Systematic review on knowledge and skills level among nurses following cardiopulmonary resuscitation (cpr) training


 Introduction:

Although millions of healthcare providers and lay providers are trained globally in CPR each year, there are major gaps in delivering adequate medical emergency care such as poor quality CPR in the hospitals or none performance of CPR in out of hospital settings for people with cardiac arrest.
Objective

This systematic review aim to highlight the effect of training on knowledge and skills retention as well as the effect of different methods of training on knowledge and skills
Methods

The review used six online databases: Scopus, ProQuest, PubMed, ScienceDirect, CINAHL, Medline and reviewed reference citations for additional studies. Systematic analysis was use in excluding articles from the database by the latest 2015 American Heart Association (AHA) guidelines, known as Grading of Recommendations, Assessment, Development and Evaluation (GRADE) for the evaluation of knowledge and skills retention.
Results

A total of 71,671 possible articles were registered in six databases, with 331 articles found meeting eligibility criteria. And only 20 were finally included that met all the qualifying criteria for this systemic analysis. Most research used multi-choice questionnaires to assess retention of both knowledge and skills. From the pretest to the posttest, studies have shown that knowledge or skill rates differ and decrease from 6 weeks to 2 years after training. Between the reviewed papers, significant variations in retaining knowledge and skills were found between instructor-led training and other forms of teaching methods.
Conclusion

Between the pre-test and post-test outcomes, knowledge and skills have shown a substantial overall improvement in acquisition and retention. However, skills were observed to decay faster than knowledge. Skills start to decay as early as 2 weeks whereas knowledge retention start to decline on average between 1 month and 6 month after BLS/CPR training. Most of the articles showed that other forms of teaching methods are also as effective as the instructor led/traditional teaching methods in knowledge and skills levels among nurses. Therefore, future research should concentrate on the ideal timeline needed after initial training for the re-certification of the BLS/CPR.


Background
Cardiopulmonary resuscitation (CPR) is the primary management for any patient with cardiac arrest. Numerous research centered on the ability of each trainee to apply the learned skills and knowledge gained through training over time while a patient undergoes CPR ((1). After the rst training course, participants usually received CPR certi cation. Subsequently, they will need to renew their certi cation every two years to remain certi ed and to retain or update their knowledge and skills in performing CPR (2,3). However, only a handful of trainees are typically able to practice their CPR knowledge and skills in actual clinical situations. Therefore, acquired knowledge and skills are taught, lost, and re-learned after two years. CPR training's primary aim is to show how a learner can successfully resuscitate patients with cardiac arrest by combining applied cognitive knowledge, psychomotor skills, and interpersonal interaction in real hospital settings. Using the skills they learned while confronting a resuscitation situation, learners will develop self-con dence (4).
Although the ability of nurses to undertake CPR could be a crucial indicator of cardiac arrest patient survival, there is convincing evidence suggesting that registered nurses across continents lack the ability to perform proper CPR. It is clear from the literature that the skills of the CPR cannot be easily acquired or retained by nurses (5).
Nevertheless, CPR seeks to save a life; therefore, there is a robust link between the degree of CPR knowledge and skills training received by healthcare professionals and the result of survival outcome among cardiac arrest patients (6). As a critical determinant of improved CPR outcome, nurses' knowledge is signi cantly correlated with their level of CPR training (7). The results of a study in Tanzania showed a signi cant difference in the overall performance on testing knowledge and skills between those who had and had not performed CPR in the past, with median test scores of 40%(IQR 28-54) and 26%(IQR 16-42) respectively (p = 0.001) (8). A similar cross sectional study in Nigeria demonstrated there is statistically signi cant changes of theoretical knowledge of CPR among clinical medical students with past CPR experience compared to those without CPR experience (P < 0.001) (9 ). Similarly, a number of variables that can also affect the level of acquisition and retention of the knowledge and skills of nurses have been demonstrated, such as age (10,11) gender (12,13,14) academic quali cation (15), clinical experience (16, 17, 18) timing of training (19), methods of training (20,21) and primary working unit of nurses (22). CPR training is crucial to nurses, and undeniably the methods of delivery could affect the level of retention of knowledge and skill of CPR among nurses (23). Instructor led training method, otherwise, refer to as traditional CPR training, have been practice for over 40 years. Nonetheless, there is evidence to show that this method of CPR training has failed to achieve the desired result in the retention of knowledge and skills level delivered to both lay people and health care providers (20). There is growing evidence that supports the application of alternative methods instead of traditional method of CPR training, such as self-directed training methods with practical skills modules (11,24,25) simulation method (26,27), concept mapping (28) as well as online computers or CD video (24 ,29 ).
Effective resuscitation training for busy health care providers may be a challenge. Time and cost for the instructor led CPR, and AED training certi cation may discourage individuals, private and public organizations from participating in CPR courses and obtaining requisite skills (25). Furthermore, if the instructors did not teach CPR course in a uniform way, this may contribute to the poor retention of CPR skills among students (30). Some studies have shown signi cant differences in effectiveness between the instructor-led training and others such as self-directed method (5,11,24), while, others reported there was no statistically signi cant difference on performance total score (skill score) between two groups (21,31,32,33).
The purpose of this systematic review is therefore to examine the present evidence for the level of acquisition and retention of knowledge and skills of nurses in CPR and the implications of training strategies towards nurses' CPR knowledge and skills. Most resuscitation organizations recommended a CPR refresher training after two year. It is not clear, however whether CPR-trained nurses can sustain an appropriate level of knowledge and skills suitable for patient care over the two-year intervals. Yet, more frequent practical training would appear to increase the workload on both trainees and instructors at a period when resources and time are required to train healthcare professionals.

