To the Editor.
—We applaud the efforts of Kellermann et al1 to elucidate the relative efficacy of first-response defibrillation on survival after out-of-hospital cardiac arrest. They have used an elegant method to study a difficult issue. Nevertheless, we would like to raise a few concerns about their study. First, the low percentage of patients in the defibrillator arm of the study actually receiving automatic defibrillation may bias the study toward finding no significant difference in survival between groups. The patients receiving automatic defibrillation had a statistically significant slower paramedic response time than those not receiving it within the treatment arm. The lack of benefit from first-responder defibrillation in Memphis, Tenn, may then reflect the negative effect of longer paramedic response times rather than inefficacy of early defibrillation. Alternatively, it may reflect the fact that firefighters clearly arrived first only 50% of the time because of the very rapid response