Finding
The meta-analysis of psychosocial interventions after myocardial infarction by Linden et al suggests mortality reductions greater than those of standard treatments, but their pooling of trials suffers from important omissions and errors, including the exclusion of studies by Naismith et al, Mayou et al, and the P.RE. COR.
Paper
Intervention after myocardial infarction.
Meta Analysis
Citations: 2
Full text
Semantic ScholarAbstract
T meta-analysis of trials of psychosocial interventions following myocardial infarction, which was published by Linden et al,1not only suggests reductions in 2-year mortality greater than those shown by aspirin, β-blocker, and thrombolysis, but also claims that these mortality reductions are over and above those to be expected in cardiac rehabilitation programs. Other investigators have reviewed the same literature and drawn rather more cautious conclusions.2,3 There are risks as well as benefits in statistical overview (or meta-analysis),4,5particularly if the overview is not rigorous or complete,5since, then, it is only masquerading as science. Linden and colleagues' recent pooling of trials of psychological rehabilitation suffers in a number of important respects. They include many extremely small trials and in the tally of 22 (trials) enter several twice. They omit several others, eg, those of Naismith et al,7Mayou et al,8and the P.RE. COR
Authors
R. West, D. Jones
Journal
Archives of internal medicine