sought for Reports assessed

Reports not retrieved Reports excluded:
retrieval for eligibility
(n = 0) Duplicates (n = 9)
(n = 121) (n = 9)

Reports assessed for Reports excluded (n = 66):

eligibility Not a CPG (n = 50)
(n = 55) Methods not described (n = 3)
Wrong patient population (n = 10)
Not published in English (n = 2)
Other (n = 1)

Studies included in review

OA (n = 20), high quality (n = 11)
Included

Reports of included studies

(n = 0)
Other conditions excluded from
this review (n = 35)

Figure 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 flow diagram for new systematic reviews that
included searches of databases, registers, and other sources. CINAHL = Cumulative Index to Nursing and Allied Health Literature;
CPG = clinical practice guideline; OA = osteoarthritis; PEDro = Physiotherapy Evidence Database.
OA CARE CORE RECOMMENDATIONS 1901

of development (range 59–96%); clarity of presentation (range index [BMI] ≥25 kg/m2) or obese (BMI ≥30 kg/m2) with hip and/or
53–100%); applicability (range 2–42%); and editorial indepen- knee OA (12,13,19,25). People with OA should be educated about
dence (range 33–100%). The mean ± SD AGREE II scores for the importance of maintaining a healthy body weight, while those
each item, domain, and overall scores across all guidelines are who are overweight or obese should be encouraged to achieve a
displayed in Supplementary Table 5, available on the Arthritis minimum weight loss target of 5.0–7.5% of body weight, with
Care & Research website at http://onlinelibrary.wiley.com/doi/ greater weight loss being linked to symptomatic benefits (12).
10.1002/acr.25101. The domain with the lowest mean ± SD Hand orthosis. Three CPGs strongly recommended the use
score was applicability (21.14% ± 15.0%), and the highest of a hand orthosis for OA of the carpometacarpal joint and condi-
mean ± SD score was for scope and purpose (88.77% ± 9.8%). tionally recommended it for OA of other hand joints (13,24,25).
Two CPGs stated that hand orthoses are suitable for both short-
Consensus recommendations (“should do”). After term and long-term use as they provide symptom relief, improve
synthesis (see Supplementary Table 4, available at http:// function, and prevent progression of degenerative changes (13,24).
onlinelibrary.wiley.com/doi/10.1002/acr.25101), 7 “should do” Patient-centered care. Two CPGs strongly recommended
recommendations were identified. The following recommenda- that care be patient centered for people with OA of the knee,
tions were all found in at least 2 CPGs, where the majority strongly hip, and/or hand (13,24). This included shared decision-making
recommended the intervention. between the patient and health professional and care that is indi-
Exercise. Eight CPGs strongly recommended strengthening, vidualized to the patient’s circumstances.
aerobic exercise, and tai chi exercise therapy for management of Surgery. Two CPGs strongly recommended considering sur-
knee, hip, polyarticular, and/or hand OA (12,13,18,19,21,24,25,27). gery for people with hip, knee, and hand OA in certain circum-
CPGs recommended several modes of exercise therapy, acknowl- stances (13,24). The patient should have radiographic evidence
edging that there is currently no consensus on the type of exercise of OA, marked disability, and reduced quality of life, and other
that elicits the greatest benefit (12). Programs should be individual- treatment modalities should have been unsuccessful in relieving
ized and progressively overloaded with frequency, duration, and pain (13,24).
intensity consistent with the patient’s preference and capability and
the availability of local facilities (12,13). Consensus recommendations (“could do”). The fol-
Education. Five CPGs strongly recommended patient edu- lowing recommendations were found in at least 2 CPGs in which
cation for managing knee, hip, hand, and polyarticular OA the majority conditionally recommended (or where there was an
(13,19,21,24,25). They recommended that education be an even number of strongly and conditionally recommended recom-
ongoing intervention that is patient centered and include informa- mendations) that these could apply in a given patient’s circum-
tion to enhance understanding about OA, its management stances: balance exercises; yoga; weight management and
options, education and training in exercise therapy, ergonomic exercise; cognitive behavioral therapy; assistive devices;
principles, and pacing and assistive devices (13,21,24). ultrasound-guided injections; duloxetine; and glucocorticoid
NSAIDs. Five CPGs strongly and 2 CPGs conditionally injections for knee and hand OA. In surgical contexts, preopera-
recommended the use of oral NSAIDs for people with knee, hip, tive physical therapy and postoperative physical therapy or exer-
hand, and/or polyarticular OA unless contraindicated (11– cise can be recommended after joint replacement surgery.
13,19,21,24,25). CPGs recommended that clinicians prescribe a General and neuraxial anesthesia and tranexamic acid could be
low dose for a short period of time and discontinue if not effective, considered during surgery (see Supplementary Table 4, available
monitoring for side effects or adverse events (12,24). at http://onlinelibrary.wiley.com/doi/10.1002/acr.25101).
Four CPGs strongly recommended the use of topical
NSAIDS for knee, hip, and/or hand OA (13,19,24,25). Two CPGs Consensus recommendations (“do not do”). After
conditionally recommended topical NSAIDS for patients with synthesis (see Supplementary Table 4), 5 “do not do” recommen-
hand, knee, hip, and/or polyarticular OA and some comorbidities dations were identified (Table 4). The following recommendations
(21,25). One CPG was unable to recommend for or against the were found in at least 2 CPGs, where the majority recommended
use of topical NSAIDS for people with knee and/or hip OA (12). against the intervention.
However, the authors stated that it might be reasonable to trial Therapeutic ultrasound and pharmacologic interventions
topical NSAIDs for a short period and then discontinue use if not (bisphosphonates, colchicine, methotrexate, diacerein). Three
effective. Topical NSAIDs are seen to be safe and effective and CPGs recommended against the use of therapeutic ultrasound
should be recommended for older adults (>75 years) with only a for people with knee, hip, and/or polyarticular OA (12,21,27).
few symptomatic joints (13). Clinicians should monitor for side Two CPGs recommended against the use of bisphosphonates,
effects or adverse events (12,21). colchicine, hydroxychloroquine, and methotrexate in people
Weight loss. Four CPGs strongly recommended weight loss with knee, hip, and/or hand OA (12,25). Similarly, 1 CPG recom-
or management for people who are either overweight (body mass mended against the use of biologic disease-modifying
 1902

