The behaviorism theory was developed based on classical conditioning by Pavlov and operant conditioning by Skinner in the early 20th century. It proposes that an action is formed through the repetition and intensification of a stimulus-reaction process; initially, actions are evoked by chance until the designated action can be purposefully evoked[9]. In conclusion, behaviorism states that changes in actions are results of education, and all actions are reactions to the stimulus[10]. This can be more easily applied to actual practice in education because the learning process can be more specific than the insight and exploration suggested by constructivism and cognitivism[11].

The repeated learning that behaviorism emphasizes helps to improve memory, attitude, self-efficacy, and skill performance, among others. Therefore, it can enhance the learning efficiency of CPR[11]. Consequently, to increase the survival rate of firefighters suffering from cardiac arrest, the training should be repeated to allow the firefighters to become habituated to applying a consistent CPR strategy<Fig. 1>.

The basic CPR training program was introduced by the International Committee on Resuscitation(ILCOR) in 2015 based on the international standard. It is the basic CPR program(KACPR-BLS program) conducted for 80 minutes and targeted at laymen, and it was developed by the Korean Association of Cardiopulmonary Resuscitation[14].

This study is about the FD-CPR program. The program is based on the previous KACPR-BLS program and developed according to the ADDIE instructional design based on the behaviorism theory and in consultation with a group of 7 experts. The training lasts for 80 minutes in total, just like the KACPR-BLS program.

To the question “If a colleague falls on the fireground, will you practice CPR?”, 199 participants out of 200 (99.5%) answered “yes” and 133 (66.8%) preferred the transfer to the nearest safe zone before performing CPR. However, to the question “Are you confident to perform CPR of high quality?”, the answers were mostly “no” (178 participants, 88.0%), and the most common reason was the fear of incompetence related to the lack of a training program (153 participants, 86.0%). To the question about periodic CPR training, the answers were mostly “no” (132 participants, 66%), and the most common reason was the lack of a CPR training program; they added that they did not know how to train for it (100 participants, 75.8%). The 47 participants out of 68 (69.1%) who answered “yes” to the question on practicing CPR answered that they performed CPR on the mannequin just several times and that was all for practicing CPR in a perfunctory way.

In conclusion, the positive attitude of firefighters toward performing CPR would be considered high, but the self-efficacy was low, and because of the absence of a CPR training program, they did not train for it or they performed it ineffectively. The need to develop a CPR training program suited for the actual fireground turned out high (194 participants, 97.0%)<Fig. 2>.

The collected documents from domestic and foreign research were 1,039 in total. Among these, 108 pieces published before 2000 were omitted because they were not satisfactorily related to the study, and 931 pieces were reviewed through a 3-step approach. The first step was to filter out the duplicate documents (n=458) by reviewing the titles. The next step was to omit the irrelevant documents (n=431) by reviewing the abstracts. Finally, 30 documents were chosen after excluding the 12 documents that did not have the statistical data, such as mean and standard deviation, among others, and a mediating variable.

For the domestic documents, there were 6 articles on the development and estimation of the effect of CPR for ordinary people. Among them, there were 3 articles for nurses, and 2 articles for teachers. The articles for firefighters were 3, however they were all for paramedics and none for fire[22]<Table 2>.

For the foreign documents, there were 16 related to an isolated or fallen firefighter on the fireground, unlike national research. To be more specific, there were 2 research papers on the challenges of the Rapid intervention team(RIT), 1 research paper on the operation of RIT, 3 papers on the apparatus of RIT, and 1 paper on the arrangement of the suburban local RIT. There were 9 papers on the training program for firefighters. Only 3 among these were related to CPR on the fireground[22]<Table 3>.

A review of the 3 documents on CPR for firefighters showed the method by Six Nations Fire & Emergency Services Training Academy in 2017, another method by Herbert & Watford working at Leland Fire Department in 2014, and the last method made by Coral Springs Fire Academy in Florida in the US in 2015 also cited in Kos (2017). However, all of these had not been examined through a hypothesis test, and they only provided suggestions based on the experiences under the local circumstances related to fire.