Materials And Methods
This systematic analysis was performed in conjunction with the latest 2015 AHA guidelines, as translated from the evidentiary review phase of the International Liaison Committee on Resuscitation (ILCOR), de ned as the Grading of Recommendations, Assessment, Development and Evaluation (GRADE). These included studies from January 2013 to December 2019. The research used six online databases: Scopus, ProQuest, PubMed, ScienceDirect, CINAHL, Medline and reviewed reference citations for additional studies.

Criteria for eligibility
The criteria followed for this systematic analysis are as follows: CPR training articles on knowledge and skills applicable to nursing, English articles written and available in full text, both observational and interventional studies, articles with a shared sample audience of nurses and nursing, and keyword articles as illustrated in the table below. Keywords included in this study were: "knowledge OR awareness OR information" OR "skills OR ability OR expertise" AND "nurses OR nursing" AND "cardiopulmonary resuscitation OR CPR OR basic life support OR BLS OR resuscitation." (Table 1) 2.2 Literature selection (screening / eligibility) Journals not meeting requirements such as those with an irrelevant title, not published in the English language, summary journal, over six-year publishing duration, as well as literature such as newspaper article, conference or conference text, book, policy report, etc. were all excluded using the ltering method process. From the selected items, the reference lists in each article were checked. First, two reviewers reviewed title, overlapping subjects, and irrelevant domain subject and information focused on inclusion and exclusion criteria (SA and LPY). The second review team (SA and KT) then screened articles abstracts, followed by full articles content.
Six databases, hand search and references reported a total of 71,671 potentials articles and excluded 59,909 based on inclusion and exclusion criteria ( Figure 1). Of 11,702 articles have been excluded, where 6,810 were irrelevant, 1,150 with irrelevant sample populations, and 3,411 with duplicated information. This screening process yielded 331 articles that matched the eligibility requirements and was further screened for the full text. Of the total, 208 did not relate to CPR knowledge and skills, and 103 articles did not speci cally identify the sample group, such as a mixed population were excluded. Therefore, 20 studies met all qualifying requirements for this systematic study.

Data extraction summary
A total of 13 articles were reported on knowledge among nurses in the review. Four of the articles are RCTs, while three consists of quasi-experimental designs, four articles use prospective repeated measures, and, two use descriptive cross-section studies. Three out of the 13 articles assessed respondents at one point in time (29,34,29) two examined at two points in time on each question (36, 37) and the other eight papers examined more than two points in time (11,13,26,28,38,39,40,41). A total of 18 articles reported on the skills of nurses. Six of the articles are RCTs, eight are quasi-experimental and four are prospective repeated measures. Only one article out of the 18 articles assessed respondents at one point in time (29), whereas two articles examined respondents at two points in time (36, 37) and the remaining other 15 papers examined more than two points in time (11, 13, 24.26, 27,, 28, 38, 39, 40, 41, 42, 43, 44, 45, 46). Several types of knowledge questionnaire were used in all 13 papers to collect data for the knowledge of participants. Five studies used the AHA questionnaire (13, 26, 28, 29.39), seven studies used self-constructed questionnaires (11,34,35,37,38,40,41 ) and one study used Resuscitation Council of South Africa (RSCA) qusetionnaire (36). Checklist was used for skill test by evaluators in 11 out 18 articles to records the skill scores during data collection by using the Laerdal manikins. Four articles used skill meter attached to the manikin (26, 39,44,46) and the remaining three articles used video to records the scores of the participants (27,41,45).

