Risk converter
Turn "doubled" or "50 percent higher" into before-and-after counts out of the same total, and get a sentence you can say aloud.
The converter does arithmetic on the numbers you give it. It does not check them, and it does not say whether they came from a sound study. Communicating uncertainty explains why to say the counts instead of the percent change. The starting values are the 1995 pill-scare example, so the page shows a result the first time it opens.
How to use it
- Find the baseline. How many people are affected without the change, out of how many? A relative change means nothing without this pair of numbers. If the source does not give it, ask the source or say you do not have it.
- Choose what you know. Either the count with the change, or the relative change in percent and whether it goes up or down.
- Read the sentence. Say the two counts out of the same total, then the difference. Add the time frame yourself: "per year," "over five years," "in a lifetime." The converter cannot know it.
- Write it down. Put the result in the absolute-form column of the numbers table, with the source.
Where the starting example comes from
In October 1995 the UK Committee on Safety of Medicines warned that third-generation contraceptive pills doubled the risk of blood clots, a rise of 100 percent. Gigerenzer and colleagues report the absolute numbers: about 1 in 7,000 users of the older pills had a thrombosis, against 2 in 7,000 for the newer ones. The absolute increase was 1 in 7,000. The paper reports an estimated 13,000 additional abortions in England and Wales the following year. The example shows what a relative figure hides; it makes no claim about any pill today. Source: Gigerenzer and colleagues, Psychological Science in the Public Interest (2007).
Assumptions behind the numbers
- New risk = baseline risk times (1 plus or minus the relative change divided by 100). Absolute difference = new minus baseline, on the same total.
- The FDA's evidence-based guide to communicating risks and benefits recommends absolute risks, not only relative ones, a constant denominator and a constant time frame, and whole numbers, using a larger denominator for small risks. See the FDA guide (2011), Chapter 7.
- The National Cancer Institute's Making Data Talk adds: three or fewer numbers, familiar units, no false precision.
- Rounding to whole counts can hide a small difference. When the counts come out below 1, the converter offers a larger total.
This page converts numbers. It does not judge whether a study, an estimate or a test is sound, and it gives no medical, safety or risk advice. The author of the estimate, an expert in the field or the agency responsible settles those. A correct count can still mislead without its time frame and its population.