The therapeutic arsenal in rheumatology has improved drastically during the last 20 years with the availability of biological and synthetic diseasemodifying antirheumatic drugs. All these drugs required randomised controlled trials (RCTs) with thousands of rheumatic patients. All these patients went through the informed consent process and signed corresponding informed consent forms (ICFs). Objectives: To assess trends in the length and readability of ICFs for industrysponsored RCTs over a 17

more »

... ar period. Additionally, to assess the health literacy (HL) and perceptions on ICFs among participants of current RCTs. Methods: The grammatical readability (GR) of six pre-defined ICF sections (global, introduction, methods, risks/benefits, rights/responsibilities, and voluntary participation statement) of pivotal ICFs from industry-sponsored RCTs conducted at an outpatient rheumatology clinic during a 17 year period was assessed by using the INFLESZ instrument. Differences in GR were assessed by company, by disease and by study phase (95% CI of the mean and proportions; statistical significance assumed if no overlap), and by the years ICFs were written (1999 -2005, 2006-2010, and 2011-2016 Kruskal-Wallis test). HL of patients was assessed with the SALHSA (inadequate, adequate) and S-TOFHLA (inadequate, marginal, adequate) instruments. Differences by age and schooling were determined by one-way ANOVA. Patient's perceptions and opinions on the ICF were assessed using a structured, self-reported, in-office questionnaire on an independent patient sample that had signed a pivotal ICF in the past 6 months. Results: Thirty-nine ICFs about 22 drugs (18 biological or targeted synthetic DMARDs; 13 currently available on the market) from 13 pharmaceutical companies were analysed. The global mean readability was 57±3 (95% CI 56-58), and all ICFs were categorised as either "somewhat difficult to read" or "average". Readability remained at these levels without significant changes from 1999 to 2016. The "somewhat difficult" reading score was significantly more frequent in the "rights and responsibilities" and in the "informed consent statement" sections (p<0.001). The mean length of the ICFs written between 1999 and 2005 was 13±5 pages, with a significant increase thereafter (22±8 pages, p<0.01). Depending on the instrument, of 95 patients participating in the HL assessment, between 18% and 44% had limited HL. Of 90 patients participating in the perceptions questionnaire, 84% reported understanding the ICF well. However, 2%-57% misunderstood basic concepts, including the study drug name, randomization and placebo. Conclusions: It seems that the considerable progress that has been made in medical research methods over time has not produced improvements in the process of informed consent in the industry-sponsored RCTs. The disparity between the readability of ICFs with patients' HL and their comprehension of ICFs continues, even after decades of attempts of regulatory agencies and numerous published suggestions.