Table 3. Clinical practice guidelines included in the systematic review*

Domains 2,3,
Domain 1, Domain 2, Domain 3, Domain 4, Domain 6, Over and 6
scope stakeholder rigor of clarity of Domain 5, editorial all assessments combined
Author, year (ref.) Title/type and purpose involvement development presentation applicability independence core value
RACGP, 2018 (12) RACGP 97 58 81 89 35 79 83 73
Bannuru et al, 2019 OARSI 89 72 59 89 8 100 58 77
(21)
Kolasinski et al, 2020 ACR 97 89 85 86 21 63 75 79
(25)
AAOS, 2020 (10) AAOS (GH joint) 94 61 90 69 15 92 75 81
AAOS, 2017 (11) AAOS (hip OA) 97 61 92 53 42 71 75 75
Ariani et al, 2019 (13) ISR 81 64 65 72 2 67 58 65
Kloppenburg et al, EULAR 78 81 77 81 6 79 67 79
2019 (24)
Van Doormaal et al, – 75 67 81 75 42 33 58 60
2020 (27)
Brosseau et al, 2018 OTTAWA 75 61 75 56 6 92 92 76
(17)
AAOS, 2015 (29) AAOS (knee OA 100 81 96 100 21 100 92 92
surgical)
AAOS, 2021 (19) AAOS (knee OA non- 94 61 96 100 35 92 92 83
arthroplasty)
* Values are the domain scores in % on the Appraisal of Guidelines for Research and Evaluation II (AGREE II) instrument. AAOS = American Academy of Orthopaedic Surgeons;
ACR = American College of Rheumatology; GH = glenohumeral; ISR = Italian Society for Rheumatology; OA = osteoarthritis; OARSI = Osteoarthritis Research Society International;
OTTAWA = Ottawa Panel; RACGP = Royal Australian College of General Practitioners; Ref. = reference.
CONLEY ET AL
OA CARE CORE RECOMMENDATIONS 1903

