Introduction

Electrocardiography (ECG) is a noninvasive diagnostic method used to examine the electrical and muscular activity of the heart [1]. Usually, carried out as a first-line diagnostic tool in patients with chest pain [2]. Electrocardiography is crucial for diagnosing patients who need time-sensitive interventions such as urgent reperfusion therapy [3]. Electrocardiography is the most frequently used cardiac diagnostic method performed to identify electrico-cardiac abnormalities and acute coronary syndrome (ACS) [4]. Cardiac arrhythmias are one of the leading causes of death worldwide [5]. Because of a scarcity of resources and medical practice, they are also a significant public health concern in low- and middle-income countries in Sub-Saharan Africa [6, 7]. Currently, ECG is the sole and most effective method for identifying ACS and life-threatening conditions, helping with the prompt delivery of emergency therapy [8]. It is also essential to examine cardiac arrest and direct clinicians to quickly begin the patient's previous therapy in emergency and critical care areas [9]. ECG recordings are becoming familiar as a result of the rise in cardiovascular events, an aging population, and ease of use [5].

Nurses play a crucial role in providing care in the emergency department and critical care settings [10]. Nurses are the most crucial and noticeable component of the healthcare system in underdeveloped countries [11]. Nurses are usually the front line to look at the ECG results and identify both normal and pathological electrocardiographic tracings that may need urgent attention and management [4]. Nurses who care for critically ill patients need to have a thorough understanding of the data offered by the ECG, which includes heart rate, regularity of the rhythm, interval measurements, and wave form pattern [12].

Numerous health care providers, such as cardiologists, paramedics, and doctors, as well as nurses, interpret electrocardiography [13]. ECG interpretation is a crucial clinical skill for emergency and critical care unit nurses, as they are the continuing nurses who monitor and care for patients during emergencies and enable the patients' anticipation of potentially fatal events [14]. In accordance with current practice, nurses may collect data and inform physicians, who make treatment decisions based on rhythm interpretation [12].

ECG misinterpretation has the potential to lead to unwanted patient outcomes, including death [15, 16], as well as cause an unpleasant burden on patients and hospitals [2]. Patients who present with symptoms of ACS need an ECG immediately and an interpretation of the ECG strip in less than 10 min, since delays in therapy may cause considerable tissue damage to the cardiac cells [17]. To decrease mortality due to cardiac emergencies, nurses’ ability to read ECGs should be assessed [18]. In fact, not all health service managers are aware of the importance of improving nurses’ ability to decrease mortality due to slow patient handling during life-threatening dysrhythmias [19]. A study was performed at the University of California, San Francisco, with 411 consecutive ED ST elevation myocardial infarction (STEMI) diagnoses, and 36% were false-positive [20]. According to Breen et al. (2019), up to 33% of ECG interpretations, including some, were inaccurate, and up to 11% led to clinical mismanagement [21].

To our knowledge, there is no previous study on adult emergency and critical care unit nurses’ knowledge and practice regarding electrocardiography interpretation in Ethiopia. To ascertain the knowledge and practice of nurses working in adult emergency and critical care units at selected governmental hospitals in Addis Ababa, Ethiopia, this cross-sectional survey was carried out.

Methods

Study design, study area and period

This study was an institutional-based cross-sectional study conducted at four randomly selected governmental hospitals in Addis Ababa, Ethiopia, from March 06, 2023, to April 05, 2023. Addis Ababa is the capital city of Ethiopia and has thirteen governmental hospitals. Among governmental hospitals, four were selected using a simple random sampling technique, namely, St. Paul’s Hospital Millennium Medical College (SPHMMC), Zewditu Memorial Hospital, Tikur Anbesa Specialized Hospital (TASH), and St. Peter’s Specialized Hospital. These hospitals have been providing emergency and critical care services. The total number of nurses working in adult emergency and critical care units at selected governmental hospitals was 359.

