“Fifty percent less” can sound impressive. Without the starting number, the actual size of the difference remains unclear. This calculator shows both perspectives using your entries. It does not evaluate a study or predict what a medicine will do for you.
Your tool
This tool does not send your entries. They remain in this open page.
Compare proportions of people for the same event and follow-up period; do not enter repeat events or person-years.
Define the event first
The input counts people who experienced a clearly defined event at least once. In a real study, that might be a favourable or an unfavourable outcome. Both groups must refer to the same event over a comparable follow-up period. Repeated events per person and person-years do not fit this simple model of proportions.
Keep the group sizes
Enter the total number of people and the number with an event for both A and B. The groups may be different sizes. People with events cannot outnumber the group, and group size cannot be zero. The calculator therefore requires whole numbers and rejects incompatible values instead of returning an apparently meaningful answer.
Take a fictional example apart
Imagine 20 people with an event among 1,000 in A and ten among 1,000 in B. The proportions are two percent and one percent. B minus A is minus one percentage point. Relative to the two percent in A, that is fifty percent fewer. Both statements describe the same invented data from different perspectives.
Read signs and the reference correctly
The absolute difference is always displayed as B minus A. A minus sign means fewer events in B; whether that is beneficial depends on the event. Relative change is proportion B divided by proportion A, minus one, multiplied by one hundred. If A has no events, that ratio is undefined. The tool does not invent a percentage in its place.
Recognise what arithmetic cannot establish
A correctly calculated difference proves neither causation nor statistical certainty. The tool does not calculate confidence intervals or assess bias, missing data, dropouts or differences between participants. It cannot turn an association from an observational study into a treatment effect. Real study interpretation still requires methods, the comparator, follow-up and uncertainty alongside the figures.
Common questions
Can I compare average weight losses here?
No. The calculator uses proportions of people with an event, not average kilograms or mean weight changes.
Why is there no relative value when A has no events?
Division by zero is undefined. The absolute proportions and their difference can still be shown.
Questions and experiences
Share a question or a useful experience about this topic. Every contribution is checked before publication. Comments may be written in different languages.
Loading reader contributions…
Opening the form sets a necessary session cookie. No email address is required. Publication is subject to moderation and is not guaranteed. For deletion requests: Contact