Fig. 1: Study design and summary result.
From: Outcome differences by sex in oncology clinical trials
a Schematic of our study design. Subpanel 1) Using text mining methods we identified trials in Trialtrove that may have a comparison of outcomes or side-effects by sex. Subpanel 2) We individually curated each candidate trial to record what was measured such as a survival difference or a difference in the frequency of a side effect such as rashes; we recorded whether there were differences or similarities between the sexes and what evidence was provided. b A flow diagram showing the breakdown of data collected. All data points originated from the 89,221 oncology clinical trials in Trialtrove on December 23, 2022. On the left we show all 532 sex comparisons found, among which 356 showed sex differences and 176 showed no difference. These comparison sets are classified by the type of evidence they present. Yellow boxes indicate categories with significant evidence. On the right we show another view of the data flow centered on the trials for which we found sex comparisons (shown on the left). Filtering for trials which present statistically significant evidence, we show the 288 trials with SOR comparisons and the 44 trials with SE comparisons. These two groups are broken down further by whether they show a preference towards males (blue), females (red), or neither (gray). Panel b may be interpreted as a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram in which the unit of analysis is a clinical trial.
