The effect of the severe acute respiratory syndrome (SARS) outbreak on the willingness of laypersons to provide bystander cardiopulmonary resuscitation (CPR) using standard CPR (SCPR) or compression-only CPR (CCPR) was evaluated.The preferred type of SCPR in the post-SARS era was assessed.

A descriptive study was conducted through telephone interviews. Persons who attended a CPR coursefrom January 2000 through February 2003 answered a structured questionnaire. The respondents' willingnessto perform SCPR or CCPR during a witnessed cardiac arrest of an average adult stranger or that of a family member in the pre-SARS and the post-SARS era was surveyed.

Data for 305 respondents were processed. For the scenario of cardiac arrest of an average stranger, more respondents would perform CCPR than SCPR in the pre-SARS era (83.6% vs. 61.3%, p <0.001) and in the post- SARS era (77.4% vs. 28.9%, p <0.001). In the scenario of the cardiac arrest of a family member, more would perform CCPR than SCPR in the pre-SARS era (92.8% vs. 87.2%, p <0.001) and in the post-SARS era (92.8% vs. 84.9%, p <0.001). After SARS, more respondents were unwilling to perform SCPR (p <0.001) and CCPR (p <0.001) on strangers. After SARS, more respondents were unwilling to perform SCPR on a family member (p = 0.039), but there was no difference in the preference to perform CCPR (p = 1.000).

Concerns about SARS adversely affected the willingness of respondents to perform SCPR or CCPRon strangers and to perform SCPR on family members.Compression-only CPR was preferred to SCPR to resuscitate strangers experiencing cardiac arrest after the emergence of SARS.