antirheumatic drugs for people with hand OA (24). Two CPGs rec- further research is warranted to determine structural effects,
ommended against the use of diacerein for people with knee patients’ suitability, and cost-to-benefit ratio (13).
and/or hip OA (12,21). Postsurgical continuous passive motion (CPM) and postsur-
Glucosamine and chondroitin combined. Two CPGs recom- gical cryotherapy devices. Two CPGs recommended against the
mended against the use of glucosamine and chondroitin for knee, use of CPM after total joint replacement for patients with knee
hip, and polyarticular OA (12,21). One CPG was unable to recom- and/or hip OA, as research found no improvement in outcomes
mend for or against their combined use in people with OA of the (27,29). One CPG recommended against the use of cryotherapy
glenohumeral joint (10). Moreover, 1 CPG conditionally recom- devices for patients after total knee arthroplasty (TKA) (29). In con-
mended this intervention for people with knee OA, noting that trast, 1 CPG conditionally recommended the use of cryotherapy
or cold packs following total shoulder replacement while acknowl-
Table 4. Consensus recommendations* edging that this decision was based on the opinion of the working
Should do group and not strong/reliable evidence (10).
Exercise therapies (strengthening, aerobics, and/or tai chi)
Education
Weight loss Recommendations with no consensus. The following
Hand orthosis were conflicting recommendations found in at least 2 CPGs:
Patient-centered care
aquatic therapy; balneotherapy; massage therapy; manual ther-
Nonsteroidal antiinflammatory drugs (oral and topical)
Surgery apy; acupuncture; dry needling; heat and cold therapy; electro-
Could do therapy; taping and braces; shoe orthotics; footwear; opioids;
Balance exercises injections; topical capsaicin; glucosamine and chondroitin individ-
Yoga
Assistive devices ually or combined for OA of the glenohumeral joint; acetamino-
Weight management and exercise phen; and nutraceuticals (see Supplementary Table 4).
Cognitive behavioral therapy
Glucocorticoid injection (knee and hand OA)
Ultrasound-guided injections DISCUSSION
Duloxetine
Preoperative physical therapy Following quality assessment, 9 CPGs were rated as low
Tranexamic acid
quality, and 11 CPGs were high quality and included in the final
Neuraxial anesthesia
Pre- and postoperative physical therapy synthesis. Overall, CPGs recorded the highest score for the
Do not do AGREE II domain “scope and purpose” and the lowest score for
Therapeutic ultrasound the domain “applicability.” This is consistent with the findings of
Bisphosphonates
Colchicine similar systematic reviews (37,45). The AGREE II “applicability”
Methotrexate domain assesses whether CPGs provide advice and/or tools for
Diacerein how to apply the guideline in practice, considers the facilitators,
Glucosamine and chondroitin combined (hip and polyarticular
barriers, and resource implications, and includes monitoring
OA)
Postsurgical continuous passive motion and cryotherapy devices and/or auditing criteria (38). Poor applicability has been identified
No consensus as a barrier to the uptake of CPG recommendations into practice
Aquatic therapy
(37). Given that developing CPGs is expensive, development of
Balneotherapy
Manual therapy fewer, higher-quality CPGs that focus on implementation
Acupuncture (as reflected in higher scoring in the applicability domain on the
Massage therapy AGREE II tool) is recommended.
Dry needling
Heat and cold therapy Recommendations from 11 high-quality CPGs were that
Electrotherapy first-line care should be patient centered and include exercise
Taping and braces therapy, patient education, and weight loss (if appropriate). These
Shoe orthotics
interventions can be beneficial in reducing pain and in improving
Footwear
Topical capsaicin function, performance, and quality of life outcomes (46–49). This
Glucocorticoid injection (hip and polyarticular OA) should be followed by pharmacologic strategies such as NSAIDs
Intraarticular hyaluronic acid injections
in oral or topical form before considering surgical interventions
Platelet-rich plasma injections
Stem cell injection as second- and third-line care. For people with hand OA, orthosis
Acetaminophen should be used for symptom relief and improved function and to
Oral opioids prevent progression of degenerative changes (13,24). This syn-
Glucosamine and chondroitin, individually or combined (GH joint
OA) thesis of recommendations provides evidence-based guidance
Nutraceuticals for clinicians on what should be delivered for best practice in OA
* GH = glenohumeral; OA = osteoarthritis. care. These recommendations could also be used as a minimum
1904 CONLEY ET AL

