Presenting the results of RCTs

Assuming that you are happy with the validity of an RCT, you need to move on to consider whether it is important. This means addressing the inevitable plethora of statistics which accompany the text.

This task should not be taken lightly. There is a lot of evidence to show that presenting the same information in a different way influences how we interpret it.

For example, one study of cancer specialists found that their decisions were influenced both by the format of presenting information (e.g. as a chart, table or icon) and by the emphasis of the information (e.g. whether failure rates were shown in red ink or not). (Elting LS et al, BMJ 1999;318:1527-31.)

So, just as researchers conducting a trial may unconsciously bias the methods towards a particular result, the way the results are presented may have a similar effect.

We need to be wary of such "framing effects" when we look at data from trials.