Reporting of Race and Ethnicity in Randomized Controlled Trials in Hematology

Background: Race and ethnicity are important social constructs that are frequently reported in clinical research. Race and ethnicity data can provide insights into the impact of racism, structural inequities, and in some cases ancestry on health outcomes. To ensure that results are generalizable, clinical trials should reflect the diverse populations affected by the disease under investigation. In addition, clear, transparent reporting on how and why race/ethnicity was or was not collected, and a discussion of findings are important. In 2021, JAMA published updated guidance on reporting race and ethnicity as part of the Inclusive Language section of the AMA Manual of Style: A Guide for Authors and Editors as subsection 11.2.3, Race and Ethnicity. We sought to determine the extent to which randomized controlled trials (RCTs) in Hematology, published after these updated guidelines, follow AMA recommendations in this space.

Methods: We conducted a systematic review of RCTs of treatments for blood disorders published between April 1, 2022, and June 27, 2024, in 5 high-impact journals: NEJM, JAMA, Lancet, Lancet Hematology, and Blood. The inclusion criteria were: Randomized controlled trials, English language, adult participants, and pertaining to either benign or malignant blood disorders. Data extraction was completed in duplicate with differences resolved by consensus. The following binary outcomes were assessed: 1) Does the methods report whether participant race/ethnicity was collected, 2) Was the rationale for collecting or not collecting race/ethnicity outlined, 3) Was race/ethnicity defined, 4) If race/ethnicity was collected, was it stated who identified the race/ethnicity of participants, 5) Was race/ethnicity of participants reported, and 6) Did the study discuss the implications of race/ethnicity on the findings. Associations between region, journal policies and race/ethnicity reporting were explored in univariate analysis using the chi-squared test for significance.

Results: Using MEDLINE, a total of 967 articles were identified. After title and abstract screening, 172 studies were selected for full-text review. Of these, 88 studies were identified for data extraction. For the purposes of this abstract, studies from Jan 1, 2023 to Jun 27, 2024 were included, totaling 61 papers. A minority of methods sections explicitly reported collecting race/ethnicity data (31.1%), why race/ethnicity was or was not collected (6.5%), how race was defined (0%), or who identified the race/ethnicity of participants (29.5%). Most papers (65.6%) reported race/ethnicity data. However, only 26.2% of papers discussed the implications of race/ethnicity with respect to their findings. In univariate analysis, trials recruiting exclusively in Europe were less likely to report race/ethnicity than trials from other regions (p <0.01). Trials published in journals with a formal policy on how and when to report race/ethnicity data had a trend to higher reporting of race/ethnicity (p=0.06) and were more likely to discuss the implications of their findings with respect to race/ethnicity (p=0.04).

Conclusion: Efforts to address racial and ethnic disparities in research and healthcare are increasing, with most high-impact studies in this preliminary sample having reported race/ethnicity of participants. Despite this, few trials reported how race/ethnicity was determined or discussed the implications of race/ethnicity with respect to their findings. Regional variation in reporting of race/ethnicity was observed. Formal journal policies may increase adherence to AMA guidance in this domain.