standard for health services to assess OA care and to provide the annually on both public and private hospital services (58–60); sim-
basis for clear consumer information about recommended OA ilar findings have been documented in the UK and the US (61,62).
management. Surgery is a successful and cost-effective intervention for people
We identified a substantial number of recommendations that with end-stage hip and knee OA, although overuse of surgery in
were inconsistent between CPGs, which may contribute to con- patients who could benefit from conservative care remains a chal-
fusion among clinicians and to varied management. For example, lenge (63–66).
manual therapy recommendations were inconsistent across the Implementation of high-value care such as exercise and
CPGs, with a majority recommending against, yet these are still weight loss is needed (67). One way is through OA management
widely used in clinical practice (50). Similarly, the majority of CPGs programs such as OA models of care that operationalize what
recommended against opioids, although 2 CPGs recommended and how recommended care should be delivered (68). In order
that opioids can be considered in particular circumstances, when to achieve better care, priorities include training/education of OA
pain is severe or if patients do not respond, are intolerant, or con- health care workers, identifying core skill sets and competencies,
traindicated to NSAIDs, or when other alternatives have been developing resources, and creating a framework to improve qual-
exhausted (13,25). Despite this, opioids are often prescribed for ity of care (69). Outcomes from models of care suggest that this
persistent musculoskeletal pain conditions, including OA, and has been an effective way to translate evidence into practice,
opioid-related harms are of increasing concern (51,52). Additional although definitive evidence for OA management is currently lack-
conflicting recommendations included acupuncture/dry needling, ing (70). Structured exercise therapies, with or without education
shoe orthotics, taping/braces, glucosamine and chondroitin, and and dietary interventions, are cost effective and clinically effective
injection therapies, e.g., platelet-rich plasma, stem cell, and intra- (71). Implementation research that operationalizes recommended
articular hyaluronic acid (for a comprehensive list of conflicting care, especially for populations who experience a higher burden
recommendations, see Supplementary Table 4, available at of OA, including low- and middle-income countries and First
http://onlinelibrary.wiley.com/doi/10.1002/acr.25101). Nations people, is a pressing future priority (72,73). We excluded
Many of these recommendations are routinely utilized in clin- CPGs that were not published in the English language and that
ical care. Further high-quality trials are needed to determine their addressed assessment and/or diagnosis of OA without manage-
efficacy and/or their suitability for certain subgroups of individuals ment or treatment recommendations. It is possible that we may
with OA in order to guide clinicians’ practice. CPG recommenda- have overlooked other CPGs containing recommendations
tions vary, potentially due to differences in evidence included related to OA care. To mitigate this risk, all authors checked the
based on the year of CPG publication, assessment of evidence list of full-text CPGs to augment the search process, including
quality, and involvement of expert panels or societies members. authors who are expert clinician researchers in the field of OA
For example, 1 CPG recommended against heat therapy, while (MMD and PC).
another CPG recommended that it can be considered as an Strengths of this systematic review include the involvement
adjunctive management option for people with hip and/or knee of a multidisciplinary team and the use of the AGREE II tool.
OA (12,27). Both CPGs acknowledged limited evidence support- The research team defined high-quality CPGs as ≥60% in the
ing this intervention; however, interpretation of evidence by the 3 domains of interest on the AGREE II instrument. These
respective guideline development groups led to conflicting domains are consistent with other high-quality musculoskeletal
recommendations. reviews (36), while 60% is supported by other arthritis and oste-
While exercise, weight management, and education were oporosis reviews (41,42). Grading of interventions and consen-
supported across the CPGs and have been recommended as sus statements (e.g., “should do,” “could do,” “do not do,” or
first-line interventions for almost 2 decades, translation into prac- “unsure”) were based on the language used in CPGs and
tice remains an issue (53). In Western health care settings includ- required interpretation by the research team. Consensus state-
ing Australia, Europe, the UK, and the US, a majority of patients ments were cross-checked by 2 authors (SB and IL) to mitigate
do not receive care consistent with CPGs (54). Conservative man- the risk of misinterpretation. It is important to acknowledge that
agement interventions are often overlooked in favor of pharmaco- the majority of the literature regarding OA management is based
logic and surgical care despite being associated with higher on hip and knee OA, often neglecting OA in other joints. For
financial costs and risks (e.g., medication side effects or surgical transparency, we have listed the affected joint for each recom-
complications) (6,30,55,56). Globally, utilization of exercise and mendation in Supplementary Tables 3 and 4, available on the
education is low, while pharmacologic therapy and surgical refer- Arthritis Care & Research website at http://onlinelibrary.wiley.
rals are common (54,57). In Australia, joint replacement surgeries com/doi/10.1002/acr.25101.
are a substantial cost to the health care system, estimated at In conclusion, 7 consistent “should do” recommendations
between $19,000 and $30,000 (Australian; between $13,000 were identified across the 11 CPGs. Exercise therapy, education,
and $20,600 US dollars) per patient for total knee or hip replace- and weight loss (if relevant) should be recommended for people
ment, resulting in an expenditure of $1.2 billion (Australian) with OA before considering pharmacologic or surgical
OA CARE CORE RECOMMENDATIONS 1905

interventions, with care being patient centered. Hand orthosis 2017. URL: https://www.aaos.org/quality/quality-programs/lower-
should be considered for those with hand OA. These core tenets extremity-programs/osteoarthritis-of-the-hip/.

