| Article Access Statistics|
| Viewed||2100 |
| Printed||18 |
| Emailed||0 |
| PDF Downloaded||70 |
| Comments ||[Add] |
Click on image for details.
|Year : 2016 | Volume
| Issue : 4 | Page : 840
Interpreting forest plots and funnel plots in meta-analysis
Sunil K Raina
Department of Community Medicine, Dr. RP Government Medical College Tanda, Kangra, India
|Date of Web Publication||5-Jul-2016|
Sunil K Raina
Department of Community Medicine, Dr. RP Government Medical College Tanda, Kangra
Source of Support: None, Conflict of Interest: None
|How to cite this article:|
Raina SK. Interpreting forest plots and funnel plots in meta-analysis. Neurol India 2016;64:840
I went through with interest the article entitled “Rheumatoid arthritis and the risk of dementia: A systematic review and meta-analysis” published in Neurology India (2016;64:56-61). Studies based on systematic reviews are an important means of summarizing the methods and results of individual studies. Systematic reviews are being increasingly used in the development of clinical practice as well as the starting and ending point of randomized trials. The authors, therefore, deserve credit for their effort. The authors have resorted to display of forest plot and funnel plot for interpretation of the results of their study, which is an accepted standard for conducting systematic reviews and meta-analyses. In this, the methodology used by the authors is correct. Herein, however, I would like to share a few concerns with the authors.
The authors in the past have emphasized the use of forest plots only when there are sufficient studies to make them of value, although this has not been defined. However, it appears (from the forest plot) that inclusion of only five studies may be slightly on the lower side to derive the benefits of drawing a forest plot for the interpretation of results.
This takes me to my second point of concern. The authors report in their 'Results' section that the visualization of the funnel plot did not provide suggestive evidence for publication bias because the graph was symmetric. The authors further resort to Egger's regression test (P = 0.65), and quite correctly so, to establish any publication bias. However, a look at the Cochrane review reveals that, as a rule of thumb, tests for funnel plot asymmetry should be used only when there are at least ten studies included in the meta-analysis. This, they state, is because when there are fewer studies, the power of the tests is too low to distinguish chance from real asymmetry. The Cochrane review further advocated that the tests for funnel plot asymmetry should not be used if all studies are of similar sizes (similar standard errors of intervention effect estimates). Importantly, the test proposed by Egger in 1997 may be used to test for funnel plot asymmetry (as has been done by the authors); however, the review suggests that the use of tests to assess the power of study with substantially fewer than 10 studies would be unwise.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| » References|| |
Ungprasert P, Wijarnpreecha K, Thongprayoon C. Rheumatoid arthritis and the risk of dementia: A systematic review and meta-analysis. Neurol India 2016;64:56-61.
Schriger DL, Altman DG, Vetter JA, Heafner T, Moher D. Forest plots in reports of systematic reviews: Across-sectional study reviewing current practice. Int J Epidemiol 2010;39:421-9.