Journal Club Summary

Methodology Score: 3.5/5              
Usefulness Score: 3/5
PEITHO Investigators. N Engl J Med. 2014 Apr 10;370(15):1402-11.
This multicentre double blind RCT focused on patients with submassive pulmonary embolism and the use of full dose tenecteplase (TNK) plus heparin versus placebo and heparin for the composite outcome of death or hemodynamic instability, Although the trial found an ARR of 3% in favour of TNK, JC attendees felt this was driven by the less important outcome of hemodynamic instability and thus was not enough to influence a change in practice. 
By: Dr. Krishan Yadav
(Presented June 2014)

Epi lesson: Composite Outcomes

Designers of randomized controlled trials will often select composite outcomes when hard clinical outcomes of interest (such as mortality) occur with low frequency. The benefit of doing this is that you can achieve greater statistical power with a lower sample size. Important rules for composite outcomes are: 1) all outcomes included should be clinically significant and sensible to combine; 2) outcomes should not be anticipated to be largely unbalanced between treatment and intervention groups; 3) all components of composite outcomes should be reported separately as well as together. 
By: Dr. Lisa Calder