The 28.3% figure is an average reduction for the 12 mg trial arm at 80 weeks under the sponsor-reported efficacy analysis. The corresponding placebo estimate was 2.2%.
A group average with an analysis definition
A percentage of baseline weight differs from the percentage of people who reached a weight-loss threshold. Both appear in trial tables, so the row labels matter.
How to interpret the evidence
Include the arm, time point, population and analysis when citing the figure. This preserves the meaning even when the number is quoted on its own.
When comparing results, record the primary outcome, the measurement week, and whether the analysis included outcomes after treatment stopped.