Most cardiac arrests occur at home; therefore, widespread training is required in order to educate as many lay people as possible in cardiopulmonary resuscitation (CPR) (Atwood et al, 2005; Isbye et al, 2007). As part of the strategy to increase bystander CPR, the training of schoolchildren has been recommended due to their ability to learn and retain knowledge and skills (Isbye et al, 2007; Toner et al, 2007;. Bollig et al, 2009; Hill et al, 2009).
As the quality of CPR is a major determinant of victim survival, it is concerning that CPR quality has been reported as poor, even in cases when it has been performed by a health care professional (Abella et al, 2005a; Wik et al, 2005; Losert et al, 2006; Jäntti et al, 2009). Although several devices which provide CPR feedback or prompts during training exist, (Isbye et al, 2007; Jones et al, 2007; Bollig et al, 2009; Hill et al, 2009) they are currently in a preliminary phase of development and their widespread use by lay people is not predicted within the next ten years.
Popular songs are easily remembered by lay people and some of them have a rhythm (beats per minute) similar to that of a recommended chest compression (CC) rate. Therefore we hypothesized that introducing the rhythm of a popular song as audio guidance would be advantageous in helping schoolchildren learn and retain an adequate rate of CC when conducting CPR.
Material and methods
This is a prospective, randomized, parallel group design study that was carried out at a primary school in Galicia (northwest Spain) and the participants, who were comprised of schoolchildren aged 13–14 years old, had received no prior CPR training before the study. Written informed consent for participation in the study was requested from the children's parents, with 59 of the 72 eligible subject's parents (82%) giving consent and returning the signed form on the day of the study. The selected pupils were then randomized into two groups: ‘Song’ and ‘Silent’, with 34 pupils in the ‘Song’, and 25 in the ‘Silent’ group.
All of the participating children received one hour of basic life support training, carried out by advanced life support certified pediatric residents who had been trained for this task by a senior instructor. For the purposes of this study, continuous CC was emphasized during instruction with Resusci Anne mannequins (Laerdal Medical, Madrid, Spain) in supine position also being used at ground level.
‘Song’ group was trained to make chest compressions with the audio guidance of the song ‘Macarena’ by Los del Río (1993, Sevilla, Spain), a very popular song at local and international level, which has a continuous rhythm and a rate of 103 beats per minute (bpm), close to the current recommendations for CC (Handley et al, 2005). The control group was trained in a traditional way within rooms located at sufficient distance as to avoid hearing the song heard by the other group.
After the training session, all participants were then asked to make continuous CC on the mannequin for two minutes, first in silence and then with the audio guidance of ‘Macarena’.
This test was then repeated one month later, with the ‘Song’ group being told to perform continuous CC throughout the two minutes while thinking of ‘Macarena’ (and so trying to follow the rhythm mentally), and the ‘Silent’ group of children being instructed to perform CC with the target of 100 per minute without any guidance. After a 20 minutes rest period, all subjects were then played the song collectively and then asked to perform another two minutes of CC, thinking of the song in order to follow its rhythm.
During the study, participants were not informed of their results or progression, and the total number of compressions made was counted by means of a manual counter, a method that was previously validated by the investigators with video recordings of CC series played and recounted ‘frame by frame’. In both groups a 10% deviation from the target was considered acceptable; therefore, to calculate the percentage of cases fitting the target rate, the limits were restricted at 180 and 226 compressions in two minutes respectively, that is, 100±10 per minute for control group and 103±10 per minute for ‘Song’ group.
Statistical analyses were performed using SPSS software 15.0. (Chicago Inc. IL, USA) and the data collected was expressed as a percentage. The Mann-Whitney U test was used to compare differences between different groups, and the Wilcoxon test was used for statistical comparisons within the same group. To compare qualitative variables, Chi-square was used. A P-value of <0.05 was considered statistically significant.
Results
A total of 59 schoolchildren (35 boys and 24 girls), aged between 13–14 years, were included in the study. Table 1 shows the total number of CC delivered in two minutes within both groups throughout the different steps of the study.
| Group | Song | Silence | |||
|---|---|---|---|---|---|
| n. compressions | % inside accepted range | n. compressions | % inside accepted range | P | |
| Initial assessment | |||||
| A. Compressions in silence just after training | 234 (215–250) | 37 | 267 (228–315) | 12 | <0.05 |
| B. Compressions while listening to song | 207 (204–207) | 96* | 206 (204–207) | 100* | NS |
| Assessment one month after | |||||
| C. Compressions in silence | 196 (183–211) | 75** | 218 (189–249) | 42*** | <0.05 |
| D. Compressions after listening to song | 206 (199–215) | 84**** | 208 (192–222) | 80**** | NS |
Intra-group comparisons:
When tested in silence on the day of the course, children in ‘Song’ group delivered less compressions than the control group (P<0.05), and although the percentage of participants who had delivered a number of compressions inside the target range was also better (37 vs 12%, P<0.05), the figure still remained low for both groups.
