RT Journal
A1 Meier CR, Derby LE, Jick SS, Vasilakis C, Jick H
T1 ANtibiotic use and risk of myocardial infarction—reply
JF JAMA
JO JAMA
YR 1999
FD December 1
VO 282
IS 21
SP 1997
OP 1999
DO 10-1001/pubs.JAMA-ISSN-0098-7484-282-21-jbk1201
UL http://dx.doi.org/10-1001/pubs.JAMA-ISSN-0098-7484-282-21-jbk1201
AB 
In Reply: Dr Haider and colleagues state that
residual confounding, chance, or bias may be alternative explanations for
our findings. We agree, as discussed in detail in our article. However, 3
reasons speak against selection bias (ie, antibiotic users differ from nonusers
with regard to cardiovascular risk factors and medical attention sought):
(1) there were no subjects with diagnosed cardiovascular diseases in our study;
(2) we adjusted the analysis for number of practice visits (and therefore
for medical attention sought) with no effect on the results; and (3) we analyzed
antibiotic use in groups and not just users vs nonusers, which would mean
that any bias had to selectively affect users of certain antibiotics but not
others.