An odds ratio is not a quitting probability

Name both parts of the comparison

For a defined event, risk is the proportion with the event; odds compare the number with it to the number without it. Cochrane explains that confusing those scales can misrepresent an intervention comparison. The difference is especially relevant when the event is common, so familiar conversational language can be misleading.

Find whether the report used an odds ratio, risk ratio, or another measure. Also locate what counts as the event: a particular abstinence definition, a survey response, or something else. The same statistical term can describe very different endpoints. Do not replace a study’s exact definition with a broader word such as success.

Cochrane: Cochrane Handbook, Chapter 6: Choosing effect measures

Use an original arithmetic example

Here is a fictional counting sheet, not a study. Group A has 30 event rows and 30 nonevent rows; Group B has 20 event rows and 40 nonevent rows. The group risks are 30/60 and 20/60, so the risk ratio is 1.5. The odds are 30/30 and 20/40, so their odds ratio is 2.

The same made-up sheet therefore supports two different numerical ratios without either calculation being wrong. Calling the odds ratio twice the probability would change its meaning. The invented event has no medical definition, and these counts provide no evidence about any vaping intervention, person, or app.

Do not convert an isolated headline yourself

A real report might give an adjusted odds ratio from a model rather than the simple comparison in the example. Its uncertainty, variables, and analysis population still matter. Do not apply the toy arithmetic to that model or invent an unreported baseline probability in order to manufacture a personal success percentage.

If a promotional graphic provides only a ratio, request the underlying report and the absolute group information where available. Keep unavailable details marked unavailable. A ratio above one also needs the event’s direction: more of the measured event cannot be labeled beneficial merely because the displayed number is larger.

Write a statement you can defend

An accurate note could say that the report presents this odds ratio for this event, in these groups, at this follow-up. If the headline says times as likely, note the scale mismatch instead of substituting your own clinical interpretation. Preserve the uncertainty interval and the report’s limitations when sharing the finding.

This exercise teaches terminology and arithmetic, not appraisal of a named treatment. A qualified professional can help interpret relevance to a health decision. A personal puff or craving record cannot supply the comparison-group outcome needed to translate a research estimate into an effect for you. Continue practical planning without borrowing a probability the report never established.