To the Editor.
—Mullen et al1 apply a meta-analysis to five epidemiologic studies to conclude that the use of inhaled β-agonists by metered-dose inhaler is not associated with the risk of death from asthma. This conclusion is in direct contradiction to the conclusions of the four largest source studies used in their meta-analysis. We submit that this apparent inconsistency is the result of serious methodological flaws in the comparison of "did vs did not use β-agonist" employed in their meta-analysis.This comparison does not correspond to the contrasts designed or performed in the four principal source studies. The three New Zealand2-4 studies were clearly comparing fenoterol with albuterol,2 finding a higher risk for the former. Our study5 focused on the dose-response analysis, which showed, for example, that in patients using any β-agonist the risk of asthma death for those using more than two canisters per month
Suissa S, Ernst P, Spitzer WO. β-Agonist Use and Death from Asthma. JAMA. 1994;271(11):821-822. doi:10.1001/jama.1994.03510350029019