Table 1 shows 11 interventions with highest ratings in the ‘Overall’ category. These ratings indicate which interventions experts perceive merit translation based on their potential to improve the health of individuals who are disadvantaged, when ease of implementation, health system requirement, universality and impact on inequities are considered. Ordered lists based on the highest rated interventions for each of the other five criteria are provided as well (Fig 1-5). The top 11 interventions prioritized overall were extracted from systematic reviews focused on the following interventions: Exercise for osteoarthritis of the knee [37], Exercise for hand osteoarthritis [38], Exercise for osteoarthritis of the hip [39], High‐intensity versus low‐intensity physical activity or exercise in people with hip or knee osteoarthritis [40], Self-management education programmes for osteoarthritis [41], Celecoxib for osteoarthritis [42], Oral herbal therapies for treating osteoarthritis [43], Chondroitin for osteoarthritis [44], and Intra-articular corticosteroid for knee osteoarthritis [45]. The top six interventions involved treatment of OA with exercise (four specifying land-based exercise) and the 7th intervention involved self-management programs for treatment of OA. On the whole experts gave higher ratings to interventions treating OA with exercise and lower ratings to pharmacological interventions including opioids, oral herbal therapies, celecoxib, glucosamine and chondroitin sulfate. Notably, aquatic exercise was lower rated compared to land-based exercise. To further elucidate which criteria may have influenced the expert’s ‘Overall’ ratings, each of the four equity-focused criteria (‘Ease of Implementation’, ‘Health System Requirements’, ‘Universality’, and ‘Impact on Inequities) was plotted against the ‘Overall’ criterion measuring expert opinion of priorities for an E4E summary. ‘Universality’ was estimated to have the greatest impact on respondents ‘Overall’ rating with a Pearson correlation of 0.96. This was greater than for the criteria ‘Ease of implementation’, ‘Health system requirements’ and ‘Impact on inequities’ which had Pearson correlation values of 0.82, 0.85 and 0.84 respectively. These values were all statistically significant (alpha<0.05).
Safety Concerns
Based on suggestions from previous priority-setting exercises used to develop E4E summaries [31], experts were invited to identify safety concerns associated with rated interventions. A total of 17 (58.6%) interventions were flagged for safety concerns by at least one reviewer. None of the top 7 prioritised interventions had any safety concerns. However, experts identified safety concerns for all pharmacological interventions as well as those included in the top 11 list of interventions rated highest ‘overall’.
Qualitative Analysis
Following qualitative coding we identified three themes overall which were further divided into subthemes.
Theme 1 - This intervention or an aspect of this intervention is unnecessary or unsafe
The most frequent concerns brought forward by experts related to the overall safety of an intervention. The most common safety concerns were that an intervention was not effective and thus unnecessary, followed by comments that an intervention could increase a patient's risk of adverse events, was not recommended by clinical practice guidelines, or required ongoing patient monitoring. Also noted were a risk of developing an addiction, infection risk from needle usage, the need to ‘pace-up’ the interventions, the need for greater quality control measures, and issues regarding patient compliance.
Subtheme 1.1: This intervention has questionable clinical efficacy
14 comments noted that interventions had questionable clinical efficacy, of which 5 came from a review on oral herbal therapies, 2 from a review on braces and orthoses, 2 from a review on acupuncture, 2 from a review on chondroitin, and the remainder from reviews on hyaluronic acid, opioids, and intra-articular corticosteroids. One expert stated “MD not clinically relevant” for 3 different oral herbal therapies, 2 different brace/orthotic interventions, and 2 different acupuncture interventions. This was important to capture as we excluded interventions with effect sizes that were not statistically significant but sought out expert opinion regarding the clinical relevance of effect sizes.
Subtheme 1.2: General safety concerns
There were 18 comments made regarding general safety concerns, of which 14 were from a review on oral herbal therapies (comprising 10 different interventions), 2 from a review on intra-articular corticosteroid for knee osteoarthritis, 1 from a review on celecoxib, and 1 from a review on high-intensity vs low-intensity physical activity. One expert expressed concerns regarding the “safety of [the] agent” for each of the 10 interventions in the oral herbal therapy review. For the review on intra-articular corticosteroids, one expert mentioned that the intervention could “damage [knee] cartilage”. When assessing the safety of high intensity exercise one expert noted there is a risk of “pain flares” and a need to administer exercise with “comorbidities” in mind. Lastly, regarding a review on celecoxib for osteoarthritis, an expert stated the risk of “adverse events” as a concern.
Subtheme 1.3: Clinical practice guidelines conditionally, are neutral or do not recommend this intervention
One expert made 13 comments related to clinical practice guidelines and whether interventions were neutral, conditionally, or not recommended. Of these comments, 4 were from a review on oral herbal therapies, 3 from a review on acupuncture for hip osteoarthritis, 2 from a review on braces and orthoses, 2 from a review on chondroitin, 1 from a review on intra-articular corticosteroids, and 1 from a review on self-management programs. Of the comments made on oral herbal therapies, Salix purpurea was listed as ‘do not recommend’ whereas Persea gratissima and Boswellia serrata were cited as neutral. Regarding acupuncture, the expert suggested “not offering acupuncture for people with knee and/or hip OA'' according to the current clinical practice guidelines. Concerning braces and orthoses for osteoarthritis, the same expert stated, “Clinical practice guidelines [are] conditional[ly] against this recommendation for lateral wedge medial tibial OA [and] neutral for lateral.” indicating that certain conditions need to be met before this intervention could be recommended. For intra-articular corticosteroids, the expert also mentioned that the clinical practice guidelines recommend that this intervention be “only [administered] as indicated when no progress [occurs] with other first line interventions.” Further, they stated, “CPGs conditional for recommendation” which again captures the idea that certain conditions should be met before administering this intervention. Further, they note that this intervention could increase inequity and requires “Simple messaging to mitigate health literacy issues (hence rating on inequity lower).”