of OA care can be used by health care providers to improve con- 12. Royal Australian College of General Practitioners. Guideline for
the management of knee and hip osteoarthritis. 2nd ed. East
sistency and quality of OA care. Melbourne (Australia): RACGP; 2018. URL: https://www.racgp.org.
au/download/Documents/Guidelines/Musculoskeletal/guideline-for-the-
management-of-knee-and-hip-oa-2nd-edition.pdf.
ACKNOWLEDGMENT 13. Ariani A, Manara M, Fioravanti A, et al. The Italian Society for Rheuma-
The authors acknowledge open access publishing facilitated by tology clinical practice guidelines for the diagnosis and management
The University of Melbourne, as part of the Wiley - The University of Mel- of knee, hip and hand osteoarthritis. Reumatismo 2019;71:5–21.
bourne agreement via the Council of Australian University Librarians. 14. Brosseau L, Taki J, Desjardins B, et al. The Ottawa Panel clinical prac-
tice guidelines for the management of knee osteoarthritis. Part one:
introduction, and mind-body exercise programs. Clin Rehabil 2017;
AUTHOR CONTRIBUTIONS 31:582–95.

All authors were involved in drafting the article or revising it critically 15. Brosseau L, Taki J, Desjardins B, et al. The Ottawa Panel clinical prac-
for important intellectual content, and all authors approved the final ver- tice guidelines for the management of knee osteoarthritis. Part three:
sion to be submitted for publication. Ms. Conley had full access to all of aerobic exercise programs. Clin Rehabil 2017;31:612–24.
the data in the study and takes responsibility for the integrity of the data 16. Brosseau L, Taki J, Desjardins B, et al. The Ottawa Panel clinical prac-
and the accuracy of the data analysis. tice guidelines for the management of knee osteoarthritis. Part two:
Study conception and design. Lin. strengthening exercise programs. Clin Rehabil 2017;31:596–611.
Acquisition of data. Conley, Gunatillake.
17. Brosseau L, Thevenot O, MacKiddie O, et al. The Ottawa Panel guide-
Analysis and interpretation of data. Conley, Bunzli, Bullen, O’Brien,
lines on programmes involving therapeutic exercise for the manage-
Persaud, Gunatillake, Dowsey, Choong, Lin.
ment of hand osteoarthritis. Clin Rehabil 2018;32:1449–71.
18. Brosseau L, Wells, GA, Pugh AG, et al. Ottawa Panel evidence-based
clinical practice guidelines for therapeutic exercise in the management
REFERENCES of hip osteoarthritis. Clin Rehabil 2016;30:935–46.
1. Charlesworth J, Fitzpatrick J, Perera NK, et al. Osteoarthritis: a sys- 19. American Academy of Orthopaedic Surgeons. Management of osteo-
tematic review of long-term safety implications for osteoarthritis of arthritis of the knee (nonarthroplasty) evidence-based clinical practice
the knee. BMC Musculoskeletal Disord 2019;20:151. guideline. 2021. URL: https://www.aaos.org/oak3cpg.
2. Fu K, Robbins SR, McDougall JJ. Osteoarthritis: the genesis of pain.
20. Ayhan FF, Sunar _I, Umay E, et al. The Turkish League Against Rheu-
Rheumatology (Oxford) 2018;57 Suppl 4:iv43–50.
matism recommendations for the management of hand osteoarthritis
3. Global Burden of Disease 2017 Risk Factor Collaborators. Global, under guidance of the current literature and 2018 European League
regional, and national comparative risk assessment of 84 behavioural, Against Rheumatism recommendations. Arch Rheumatol 2020;35:
environmental and occupational, and metabolic risks or clusters of 309–20.
risks for 195 countries and territories, 1990–2017: a systematic anal-
21. Bannuru RR, Osani MC, Vaysbrot EE, et al. OARSI guidelines for the
ysis for the Global Burden of Disease Study 2017 [correction pub-
non-surgical management of knee, hip, and polyarticular osteoarthri-
lished in Lancet 2019;393:132] [correction published in Lancet 2019;
tis. Osteoarthritis Cartilage 2019;27:1578–89.
393:e44]. Lancet 2018;392:1923–94.
4. Safiri S, Kolahi AA, Smith E, et al. Global, regional and national burden 22. Bruyère O, Honvo G, Veronese N, et al. An updated algorithm recom-
of osteoarthritis 1990–2017: a systematic analysis of the Global mendation for the management of knee osteoarthritis from the
Burden of Disease Study 2017. Ann Rheum Dis 2020;79:819–28. European Society for Clinical and Economic Aspects of Osteoporosis,
Osteoarthritis and Musculoskeletal Diseases (ESCEO). Semin Arthritis
5. National Institute for Health and Care Excellence. Clinical guideline.
Rheum 2019;49:337–50.
Osteoarthritis: care and management. 2014;CG177:1–27. URL:
https://www.nice.org.uk/guidance/cg177. 23. Eymard F, Ornetti P, Maillet J, et al. Intra-articular injections of platelet-
6. Hunter DJ, Schofield D, Callander E. The individual and socioeco- rich plasma in symptomatic knee osteoarthritis: a consensus state-
nomic impact of osteoarthritis. Nat Rev Rheumatol 2014;10:437–41. ment from French-speaking experts [correction published in Knee
Surg Sports Traumatol Arthrosc 2021;29:3211–12]. Knee Surg
7. Graham R, Mancher M, Miller Wolman D, et al, editors. Institute of
Sports Traumatol Arthrosc 2021;29:3195–210.
Medicine Committee on standards for developing trustworthy clinical
practice guidelines: clinical practice guidelines we can trust [review]. 24. Kloppenburg M, Kroon FP, Blanco FJ, et al. 2018 update of the
Washington (DC): National Academies Press; 2011. EULAR recommendations for the management of hand osteoarthritis.
Ann Rheum Dis 2019;78:16–24.
8. Ng JY, Mohiuddin U, Azizudin AM. Clinical practice guidelines for the
treatment and management of low back pain: a systematic review of 25. Kolasinski SL, Neogi T, Hochberg MC, et al. 2019 American College
quantity and quality. Musculoskelet Sci Pract 2021;51:102295. of Rheumatology/Arthritis Foundation Guideline for the Management
9. Romão VC, Talarico R, Scirè CA, et al. Sjögren’s syndrome: state of of Osteoarthritis of the Hand, Hip, and Knee [correction published in
the art on clinical practice guidelines. RMD Open 2018;4 Suppl 1: Arthritis Rheumatol 2021;73:799]. Arthritis Rheumatol 2020;72:
e000789. 220–33.
10. American Academy of Orthopaedic Surgeons. Management of 26. Rillo O, Riera H, Acosta C, et al. PANLAR consensus recommenda-
glenohumeral joint osteoarthritis evidence based clinical practice tions for the management in osteoarthritis of hand, hip, and knee.
guideline. 2020. URL: https://www.aaos.org/globalassets/quality- J Clin Rheumatol 2016;22:345–54.
and-practice-resources/glenohumeral/gjo-cpg.pdf. 27. Van Doormaal MC, Meerhoff GA, Vliet Vlieland TP, et al. A clinical
11. American Academy of Orthopaedic Surgeons. Management of practice guideline for physical therapy in patients with hip or knee
osteoarthritis of the hip evidence based clinical practice guideline. osteoarthritis. Musculoskeletal Care 2020;18:575–95.
1906 CONLEY ET AL

