Blind spots in reporting and representation in knee osteoarthritis research: A systematic review of the sociodemographic and clinical characteristics of 22,443 participants in randomised controlled trials of first-line care

Objectives Knee osteoarthritis affects millions globally, and understanding characteristics of participants included in randomised controlled trials (RCTs) that inform current treatment recommendations is required to assess their global applicability. We aimed to comprehensively describe the sociodemographic and clinical characteristics of participants reported in knee osteoarthritis RCTs of first-line care. Methods Nine databases, guideline repositories, and professional associations' websites were searched from inception to March 2025 for higher-quality systematic reviews (determined by A MeaSurement Tool to Assess systematic Reviews-2) and clinical practice guidelines (CPGs) (determined by Appraisal of Guidelines for Research and Evaluation II). Sociodemographic (guided by the PROGRESS-Plus framework) and clinical characteristics reported in RCTs included within these systematic reviews and CPGs were extracted and summarised descriptively. Study registration: PROSPERO #CRD42023463878. Results We included 241 RCTs involving 22,443 participants, representing 17% of countries (34/198). 95% were (n=229/241) of moderate-to-high quality (PEDro). 55% (n=133/241) were from high-income countries. None were from low-income countries. No RCTs originated from Eastern Europe, Central Asia, the Caribbean, East and Central Africa, Melanesia, Micronesia, nor Polynesia. Most sociodemographic variables were reported in less than 15% of RCTs, including residence (2%, n=6), primary language (0%, n=0), employment (14%, n=33) and socioeconomic status (7%, n=17). Only 15% (n=36) of RCTs reported race/ethnicity; of which 32 originated from the United States. Reporting of sociodemographic variables did not improve over time. Randomised controlled trial participants were typically older (average=63 years), white women/female, tertiary educated, overweight (average BMI=30 kg/m 2 ), multimorbid (range=2-14 comorbidities), and experiencing moderate pain (average=58/100). Conclusion Limited reporting and global population representation in knee osteoarthritis RCTs reduces applicability of findings, undermines treatment recommendations, and risks exacerbating existing global health inequities. Addressing reporting and representation blind spots is urgently needed in RCTs informing first-line care for knee osteoarthritis.