We used content analysis, a methodological approach widely used to analyze the content of written or visual materials [25,26]. More specifically, we applied manifest content analysis, which is a way to describe explicit content without theoretical interpretation. This technique differs from thematic coding typically used to analyze transcripts of interviews or focus groups and involves both quantitative (eg, counting the number of documents that mention a concept) and qualitative (eg, extracting content relevant to a concept) methods. To optimize rigor, we complied with content analysis methods [25,26] and criteria for reporting qualitative research [27]. The research team informed data collection, analysis, and interpretation to further support rigor. The research team was comprised of a 12-member advisory group of diverse women with osteoarthritis; 6 coinvestigators including health care professionals (family physician, rheumatologists, physiotherapist, and pharmacist), and health services researchers with expertise in the topics of osteoarthritis, person-centered care, equity, and women’s health; and 4 collaborators who represented organizations that could apply the findings: 2 advocacy groups for persons with osteoarthritis and 2 health care professional societies. In particular, women advisors, who varied by type of osteoarthritis, age, length of time since osteoarthritis diagnosis, ethnocultural group, and location in Canada, helped us to search for osteoarthritis materials, develop and test the use of analytic frameworks, and summarize the findings.

Multimedia Appendix 1 provides detailed eligibility criteria. In brief, we included publicly available, online, English-language materials (eg, video, infographic, checklist, booklet, and web page featuring text with or without other materials) that support osteoarthritis first-line management for persons with osteoarthritis or health care professionals who care for such persons; and were developed by Canadian nonprofit organizations (eg, government, academic groups, professional societies, and charities) in 2012 or later, when the Arthritis Alliance Canada issued a framework to improve arthritis prevention and care [28]. These types of organizations are inclusive of multiple sectors, represent a comprehensive list of types of organizations most likely to generate develop and share osteoarthritis materials on the internet, and because they are nonprofit, materials are likely to freely accessible. We included osteoarthritis materials developed or shared by only Canadian organizations for reasons, that (1) this study was 1 part of a multicomponent research study focused on osteoarthritis care in the Canadian setting, (2) the study was funded by the Arthritis Society Canada, and (3) the study aimed to inform policies and practice needed in the Canadian setting to enhance equitable access to quality of osteoarthritis care. Furthermore, restricting to only Canadian osteoarthritis materials enhanced study feasibility; it required considerable time and effort to evaluate the eligible Canadian osteoarthritis materials—we would not have been able to do so for the potentially hundreds or thousands of osteoarthritis materials developed by organizations around the world. We excluded clinical practice guidelines and resources that focused solely on second-line management (eg, joint replacement) given the importance of first-line management in preventing further decline and optimizing the quality of life [9,10].

SM (woman of color, Master of Public Health candidate) and SD (woman of color, Master of Public Health research coordinator) conducted searching and independently screened all items in duplicate between April 12 and May 23, 2023. ARG (woman, PhD-trained principal investigator) provided training and oversight and resolved discrepancies by independent review, and then discussion with SM and SD. SM and SD first conducted searches, capturing the organization, title, and address of potentially relevant items in a Microsoft Excel file; then revisited each item to confirm eligibility. Given no standard method of searching for “gray literature” [29,30], we used a multipronged strategy that included (1) using Google, searched for “osteoarthritis AND Canada OR <each of the provinces>,” and scanned search results on up to 6 pages of 10 results per page, or before relevancy dropped off; (2) consulting members of the aforementioned research team including women advisors, coinvestigators and collaborators plus other professional societies representing health care professionals who care for persons with osteoarthritis (eg, family physicians, physiotherapists, chiropractors, occupational therapists, and pharmacists); (3) scanning the websites of the Arthritis Society, advocacy groups and professional societies; (4) reviewed the content of Canadian guidelines and policies related to osteoarthritis identified in our previous research [31,32]; and (5) Searched MEDLINE for “osteoarthritis AND Canada AND Internet AND health resources OR education, medical, continuing” for articles that reviewed or developed osteoarthritis information.

SM extracted data from osteoarthritis materials for persons with osteoarthritis, SD extracted data from osteoarthritis materials for health care professionals, and then each checked the others’ work, consulting with ARG to resolve discrepancies through discussion to achieve 100% agreement. From all materials, we extracted data on osteoarthritis material characteristics (developer name, type of organization, year published, title of osteoarthritis material, web address, development approach, osteoarthritis topic, and objective) and design according to the Workgroup for Intervention Development and Evaluation Research framework (content, format, and delivery) [33].

