My watch list
my.bionity.com  
Login  

Forest plot



A forest plot is a graphical display that shows the strength of the evidence in scientific studies. It is used in medicine to combine all the studies of a treatment (in a meta-analysis) to show whether the treatment is effective. Forest plots are used frequently in evidence-based medicine.

For example, a forest plot can list all the randomized, controlled studies of drugs for hypertension in the left-hand column. The right-hand column contains a square for each study. The area of the square is proportional to the statistical strength of the evidence in the study. The more subjects in the study, the larger the square.

The right-hand column contains horizontal lines. If the end-point of the studies is deaths, then the horizontal lines might have ruler markings with the average number of deaths (100%) of the placebo group in the middle, 50% of the average deaths on the left, and 150% of the average deaths on the right. If a study show, for example, that the drug has 50% fewer deaths, the square for that study is placed on the 50% deaths mark.

Then a thicker horizontal line is drawn to illustrate the confidence interval of the study. If the confidence interval of the hypertension study is statistically significant (P<.95), then the horizontal line will start and end on the 50% side of the line. If the confidence interval is not statistically significant, then the horizontal line will cross the 100% mark.

Finally, under the squares, the studies are combined, and represented as a diamond. The statistical strength of the studies is represented by the area of the diamond. The confidence interval is represented by the width of the diamond. The narrower the diamond, the stronger the evidence. If the end of the diamond crosses the 100% line, then the studies are not statistically significant.

Forest plots date back to at least the 1970s, although the first use in print may be 1996.[1] The name refers to the forest of lines produced. In September 1990, Richard Peto joked that the plot was named after a breast cancer researcher called Pat Forrest and the name has sometimes been spelt "forrest plot".[1]

See also

  • Galbraith plot

Sources

  1. ^ a b Steff Lewis & Mike Clarke (June 2001). "Forest plots: trying to see the wood and the trees". BMJ 322: 1479–1480. doi:10.1136/bmj.322.7300.1479.
 
This article is licensed under the GNU Free Documentation License. It uses material from the Wikipedia article "Forest_plot". A list of authors is available in Wikipedia.
Your browser is not current. Microsoft Internet Explorer 6.0 does not support some functions on Chemie.DE