I am mystified by the data calculations in the table accompanying
Finlay McAlister's letter regarding needed trials of ACE inhibitors plus
Beta Blockers in survivors of myocardial infarction (BMJ 1998;317:751[12
Using the data given, I believe the Odds Ratio for the TRACE study
should be 0.64, not 0.56 as indicated. For the Total, the OR calculates to
0.93, not 0.86. It does not appear such discrepancies would be the result
of number rounding.
While the apparent errors in simple calculations would not alter
McAlister's conclusions, this unfortunate situation does tend to shake
one's faith in the accuracy and reliability of reporting in BMJ.
Competing interests: No competing interests