Quality assessment of the literature
A 2015 guideline recommendation used the current edition of AHA de nitions for evidence level and guidance class. Evidence level is as follows; 'LOE A' is high-quality evidence from more than one RCTs; 'LOE B (randomized)' is moderate-quality evidence from one or more RCTs; 'LOE B-NR (nonrandomized)' is moderate-quality evidence from one or more well-designed, well-executed nonrandomized studies or observational studies; "LOE C-LD (limited data)' is a randomized or nonrandomized observational study with limitations of design or execution; "LOE C-EO (expert opinion)" is a consensus of expert opinion based on clinical experience (.47).
The recommendation class is the strength of treatment or training that bene ts patients or learners and graded as follows; for example "Class 1 (strong)", such as strategy A is recommended/indicated in preference to treatment B the bene t is three folded than the risk; "Class IIa (moderate)", such as strategy A is possibly recommended/indicated in choice to treatment B the bene t is two folded than the risk; "Class IIb (weak)", such as effectiveness is unknown/unclear/uncertain or not well-established than B or the bene t is equal or more than the bene ts; "Class III No Bene t / Moderate", such as strategy A equals B and equal bene ts. "Class III harm / strong strategy A harms B more than the bene t (48). Table 2 Summary of levels of evidence and class of recommendation in training CPR.    2015) study also showed that there were signi cant variances in the knowledge score obtained during the instructor-led pretest at 65% and 90% after training and 80% after three months. Likewise, the student-directed knowledge score obtained at pretest was at 64% and 99% at posttest and 99% at the end of three months (P<0.001). Aliyari et al., (2019) conducted a comparative one-month follow-up study between instructor-led and concept map training method which con rmed the session effectiveness of the two methods and the result showed signi cant changes were found between the mean scores of knowledge of the two methods i.e. instructor-led at pretest were 5.82 and one month at 17.24(P<0.001) and concept map pretest at 6.6 and one months at 17.85 (P<0.001). However, the comparison of mean scores of knowledge was not statistically signi cant in the two groups in the post-test (P > 0.05).

Acquisition and retention of skills after CPR training
Eighteen studies reported on skills outcomes during CPR training in this systematic review study. The study population is the registered nurses (13 articles) and student nurses (5 articles). 13 studies were done in the tertiary hospital setting and ve were in the educational environment, such as the university nursing school. Fourteen articles used a one-day training period, while the remaining four used between three to seven days and all of them comprising either on pretest, posttest or follow-up. The review on level of acquisition and retention of skills have a combined sample size of 1,422 respondents with 37 minimum sample size and 404 maximum sample size. All eighteen papers use instructor-led assessment approaches. Eleven studies had a participant's response rate above 75%. All articles except Shahrakivahed et al., (2015) reported sex and age. Women made up most respondents except in a study by Aliyari et al., (2019). Twelve articles have a representative population of registered nurses, and only one article has graduate nurses as the participants in the study (44). Years of work experience for registered nurses in 15 studies range from 2 to 25 years, averaging 8.5±1.37 years. Most respondents come from medical or surgical wards in hospitals where they work.
Of the 18 articles evaluating the skills of the participants, only six tested skills only (24,27,42,44,45,46) and were evaluated in their assessment procedures, while the remaining 12 articles (11, 13, 26, 28, 29, 36, 27, 38, 39, 40. 41, 43)  In all 13 studies in this analysis, four uses repeated measures, while three are quasi-experimental research designs and four were randomized controlled trials. All these design uses tools that reduces bias and examine cause-effect relationships which is good at answering the question about knowledge level and retention among the nurses in this study. Equally, 11 out of 13 articles have a sample size of less than 200, ranging from 28-177 people, involving a single institution in hospitals or nursing schools.
The ndings of 10 articles in this analysis demonstrated the impact of CPR training in acquiring CPR knowledge level among nurses, comparing the level of knowledge acquired before training (pre-test) to signi cant level of knowledge acquired after training (post-test). (11,13,26,28, 36,37. 38, 39,40,41). Similar studies have been recorded between the pre-and post-tests performance by (5,16,52) that support the vital role of CPR training in ensuring that nurses are skilled, up-to-date and con dent in the event of cardiac arrest.
On the other hand, seven articles in this review also reported on the retention of CPR/BLS knowledge level on critical skills after the post training test. Their result also showed a gradual decline in knowledge retention among nurses after the post training test which appears to occur, on average between 1 months and 6 month after CPR training (one month -28, three months-26, 38, 39 ,41 and six months-11, 13 ). The above result is similar to the results of previous studies conducted by (49,50) which looked at CPR knowledge and skills retention. It should be noted that the less frequent an individual updates his or her knowledge, the lower the retention of knowledge/skills is expected, as theoretical knowledge and skills tend to decline over time (51,52). It is therefore suggested that training and retraining of CPR participants should be carried out at least within six months for all health professionals instead of the recommended 2-year period, as outlined in the 2015 ILCOR recommendation, and should be in keeping with the reality of each hospital training needs or environments.
Four studies in this review compared two different CPR/ BLS teaching methods among nurses between the instructor-led as a control group and others, such as video self-directed, concept map, student-directed and high-delity simulation as intervention group (