None of the CPR training programs were tailored to enhancing the survival rate of the firefighters suffering from cardiac arrest. There were ongoing integral studies on the role, apparatus, education, and training of the separately arranged RIT in the foreign countries. However, those were focused on the rapid rescue methods, and the lack of CPR studies applicable to fireground was still existent. There was a limitation related to the structure of the firefighting object, the management of fire officials, the procedure and apparatus of the operations on the fireground; they were different from ours.

The analysis of the training requirement survey and research review of the domestic and foreign documents concluded on the necessity of CPR training program applicable to the actual fireground; the learning objective was set as ‘in the event of cardiac arrest of firefighters at the fireground, high-quality CPR can be provided after moving to a safe zone.’

Based on the learning objective, 7 experts discussed the arrangement of the contents for the training and the instructional strategy. As a result of the discussion, included the 4 essential goals(rapid and precise check of cardiac arrest, prompt report to the station and request for help, high-quality chest compression, and the right use of AED) of the KACPR-BLS program, and some additional methods were recommended to overcome the difficulties on the fireground outlined during the analysis step. Specifically, the added methods included the transfer of the colleague to a safety zone, request for help via radio, application of CPR, removal of private protective apparatus, and the efficient exchange of roles.

To accomplish the learning objective, the following 2 factors were recommended for inclusion in the instructional strategy.

First, focus on learning by practice. The knowledge of CPR is crucial but consistent practice is also important, so the portion of practicing time sets 60 minutes (75.0%) which deems high comparing with the whole percentile.

Second, develop the scenario and program with the actual situation. To enhance the skill performance, focusing on reflecting the scenarios of the fireground was a top priority for the newly developed program.

The FD-CPR program adopted the PWW method to prevent distortion of the contents of training and minimize the intervention of the instruct.

It was based on the assessment table for the practice test of the KACPR-BLS program, the assessment table for utilizing the AED for the paramedics for their second-degree license, which was launched at Korea health personnel licensing examination institute, controlling fire tactics I among the main texts at the fire academy, and searching the urgent target required to be saved and rescue among items for assessing the reactive capability against fire. Likewise, the 16 questions were developed and assessed using the difficulty and discrimination test. The average score of the difficulty test was 41.16% (even)<Table 4>. and that for the discrimination test was 0.32 (high)<Table 5>.

To secure the CVI of the developed questions, I had it estimated from the group of 6 experts consisting. In this study, the CVI for all the items was estimated as 1.0. However, there were revisions. Question no. 15: ‘Hands-off time’ was revised to ‘time to cease chest compression’ to facilitate its integration with the other appraisal item<Table 6>.

After the discussion by the group of 7 experts consisting of 1 emergency medicine professor, 1 emergency medical technology professor, 2 fire service academy professors, 1 education professor, 1 paramedic, and 1 firefighter, the learning objective, training contents, instructional strategy, selection of the media were decided for developing the scenario for the video for the FD-CPR program. It was divided into 7 steps, and each step was integrated for continuous training.

Based on the initially developed scenario, the FD-CPR program was developed. The intervention program for the control group It was conducted for 80 minutes just as the KACPR-BLS program. For the training to focus on learning by practice, as emphasized by ILCOR, the practicing time was set to 60 minutes (75.0%), which was considered high<Table 7>.

As a pilot test, the training program was conducted according to the planned procedure, but for the effect on skill performance, there were some problems.

First, too much time passed in checking the status of a fallen colleague on the dark fireground. It is difficult to determine the status of a colleague due to the smoke and impossible to apply CPR due to the flames, collapse, burial, and isolation, among others. Nevertheless, the firefighters tended to report the status of being capable of breath, any symptom of abnormal action, and any traces of trauma precisely, and it rather delayed the performance of CPR.

A revision of step one was made from ‘check the mental status’ of the scenario to ‘check the mental status while tapping the shoulder (“OO chief, are you okay?”)’ as fast and clear yet to act the, accordingly, the related video was altered<Table 8>.

