Forest plots

Question 1

What is a forest plot?

Answer 1

It is a graphical representation of a number of results from different studies.

And is the main method for illustrating the results of a meta-analysis

Question 2

• Name of the trials is on the left - in chronological order.
• Right side - squares centred on the point estimate of the result in each trial.
• Size of the square is proportional to the weight of the study.
• The line running through the square shows the confidence interval (usually 95%)
• Large vertical line - the line of no effect - if the line crosses this, it means the study could be insignificant.
• Diamond underneath the studies- summary result.

Answer 2

Question 3

A meta-analysis examines whether giving a new dietary supplement to patients who’ve recently had a myocardial infarction can help prevent a further attack. The meta-analysis consists of four randomised controlled trials and is summarised below. What is the most appropriate interpretation of the data?

• There is publication bias in studies looking into this question
• There is a non-significant trend that taking the supplement reduces the chance of a further myocardial infarctions
• There is a non-significant trend towards no benefit from taking the supplement in terms of reducing further myocardial infarctions
• Taking the supplement reduces the chance of a further myocardial infarctions
• Taking the supplement increases the chance of a further myocardial infarction

Answer 3

Taking the supplement increases the chace of further myocardial infarction.

The meta-analysis of the results, represented by the diamond, is clear from the no effect line (odds ratio of 1) and shows a significant increase in the chance of a further myocardial infarction.