What the study found
The study found that as the number of matched data pairs increased, the win ratio tended to increase, though not in a regular pattern. It also found that the lower and upper limits of the win ratio confidence interval moved closer together as paired data increased, again not regularly.
Why the authors say this matters
The authors conclude that the increase in paired data affects the number of significant win ratios and the maximum win ratio. They also suggest that the win ratio confidence interval is influenced more by the data structure than by the size of the paired sample.
What the researchers tested
The researchers used a simulation study to examine how the number of matched data pairs affected the win ratio statistic and its 95% confidence intervals for two composite endpoints. They generated data for 35 scenarios with sample sizes from 10 to 1000 using the Python random library, and analyzed 34,101 data sets in total.
What worked and what didn't
As paired data increased, the win ratio increased overall, but the change was not regular. The number of cases where the win ratio was greater than one did not change with the number of paired data, while the number of win ratio values considered important increased as paired data increased. The confidence interval limits moved closer together as paired data increased, but not consistently.
What to keep in mind
This was a simulation study, so the findings come from generated data rather than observed clinical data. The abstract does not describe additional limitations beyond the note that the effects were not regular.
Key points
- The study examined the effect of matched data pair number on the win ratio statistic and its 95% confidence intervals.
- More paired data were associated with higher win ratio values, but the pattern was not regular.
- The number of cases with win ratio greater than one did not change with paired data size.
- The lower and upper confidence limits tended to move closer together as paired data increased.
- The authors conclude that confidence intervals were affected more by data structure than by sample size.
Disclosure
- Research title:
- More matched pairs changed win ratio and confidence intervals
- Authors:
- İsmet Doğan
- Institutions:
- Afyonkarahisar Sağlık Bilimleri Üniversitesi, Sağlık Bilimleri Üniversitesi
- Publication date:
- 2026-04-25
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with scholarly research without reading academic papers — one filtered digest, every Friday.