The Cochrane Collaboration open learning material  The Cochrane Collaboration web site
 
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 A1 A2
Module contents:
Collecting data from relevant studies
Learning objectives
Collecting the data you want, in a format you want it
Review, reviewer and study information
Study eligibility
Table of included studies
Study quality
Data for results
Testing the form
Next module

Part Five of a Data Collection Form:
Data for Results

The final section of your data collection form is where you record the results of the included studies in a format to allow later entry into RevMan. It is this part of the form that may be transferred electronically into RevMan if it is set up as a spreadsheet, although many reviewers prefer a paper based system with later data entry.

Before we go on to discuss the design and use of this part of your data collection form, you need an understanding of the type of data you will need to extract in your review. Go to Module 10: Planning the Analysis and complete it, along with the Additional Module A1 on continuous data if any of the outcomes in your review are measured on a continuous scale (for example blood pressure, pain on a visual analogue scale), then return to here and complete this module.
Skip to Module 10 before advancing further
Module 10: Planning the Analysis

To design the rest of your data extraction form you will need to formulate a table to allow recording of data available from the study for each outcome under each comparison. Your table may look something like this:

COMPARISON 1
  Intervention A Intervention B
Outcome 1    
Outcome 2    
COMPARISON 2
  Intervention A Intervention B
Outcome 1    
Outcome 2    

If you are performing subgroup or sensitivity analyses, you may also want to include room in the table to collect information about which group the study belongs to.

You should by now have an idea about the type of data you are likely to be collecting from the results sections of your included studies. This will relate to your outcomes. Think about whether your results are likely to be in the form of dichotomous (for example dead/alive, smoking/not smoking) or continuous (for example blood pressure, pain on a visual analogue scale) outcomes. For dichotomous outcomes you will need to extract the number of participants experiencing the outcome and the total number in the group. For continuous outcomes you will need to extract the number of participants, the mean and the standard deviation for each group.

Try designing the results part of your data collection form

A useful hint is to allow sufficient space on your data collection form to record any conversion calculations (for example calculating a standard deviation from a standard error) or notes.

Now have a try at designing this section of the form for your review.
© The Cochrane Collaboration 2002   Next: Testing the form