Article Text

Download PDFPDF

Basic life support skill retention of medical interns and the effect of clinical experience of cardiopulmonary resuscitation
  1. Ji Ung Na,
  2. Min Seob Sim,
  3. Ik Joon Jo,
  4. Hyoung Gon Song,
  5. Keun Jeong Song
  1. Department of Emergency Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
  1. Correspondence to Professor Min Seob Sim, Department of Emergency Medicine, Samsung Medical Center, Sungkyunkwan, University School of Medicine, 50 Irwon-dong, Gangnam-gu, Seoul 135-710, South Korea; coldco2{at}


Objective To investigate the level of basic life support (BLS) skill retention of medical interns 6 and 12 months after BLS education and analyse the correlation between clinical experience of cardiopulmonary resuscitation (CPR) and BLS skill retention.

Materials and methods The baseline performance of BLS skills in medical doctors during their internship was tested immediately after the BLS provider course. The subjects were divided into two groups, which were tested using the same method after 6 months or after 12 months. Data on the subjects' CPR experience were collected through CPR records—specifically, the number of CPR experiences and the feedback given by the CPR team leaders. To evaluate BLS skill retention, baseline BLS skill performance was compared with the skill performances measured after 6 or 12 months.

Results Fifty-six subjects were enrolled in the 6 month group and 36 in the 12 month group. For non-compression skills, the points for skills declined from 12 to 6 points in the 6 month group and from 12 to 6 points in the 12 month group and the declines in both groups were statistically significant. For compression skills, in the 12 month group, the hands-off time improved from 9.9 s to 8.7 s, with statistical significance. In the multivariate linear regression test, the number of times feedback was given had a statistical relationship with improvement in hands-off time in the 12 month group (coefficient 0.58, 95% CI 0.12 to 1.05).

Conclusions In medical doctors, the compression skills were well preserved, but the retention of non-compression skills was poor.

  • Life support
  • skill retention
  • healthcare professional
  • resuscitation
  • intensive care

Statistics from


Basic life support (BLS) is an essential emergency care component that should be provided for all cardiac arrest victims with no definite contraindications such as a do not attempt resuscitation order.1 Immediately performing good quality BLS improves the survival rate.2–4 Because cardiac arrest usually occurs unexpectedly, it requires immediate action by witnesses or nearby rescuers and BLS is therefore an essential skill for all healthcare professionals. Most healthcare professionals learn this basic skill and know how to perform it, but in real clinical situations it seems that it is often performed sub-optimally.5–7

Several factors may affect the quality of cardiopulmonary resuscitation (CPR), such as education, feedback and monitoring and it has been emphasised that these should be developed together in order to improve quality.8 However, if we just focus on education, the retention of, and deterioration in BLS skills may be an important problem.

Although one may pass a skill test and be considered to have an adequate skill level immediately after receiving a BLS education course, BLS skills deteriorate over time. Several studies have shown that for a layperson it is difficult to maintain BLS skills for more than 1 year.9 ,10 Other studies have shown that CPR skills deteriorate to the pre-training level within a 3–6 month period even in healthcare professionals.11 ,12 Despite these studies, this problem has not been fully investigated in doctors, who are always likely to need to carry out CPR. In addition, the correlation between real CPR experiences and BLS skill retention has not been not fully examined.

We hypothesised that BLS skill deterioration might be delayed in doctors who had had real CPR experience. To test this, we evaluated the level of BLS skill retention among medical interns at 6 months and 12 months after BLS education and analysed the correlation between clinical experience of CPR and BLS skill retention.

Materials and methods

Study design

A prospective randomised observational study was carried out between February 2009 and February 2010. At the start of our study, all participants took an American Heart Association (AHA) BLS provider course. Baseline BLS skills were measured immediately after the BLS training. We randomly divided the subjects into two groups to re-measure their BLS skills after different periods of time: at 6 or 12 months after training. The study design and flow of participants are shown in figure 1.

Figure 1

Flow diagram of the study.


The participants in this study were doctors undergoing internship training at Samsung Medical Center, a tertiary-care teaching hospital in Seoul, Korea. The one day AHA BLS provider courses are conducted for new interns and are held at our BLS training site for four consecutive days. We recruited subjects who had passed both the skills and the written tests taken after each BLS provider course. We gave the subjects an outline of the study, including the primary themes, the methods and the range of data collection. However, they were not informed about the detailed schedule and when they would have the follow-up test.

Sample size

To date, there has been no study on the relationship between the retention and deterioration of BLS skills and the clinical experience of medical interns. Therefore, we used preliminary data from a pilot study conducted with 10 subjects to calculate the sample size. The preliminary pilot study showed that the BLS skill score of the modified Cardiff test in medical interns decreased from 94.8% to 75.5% after 6 months. From these data, we assumed that the performance of BLS skills would decrease 20% by 6 months after education and by more than 20% by 12 months. Thus, we determined that 55 subjects were needed in each group to achieve a power of 0.9 at an α level of 0.05.