Samples

All nurses who were working in adult emergency and critical care units (cardiac care unit and intensive care unit) at the participating hospitals who were available during the data collection period and who volunteered to participate in the study were included. Nurses who had less than six months of work experience were excluded from this study.

The largest sample size calculated was 384. After adding a non-respondent rate of 5%, the final sample size was 403. Since our study populations were less than the calculated sample size, all nurses who fulfilled the inclusion criteria were included.

Variables

Dependent variable

Knowledge and practice regarding electrocardiography interpretation.

Independent variables

Socio-demographic variables: gender, age, working area, educational qualification, work experience, and current working unit experience.

Institutional-related variables: availability of monitors or ECG machines, availability of ECG manuals, discussion of ECG for patients, in-service training, types of training attended, mode of training course, months since the respondents took the last ECG course, current working unit in the hospital, and nurse-to-patient ratio.

Operational definitions

Satisfactory Knowledge: Study participants who scored ≥ 75% on the knowledge questions were considered to have satisfactory knowledge about electrocardiography interpretation [22].

Unsatisfactory knowledge: Those nurses who scored below 75% on knowledge items were considered to have unsatisfactory knowledge about electrocardiography interpretation [22].

Good practice: Those study participants who scored ≥ 75% were considered to have good practice in electrocardiography interpretation [22].

Poor practice: Those nurses who scored below 75% on the practice-related questions were considered to have poor practice in electrocardiography interpretation (2

Critical care unit: An organized system for providing care to critically ill patients who offers intensive and specialized medical and nursing care, including intensive care units and cardiac care units [23].

Data collection

Five Bachelor of Science (BSc) nurses were trained and assigned to collect data, and one of them was a supervisor. Data were collected by using a pretested, self-administered questionnaire adopted from previous similar studies [2, 8, 12, 22]. The questionnaire had four parts: part one included nurses’ demographic information, part two presented institutionally related variables, and part three consisted of 12 multiple-choice questions about the nurses’ knowledge of ECG interpretation. The last part of the data collection tool consisted of 12 ECG strips to assess nurses’ ECG interpretation practices. For each strip, nurses were requested to recognize (interpret) the displayed rhythm by selecting one answer only out of thirteen possible answers. One score was assigned to each right answer, and no score was allocated to the wrong answers or to the choice of "I do not know" for both knowledge and practice questions.

A pretest was performed on 5% of the total respondents two weeks prior to the actual data collection at Yekatit 12 Hospital to test the clarity, acceptability, and understandability of the questionnaire. Data collectors were closely followed by the supervisors and principal investigator. The data collection instrument was assessed for its completeness and accuracy at every shift and day during the data collection period. Cronbach’s alpha value was calculated to check the internal reliability of the 2 tools used in the current study as follows: knowledge item scored (α = 0.724) and practice item scored (α = 0.774).

Data analysis

The collected data were coded and cleaned using Epi-data version 4.6 and transferred to Statistical Package for the Social Science (SPSS) version 25 for further analysis. The results were presented using descriptive statistics and numerical summary measures such as frequency distribution tables and graphs. Both bivariate and multivariate logistic regression analyses were performed. Those variables with P ≤ 0.25 in the bivariate logistic regression were selected for the multivariate logistic regression analysis. This could help to control the confounders of the independent predictors of ECG interpretation knowledge and practice. The Hosmer and Lemeshow goodness of fit test was used to test the fitness of the final model. To declare a statistically significant association between the dependent and independent variables, P values of < 0.05 in the multivariate analysis were used as the cut-off point. The strength of association was measured using odds ratios (crude and adjusted) with corresponding 95% confidence intervals (CIs).

Ethical considerations

Ethical clearance was obtained from the Institutional Review Board (IRB) of St. Paul ‘s Hospital Millennium Medical College. Supportive official letters were also obtained from the respective hospitals to collect data from the participants. Written informed consent was obtained from each study participant according to the principles of the Helsinki declaration. Confidentiality of the participants’ information was maintained by not writing their names in the questionnaire.

