Sex-Specific Analyses and Reporting in Clinical Trials The Health Researcher’s Toolkit: Why Sex & Gender Matter

  • Introduction
    • The Health Researcher’s Toolkit: Why Sex & Gender Matter
    • Introduction
    • Module Objectives
    • Presenter Profile
  • Content
    • Sex & Gender
    • Module Introduction
    • Module Prerequisites
    • Key Terms & Concepts
    • Dr. Peter Jüni Part I: Lecture Video
    • Module Summary
    • Dr. Peter Jüni Part I: Clinical Trials and Qualitative Studies
    • Dr. Peter Jüni Part I: Risk
    • Dr. Peter Jüni Part I: Interpretation of Sub-Groups
    • Dr. Peter Jüni Part I: Meta-Analysis for Sub-Group Differences
    • Knowledge Check-in
    • Dr. Peter Jüni Part II: Lecture Video
    • Dr. Peter Jüni Part II: Sub-Group Analysis and Interpretation
    • Dr. Peter Jüni Part II: Sub-Group Reporting
    • Dr. Peter Jüni Part II: Design Recommendations
    • Module Summary II
  • Conclusion
    • Module Quiz
    • Quiz Results
    • Reflection
    • Resources
    • About Women's Xchange
Auto-play audio transcript  

The Health Researcher’s Toolkit: Why Sex & Gender Matter

Developed by the Women's Xchange

  • Define sex and gender and know how to correctly apply these terms
  • Explain why sex and gender matter in health research
  • Identify and apply methods for integrating sex and gender in different types of studies
Auto-play audio transcript  

Introduction

Sex-Specific Analyses and Reporting in Clinical Trials

Presenter: Dr. Peter Jüni

Auto-play audio transcript  

Module Objectives

In this module, you will learn

  • How to appraise results of sex-specific clinical trial data
  • What to consider when designing, interpreting and reporting
  • Why clinical trials should report all relevant outcomes by sex
Auto-play audio transcript  

Presenter Profile

Dr. Peter Jüni
Director, Applied Health Research Centre (AHRC)
Li Ka Shing Knowledge Institute, St. Michael’s Hospital
Professor, Department of Medicine
University of Toronto

Download Dr. Jüni’s biography

Auto-play audio transcript  

Sex & Gender

Not interchangeable terms

Sex

  • Biological attributes
  • Associated with physical and physiological features
  • Often conceptualized as binary: female or male
    • However biological attributes and expression can vary as individuals may be born with reproductive anatomy that doesn’t fit the typical definition of female or male
  • Commonly understood as what was assigned at birth

Gender

  • Socially constructed and fluid
  • Culturally specific
  • Roles, behaviours, expressions, identities of girls, women, boys, men, gender diverse people
  • Gender identity
    • One’s innermost concept of self
    • May be same or differ from sex assigned at birth
Auto-play audio transcript  

Module Introduction

Transcatheter Aortic Valve Implantation (TAVI)

  • Treatment for high-risk patients with severe aortic stenosis

Clinical trials

  • Rarely identified sex of participants
  • Assumed outcomes same for each sex

Sex as sub-category

  • Consider differences in sexes’ underlying health when interpreting data
Auto-play audio transcript  

Module Prerequisites

This module assumes the learner is familiar with:

  • Clinical trials
  • Associated terminology
  • Forest plots for displaying data

Unfamiliar with these concepts?

  • Peter M Rothwell, “Subgroup analysis in randomised controlled trials: importance, indications, and interpretation”, The Lancet, 2005, Volume 365, No. 9454
Auto-play audio transcript  

Key Terms & Concepts

Sub-group analysis: evaluation of treatment effect on a small group of patients defined by baseline characteristics

Pre-specified sub-group analysis: a hypothesis that is planned and documented before any examination of the data, preferably in the study protocol

Post-hoc analysis: a hypothesis tested that was not specified before any examination of the data, rather it was suggested after doing the analysis

Baseline risk: the rate of occurrence of an event when standardized treatment is used

Relative risk: the risk of an event occurring in one group compared to the risk of the same event happening in another group. A relative risk of 1 means there is no difference between the groups. A relative risk greater than 1 means that being exposed to a certain substance increases the risk of an event. A relative risk less than 1 means being exposed to a substance reduces the chance of an event. This is also called a risk ratio.