28. Zhang Z, Huang C, Jiang Q, et al. Guidelines for the diagnosis and better effects when combined with exercise therapy: a systematic
treatment of osteoarthritis in China (2019 edition). Ann Transl Med review. J Physiother 2021;67:177–89.
2020;8:1213. 47. Goh SL, Persson MS, Stocks J, et al. Relative efficacy of different
29. American Academy of Orthopaedic Surgeons. Surgical management exercises for pain, function, performance and quality of life in knee
of osteoarthritis of the knee: clinical practice guideline. 2015. URL: and hip osteoarthritis: systematic review and network meta-analysis.
https://www.aaos.org/quality/quality-programs/lower-extremity- Sports Med 2019;49:743–61.
programs/surgical-management-of-osteoarthritis-of-the-knee/. 48. Wellsandt E, Golightly Y. Exercise in the management of knee and hip
30. Basedow M, Esterman A. Assessing appropriateness of osteoarthritis osteoarthritis. Curr Opin Rheumatol 2018;30:151–9.
care using quality indicators: a systematic review. J Eval Clin Pract 49. Gay C, Chabaud A, Guilley E, et al. Educating patients about the ben-
2015;21:782–9. efits of physical activity and exercise for their hip and knee osteoarthri-
31. Correa VC, Lugo-Agudelo LH, Aguirre-Acevedo DC, et al. Individual, tis: systematic literature review. Ann Phys Rehabil Med 2016;59:
health system, and contextual barriers and facilitators for the imple- 174–83.
mentation of clinical practice guidelines: a systematic metareview.
50. Xu Q, Chen B, Wang Y, et al. The effectiveness of manual therapy for
Health Res Policy Syst 2020;18:74.
relieving pain, stiffness, and dysfunction in knee osteoarthritis: a sys-
32. Ferreira de Meneses S, Rannou F, Hunter DJ. Osteoarthritis guide- tematic review and meta-analysis. Pain Physician 2017:20;229–43.
lines: barriers to implementation and solutions. Ann Phys Rehabil
51. Ashaye T, Hounsome N, Carnes D, et al. Opioid prescribing for
Med 2016;59:170–3.
chronic musculoskeletal pain in UK primary care: results from a cohort
33. Swaithes L, Paskins Z, Dziedzic K, et al. Factors influencing the imple- analysis of the COPERS trial. BMJ Open 2017;8:e019491.
mentation of evidence-based guidelines for osteoarthritis in primary
52. Van den Driest JJ, Schiphof D, de Wilde M, et al. Opioid prescriptions
care: a systematic review and thematic synthesis. Musculoskeletal
in patients with osteoarthritis: a population-based cohort study.
Care 2020;18;101–10.
Rheumatology (Oxford) 2020:59;2462–70.
34. Allen KD, Golightly YM, White DK. Gaps in appropriate use of treat-
53. Meneses SR, Goode AP, Nelson AE, et al. Clinical algorithms to aid
ment strategies in osteoarthritis. Best Pract Res Clin Rheumatol
osteoarthritis guideline dissemination. Osteoarthritis Cartilage 2016;
2017;31:746–59.
24;1487–99.
35. Nelson AE, Allen KD, Golightly YM, et al. A systematic review of rec-
54. Hagen KB, Smedslund G, Østerås N, et al. Quality of community-
ommendations and guidelines for the management of osteoarthritis:
based osteoarthritis care: a systematic review and meta-analysis.
the chronic osteoarthritis management initiative of the U.S. Bone and
Arthritis Care Res (Hoboken) 2016:68;1443–52.
Joint Initiative. Semin Arthritis Rheum 2014;43:701–12.
36. Lin I, Wiles L, Waller R, et al. What does best practice care for muscu- 55. Lützner J, Kasten P, Günther KP, et al. Surgical options for patients
with osteoarthritis of the knee. Nat Rev Rheumatol 2009:5;309–16.
loskeletal pain look like? Eleven consistent recommendations from
high-quality clinical practice guidelines: systematic review. Br J Sports 56. Ackerman IN, Bohensky MA, de Steiger R, et al. Substantial rise in the
Med 2020;54:79–86. lifetime risk of primary total knee replacement surgery for osteoarthritis
