The Inclusion of Minority Groups in Clinical Trials: Problems of Under Representation and Under Reporting of Data

• Published in: Accountability in research Vol. 11; no. 3-4; pp. 215 - 223
• Main Authors: Rochon, Paula A., Mashari, Azad, Cohen, Ariel, Misra, Anjali, Laxer, Dara, Streiner, David L., Clark, Jocalyn P., Dergal, Julie M., Gold, Jennifer
• Format: Journal Article
• Language: English
• Published: United States Taylor & Francis Group 01-07-2004
• Subjects: Bioethics; clinical trials; Continental Population Groups; Ethnic Groups; ethnicity; Humans; medical journals; Minority Groups; National Institutes of Health (U.S.); Periodicals as Topic; Publishing; Randomized Controlled Trials as Topic; Research Design; systematic review; United States; United States; Empirical Approach; Biomedical and Behavioral Research
• ISSN: 0898-9621; 1545-5815
• DOI: 10.1080/08989620490891412

Objective: To evaluate the representation of minority groups in randomized control trials (RCTs), and the frequency with which this information is reported.
Study Design: Reviewers collected data on the racial/ethnic composition of study samples from all RCTs published in six leading medical journals in 1999.
Results: Of the 280 RCTs, most (204, 71.3%) provided no information on the race/ethnicity of participants. Of the 89 U.S.-based RCTs, 50 (56.1%) reported their minority distribution. Relative to other trials, those funded by the National Institute of Health (NIH) (n = 52) were more likely to report race/ethnicity data (55.8% vs. 23.7%; χ
2
= 20.9, p ≤ 0.001) and to include nonwhite participants (13.5% vs. 12.5%; χ
2
= 22.7, p ≤ 0.001).
Conclusion: Minority groups are currently under-represented in clinical trials. Information on the race and ethnicity of clinical trial participants is currently underreported in six leading medical journals. Reporting of minority group information was significantly better only in NIH funded trials, which also were more likely to include nonwhite participants. This suggests that mandatory reporting policies may have a positive effect on both reporting and representation.