When participants were tested while listening to the song, there was no significant difference between groups, and the percentage of subjects achieving the target compression rate was found to be very high (96% and 100% respectively).
One month later and in silence, the rate of the ‘Song’ group was lower and the percentage of cases that achieved the target rate compression was higher than in the ‘Silent’ group (75% compared to 42%, P<0.05). When both groups were re-tested after listening one more time to the song, no differences were observed and again, the percentage of cases inside the desired compression rate remained quite high (84% and 80%, respectively).
Discussion
Adequate CC rate is an essential element of CPR quality but (Kern et al, 1992; Berg et al, 1994; Abella et al, 2005b; Losert et al, 2006; Jäntti et al, 2009) lay people, and indeed many professionals, still have difficulties in ascertaining and maintaining the recommended CC rate (Abella, 2005; Wik, 2005; Losert, 2006; Jäntti, 2009). Therefore, innovative educational and technical efforts are needed in order to solve this problem (Leary and Abella, 2008). Several audiovisual feedback systems developed for improving CPR quality (Sutton et al, 2007; Abella et al, 2007; Dine et al, 2008; Sutton et al, 2009) exist, though unfortunately they are rather expensive and this may be a significant drawback to their widespread use, specifically by lay people.
School children are considered to be the ideal target group to be trained in basic life support due to their ability to retain knowledge and skills (Isbye et al, 2007, Jones et al, 2007; Toner et al, 2007; Bollig et al, 2009; Hill et al 2009)
In theory, any song with an adequate beats per minute (bpm) rate that can be immediately recognized and remembered would be a good guidance for CC. The well know song ‘Staying alive’ by the Bee Gees (103 bpm) was tested with good results in a reduced group of physicians by D. Matlock (Matlock et al 2008). Another recent study demonstrated that listening to the popular English song ‘Nellie the Elephant’ significantly increased the proportion of adult lay people delivering compression rates at close to 100 beats per minute, but did so at the cost of increasing the proportion of compressions delivered at inadequate depth (Rawlins et al, 2009).
Our study was the first to assess the effects of a Spanish popular song as a guidance for CC by young lay people, in our case 13–14 year-old schoolchildren. We chose the song ‘Macarena’ by Los del Río, because it has a rate of 103 bpm at a continuous rhythm (with no change in rhythm pattern or pauses), and remains an internationally recognized and memorable tune. Our results clearly indicate that the song's guidance is an effective method that helps schoolchildren with no prior instruction in CPR to achieve the CC rate target.
In addition to skill acquisition, another main objective of life support courses is the retention of CC regularity and skill (Isbye et al, 2007). Our results, although limited to short term skill retention, show that one month after the training 75% of children of the ‘Song’ group performed CC within the target rate, and that a brief ‘refresher’ with the song permitted an increase of that percentage up to 84% within the song group and from 42% to 80% in the ‘Silent’ group. In order to limit any skill decay over time, frequent retraining has been recommended, but this can very difficult to implement on a large scale, both for professionals as well as lay people. A novel and effective approach for re-training health providers in CC skill is the programme of frequent bedside refresher psychomotor training, referred to as ‘rolling refreshers’ (Niles, 2009), and based on our results, we suggest that a comparable programme including popular song audio guidance could be effective for schoolchildren and should be trialled.
Limitations
There remain some limitations to this study. We have only assessed one of the components of CC quality, as other aspects such as hand placement, compression depth and chest recoil are lacking. Studies carried out by health providers have showed that although audio guidance can improve CC rate accurately, it may decrease compression depth (Oh et al, 2008; Rawlins et al, 2009), though it can be argued that the use of a well know song could serve as a kind of automatic guidance which could free the rescuer's mind in order to concentrate on the other aspects of CC. Furthermore, it must be added that participants in the study were tested throughout a two minute period, and therefore, no conclusions can be driven from the potential effects of rescuer fatigue on chest compression rate (Jäntti et al, 2009; Sutton et al, 2009). We cannot discard a possible transference of information between children in the study groups during the month that elapsed from the initial to the re-test phase, futhermore, this is a mannequin study and direct extrapolations to real CPR cannot be made.