METHODS

Study design

Ethical approval was obtained from the Alfred Hospital Ethics Committee. The study was a cross sectional survey of Melbourne (population about 3.5 million) and the rest of state of Victoria, which have a central emergency phone number of 000.

The survey was administered over the telephone through a random selection of telephone numbers from a CD-ROM containing all listed phone numbers for the State. According to the Australian Bureau of Statistics there is a high percentage (98%) of telephone coverage in Victorian Households.⁹

Survey instrument

The questionnaire contained sections regarding participant demographics, CPR training, knowledge of CPR, and the emergency number and potential barriers to performing chest compressions and mouth to mouth. A split sample technique was used for one question on the survey, which involved a scenario where a person collapses and participants were asked unprompted to describe step by step what they think they would do. In half of the questionnaires administered, the person described in the scenario was a stranger in the street and in the other half a family member. The survey instrument has not been validated, however the general content was similar to a survey used to investigate public knowledge in Canada.¹⁰

Data analysis

The sampling frame was designed so that the final sample would consist of 750 participants from within metropolitan Melbourne and 750 from other areas of Victoria. A total sample of 1500 people allowed for a sampling error of +/−2.6% in 95 of 100 samples.

Survey responses were weighted to adjust for the specific sampling design.¹¹ All data were analysed using the statistical package STATA.¹² Categorical data were expressed as percentages and odds ratios with 95% confidence intervals. Statistical significance was analysed by a design based χ² statistic. Continuous data were described as a mean and 95% confidence intervals and analysed by an unpaired two tailed Student t test. Because multiple comparisons were made from a single dataset the level for statistical significance was set at <0.01.

RESULTS

A total of 2889 households were contacted between September 2000 and August 2001. Participants were excluded if the phone number was invalid (n=590), there was no answer after six telephone calls (n=174), or the participants had insufficient English or competency to participate (n=254). Of the 1871 households with eligible respondents at home at the time of telephone contact, the questionnaire was completed in full by 1489 (79%) of respondents. This did not change the power of the study as originally calculated.

When asked if they knew the emergency number 98% of participants correctly responded “000”. The majority (86%) of people stated that they knew what CPR involved. When asked to then describe CPR most (65%) respondents correctly described it as “pushing down on the chest and breathing into the mouth”.

Only 52% of the study sample, were trained in CPR and only 11% within the past 12 months. Of those CPR trained, 55% said they still felt confident about their CPR skills. Respondents were significantly more likely to have received CPR training if they were younger than 56 years of age (OR 3.3, 95% CI 2.6 to 4.3, p<0.001) and if they had training in a trade (OR 2.1, 95% CI 1.5 to 3.0, p<0.001) or a tertiary education (OR 1.9, 95% CI 1.4 to 2.5, p<0.001) compared with high school only.

Respondents were asked unprompted to name actions they believed they would take if faced with a collapse situation at home or in the street. Most participants stated that they would telephone 000. Significantly more respondents believed that they would give mouth to mouth to a family member compared with a stranger. Women were less likely than men to say that they would give chest compressions (regardless of the scenario) (table 1).

Respondents who had received CPR training were more likely to say that they would give chest compressions, mouth to mouth, and check for “Danger, Response, Airway, Breathing and Circulation” than non-trained respondents (p<0.01) (table 1).

A bleeding victim and fear of not having the skills were the most common barriers that reduced the participants perceived willingness to perform chest compressions and mouth to mouth (table 2).

DISCUSSION

A significant finding from this study is the low level of people who have current CPR training. Only 11% of participants had trained in CPR within the previous 12 months. The American Heart Association has suggested that to reduce morbidity and mortality rates from out of hospital cardiac arrest at least 20% of adults need to be currently trained in CPR.¹³

Respondents were more likely to have received training if they were younger than 56 years of age. The association between age and likelihood of being trained in CPR has been reported in previous studies.^14,15 CPR courses attract a predominance of young, healthy adults.^14–16 This contrasts strongly with the type of person most likely to witness a cardiac arrest.⁷

Participants who had CPR training were no more likely to say that they would call 000. This response is disappointing as the concept of “phoning first” in a cardiac emergency (where the victim is aged over 8 years) is part of the current Australian Resuscitation Basic Life Support Guidelines.¹⁷ It seems that the public would waste valuable time before activating the EMS.

The fear of incorrectly performing CPR and fear of catching a disease, seen in this study have been reported previously.^10,18 Disease transmission in particular has been proposed as a reason that CPR is not more commonly started.^18,19 In Sweden 94% of trained lay rescuers interviewed believed that there was a small to large risk of disease transmission from performing CPR.²⁰ This fear seems to be unfounded as the risk of acquiring an infection from CPR is extremely low.²¹ Rowe et al suggest that barriers identified in studies such as this, need to be tackled by CPR instructors to allay fears regarding basic life support.¹⁰

This study has several limitations. The possibility of a volunteer sampling effect means that the data presented may overestimate the public’s knowledge of the appropriate actions in a cardiac arrest situation. Also, the respondents were not tested on their actual performance or behaviour, which again may cause an overestimation of preparedness.

This lack of knowledge observed in the public, may contribute to the poor survival rates that have been reported for Melbourne (<5%) and in other areas of Australia.^7,22–24 Innovative CPR training delivery methods (such as video training) and public awareness campaigns require exploration.^25–28 It may also mean that awareness programmes should target specific components of the full CPR training courses such as, recognising an arrest, dialling 000, and providing chest compressions alone.