 |
 |
Hi Dominic, You bring up an important point regarding inclusion/ exclusion criteria for a systematic review. To makes things more clear I will replace your population x and y with two discrete populations (e.g. children and adults). Therefore the PICO would be for example: In children randomized to intervention A versus intervention B, what is the effect on clinical outcomes. By defining your population as children only, therefore you are not interested in details of how these interventions compare in adults, but rather only in children. Therefore with the two study populations you mentioned (one group only testing the interventions in children, while the other group investigating in both a mix of children and adults) you will have to make an a priori decision on how to include/ exclude these trials. You have several options, the first option is to be exclusive and state that you are only including studies exclusively in children and therefore all mixed populations are excluded. Option 2 is to create an arbitrary cut-off (e.g. 80%) by which the majority of participants must be from your intended population. This is to minimize any contamination from the other populations, a subgroup analysis excluding these studies would probably be justified. The third option (which I prefer) is to contact the authors for specific subgroup data (if not already presented in manuscript) on the intended population (e.g. children). In all cases, you should put everything down in your protocol and note any changes to the protocol. The important thing is that you can justify why you make decisions. Transparency in systematic reviews is essential. Good luck. Ahmed Ahmed M. Abou-Setta, MD, PhD Post-doctoral Fellow/ Project Co-ordinator University of Alberta Evidence-based Practice Centre (UA-EPC) Alberta Research Centre for Health Evidence (ARCHE) University of Alberta Aberhart Centre One, Room 8412 11402 University Avenue Edmonton, Alberta CANADA T6G 2J3 Tel: (780) 492-6248 Fax: (780) 407-6435 E-mail: [log in to unmask] Website: http://www.ualberta.ca/ARCHE/ -----Original Message----- From: Evidence based health (EBH) [mailto:[log in to unmask]] On Behalf Of Dominic Hurst Sent: January 13, 2011 8:32 AM To: [log in to unmask] Subject: Combining results for meta-analysis Hi, I'd appreciate some help on the following: A systematic review looks to compare the effectiveness of two interventions, A and B, in a particular population, X. The interventions, though, are commonly used in a discrete population Y also. Some of the studies retrieved compare A and B just in the desired population X, but others compare the interventions in a mix of populations X and Y. In the latter there may not have been block randomisation so the proportions of X and Y receiving A or B may be unbalanced. In doing a meta-analysis of these studies, should one be cautious in looking to combine the results from the X-only studies with those extracted from the X-Y mixed studies? Does it matter that in removing the subgroup X from the mixed study the original randomisation has been disrupted and does it matter that the A and B intervention groups may be then be unbalanced? Would it be reasonable to test for the significance of this with sensitivity analysis by removing the results from the mixed studies after the meta-analysis? Thanks, Dominic