37. Lin I, Wiles LK, Waller R, et al. Poor overall quality of clinical practice from 2003 to 2013: an international, population-level analysis. Osteo-
guidelines for musculoskeletal pain: a systematic review. Br J Sports arthritis Cartilage 2017:25;455–61.
Med 2018;52:337–43. 57. Brand CA, Harrison C, Tropea J, et al. Management of osteoarthritis in
38. Brouwers MC, Kho ME, Browman GP, et al. Appraisal of Guidelines general practice in Australia. Arthritis Care Res (Hoboken) 2014:66;
for Research Evaluation (AGREE) II. The AGREE II instrument. 551–8.
2017. URL: https://www.agreetrust.org/wp-content/uploads/2017/ 58. Independent Health and Aged Care Pricing Authority. National hospi-
12/AGREE-II-Users-Manual-and-23-item-Instrument-2009-Update- tal cost data collection public hospitals report - round 19 (financial
2017.pdf. year 2014–15). 2017. URL: https://www.ihacpa.gov.au/resources/
39. Murad MH. Clinical practice guidelines: a primer on development and national-hospital-cost-data-collection-nhcdc-public-hospitals-report-
dissemination. Mayo Clin Proc 2017;92:423–33. round-19-financial-year-2014-15.
40. AGREE Research Trust. AGREE II tutorial and practice exercise. 59. Australian Institute of Health and Welfare. Disease expenditure in
AGREE Enterprise. URL: http://www.agreetrust.org/resource- Australia 2018–2019: Australian Burden of Disease conditions.
centre/agree-ii-training-tools. 2021. URL: https://www.aihw.gov.au/reports/health-welfare-expenditure/
disease-expenditure-australia/contents/australian-burden-of-disease-
41. Smith CA, Toupin-April K, Jutai JW, et al. A systematic critical
conditions.
appraisal of clinical practice guidelines in juvenile idiopathic arthritis
using the Appraisal of Guidelines for Research and Evaluation II 60. Ackerman IN, Bohensky MA, Zomer E, et al. The projected burden of
(AGREE II) instrument. PLoS One 2015;10:e0137180. primary total knee and hip replacement for osteoarthritis in Australia to
the year 2030. BMC Musculoskelet Disord 2019:20;90.
42. Hayawi LM, Graham ID, Tugwell P, et al. Screening for osteoporosis: a
systematic assessment of the quality and content of clinical practice 61. Healey EL, Afolabi EK, Lewis M, et al. Uptake of the NICE osteoarthri-
guidelines, using the AGREE II instrument and the IOM standards for tis guidelines in primary care: a survey of older adults with joint pain.
trustworthy guidelines [correction published in PLoS One 2019;14: BMC Musculoskelet Disord 2018;19;295.
e0216196]. PLoS One 2018;13:e0208251. 62. Hunter DJ, Neogi T, Hochberg MC. Quality of osteoarthritis manage-
43. Landis JR, Koch GG. The measurement of observer agreement for ment and the need for reform in the US. Arthritis Care Res
categorical data. Biometrics 1977;33:159–74. (Hoboken) 2011:63;31–8.
44. Fleiss JL, Paik MC. Statistical methods for rates and proportions. 2nd 63. Kamaruzaman H, Kinghorn P, Oppong R. Cost-effectiveness of surgi-
ed. New York (NY): John Wiley & Sons; 1981. cal interventions for the management of osteoarthritis: a systematic
45. Armstrong JJ, Rodrigues IB, Wasiuta T, et al. Quality assessment of review of the literature. BMC Musculoskelet Disord 2017:18;183.
osteoporosis clinical practice guidelines for physical activity and safe 64. Hunter DJ, Bierma-Zeinstra S. Osteoarthritis. Lancet 2019:393;1745–59.
movement: an AGREE II appraisal. Arch Osteoporos 2016;11:6. 65. Cronström A, Nero H, Lohmander LS, et al. On the waiting list for joint
46. Goff AJ, De Oliveira Silva D, Merolli M, et al. Patient education replacement for knee osteoarthritis: are first-line treatment recom-
improves pain and function in people with knee osteoarthritis with mendations implemented? Osteoarthr Cartil Open 2020:2;100056.
OA CARE CORE RECOMMENDATIONS 1907