Second, all the methods for carrying colleagues from the fireground to the safety zone varied. In the case of no separate equipment, the methods for carrying were multiple (raise and carrying, crawl forward and carrying, crawl backward and carrying, pull both armpits and carrying, hang on the shoulder and carrying, crawl carrying on the back and carrying, pull the collar and carrying etc.). However, these were for saving the ordinary people on the fireground, and there was no standardized way to save the firefighters wearing personal protective equipment as heavy as 23kg. Therefore, after the discussion of experts, the method of ‘carrying by pulling the collar’ was chosen due to the low possibility of musculoskeletal injury and the ability to move fast. Yet, the firefighter would be wearing the protective clothing and SCBA, and it would be better to apply the ‘drag by pulling the shoulder string’ of the SCBA-frame back carrier rather than the collar of the protective clothing. Additionally, using the shoulder strings with 2 people on a dark fireground would make it difficult to check the precise status of a fallen colleague out of the narrow aisle and cause a delay. Consequently, the roles were specified. Specifically, firefighter 1 would move toward the exit staying down and remove the obstacles in the way and open the exit. Firefighter 2 would carry the fallen colleague by applying the ‘drag by the shoulder string method’ to move to the safety zone out of the exit.

Therefore, step 2 of the scenario, ‘carry to safety zone’ by 2 firefighters was revised to Firefighter 1 should remove the obstacle in the way and open the exit, and the other firefighter should use the method of dragging the shoulder string to carry to the safety zone (5 meters outside the exit). Subsequently, the related video was altered<Table 8>.

Third, because of the protective clothing the firefighters wear, a colleague suffering from cardiac arrest could not be checked properly or receive high-quality CPR. When firefighters wear the mask, fire hood, and helmet, they can hear sounds of their breathlessness loudly, but it also makes it difficult to listen to the breathing sound of a fallen colleague, given the shutting out of external noise and narrowing of the visual field, for abnormal actions to be recognized. The heavy weight of the SCBA interferes with the relaxation of chest compression, and the thick gloves make it difficult to pinpoint the compression position.

Accordingly, step 3 of the scenario for firefighter 1 was modified to add ‘remove the own SCBA and gloves’ before ‘remove the helmet of the fallen colleague;’ that for firefighter 2 was revised to add ‘remove the own SCBA and gloves’ before ‘remove the belts of chest and waist of fallen colleague’. Subsequently, the related video was altered <Table 8>.

Fourth, the quality of CPR depended on the physique of the firefighter. For a fallen colleague wearing SCBA, the victim was over approximately 30cm high. If the physique of the firefighter is small, the posture typically used for chest compression is tilting to one side, which does not keep the arms at right angles. However, raising the knees could keep the arms at right angles and facilitate high- quality CPR.

Step 5 of the scenario was modified: ‘the compressing posture by firefighter 2’ was revised to add ‘alternative 1) kneeling beside the fallen colleague’ and ‘alternative 2) straighten knees beside the fallen colleague’. Accordingly, the related video was altered<Table 8>.

Fifth, the method of pulling both strings of the inner side of the turnout coat and shoulder was not efficient. Pulling both strings of the inner side of the turnout coat and shoulder would cause the unstrung elbow to get caught by the clothing so that it would not be removed. Therefore, it would require several rounds of pulling to remove it, and in the end, it took a long time. Consequently, the pausing time to chest compression would take over 10 seconds, and the survival rate of a colleague would decrease.

Likewise, step 6 of scenario, ‘alternative 1- ready to pull both strings of the inner side of turnout coat and shoulder,’ was removed, and the related video was altered accordingly<Table 8>.

Likewise, the difficulties associated with assessments of the skill performance were all revised to reflect the scenario and video of the intervention program, and this completed the development of the final program<Table 9>. The feasibility of the revised FD-CPR program was tested by the group of 7 experts, and the CVI of all the items was measured as 1.0. therefore, the program was used without further revision<Table 10>.