Explanation of different types of spin and tips on how to avoid them
| Type of spin | Explanation | How to avoid |
| Focus on statistically significant effects for within-group comparisons | Highlighting significant improvements from baseline in the intervention group, without acknowledging the lack of difference in effect between groups | Focus on between-group effects, even if they are not statistically significant |
| Focus on statistically significant effects for secondary outcomes | Using statistically significant intervention effects on secondary outcomes to highlight the benefit of an intervention without acknowledging non-significant effects on the primary outcome(s) | Focus on between-group effects for the primary outcome(s), even if they are not statistically significant |
| Focus on statistically significant effects for subgroup analyses | Using statistically significant intervention effects in a subgroup of the study population to highlight the benefit of an intervention without acknowledging non-significant effects in the primary population of interest | Focus on between-group effects for the primary outcome(s) in the primary population of interest, even if the effects are not statistically significant |
| Focus on statistically significant effects for modified population of analyses (eg, intention to treat analyses) | Using statistically significant intervention effects in per-protocol analyses to highlight the benefit of an intervention without acknowledging non-significant effects in the intention to treat analyses. | Focus on between-group effects from the intention to treat analyses, even if they are not statistically significant |
| Interpret statistically non-significant effects for the primary outcome(s) as showing treatment equivalence or comparable effectiveness | Stating that both the intervention and control intervention were beneficial despite a non-significant between-group effect on the primary outcome | Do not report that an intervention was effective if the between-group difference was not statistically significant |
| Highlighting the benefit of an intervention despite statistically non-significant effects | Highlighting differences in outcome values between the intervention and control without acknowledging there was no statistically significant difference between these values. Infographics often did this through a figure that displayed a difference in outcome values between groups, without mentioning the non-significant p value | Present differences in outcome values with accompanying effect sizes and measures of precision |
| Focus on statistically significant effects for one primary outcome while ignoring non-statistically significant effects for other primary outcomes | Using statistically significant intervention effects for one primary outcome to highlight the benefit of the intervention without acknowledging non-significant effects for other primary outcomes | Mention between-group effects for all primary outcome(s), even if some are not statistically significant |
| Focus on statistically significant effects for the primary outcome at a non-primary time-point | Using statistically significant intervention effects for the primary outcome at a non-primary time point (eg, 8 weeks) to highlight the benefit of the intervention without acknowledging non-significant effects at the primary time-point (eg, 4 months) (when a primary time point is specified) | Focus on between-group effects for the primary time point, even if it was not statistically significant |
| Focus on statistically significant effects for the primary outcome at one time point while ignoring non-statistically significant effects for the primary outcome at other time points (when no primary time point is specified) | Using statistically significant intervention effects for the primary outcome at some time points to highlight the benefit of the intervention without acknowledging non-significant effects at other time points | Acknowledge when between-group effects for a primary outcome were significant at some time points but not others |