Read the numbers behind a claim

Put percentages, uncertainty and patient relevance in the same frame before deciding what a result means.

A headline saying risk fell by fifty percent leaves several essential questions unanswered. Fifty percent of what starting risk, in which people, over how long? Absolute risk describes how often an event occurs in a defined population and period. Relative risk compares event probabilities between groups. Both can be useful, but they answer different questions. The National Cancer Institute explains this distinction in its risk information. Before interpreting a peptide-related claim, identify the event, the group sizes, the comparison and the time window.

Consider an invented study, used only to demonstrate arithmetic. Over one year, an event occurs in twenty of one thousand comparison participants and ten of one thousand intervention participants. The risks are two percent and one percent. The relative reduction is fifty percent because the risk was halved. The absolute reduction is one percentage point, or ten fewer events per thousand people. Both descriptions are mathematically correct. Neither establishes whether the hypothetical intervention is worthwhile without information about harms, uncertainty, costs and the importance of the event.