Dear Colleagues,
Hope all is well. I hope someone will be able to assist me in scientifically
resolving an issue that has arisen regarding the use of standardized mean
difference (SMD) in meta-analyses. Of course any citations on the subject
would be a great help. The situation is as follows:
Different studies provide their results for continuous outcomes using
different parameters (e.g. percentage change from baseline, post-treatment
means and/or post-treatment change scores). Since percentage change from
baseline can not be combined with post-treatment means/ change scores using
the weighted mean difference (WMD) because the former uses a 1 - 100% scale
while the latter uses a physical measurement (e.g. mg/ dL, mmol/ L, etc.),
can the two outcome measures be combined in the same meta-analysis using
SMD?
In other words would it not be possible to use SMD to pool both the
percentage change from baseline from one trial and the post-treatment mean
from another trial in the same analysis? This would allow reviewers to
streamline the analyses by merging both analyses into one analysis, each
using a different scale to measure the same effect. Most of the articles I
have found discuss measurements of pain, but I have yet to read extensively
about other applications for this type of meta-analysis. Therefore any and
all help will be highly appreciated.
Thank you for your time and consideration.
Please accept my regards.
Sincerely,
Ahmed
_______________________________________
Ahmed M. Abou-Setta, MD, PhD
Independent researcher/ systematic reviewer
Consulting Senior Systematic Reviewer,
Chalmers Research Group, CHEO Research Institute (CHEO RI)
Children's Hospital of Eastern Ontario, University of Ottawa
Member, Editorial Advisory Board, The Open Medical Devices Journal
Member, Geneva Foundation for Medical Education & Research
Member, Menstrual Disorders & Subfertility Subgroup, Cochrane Collaboration
5, El-Ashwal Street, from Pyramids Road
Maashal Station, Pyramids
Giza, Egypt
Mobile: (+20-11) 852-0067
E-mail: [log in to unmask]
|