Individual Patient Data or IPD meta-analysis: To conduct an IPD meta-analysis, original data from two or more studies is re-analysed and, if appropriate, combined.

References:

https://fhs.mcmaster.ca/anesthesiaresearch/documents/SubgroupandPost-hocAnalyses_BenefitsandChallenges.pdf

http://methods.cochrane.org/ipdma/about-ipd-meta-analyses

Dr. Peter Jüni Part I: Lecture Video

Auto-play audio transcript  

Module Summary

Improve health outcomes for everyone

  • Sub-grouping by sex in clinical studies
    • Design, report and interpretation implications

Clinical trials can signal sex differences but they may not be powered to assess absolute effect

  • Meta-analysis is more effective for comparing sex differences across multiple clinical trials
    • Clinical researchers should report all sub-group outcomes, even null findings

When appraising sex-specific results, consider

  • Baseline differences compared to relative risk differences
  • Pre-specified (i.e., defined a priori) or post-hoc
Auto-play audio transcript  

Dr. Peter Jüni Part I: Clinical Trials and Qualitative Studies

Clinical trials

  • Compare options
  • To explore “why”, nest qualitative study within the trial
Auto-play audio transcript  

Dr. Peter Jüni Part I: Risk

Baseline risk

  • The rate of occurrence of an event when standardized treatment is used

Relative risk

  • The risk of an event occuring in one group compared to the risk of the same event happening in another group
Auto-play audio transcript  

Dr. Peter Jüni Part I: Interpretation of Sub-Groups

Interpreting sub-group differences

  • Consider point estimate
  • Consider confidence intervals
  • Test for interaction of treatment effect between groups
Auto-play audio transcript  

Dr. Peter Jüni Part I: Meta-Analysis for Sub-Group Differences

Clinical trials often unable to detect/prove sub-group differences

  • Meta-analysis is more successful

Knowledge Check-in

Clinical trials are best suited for:

True or False: When interpreting this finding, the researchers must consider the risk difference in males and in females for the underlying health problem

Which of the following are correct statements about meta-analyses?

Auto-play audio transcript  

Dr. Peter Jüni Part II: Lecture Video

  • Interpretation
  • Reporting
  • Implications for designing clinical research of sub-groups
Auto-play audio transcript  

Dr. Peter Jüni Part II: Sub-Group Analysis and Interpretation

Interpretation

  • Was the sub-group analysis pre-specified or post-hoc?
  • What is the strength of the evidence against the null hypotheses?
  • Is the finding biological plausible?
Auto-play audio transcript  

Dr. Peter Jüni Part II: Sub-Group Reporting

Reporting

  • Minimum: all outcomes with crude numbers and hazard ratios
  • Better: baseline and procedural characteristics
  • Optimal: time to event curves
Auto-play audio transcript  

Dr. Peter Jüni Part II: Design Recommendations

Design

  • Pre-specify subgroup analysis
  • Achieve a power of 80% for treatment X sex interaction
  • Collect explanatory baseline and procedural characteristics
  • Consider prospective individual patient data (IPD) meta-analysis
Auto-play audio transcript  

Module Summary II

Improve health outcomes for everyone

  • Sub-grouping by sex in clinical studies
    • Design, report and interpretation implications

Clinical trials can signal sex differences but they may not be powered to assess absolute effect

  • Meta-analysis is more effective for comparing sex differences across multiple clinical trials
    • Clinical researchers should report all sub-group outcomes, even null findings