Skills among nurses following cardiopulmonary resuscitation (CPR) training
Four study were repeated-measure studies, while eight were quasi-experimental study designs and six were randomized controlled studies. Like most CPR / BLS training, it has been repeatedly shown in many previous studies that skills tend to decline faster than knowledge after CPR training (10,46). For example, two study results in this review showed a poor level of skills after initial training and follow-up compared to knowledge (13,36). This is similar to (61) ndings in their study, where the skills appear to be declining at a faster rate than knowledge, decreased as early as 2 weeks after training, and up to the pre-training levels by 1-2 years later. Similar studies by (62, 63, 64, 65 66) on varying pace of skills deterioration compared to knowledge was also identi ed.
On the contrary, four studies in this review reported opposite results where retention of skills was found to be better maintained at post-training than knowledge (11,24,26,44). Two studies among them reported previous half a year professional training in the BLS / CPR course seemed to enhance the actual memory of learning among participants (24,44). Furthermore, two studies by (67 68 ) also revealed a similar conclusion that previous training can enhance skills mastery.
Six articles in this reviewed study that tested skills only, revealed an overall signi cant increase in skill acquisition and retention between pre-test results and post-test outcomes using a deliberate practice approach (24,27,42,44,45,46). This nding is similar to what has been reported in a study among a group of lay people (21,69). Similarly, all 18 studies that tested skills competency used the checklist as a key instrument for testing skills using the lead manikins, except (43), which opted for the use of the MCQ. Again, six of the 18 articles were evaluated by a single evaluator, while ve were evaluated by two teams of evaluators, and the remaining seven use skill recording manikins and video recorders to evaluate the skills of the participants.
While comparing two different CPR/ BLS teaching methods between the instructor-led as a control group and other forms of teaching, such as video selfdirected, concept map, student-directed, and high-delity simulation as an intervention group, ve studies revealed a signi cant result in this review. This is comparable to the conclusion reported by (70), which revealed that participants who used a mix of case study and concept map rather than a case study alone had a superior knowledge. According to a study by (26, 49), those who have previous cardiac simulation experience learn and retain CPR skills better than those who learn and retain CPR knowledge.
When comparing student and teacher learning techniques, participants' BLS/AED knowledge and skills are observed to decline signi cantly between posttest and retention-test (41). This is also in line with previous research (71,72). However, Mardegan et al., (2015) observed no statistically signi cant changes in BLS skills of novice and practicing nurses at one week and eight weeks after training between CD and traditional instructor led BLS training approaches (P=0.41).

Limitation
Included studies that employ quasi-experimental procedures without blinding participants may raise doubts about their dependability, which is one of the drawbacks noted. In the same way, respondents' beliefs could impact the self-reported questionnaires employed in the identi ed studies.

Conclusions
Thirteen articles analyzed knowledge of nurses in this study, of which four are repeated studies, three quasi-experimental research designs and four are randomized controlled trials, only two uses cross-section descriptive studies. Likewise, 18 articles reported performance of skills four repeated measures studies, eight quasi-experimental study designs and six are randomized controlled studies. In this analysis, acquisition and retention of knowledge and skills have been shown to signi cantly increase between pre-test and post test results. However, skills were observed to decay faster than knowledge. Skills start to decay as early as 2 weeks whereas knowledge retention start to decline on average between 1 month and 6 month after BLS/CPR training. Most of the articles showed that other forms of teaching methods are also as effective as the instructor led/traditional teaching methods in knowledge and skills levels among nurses. Future studies should therefore focus on the optimal timeframe required for the re-certi cation of the BLS/CPR after initial training.
This will help to achieve the desired outcome of the reduction of cardiac arrest mortality.