Subtheme 1.4: This intervention requires ongoing monitoring (bp, side effects)
There were 5 comments made regarding the need for ongoing patient monitoring or broader health implementation requirements to support OA interventions: 2 from a review on opioids, 2 from a review on celecoxib, and 1 from a review on oral herbal therapies. One expert observed that there is “need [for] systems in place for safe use of opioids”. Regarding the use of celecoxib, experts commented that there is a “need to manage comorbidities risk” as well as a “need for monitoring cr [creatine rise] BP”. For oral herbal therapies, one expert mentioned that “[for] diclofenac - assess for GI bleeding risk, CVD risk and renal risks.”
Subtheme 1.5: There is a risk of addiction associated with this intervention and/or drug trafficking
All 4 comments concerning the risk of addiction were made regarding a review on the use of opioids for osteoarthritis. Experts commented, “Risk of opioid addiction & traffic”, “Addiction [and] adverse events”, “Adverse outcomes with prolonged use; tolerance and addiction risks”, and simply, “Addiction.”
Subtheme 1.6: Interventions that require use of needles run a risk of infection
Of the 5 comments made concerning the risk of infection from needles, 2 were from a review on acupuncture for hip osteoarthritis, 2 from a review on intra-articular corticosteroid, and 1 from a review on hyaluronic acid and other conservative treatment options for osteoarthritis of the ankle.
Subtheme 1.7: This intervention may need to be 'paced up' or implemented gradually
All 5 comments regarding the need to ‘pace-up’ interventions were made by one expert concerning three different reviews on the use of exercise to treat osteoarthritis: Exercise for osteoarthritis of the knee, Exercise for osteoarthritis of the hip, and High‐intensity versus low‐intensity physical activity or exercise in people with hip or knee osteoarthritis.
Subtheme 1.8: Certain interventions require quality control measures to ensure integrity and safety of components
There were 20 comments made regarding the need for better quality control measures, of which 18 were from a review on oral herbal therapies and 2 from a review on chondroitin. Experts frequently mentioned either “quality control” or “consistency of components” for herbal therapies, referring to the need to ensure rigorous quality control, which is particularly important for herbal therapies regulated as natural health products.
Subtheme 1.9: When implementing this intervention, there may be difficulties associated with compliance
Finally, there was one comment regarding issues with patient compliance from a review on exercise for osteoarthritis of the knee.
Theme 2 - This intervention or an aspect of this intervention may increase health inequities
Following safety concerns, experts also frequently commented that certain interventions may potentially increase health inequities. The high cost of certain interventions in addition to a lack of access to the materials and/or expertise required to utilize such interventions were cited as the main sources driving inequities. Some experts also mentioned that certain interventions required adaptations to the cultural and social contexts in which they are found.
Subtheme 2.1: This intervention requires specific expertise which may not be accessible and may only be available in certain regions
Experts made 13 comments regarding the need for specific expertise in order to utilize interventions, of which 7 came from 5 different reviews on exercise for osteoarthritis, 2 from a review on braces and orthoses for treating osteoarthritis of the knee, and one comment from reviews on oral herbal therapies, acupuncture, hyaluronic acid, and self-management education programs respectively. Regarding exercise for osteoarthritis, one expert stated that “[the patient] could attend community-based centers or gyms” which may require the use of personal trainers and other expertise that may not always be available in certain regions, such as rural and remote communities. Concerning the use of braces and orthoses for osteoarthritis, one expert mentioned that there is a need for the “Skill of [a] practitioner and availability of equipment.”
Subtheme 2.2: The material resources for this intervention may be difficult to access and/or is only available in certain regions
Experts made 12 comments regarding the requirement for materials only available in certain regions and/or difficult to source. Half of these comments were from 3 different reviews on exercise for osteoarthritis, 2 from a review on acupuncture, and the remainder from reviews on oral herbal therapies, braces and orthoses, and opioids.
Subtheme 2.3: This intervention is costly, may not be feasible in LMICs and may not be covered by insurance
Experts made 19 comments regarding the high cost of certain interventions. 10 of these comments were associated with a review on oral herbal therapies, 2 from a review on celecoxib, 2 from a review on acupuncture for hip osteoarthritis, 2 from a review on braces and orthoses, 2 from a review on chondroitin, and 1 from a review on hyaluronic acid. For 9 of the 10 interventions assessed in the review on oral herbal therapies, one expert commented that “Herbal therapies conditioned are costly in most countries”, highlighting the cost-inequities that many LMICs experience.
Subtheme 2.4: This intervention requires adaptations to cultural and social context
Experts made 3 comments regarding the need to adapt certain interventions to the social and cultural context in which they are used. 2 of these comments came from a review on self-management educational programs for osteoarthritis while the other came from a review on braces and orthoses for treating osteoarthritis of the knee. Regarding self-management educational programs, an expert stated “Need SMP adapted to cultural and social context”, while another commented “Simple messaging to mitigate health literacy issues (hence rating on inequity lower); digital or community-based dissemination.” One expert mentioned that “insoles [are] not practical or effective in countries where people go barefoot or in sandals.”
Theme 3 - Experts noted difficulties completing rating exercise
Experts also brought up that it was difficult to complete the rating exercise for certain interventions due to concerns regarding their clinical efficacy as well as potential adverse outcomes. Regarding opioids, one expert commented that it was “Difficult to rate when outcome worse.” as this systematic review reported a greater number of adverse events among patients administered opioids compared with those administered a placebo.