SM and SD independently scored all osteoarthritis materials, then met to resolve discrepancies through discussion. ARG resolved the remaining uncertainties, and discrepancies, and checked all data. DISCERN scoring involved a 5-point scale with each of the 15 items marked as met (5), partially met (2-4), or not met (1). We deemed items that were not relevant to the materials as not applicable and did not record a score (eg, not all items included treatment information). We applied a similar scoring system for all other frameworks. For all instruments and frameworks, we derived an overall score based on the total number of “met” items divided by the total number of applicable items. Applying the same rubric as DISCERN, we categorized the overall score as low (<50%), moderate (50%-69%), or high (≥70%+). We used summary statistics, referring to simple counts of numbers and percentages, to report osteoarthritis material characteristics, design, quality, and cultural safety, and used tables to summarize data along with specific examples of osteoarthritis materials.

No ethics approval was required because all data were publicly available.

We identified 176 potentially eligible items: 122 for persons with osteoarthritis and 54 for health care professionals. We excluded 129 items that did not meet eligibility criteria. This included 79 items for persons with osteoarthritis: content not relevant to osteoarthritis (n=29), duplicate (n=27), no educational content (n=17), focused solely on joint replacement (n=5) or not produced in Canada (n=1); and 50 items for health care professionals: not relevant to osteoarthritis (n=23), blank template or form to record details for patient records (n=10), duplicate (n=8), clinical guideline or standard (n=5), or focused solely on joint replacement (n=4). Ultimately, we included 47 items: 43 for persons with osteoarthritis [47-89] and 4 for health care professionals [90-93].

Multimedia Appendices 2 and 3 provide data on the characteristics of included osteoarthritis materials for persons with osteoarthritis and for health care professionals, respectively. Table 1 summarizes the characteristics of the included items.

Multimedia Appendices 4 and 5 provide data on the quality of included osteoarthritis materials for persons with osteoarthritis and for health care professionals, respectively. Table 1 summarizes the quality of the included items. There did not appear to be associations between osteoarthritis material quality and other characteristics.

Multimedia Appendices 6 and 7 provide data on the cultural safety of included osteoarthritis materials for persons with osteoarthritis and for health care professionals, respectively. Table 1 summarizes the cultural safety of the included items. There did not appear to be associations between osteoarthritis material cultural safety and characteristics.

We assessed the quality and cultural safety of 47 osteoarthritis materials: 43 for persons with osteoarthritis [47-89] and 4 for health care professionals [90-93]. Most were developed by charities, focused on osteoarthritis in general rather than osteoarthritis affecting a specific body part, and in the format of booklets. Many did not describe how they were developed; those that did were most often based on expert advice. Only 30% (13) patient osteoarthritis materials and 50% (2) health care professional materials scored high on the DISCERN scale for quality. Only 26% (11) of patient osteoarthritis materials scored high for cultural safety, and all health care professional osteoarthritis materials scored low for cultural safety. There did not appear to be associations between osteoarthritis material quality or cultural safety with other characteristics.

This study featured several strengths. We used rigorous content analysis methods and complied with qualitative research reporting criteria [25-27]. To ensure reliability, multiple individuals independently performed searching and screening, and data extraction and analysis. We also used multiple frameworks or instruments to describe osteoarthritis material characteristics [33], quality [4,34,42,43], and cultural safety [35-41,44-46]. An interdisciplinary research team that included ethno-culturally diverse women with osteoarthritis guided all aspects of the study. We must also acknowledge several limitations. The search strategy may not have identified all relevant osteoarthritis materials. Also, because we assessed only publicly available osteoarthritis materials, we may have missed osteoarthritis materials available to health care professionals through individual or organizational subscriptions. We excluded osteoarthritis materials available only in French, Canada’s second official language. While we included only English-language materials, we assessed whether those materials were available in languages other than English, which was not the case for any included in this study. We included only osteoarthritis materials developed in Canada. This enhanced study feasibility, but possibly excluded osteoarthritis materials of high quality or cultural safety that may be widely accessible on the internet. However, others could use the frameworks and instruments used here to assess osteoarthritis materials developed in their own countries. With no validated instruments to assess the quality of online materials for health care professionals, and no validated instruments to assess cultural safety of materials for either patients or health care professionals, we relied on sparse available research to compile frameworks by which to evaluate included materials. Future research could expand on and validate these frameworks.