Results

Socio-demographic and institution-related factors

From the total 359 samples, 334 volunteered to participate in the study with a response rate of 93%. More than half (184, 55.1%) of the respondents were males. The median age of the study participants was 28 years, with an IQR of 26–30. Almost two-thirds of the participants (231, 69.2%) were bachelor’s degree holders. Approximately half (170, 50.9%) of the study participants had 1–5 years of clinical experience in the nursing profession, and more than one-third (124, 37.1%) of them had current unit job experience ranging from two to four years. Of the total participants, 34 (10.9%) were working in the CCU (Table 1).

Table 1 Socio-demographic and institution-related factors towards ECG interpretation among adult emergency and critical care unit nurses at selected public hospitals in Addis Ababa, Ethiopia, 2023 (n= 334)
Full size table

In terms of past ECG-related training, only 116 (34.7%) had been exposed to in-service training programs, and 218 (68.3%) of the nurses had not received any ECG-related training (Fig. 1).

Fig. 1
figure 1

The participating nurses related to their ECG training in adult emergency and critical care units at selected public hospitals in Addis Ababa, Ethiopia, 2023 (n = 334)

Full size image

Nurses’ knowledge in electrocardiography interpretation

Satisfactory knowledge regarding ECG interpretation was determined to be a minimum score of 9 out of 12 points. Nurses who scored less than 9 were considered to have unsatisfactory knowledge of ECG interpretation. Out of the twelve knowledge-related questions distributed to the respondents regarding ECG interpretation, only 53 (15.9%) nurses had satisfactory knowledge and scored ≥ 75% (Fig. 2).

Fig. 2
figure 2

The level of knowledge towards ECG interpretation among adult emergency and critical care unit nurses at selected public hospitals in Addis Ababa, Ethiopia, 2023 (n = 334)

Full size image

In this study, the highest knowledge score was 11 (91.6%), and the lowest score was 0. The results of this study revealed that the knowledge level of nurses regarding ECG interpretation was generally unsatisfactory. In our study, respondents’ knowledge of the specific item was variable. The most frequently correctly answered item by nurses, according to the results of the current study, was that the correct order of ECG waves and intervals was “P wave, PR interval, QRS complex, T wave, ST interval, and U wave” 248 (74.3%). However, in a normal ECG (augmented vector right), 74 (22.2%) aVR leads were “Negative waves”, and 85 (25.4%) T waves were positive in normal ECGs except “in leads aVR and VI”, which were the two questions most frequently wrongly answered by the study participants. Over half (190, 56.9%) of the study participants correctly answered the item related to ST depressions in ECG indicating “myocardial infarction” (Table 2).

Table 2 Nurses’ knowledge of ECG interpretation among adult emergency and critical care unit nurses at selected public hospitals in Addis Ababa, Ethiopia, 2023(n = 334)
Full size table

Nurses’ practice in electrocardiography interpretation

The nurses’ practice level regarding ECG interpretation was determined using twelve ECG stripe questions, including patient scenarios. Based on the category specified in the operational definition, those nurses who answered ≥ 75% of practice-related questions said they had good practice. The results indicated that 304 (91%) of the study participants had poor practice regarding ECG interpretation, and they scored < 75% of the practice questions correctly (Fig. 3).

Fig. 3
figure 3

The level of practice towards ECG interpretation among adult emergency and critical care unit nurses at selected public hospitals in Addis Ababa, Ethiopia, 2023(n = 334)

Full size image

In this study, the highest practice score was 10 (83.3%), and the lowest score was 0.

Regarding the interpretation of the ECG strips, the results were diverse. The majority (271, 81.1%) of the study participants scored a high level of practice on the identification of asystole. On the other hand, a significant number of nurses had difficulty recognizing third-degree heart block (306, 91.6%), first-degree heart block (291, 88%), sinus bradycardia (288, 86.2%), atrial flutter (283, 84.7%), and atrial fibrillation (259, 83.7%). More than one-third of nurses correctly identified ventricular tachycardia (132, 39.5%), sinus tachycardia (113, 33.8%), STEMI (142, 42.5%), and nearly half (156, 46.7%) of nurses correctly identified PEA in a given ECG strip (Table 3).