Study protocol

All enrolled subjects performed the BLS skill test twice with the same protocol. After the baseline test, which was performed immediately after the BLS training, the follow-up test was performed 6 or 12 months after the training according to the group allocation. The allocation of the participants was randomised using the table of random numbers available in Microsoft Office Excel 2007 (Microsoft, Redmond, USA).

To reduce information bias, the follow-up tests were performed without notice, within 4 days of the designated date for each group. One blinded investigator checked the participant's schedule, decided the time for the test and made direct contact individually. All participants took the test immediately after receiving the call at a specific prepared location. Even when the participant could not take the test immediately owing to work commitments, the test was performed within 2 hours. Both the baseline test and the follow-up test comprised a simple simulation of sudden cardiac arrest and all tests were performed in an isolated room with one participant at a time. All the test scenes were recorded with a video camera for review. After finishing the test, subjects were required to complete a questionnaire and provide additional information.

Outcome measurement

The BLS skills were measured with a modified Cardiff test and skill-reporting manikins. According to the objectivity of measurement and the clinical significance of each performance, we divided the BLS skills into compression skills and non-compression skills.

The compression skills, which comprised compression rates, compression depth and hands-off time, were measured using the skill reporting system of Resusci Anne Skill Reporter (Laerdal Medical, Stavanger, Norway). The non-compression skills were assessed with a modified Cardiff test.13 The non-compression skills consisted of 12 different skills: (1) check responsiveness; (2) check responsiveness: activate EMS system; (3) airway and breathing: initial airway opening; (4) airway and breathing: initial breathing check; (5) first rescue breaths: open airway; (6) first rescue breaths: two initial rescue breaths; (7) check signs of circulation: open airway; (8) check signs of circulation: pulse check; (9) ventilation: open airway; (10) ventilation: appropriate breaths; (11) compression to rescue breath ratio 30:2; and (12) entire sequence of steps performed in correct order.

Three blinded BLS instructors measured the non-compression skills by reviewing the recorded videos. When there was a divergence of opinion, a consensus was reached after discussion. The three BLS instructors used a checklist containing the 12 items that could be checked simply with ‘yes’ or ‘no’: thus the maximum score was 12 points.

To analyse the correlation between real CPR experience and BLS skill retention, we reviewed all in-hospital cardiac arrest registries during the study period. From the registries, we collected data about the subjects' experiences of real CPR, either as a CPR team member or as a first responder to in-hospital arrest, and checked for records of any feedback given by the CPR team leaders. We also collected data about the subjects' basic information and out-of-hospital BLS experience after each follow-up BLS skill test.

Statistical analysis

A paired Student t test was used to compare the mean values of the compression skills, and the Wilcoxon matched-pairs signed-ranks test was used to compare the non-compression skill scores at baseline with those at 6 or 12 months. Multivariate linear regression was also used to analyse the correlation between real CPR experience and BLS skill retention. In the regression test, we used the number of experiences of BLS, advanced cardiac life support (ACLS) and feedback as the main factors and gender, age and experience of previous BLS training within 2 years as confounding factors. Data are presented as means with 95% confidence intervals. A p value <0.05 was considered statistically significant. All statistical analyses were performed using STATA V.11.0 for Windows (StataCorp)

Ethical issues

All subjects participated in the study voluntarily and provided written consent. This study was approved by the institutional review board of Samsung Medical Center.


A total of 112 medical interns were enrolled at the initial stage of the study: 56 subjects were randomised to each group (6 month group and 12 month group). Twenty subjects dropped out of the 12 month group for various reasons: 11 participants refused the test because they were too busy or failed to adjust to the schedule and nine left the hospital earlier than expected (figure 1). Therefore we compared 56 subjects in the 6 month group with the 36 remaining subjects in the 12 month group.

Participant characteristics

Characteristics of the subjects and their experience of CPR according to the follow-up period are summarised in table 1.

Table 1

Characteristics of the study participants

BLS skill retention in the 6 month group

For all compression skills, there was no statistical difference between the baseline skill performance and that measured 6 months later. However, for non-compression skills, the median score declined from 12 points to six points with statistical significance (table 2).

Table 2

Comparison of basic life support performance at baseline and after 6 months

BLS skill retention in the 12 month group

For compression skills, the rate and depth of compression were retained well after 12 months. Furthermore, the hands-off time actually improved from 9.9 s (95% CI 9.3 to 10.5) to 8.7 s (95% CI 7.8 to 9.6), with statistical significance. However, for non-compression skills, the median score of the 12 month group declined from 12 points to 6 points, with statistical significance (table 3).

Table 3

Comparison of basic life support performance at baseline and after 12 months

Retention of the 12 non-compression skills

Figure 2 and figure 3 show the percentages of subjects who performed correctly, according to the 12 components of non-compression skills, in the 6 month and 12 month groups. Although the check responsiveness and ventilation skills were relatively well preserved, the retention of other skills was poor.

Figure 2

The percentage of subjects in the 6 month group who performed correctly according to the 12 parts of non-compression skills.

