Regulatory Science Virtual Symposium: “Make Informed Decisions: Key Statistical Principles to Clinical Trial Design” Session 6: Gender, Race, and Ethnicity in Clinical Trials (2022)

Research & Study Design
Communication, Dissemination, & Teamwork
Wendy Mack, PhD

Director, Biostatistics, Epidemiology, and Research Design

Course Syllabus/Topics

  1. Objectives
    1. NIH, FDA policies on inclusion and reporting by sex, race, ethnicity
    2. Rationale for inclusion and reporting by sex, race ethnicity
    3. Examples of heterogeneity of intervention effects by sex
    4. Implications for trial design and reporting
  2. NIH Policies on Inclusion
    1. Must provide justification if clinical trial does not include both women and men
    2. If prior research suggests that there might be differences for intervention effects by sex, race, ethnicity, then the trial should be designed to test efficacy within relevant subgroups
  3. FDA Policies
    1. Example of FDA Guidance: Collection of Race and Ethnicity Data in Clinical Trials
    2. Standardized and consistent collection and reporting of sex, race, and ethnicity data
    3. Be sure to include plans for retention and recruitment for specific subgroups
    4. Two-Question Format
      1. Question 1: Do you consider yourself Hispanic/Latino or not Hispanic/Latino?
      2. Question 2: Which of the following five racial designations best describes you? More than one choice is acceptable.
  4. NIH and FDA policies are recommendations, not mandates
  5. Rationale
    1. Incidence and survival of many diseases as well as drug effects vary by sex, race, and ethnicity
    2. Disparities are related to genetic differences, lifestyle, environment, and socioeconomic factors
    3. Clinical trial participation is historically highly education white men
    4. Inclusion of sub-groups needs to start with early phases of study
  6. Subgroup Reporting
    1. FDA demographic rule
    2. Population level PK studies
  7. Sex and Clinical Trials
    1. Studies were historically limited to men because women can become pregnant; hormonal fluctuations; and perceptions that certain illnesses are “male” diseases; women typically are smaller than men
  8. Sex vs. Gender in Clinical Trials
    1. Sex is biological vs. gender is socially constructed
    2. Clinical trials often use these terms interchangeably
  9. Sex and Gender in Clinical Trials
    1. Both factors influence how/what particular treatment a person selects, how they adhere to it, and how they metabolize a drug
    2. Ideally, trials should include equal numbers of men and women, because in cholesterol-lowering treatments, there were differences between the two groups
  10. Race (Mis)reporting and Participation in Cancer Clinical Trials
  11. Other Examples of Race/Ethnic Differences in Clinical Trials
    1. Race-focused trials showed that heart failure, kidney disease, and hypertension differ between black and white populations
  12. Implications for Trial Design and Analyses
    1. To address and understand health disparities, clinical trials must enroll enough participants in these subgroups to look for differences in efficacy, safety, and pathophysiology
    2. Reporting results by these sub-groups will assist in the analysis of data and identification of differences after the trials have ended
    3. Previous studies have recruited homogeneous groups to reduce statistical variability, but it is important to this also reduces the generalizability of research to other groups
    4. Groups that would be treated by this medical intervention should be the ones participating in the research studies
    5. Important to identify and respond to recruitment and retention
      1. Moving clinical trials out of institutions and into communities
      2. Having minority patients engage with professionals of similar backgrounds
      3. Multi-site trials
      4. Use of technology in clinical trial design
  13. Subgroup Statistical Analyses
    1. If a known gender/race/ethnicity effect of an intervention exists, then trial design must have sufficient sample size in order to conduct valid statistical analyses within each relevant group.
    2. If unknown, sufficient sample size in subgroups remain necessary to obtain reasonably precise estimates of effects
    3. Thank you!

​Acknowledgement

Accompanying text created by Annie Ly RKS Project Administrator, SC-CTSI  lyannie@usc.edu

NIH Funding Acknowledgment: Important - All publications resulting from the utilization of SC CTSI resources are required to credit the SC CTSI grant by including the NIH funding acknowledgment and must comply with the NIH Public Access Policy.