Using information about validity in your review

So what should you do with this information you collected?

There are four approaches

• as a threshold for inclusion of studies
• as a possible explanation for differences in results between studies
• in sensitivity analyses
• as weights in statistical analyses

These are discussed in detail in section 6.10 of the Reviewers' Handbook, which you should now read.

Everyone uses some sort of threshold for inclusion of studies, and this was set when we chose the selection criteria for the review. Sensitivity analyses are commonly undertaken in reviews to see the effect of aspects of quality. Because of the problems in deciding the weight to allocate to different aspects of quality, this is rarely done in Cochrane reviews. As a minimum, you should discuss the range of quality found in your included studies, and whether you think this may have any influence on the results of these studies, and any meta-analysis you performed. We will return to this topic in a later module, when we look at the investigation of heterogeneity.

Finally, check with your review group whether they have a policy on how to incorporate study quality in reviews.

Read the trial about a nursing intervention for treating heart failure. You will find the report of the trial at:

http://www.bmj.com/cgi/content/full/323/7315/715

Appraise this article as you plan to for your review. Ensure you extract information about allocation concealment, blinding and loss to follow up.

Test yourself by comparing your answers to the suggested responses on the next page.