When appraising sex-specific results, consider

  • Baseline differences compared to relative risk differences
  • Pre-specified (i.e., defined a priori) or post-hoc
Auto-play audio transcript  

Module Quiz

Test Your Understanding

  • Five questions
  • Select the best response
  • When you have answered all questions, click submit

1. True or False: A post-hoc analysis of study participants by sex is just as valid as a pre-specified analysis.

2. When interpreting results with possible sex differences, it is important for the research to consider:

3. The analysis of individual-level data in multiple clinical trials that address similar research questions is known as:

4. Researchers should report data by sex in clinical trials:

5. True or False: If the baseline risk for females is much lower than for males, this will impact how the findings should be interpreted.

Quiz Results

You got out of questions correct.



1. True or False: A post-hoc analysis of study participants by sex is just as valid as a pre-specified analysis.

True

False

2. When interpreting results with possible sex differences, it is important for the research to consider:

Biological plausibility

Statistical significance

If the analysis was pre-specified

All of the above

3. The analysis of individual-level data in multiple clinical trials that address similar research questions is known as:

Post-hoc subgroup analysis

Sociodemographic analysis

Pre-specified subgroup analysis

IPD meta-analysis

4. Researchers should report data by sex in clinical trials:

Only if there are positive findings

Only when sex differences are noted

Only if the participants’ sex is relevant to the research question

At every opportunity; this information may be useful for future meta-analysis

5. True or False: If the baseline risk for females is much lower than for males, this will impact how the findings should be interpreted.

True

False

Auto-play audio transcript  

Reflection

How are sub-groups in clinical trials defined?

What sort of assumptions are made about participants?

  • See "Inclusive Qualitative Data Collection" in The Health Researcher’s Toolkit

In a meta-analysis

  • What if the same sub-group has been defined differently across various clinical trials?
  • What if a sub-group is not measured?

How will a change in baseline risk for a sub-group impact interpretations of relative risk?

Resources

Please See Additional Resources Below:

Fabien Praz, et al., “Latest evidence on transcatheter aortic valve implantation vs. surgical aortic valve replacement for the treatment of aortic stenosis in high and intermediate-risk patients”, Current Opinion in Cardiology, 2017 Mar;32(2):117-122. doi: 10.1097/HCO.0000000000000379.

Philippe Pibarot, et al., “Incidence and Sequelae of Prosthesis Patient Mismatch in Transcatheter Versus Surgical Valve Replacement in High-Risk Patients with Severe Aortic Stenosis”, Journal of the American College of Cardiology, Vol. 64, No. 13, 2014.

Patricia Guyot, et al., “Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves”, BMC Medical Research Methodology, 2012, 12:9.

Sara T. Brookes, et al., “Subgroup analyses in randomized trials: risk of subgroup-specific analyses; power and sample size for the interaction test”, Journal of Clinical Epidemiology, 57, 2004.

Susan Phillips, “Defining and Measuring gender: A social determinant of health whose time has come”, International Journal for Equity in Health, 2005, 4:11.

Peter M. Rothwell, “Subgroup analysis in randomised controlled trials: importance, indications, and interpretation”, The Lancet, 2005, Volume 365, No. 9454.

Peter M. Smith and Mieke Koehoorn, “Measuring gender when you don’t have a gender measure: constructing a gender index using survey data”, International Journal for Equity in Health, 2016, 15:82.

About Women's Xchange

Based at Women’s College Hospital, Women’s Xchange is a women’s health knowledge translation and exchange centre, designed to promote women’s health research across the province. Funded by the Ministry of Health and Long-Term Care’s Health Service Research Fund (HSRF), the centre supports women’s health research in both academic and community settings. In addition to supporting research, Women’s Xchange also provides women’s health researchers and trainees across the province with opportunities to gain new skills and develop new collaborations.

The Health Researcher’s Toolkit: Why Sex and Gender Matter

Developed by


With support from: