¿Can a meta-analysis include data from the same study twice, that is, in the same forest plot? I have seen this in a recently published investigation that included data with two different cutoff points, but I think it induces bias and affects heterogeneity. Is there a bibliography on the matter?
What do you mean "include data from the same study twice"? A study in a meta-analysis can include multiple outcomes, multiple time points, multiple comparisons, et cetera. Also, if a study has more than one part (e.g., a Part A and a Part B) with different subjects in each part, you can treat the study as if it is more than one study; each part can be its own study.
Including data twice from the same study in the same forest plot can create a unit-of-analysis issue unless the data is derived from independent arms. That is to say the same arms can not be repeated twice in a single analysis. See the link below for more information on how to deal with unit-of-analysis issues.
https://training.cochrane.org/handbook/current/chapter-23#section-23-3
Thank you all very much for your responses.
Gordon L Warren I am referring to data on the same variable, in the same subjects, in the same forest plot, in a meta-analysis of observational studies.
For example: include the visual acuity data of the eye with better vision and those of the eye with worse vision separately, as if it were another study, although the subjects are the same. It is my understanding that these data are not independent and that there is an assumption of independence for these analysis.
I have also seen that they include the same study twice (same subjects) because it reports two different cut-off points.
Hello
Sometimes an article reports data for two demographic groups (eg men and women, etc.) or two diagnostic tools. Therefore, in such cases, an article appears twice in a table or chart.
For example, once for women group data and once for men group data
.
The articles you say are probably like this. The diagnostic tool or population studied varies. Good luck
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis