Prompt response in giving effective chest compressions and early defibrillation is of utmost importance in optimizing the chance of survival of a cardiac arrest victim [7]. However, from our survey, only about half of our students said that they would be willing to perform CPR when they witnessed a cardiac arrest in front of them. Our argument is, if our students are not willing to give CPR despite that they had just completed the BLS and ACLS training, how much then can we expect them to be willing to perform CPR in months or years ahead after the completion of the BLS and ACLS? Studies had shown that the retention of knowledge and skills of CPR declines over time [23–25]. Furthermore, the students had been exposed to the practices and skills of BLS at least twice in the medical school. This is a disturbing trend to us as these students represent the immediate future doctors and dentists in our country. And often, these new doctors are the first ones to be called to attend to a cardiac arrest [12, 26, 27]. Many of these newly graduated doctors had not been shown to be competent enough to perform adequate resuscitation [27–30].

Except for victims who are their own family members or close friends, most students in our survey gave a negative response to performing CC + MMV. This has been demonstrated in similar previous surveys [1–5]. However, what had often not been studied in other surveys, but which we found to be additional deterrents in our community to performing CC + MMV, are sociocultural and gender differences. As mentioned, Malaysia is rich in her diverse and unique cultural heritage that is mutually respected due to her multi-ethnic population makeup. But these may ironically form some sociocultural reservations resulting in the reluctance among our students to perform CC + MMV on a victim of different gender and/or race. This is especially the case regarding the gender factor, particularly for a female student responding to a male victim (Table 7).

Although it is true that using a pocket mask for ventilation also helps to significantly improve the willingness of our students to give CPR, pocket masks are not easily available to the general public in Malaysia. The most feasible and practical way is, of course, performing chest compression-only CPR. In fact, realizing the importance of prompt action and minimizing the delay in initiating chest compressions, the AHA recently issued a scientific statement in April 2008 calling for a bystander rescuer (regardless of whether trained or untrained) to immediately initiate chest compressions by pushing hard and fast in the center of the chest of an adult victim who suffered sudden cardiac arrest in an out-of-hospital setting [31]. Delays and interruptions should be minimized, and a rescuer should provide immediate chest compressions only unless he or she is confident and willing to give mouth-to-mouth ventilation without jeopardizing effective compressions. Chest compressions only have been demonstrated to be as effective as chest compressions plus ventilation [32–36] in the case of a sudden cardiac arrest of an adult in the out-of-hospital setting.

Various reasons have been elucidated on why bystanders refused to provide CPR, and these include a fear of disease transmission through MMV [31, 37–40] as well as panic and fear of causing further harm [31, 40]. Nevertheless, ultimately the decision whether the respondent would respond depends on whether the bystander perceives the event as an emergency situation that demands a moral responsibility to act upon it [41]. A bystander who believes that there are more capable people around will less likely offer his or her help—and this is often termed bystander apathy [42, 43]. A bystander who lacks self-confidence in his or her ability will also be less likely to offer aid [3, 5]. It has also been shown that a bystander who perceives serious consequences of not helping and a low risk of helping a particular victim will be more likely to help [2]. This explains why most bystanders are willing to help their own family members but not someone who appears like a drug addict with a high risk of human immunodeficiency virus (HIV) infection. These factors are possibly some of the most likely factors that contribute to the unwillingness among our students to perform bystander CPR. However, these remain yet to be validated in this study but they should serve as impetus for future works to be explored.

Several other limitations in this study are inevitable, as many of these are limitations inherent in the design of the study itself. Responses in hypothetical situations may not necessarily mirror actual clinical behavior. As mentioned earlier, several other factors (like the emotional makeup at that spur of the dire moment and the perceived self-confidence) may as well influence a bystander’s choice on whether to respond or not. Nevertheless, although intention or willingness to perform may not be the ideal indicators of future behavior, they are still good, practical survey tools that have been used in many other studies of health behaviors like dietary habits, HIV risk behaviors, etc. [4, 43, 44].

Secondly, we admit that the ten hypothetical scenarios in the survey are too artificial and mechanically dissected. In actual situations, complexity is the rule and very often there would be overlaps of the scenarios. For example, although we gave the students two different scenarios–victim who is a family member and victim who is involved in a motor vehicle accident with facial trauma/bleeding—in actual fact, how would one respond to performing CPR + MMV for a family member who sustained facial trauma/bleeding?

Lastly, the fact that we allowed the students to complete the survey form in an unmonitored environment may actually have encouraged them to discuss with one another rather than revealing their own true intentions.