Figure 3

The percentages of subjects in the 12 month group who performed correctly according to the 12 parts of non-compression skills.

Clinical experience and skill retention

A multivariate linear regression analysis was performed to confirm whether there was any correlation between clinical experience, taken as the number of BLS, ACLS and feedback experiences and skill retention. In the 6 month group, there was no statistical relationship between clinical experience and skill retention for either compression skills or non-compression skills. In the 12 month group, the level of feedback correlated with the improvement in hands-off time with statistical significance (coefficient 0.58, 95% CI 0.12 to 1.05).


Although most hospital-based BLS training sites for healthcare professionals have a policy of retraining in a cycle of 2 years, growing evidence suggests that shorter renewal intervals are required. A study by Smith et al showed that the performance of skills deteriorated quickly in nurses: the scores of a test taken after 3 months decreased 33.3% from those of the post-training test.11 Another study of doctors and nurses by Curry and Gass also showed that CPR skills had deteriorated nearly to pre-training levels 6 months after training,12 and a study by Gass and Curry showed a similar result within 12 months after training.14 These results are not much better than those of a layperson. In our study, the retention rates of non-compression skills were just as poor as those from other previous studies, although the compression skills were relatively well preserved.

Many factors, such as programme quality, instruction methods, practice and feedback, might have affected the acquisition and retention of skills. Specifically, frequent practice and meaningful feedback are thought to have a positive effect on CPR skill retention.15 According to Kolb's experiential learning theory, in order not to repeat the same mistake a concrete experience has to be had through reflective observation of learning.16 Such reflective observation can be achieved by debriefing or feedback from a team leader. The finding that feedback is related to the improvement in CPR skills suggests that experience is important.

In this study, almost all subjects had experienced more than 10 real CPR situations. According to our hospital CPR team constitution, interns are usually responsible for chest compression during CPR. These experiences might have significantly reinforced retention of the compression skills.

We could not clearly prove the relationship between compression skills and the level of feedback; however, there was an association between increased chest compression rate and the improvement in hands-off time and the level of feedback in the 12 month group. In the CPR situations in our hospital the CPR team leaders emphasise ‘push fast and push hard’ to the compressor and try to minimise the hands-off time through the rapid change of compressors. Afterwards they give feedback on the effectiveness of chest compression and the length of hands-off time. The content of this feedback might have influenced this relationship.

Non-compression skills are a large part of BLS and are more complicated than compression skills. When we compared the non-compression skill scores, the scores of both groups were approximately 50% lower than at baseline. There was no statistical difference between the groups. Also, on the multivariate regression analysis, no relationship could be found between non-compression skills and the BLS experience for either group. However, because only 9–14% of the participants had BLS experience and because in most cases proper feedback was not given on the actual BLS, we can assume that the small sample size and lack of proper feedback may be responsible for the absence of positive effects on the retention of non-compression skills. This is a similar result to that of previous studies on BLS skill retention which showed significant skill deterioration.

In this study we divided CPR skills into two categories, compression skills and non-compression skills and observed a difference in the retention of these two types of skill. We assume that this is because more meaningful feedback was given for compression skills owing to their simplicity and also because interns gain extensive experience through repeated practice in real situations. If it were possible to give more meaningful feedback on non-compression skills, we might expect better results for these skills to be achieved, as they were for the compression skills.

This study was based on the 2005 AHA guidelines. However, the new 2010 AHA guidelines simplified the BLS procedure—for example, by dispensing with airway and breathing check procedures, such that the non-compression skills are now less complicated.17 Considering that four of the 12 parts of the non-compression skills were associated with airway and breathing checks and these parts showed the worst skill retention scores, we expect that this change in guidelines will have positive effects on BLS skill retention.

This study has some limitations. First, we did not consider educational experience before the study, so we cannot know for sure whether repeated training influenced skill retention. However, we suspect that this effect would not be significant because only a few subjects had had previous related educational experience. Furthermore, the education level of all the subjects was standardised by the training performed at the start of the study.

Second, because this research was performed at a single centre there might have been a selection bias. Accumulation of research results from different organisations would allow a more objective evaluation to be made. Third, there was a disparity in numbers between the two groups because 20 members of the 12 month group dropped out of the study. Nevertheless, we were still able to obtain statistically meaningful results.

Lastly, we initially planned to assess the potential correlation between clinical experience and BLS retention, but we were unable to identify a clear relationship because the study sample was homogeneous. We could not verify whether the high retention rate of compression skills was a function of the compression skill itself or a result of clinical experience, although compression skills were also found to be poorly retained in previous studies.9–12


The interns who received feedback on compression skills through ACLS retained their skills even at 12 months after BLS training. In contrast, their non-compression skills deteriorated significantly after 6 months. Therefore, clinical experience with proper feedback helps to maintain CPR skills longer, especially chest compression skills.



  • Competing interests None.

  • Patient consent Received.

  • Ethics approval Institutional review board of Samsung Medical Center.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.