66. King LK, Marshall DA, Faris P, et al. Use of recommended non- participatory health research approach to implementing high value
surgical knee osteoarthritis management in patients prior to total knee care. BMC Musculoskelet Disord 2022:23;643.
arthroplasty: a cross-sectional study. J Rheumatol 2020:47;1253–60.
71. Mazzei DR, Ademola A, Abbott JH, et al. Are education, exercise and
67. Briggs AM, Page CJ, Shaw BR, et al. A model of care for osteoarthritis diet interventions a cost-effective treatment to manage hip and knee
of the hip and knee: development of a system-wide plan for the health osteoarthritis? A systematic review. Osteoarthritis Cartilage 2021:29;
sector in Victoria, Australia. Healthc Policy 2018:14;47–58. 456–70.
68. Allen KD, Choong PF, Davis AM, et al. Osteoarthritis: models for
72. Lin IB, Bunzli S, Mak DB, et al. Unmet needs of Aboriginal Australians
appropriate care across the disease continuum. Best Pract Res Clin
Rheumatol 2016:30;503–35. with musculoskeletal pain: a mixed-method systematic review. Arthri-
tis Care Res (Hoboken) 2018:70;1335–47.
69. Eyles JP, Hunter DJ, Bennell KL, et al. Priorities for the effective imple-
mentation of osteoarthritis management programs: an OARSI interna- 73. Eyles JP, Sharma S, Telles RW, et al. Implementation of best-
tional consensus exercise. Osteoarthritis Cartilage 2019:27;1270–9. evidence osteoarthritis care: perspectives on challenges for, and
70. Toomey CM, Kennedy N, MacFarlane A, et al. Implementation of clin- opportunities from, low and middle-income countries. Frontiers Reha-
ical guidelines for osteoarthritis together (IMPACT): protocol for a bil Sci 2022:2;826765.