Table 3 Nurses’ practice towards ECG interpretation among adult emergency and critical care unit at selected public hospitals in Addis Ababa, Ethiopia, 2023 (n= 334)
Full size table

Factors associated with nurses’ knowledge towards interpretation of ECG

In bivariate analysis, 9 variables showed a P value of < 0.25; respondents’ age, working area, educational qualifications, nursing work experience, current working unit experience, availability of ECG manuals, in-service training, time since the respondents took the last ECG training course, and current working unit in the hospital were associated with nurses’ ECG interpretation knowledge. However, in multivariate analysis, only four variables, educational qualification, current working unit experience, in-service training and current working unit in the hospital, were significantly associated with nurses’ ECG interpretation knowledge.

Among the study participants, master of science (MSc) nursing holders were 1.99 times more likely (AOR = 1.99 (1.09, 4.43)) to have satisfactory knowledge than bachelor’s degree holders. Nurses who had 2–4 years of experience in the current unit were 3.13 times more likely (AOR = 3.13 (1.23, 7.92)) to have sufficient knowledge than nurses who reported less than 2 years of experience in the current working unit. The results revealed that those participants who received in-service ECG training were 7.75 times more likely to be knowledgeable in ECG interpretation than those who did not receive the training before (AOR = 7.75 (3.76, 15.96)). Moreover, nurses who worked in the cardiac care unit had 4.92 times higher odds (AOR = 4.92 (1.81, 13.33)) of having satisfactory knowledge than those who worked in the intensive care units (Table 4).

Table 4 Variables associated with knowledge regarding ECG interpretation among Nurses’ Working at Selected Public Hospitals in Addis Ababa, Ethiopia, 2023(n = 334)
Full size table

Factors associated with nurses’ practice towards interpretation of ECG

In the bivariate analysis, seven variables, respondents’ age, working area, educational qualification, nursing work experience, in-service training, current working unit in the hospital, and knowledge, were associated with nurses’ ECG interpretation practice. Only educational qualifications, in-service training, current working unit, and knowledge were significantly associated with nurses’ ECG interpretation practice scores in the multivariate analysis.

Among the study participants, MSc holders were 5.79 times more likely (AOR = 5.79(2.02, 16.6)) to have a good practice compared with degree holders. Those nurses who received in-service ECG-related training had 13.79 times higher odds of having good practice (AOR = 13.79(4.63, 41.1)) in ECG interpretation compared to those who did not receive the training before. Nurses who worked in cardiac care units were 3.45 times more likely to have good practices than those who worked in intensive care units (AOR = 3.45 (1.02, 13.73)). Study participants who had satisfactory knowledge of ECG interpretation were 3.52 times more likely to have good practice than those who had unsatisfactory knowledge (AOR = 3.52 (1.56, 8.13)). In other words, nurses who were more knowledgeable about ECG interpretation had better practice regarding the interpretation of ECG stripes (Table 5).

Table 5 Variables associated with practice about ECG interpretation among Nurses Working at Selected Public Hospitals in Addis Ababa, Ethiopia, 2023(n=334)
Full size table

Discussion

Nurses are the frontline workers in emergency and critical care units, facing different types of patient problems, particularly those with life-threatening ECG abnormalities that require more attention. As a result, the present study aimed to assess the level of nurses’ knowledge and practice of electrocardiography interpretation and the associated demographic and institutional variables at selected hospitals in Addis Ababa.

The current study findings showed that only 53 (15.9%) nurses had a satisfactory level of knowledge regarding electrocardiography interpretation. Our findings are comparable with those of a study performed at Assuit University Heart Hospital in Egypt (17.5%) [8]. However, the current study findings are identical to those of studies in Turkey, India, and Malaysia, which found values of 67%, 97.1%, and 38%, respectively [2, 10, 24]. This disparity might be due to the difference in nursing curriculum regarding ECG, variations in technological advancement with the countries, differences in the operational definition of satisfactory knowledge from study to study, and differences in exposure to ongoing and sustainable on-the-job training programs. For example, in this study setting, the majority (68.3%) of nurses did not receive any ECG-related training, whereas in Turkish and Indian studies, 61% of studied nurses were trained [2, 24].

The lowest percentage of correct answers in this study is the item related to in normal ECG aVR leads are “negative waves” (22.2%), and T waves are positive in normal ECG except in “ leads aVR and VI” (25.4%). This finding was refuted by research conducted in Malaysia, which found that 81% and 76.2% of studied nurses correctly answered those items, respectively [10]. This difference could be due to differences in the study setting. However, in this study, 45.8% of nurses answered that T long waves and QRS wide waves were seen in cases of “hyperkalaemia”. This finding is in line with a study reported by Tahboub and Yılmaz (44.6%) [2].

In the present study, the most frequently known and correctly responded knowledge item was the correct order of ECG waves, and the intervals were “P wave, PR interval, QRS complex, T wave, ST interval, and U wave” (75.4%). Our finding is higher than that of a study conducted in Malaysia, in which 26.7% of nurses answered that item correctly [10]. The discrepancy might be due to differences in educational level. In the Malaysian study, 38.1% of the studied nurses were diploma holders; however, in this study, there were no diploma nurses. This finding is inconsistent with a study performed in Spain that revealed that all study participants correctly answered items related to the correct order of ECG waves and intervals [22].

In the current study, the proportion of nurses who had good ECG interpretation practices was found to be 9%. Our finding is similar to a study conducted in Egypt, which found that 7.5% of nurses had good practice regarding electrocardiography interpretation [8]. The practice level of nurses in this study was higher than that in a study conducted in Iran, which showed that only 4.6% of emergency medical service (EMS) personnel had good practice towards electrocardiography interpretation [14]. The possible justification for this might be due to a difference in sample size (i.e., the sample size for this study was 334, whereas the sample size for the Iran study was 65 EMS personnel).

The findings of this study are lower than those of a study conducted in Tanzania, Spain, Hong Kong, and Malaysia, which revealed that 60%, 97%, 64.2%, and 76.2% of nurses had good practice levels regarding ECG interpretation, respectively [10, 12, 22, 25]. Moreover, the possible justification could be due to the difference in the cut-off point of the scores regarding the practice-related tool, practice misconception of nurses regarding their role in ECG interpretation, study population, study setting, differences in nursing curriculum, low training exposure in this study population, as well as the study tools (instrument) used. For example, in this study, the cut-off point for a good practice was ≥ 75%, whereas the cut-off point for a good practice in a Malaysian study was ≥ 50% [10].

According to the current study findings, 81.1% of the respondents identified Asystole from the ECG strip. Our finding is similar to that of a study performed in Tanzania by Ruhwanya, which found that 82.3% of study participants correctly identified asystole [12]. In the current study, a considerable number of nurses had difficulties identifying third-degree heart block (306, 91.6%), first-degree heart block (291, 88%), sinus bradycardia (288, 86.2%), atrial flutter (283, 84.7%), atrial fibrillation (279, 83.7%), and normal sinus rhythm (237, 71%). This finding was inconsistent with other studies [2, 4, 12, 25]. The possible justification for this could be due to differences in the study setting and previous ECG training.

In this study, 42.5% of nurses correctly identified ST elevation myocardial infarction (STEMI). Our study findings are comparable to research performed in Iran, which found that 46% of studied nurses correctly identified STEMI [14]. In the current study, more than half (39.5%) of the study participants correctly recognized ventricular tachycardia. This finding is in line with a study conducted at Addis Abebe University and Haramaya University College of Health Science, which found that 41.1% of study participants correctly identified ventricular tachycardia [9].

This study identified that MSc. degree holders were 1.99 times more likely to have satisfactory knowledge (AOR = 1.99 (1.093, 4.435)) than BSc holders. The possible explanation for this might be that as nurses’ educational level increases, they could have a higher chance of getting up-to-date information regarding ECG interpretation. This finding is similar to a study performed in Malaysia that showed that there was a significant correlation between nurses’ educational qualifications and level of knowledge about ECG interpretation [10], while a study conducted in India and Turkish reported that there was no statistically significant correlation between nurses’ educational level and nurses’ knowledge score regarding ECG interpretation [2, 24].

Nurses who had 2–4 years of current unit work experience had 3.13 times more satisfactory knowledge than nurses who reported less than 2 years of current work experience (AOR = 3.13 (1.23, 7.92)). The possible explanation for this might be that when the nurses’ years of experience increase, they might acquire knowledge regarding ECG interpretation.

Nurses who received ECG-related training had 7.75 times higher odds of having satisfactory knowledge (AOR = 7.75 (3.76, 15.96)) than those who had not received training. The findings of this study were supported by studies conducted in Turkey, Malaysia, Spain, and India, which showed that training improved nurses’ understanding of ECG interpretation [2, 10, 22, 24]. This could be because updating the knowledge of nurses about the interpretation of ECGs could have resulted in good scores on the knowledge question.

Nurses who worked in the cardiac care unit were 4.92 times higher in knowledge (AOR = 4.92 (1.81, 13.33)) than those who worked in intensive care units. This finding is supported by a study conducted in Turkey and Malaysia [3, 23].

MSc holders were 5.79-fold more likely (AOR = 5.79 (2.02, 16.6)) to have good practice than degree holders. This study finding was consistent with the study conducted in Tanzania, which revealed that educational qualifications were significantly associated with nurses’ ECG interpretation practices [12].

Nurses who received in-service ECG training had 13.79 times higher odds of having good practice in ECG interpretation than those who did not receive the training before (AOR = 13.79 (4.63, 41.1)). This finding is supported by studies conducted in India [24], Tanzania [12], Hong Kong [25], Spain [22], Turkey [2] and Malaysia [10]. The current study finding was incongruent with those studies conducted in Saudi Arabia, which reported that nurses’ practice did not correlate with training [4, 8].

Nurses who were working in the cardiac care unit were 3.43 times more likely to have good practice than those who were working in the critical care unit (AOR = 3.43 (1.02, 13.73)). This could be because nurses working in the CCU only care for patients with heart diseases and are thus continuously involved in the interpretation of ECG. This finding was similar to those of a study performed in Turkey and two studies in Saudi Arabia and Malaysia [2, 4, 8, 10]. This study revealed that nurses with a satisfactory level of knowledge improved their practice level by 3.52 times more than those with an unsatisfactory level of knowledge (AOR = 3.52 (1.56, 8.43)). In other words, nurses who were more knowledgeable about ECG interpretation were more likely to have good practice regarding the interpretation of ECG stripes. This finding is supported by studies conducted in Egypt and India [8, 24].

Limitations

This study had several limitations that were identified. First, a cross-sectional study was used, and it did not show a cause and effect relationship. Second, the use of self-administered questionnaires to assess nurses’ practical skills during ECG interpretation may decrease data quality; the participants had the option of seeking assistance or consulting additional materials.

Conclusion

Generally, the results of the present study showed that the majority of nurses had unsatisfactory knowledge and poor practice regarding ECG interpretation. The identified factors that increased the likelihood of satisfactory knowledge and good practice of electrocardiography interpretation were educational qualifications, in-service training, and current working units. Nurses’ electrocardiography interpretation practice was significantly associated with knowledge level. Participation in ECG-related training was found to be strongly linked to nurses’ knowledge and practice. The concerned body must provide continuous on-the-job training related to ECG interpretation to all nurses working in adult emergency and critical care units, and nurses should be upgraded with continuing education. We also recommend that further observational studies using a checklist might also be useful to evaluate the actual practice among the nurses on the bedside.