The study was approved by the Regional Ethics Committee in Uppsala, Sweden (2006/201), and all participants gave their written consent after verbal and written information. The first test of the skills was accomplished in August 2006 to January 2007. The new European guidelines were presented at a National congress in November 2006 and the guidelines were in use at the studied hospital in May 2007. All data collected before education were evaluated according to guidelines 2000 [
7] and all data collected after education were evaluated according to guidelines 2005 [
4].
Study participants
Participants were selected by their working managers at a hospital in Sweden, and with respect to their working schedules. The aim was to include 30 nurses, 30 physicians and 30 assistant nurses, physiotherapists and/or occupational therapists, who worked on ordinary wards and ICU, and emergency, medical, and surgical departments. The numbers of participants are illustrated in table
1.
Table 1
Demographic characteristics of the study participants and number of participants before and after education
| n | % | n | % | n | % | n | % |
Before education
| | | | | | | | |
Participants | 28 | - | 31 | - | 29 | - | 88 | - |
After education
| | | | | | | | |
Participants | 23 | 82 | 26 | 84 | 25 | 86 | 74 | 84 |
Gender
| | | | | | | | |
Male | 16 | 70 | 3 | 12 | 3 | 12 | 22 | 30 |
Female | 7 | 30 | 23 | 88 | 22 | 88 | 52 | 70 |
Age before education
| | | | | | | | |
Median (range) | 41 | (28-71) | 38 | (25-57) | 50 | (21-62) | 42.5 | (21-72) |
Working experience
| | | | | | | | |
0-5 years | 9 | 39 | 7 | 28 | 6 | 25 | 22 | 31 |
6-20 yeras | 6 | 26 | 13 | 52 | 5 | 21 | 24 | 33 |
>20 years | 8 | 35 | 5 | 20 | 13 | 54 | 26 | 36 |
Missing | 0 | 0 | 1 | 0 | 1 | 0 | 2 | 0 |
A precision calculation estimate based on previous studies [
8‐
12] was used as a guide for sample size calculation.
Test protocol and data collection
In Sweden the use of AED is taught to healthcare professionals in a 4-hours course that also includes theory and practical training in basic CPR, use of oxygen and ventilation with mouth-to-mask technique, and use of suction devices for clearing of the airways. Study participants performed D-CPR on a manikin (baseline), attended the course, and performed D-CPR in the same set up 1-2 months after training (follow-up).
CPR training for the study participants was conducted during February 2007 to June 2007. The time from baseline evaluation until time for education varied between 1 to 5 months.
The follow-up evaluation was undertaken from March 2007 to July 2007. A period of 4-8 weeks elapsed between training and follow-up. Before the participants were asked to perform CPR on a manikin, they were asked to read a set of instructions:
"Imagine that you are somewhere in the hospital, and the person you are talking to suddenly becomes unconscious. You suspect a cardiac arrest. Perform and act as if you were at your own department. The instructor is in the room, but you cannot obtain any help from her, but you have to imagine her when you consider the safety of this situation. In the room, you can see an alarm switch; this is the only way to get help. You cannot go out of the room and ask for help. You decide by yourself if you want to perform mouth-to-mouth ventilation or use a ventilation mask, if you find it essential to perform ventilation. Do not move the manikin to the floor. The scenario takes about 5-10 minutes, this may appear a long time, but please continue to treat the person until the instructor tells you to stop. Thank you for your participation and good luck when the instructor gives you a sign to start".
Both at the baseline and follow-up evaluation, the scenario started with the manikin (Laerdal Skillmeter Resusci Anne, Laerdal Medical AS, Norway) in a hospital bed. A training AED (Laerdal Heartstart FR2, Laerdal Medical AS, Norway) and a ventilation/pocket mask was visible in the room, and could be used at the discretion of the participant. If participant choose to use the AED the first rhythm was ventricular fibrillation. The participants performed single rescuer CPR on the manikin and the entire scenario was video recorded (cf.: Appendix). The scenario was terminated after four minutes of compressions, measured from the first performed compression.
The Laerdal Skillmeter Resusci Anne includes a software program, a PC Skill reporting System for measuring vital functions during the simulated CPR situation. The measurements have a tolerance of ±15% for the variables compression depth and inflation volume. A rescue breath of minimum 250 ml was detected by the software as "ventilation", and a chest compression of at least 10 mm was detected as "compression". Correct compression depth was defined as 40-50 mm, and correct compression rate as 90-110/min. Correct ventilation volume was defined as 800-1200 ml before education (Guideline 2000) and 500-600 ml after education (Guideline 2005). The software program calculates a variable "compressions without error", which contains compressions with correct hand placement on the sternum, complete release and a compression depth of 40-50 mm.
Three experienced instructors evaluated the videotapes of the participants performing D-CPR in order to evaluate aspects of CPR not registered by the software program. The evaluation was accomplished according to the Cardiff test protocol [
13]. To secure reliability each instructor received approximately 30 minutes of training in the use of the Cardiff test protocol. In addition, as a test, they separately evaluated one video-recorded participant performing D-CPR. Thereafter, the three instructors evaluated the same part together, to come to an understanding of the protocol. Then, two of the instructors evaluated all video recordings separately, and after 12 weeks they re-evaluated the video films. The re-evaluations were accomplished in order to estimate intra-observer variability and inter-observer variation. The third instructor served as a master control by evaluating a random sample of 10% of the recorded tapes in order to minimize the risk for bias in the evaluations.
Statistics
Paired statistical methods were used for the analyses of before and after intervention within each profession and for all participants. Based on the assumptions for the tests, both parametric and nonparametric tests were considered. For parametric tests, the mean value with its standard deviation (SD) was calculated, and for nonparametric tests, the median with inter-quartile-range (IQR) was used. In order to compare the results between different professions, unpaired tests were applied. Inter-observer variability of video evaluations was assessed with Friedman's test. In the Cardiff protocol, in which the observed measure was in ordinal scale, the data were analyzed according to ordinal invariant measures for individual and group changes [
14]. All tests were two-sided and statistical significance was considered as
P < 0.05. All analyses were with the software program SAS version 9.2.