Trends in the content and format of main and optional consent forms in oncology clinical trials.

Abstract

153 Background: Informed consent forms (ICFs) should provide prospective subjects with an opportunity to balance the risks and benefits of study participation. However, there are growing concerns about the quality of ICFs. Optional ICFs are also increasingly used as the number of companion studies and biomarker evaluations requiring additional tests become more frequent. We examined trends in the content and format of main and optional ICFs. Methods: ICFs from clinical trials at a tertiary cancer center in British Columbia from 2000 to 2015 were reviewed. We focused on breast or gastrointestinal (GI) cancer studies. Readability was evaluated with the Flesch Reading Ease Score (FRES) and Flesch Kincaid Grade Level (FKGL) where a higher FRES (maximum 100) and a lower FKGL (maximum 12) indicated easier readability. We applied t-tests and linear regressions to examine variations among clinical trials and changes over time. Results: We identified 133 main ICFs of which 70% had optional ICFs and where 57% and 43% were breast and GI cancer studies. Phase III trials (44%), industry funded investigations (70%), and studies involving palliative therapies (72%) were most common. Trials from recent years were more likely to have optional ICFs than those from earlier years (p < 0.001). The median length and median word count in main and optional ICFs were 16 and 6 pages and 6183 and 1862 words, respectively. These changed significantly over time whereby main ICFs increased approximately by 1 page and 364 words per year over the 15 year period (p < 0.001). Industry funded trials also had longer ICFs (p = 0.006). Study methods, risks, and confidentiality occupied 29%, 20%, and 11% of the content on ICFs, respectively. Sections pertaining to eligibility (p < 0.001) and screening procedures (p = 0.007) also increased with time, particularly for industry funded studies (p = 0.006). In terms of readability, optional ICFs were generally more difficult to read than main ICFs (FRES 48.3 vs 50.0, p = 0.024; FKGL 11.8 vs 11.1, p < 0.001), especially in recent years (p < 0.001). Conclusions: This is one of the first analyses to include optional ICFs. Length of ICFs is increasing and readability is discordant with the average reading level of potential trial participants.