Little previous research on support for first-line osteoarthritis management examined the characteristics of online osteoarthritis materials for patients or health care professionals [22-24]. One study that assessed 125 web pages on shoulder arthroplasty reported a low mean of 51.4 (SD 10.7) and a median of 51.5 (IQR 30.5-71.5) DISCERN score [94]. However, the study did not specify that arthroplasty was specific to second-line therapy for advanced osteoarthritis, and did not report which materials were for patients or health care professionals [94]. A meta-analysis of 8 randomized controlled trials showed that digital (eg, telephone, video, internet, and mobile app) self-management programs for persons with osteoarthritis significantly reduced pain and improved physical function at 12 months [95]. While this meta-analysis underscores the benefit of online osteoarthritis materials, it did not assess the quality or cultural safety of those materials. Similarly, another review of 11 randomized controlled trials found that digital (eg, internet or telephone app) self-management support programs for persons with osteoarthritis ranged from 6 to 9 months in duration, largely focused on physical activity, and resulted in program adherence of >70% in studies that reported adherence but did not assess the quality or cultural safety of programs [96]. A systematic review of 25 articles on mobile health technologies (mobile phones or other devices) and an assessment of 23 apps relevant to osteoarthritis self-management focused only on categorizing those technologies by role: self-management, decision support, or shared decision-making [97]. A previous review of online health information for the public found that most resources were of low quality but did not assess cultural safety [6]. Hence, this study is unique because we explicitly assessed online osteoarthritis materials for patients and health care professionals and found the majority of osteoarthritis materials scored low or moderate for quality and cultural safety.

Given the high prevalence of osteoarthritis and its debilitating effects [7,8], lack of access to osteoarthritis care among high-risk racialized immigrant women [11-18], lack of high quality, culturally safe osteoarthritis materials for persons with osteoarthritis or health care professionals revealed by this study, and pan-Canadian recommendations for improved access to publicly available osteoarthritis materials [20,21], greater efforts are needed to develop such materials. With respect to development methods, few osteoarthritis materials were explicitly based on published research, relying instead on expert advice, but it was not clear in all cases whether “experts” included persons with osteoarthritis. This knowledge could be applied to the improvement of existing osteoarthritis materials or the development of new materials. Because of the long-standing recognized importance of evidence-informed decision-making, the rubrics we used to assess quality (eg, DISCERN for patient materials and frameworks we compiled for health care professional material) included explicit identification of the underlying evidence and sources of data [4,34,42,43]. Furthermore, meaningfully engaging patients, family or caregivers, or members of the public in planning or developing health care policy, services, interventions, or tools is increasingly encouraged and common [98], and numerous resources are available to provide guidance on how to optimize their engagement [99]. Additional insight could be borrowed from methods used to engage persons with lived experience in developing other types of tools such as decision aids for patients and clinical guidelines for health care professionals [100,101]. Developers may want to give extra attention to quality domains that were frequently not met by osteoarthritis materials included in this study; for example, sources of information and benefits and harms of treatment for patient materials; and case-based and interactive for health care professional material. While we compiled cultural safety frameworks for both patient and health care professional osteoarthritis materials based on our research and that of others [35-41,44-46], none were specific to osteoarthritis, therefore further research is needed to establish what constitutes cultural safety in the context of osteoarthritis.

We aimed to assess the quality and cultural safety of online osteoarthritis informational or educational materials for persons with osteoarthritis and health care professionals. Of the 47 osteoarthritis materials (43 for persons with osteoarthritis, 4 for health care professionals), most were developed by charities, focused on osteoarthritis in general rather than osteoarthritis affecting a specific body part, and in the format of booklets. Many did not describe how they were developed; those that did were most often based on expert advice rather than explicit evidence. Few osteoarthritis materials for persons with osteoarthritis or health care professionals scored high for quality. Few osteoarthritis materials for persons with osteoarthritis and none for health care professionals scored high for cultural safety. Quality and cultural safety did not appear to be associated with osteoarthritis material characteristics. These findings reveal the need for further efforts to improve existing or develop new osteoarthritis materials for both persons with osteoarthritis and health care professionals. Further research is needed to assess the quality and cultural safety of osteoarthritis materials developed by organizations outside of Canada and to establish a framework or instrument to assess cultural safety in the osteoarthritis context.