Research methods guide
How to report logistic regression results
Report the adjusted odds ratio and confidence interval, category contrast or unit increment, adjustment variables, analysis sample, and model.
Suppose Table 2 gives β=0.1823 per 10 exposure units, adjusting for age and sex in 428 participants. Convert the coefficient to a traceable results sentence.
From a logistic model table to a manuscript sentence
Values are synthetic. Use the interval from the same fitted analysis.
| Item | Worked example | Verification |
|---|---|---|
| Convert the scale | exp(0.1823)≈1.20. If the table already gives aOR 1.20, report it directly. | Verify whether the column is a coefficient or an odds ratio before exponentiating. |
| State the exposure and event | “Per 10 exposure units, the adjusted OR for 30-day readmission was 1.20 (95% CI 0.98–1.47).” | Verify units, event coding, reference category, and follow-up window. |
| Record the sample and adjustments | “The model adjusted for age and sex and included 428 participants.” | Verify the formula, analysis n, and handling of missingness. |
Apply the method to your materials
Create a traceable results row
Record outcome, coding, increment or reference, model, n, effect, interval, and source. Name groups for categorical exposures and the increment for continuous exposures.
Reconcile text and table notes
Check that the abstract and manuscript use the same model and rounding. Label crude, adjusted, and sensitivity estimates distinctly.
Questions that arise in practice
Can OR 1.20 be reported as 20% higher risk?
Report the odds ratio as 1.20. Estimating a probability difference requires a suitable method, target population, and baseline information.
Verified sources
The official sources below were checked on 2026-11-12. Access conditions and update dates are rechecked before publication.
RigorMed workspace
Start with your research materials
Bring your research question, available materials, and the steps you want to check.
