Introduction
From the limited data available in Africa, up to 50% of patients in intensive care units (ICUs) have been reported to acquire healthcare-associated infections (Mahomed et al., 2017; Ndadane & Maharaj, 2019). Furthermore, there is a paucity of data on the burden of sepsis in adult critical care units in the public sector in South Africa. Clinical practice guidelines for sepsis management such as the 2016 surviving sepsis campaign (SSC) guidelines (Rhodes et al., 2017) have been developed to eradicate sepsis globally. However, no evidence has yet been published related to the implementation of sepsis guidelines in public adult ICUs in South Africa. Furthermore, it is unclear from observation whether nurses have sufficient knowledge regarding sepsis and related guidelines and whether their practices related to managing sepsis in adult mechanically ventilated patients are evidence based.
Globally, sepsis is a major public health problem for healthcare professionals. Sepsis has been associated with approximately 28.6%-30% of mortality in the United States, although there are no information estimates related to sepsis in ICUs (Saldanha & Messias, 2017; Sinha et al., 2018; Vincent et al., 2019). The number of sepsis cases per annum was approximately 15-19 million cases worldwide, whereas, in Europe, approximately 30%-38% of patients in critical care units have had at least one healthcare-associated infection (Vincent et al., 2019).
Both developed and developing countries are affected by sepsis. However, with improved technologies (such as digital alerting and electronic warning systems) being used for at-risk patient detection in the ICU, appropriate hygiene protocols, and adequate nutrition, sepsis has been successfully minimized, especially in developed countries (Downing et al., 2019; Joshi et al., 2019; Westphal et al., 2018).
Sepsis-related morbidity and mortality in developing countries are believed to be disproportionately high compared with developed countries (Westphal et al., 2018). Although a number of research studies of sepsis have been conducted in sub-Saharan countries such as Kenya, Uganda, and Tanzania, its epidemiology in these countries remains poorly described, especially for adult ICUs (Tupchong et al., 2015).
Clinical practice guidelines are described as statements that include recommendations intended to optimize patient care. These guidelines should be informed by a systemic review of the evidence and an assessment of the benefits and harms of alternative care options (Kredo et al., 2016). Guidelines have a range of purposes, including improving effectiveness and quality of care. In addition, guidelines aim to decrease variations in clinical practice and decrease costly and preventable mistakes and adverse events, thus increasing patient safety (Jordan, 2011; Kredo et al., 2016).
Patients on mechanical ventilators are at a greater risk of sepsis because of the interference with normal processes during intubation and artificial/assisted ventilation. Patients managed through the use of clinical practice guidelines are more likely to have reduced incidences of developed sepsis. Reducing incidences of sepsis further decreases the incidences of other complications such as sepsis-induced acute respiratory distress syndrome, prolonged hospital stays, and mortality (Kim & Hong, 2016).
According to Mpasa (2017), noncompliance by nurses to clinical practice guidelines is a barrier to the management of mechanically ventilated patients in critical care units. Nurses are at the bedside of these patients and must therefore be knowledgeable in their field and associated practice, according to the latest evidence related to sepsis. Nurses have a fundamental role in surveilling sepsis, recording data correctly, and reporting information to provide relevant data that will allow fellow nurses to improve their practice and more effectively monitor their actions (Kleinpell et al., 2013).
Purpose
The aim of this study was to investigate the effect of an educational intervention on nurses' sepsis-related knowledge and practices in mechanically ventilated adult patients in public ICUs in Eastern Cape Province, South Africa.
Methods
Study Design
An intervention study using a quasi-experimental, two-group, pretest-posttest design (Gray et al., 2016; Moule et al., 2017; Sidani, 2015) was conducted in three phases. These phases included Phase 1, the pretest questionnaire (collecting baseline data before the implementation of the educational intervention); Phase 2, the educational intervention (development, review, and implementation of the intervention, based on SSC guidelines); and Phase 3, the posttest questionnaire (evaluation of the educational intervention implemented in Phase 2 of the study). Phases 1 and 2 were conducted between June and September 2018, whereas Phase 3 was conducted between the second week of September to the third week of October 2018.
Study Setting
This study was conducted in ICUs in five tertiary public hospitals in Eastern Cape Province (two public hospitals in Nelson Mandela Bay, two in Buffalo City, and one in Oliver Tambo District). The total bed capacity of these ICUs was 56 beds, providing care for adult patients only. In these five ICUs, there were more nurse specialists (with formal college training) than nurses with in-hospital training.
Population and Sampling
Nurses working in the five targeted ICUs were purposively allocated to three groups: Intervention Group 1 (ICUs 1 and 2), which received the full intervention; Intervention Group 2 (ICUs 3 and 4), which received part of the intervention; and the control group (ICU 5), which received no intervention. Group allocation was done based on geographical location. To recruit the largest sample possible, convenience sampling was used to invite all nurses (diploma and degree nurses) working in the selected adult ICUs who were available and willing to participate.
Data Collection Instrument
As no previous validated questionnaire was available, self-administered, structured pretest and posttest questionnaires were developed based on the latest SSC guidelines (Rhodes et al., 2017). The questionnaires measured the items discussed in the following sections.
Demographic data of the participants (section A)
Demographic data were collected on participants' gender, age (in years), number of years worked in ICUs, position held by the participant in the unit, type of employment, specialized ICU training as an additional qualification, and highest professional qualification. Closed-ended questions were used to collect these data.
Knowledge related to sepsis in mechanically ventilated adult patients in intensive care units (section B)
This item included four subsections, of which three explored the availability and use of the guidelines. Subsection 4 explored the knowledge of nurses related to the guidelines. "True" or "false" responses were solicited.
Practices related to sepsis in mechanically ventilated adult patients in intensive care units (section C)
This item included three subsections. Subsection 1 had 12 questions, whereas the other two subsections had four questions each. Subsection 1 assessed nursing practices related to the guidelines in mechanically ventilated adult patients admitted in ICUs, Subsection 2 assessed the diagnosis of sepsis, and Subsection 3 assessed the practices of the guidelines. A 5-point Likert scale, ranging from (1) never to (5) always, was used.
The Intervention
The educational intervention was developed by the first author based on 2016 SSC guidelines (Rhodes et al., 2017). These guidelines include "The Sepsis in Resource Limited Nations Initiative," which was deemed by the first author to be suitable for inclusion in the educational intervention because of the limited resources available to the public health system in South Africa. The developed educational intervention consisted of a 20-minute PowerPoint presentation, printed materials based on the 2016 SSC guidelines, and, for Intervention Group 1, monitoring visits conducted twice per month over the 3 months after the implementation of the educational intervention. The purpose of the monitoring visits was to check if the guidelines were being used by the participants and whether any clarification regarding its use was required. Intervention Group 2 received the same educational intervention, but no monitoring visits were made. The control group did not receive any intervention.
Data Collection Process
After obtaining relevant ethical clearances and permissions, all of the eligible nurses at the five selected ICUs were invited to participate. The study was explained to these nurses to provide a broad understanding of the research and to enable them to make informed decisions. All of the nurses who agreed to participate completed consent forms before completing the pretest questionnaire, which was distributed by the first author. Participants were asked not to discuss the answers, and the questionnaires were collected by the first author immediately after completion.
The process of collecting the pretest data in all of the ICUs took 3 weeks, after which the educational intervention was immediately implemented, lasting for a period of 3 months. Posttest data were collected in the same manner as the pretest data upon completion of the intervention.
Data Analysis
A Microsoft Excel template was used by the first author to capture pretest/posttest results to obtain both knowledge and practice scores. A statistician assisted with the descriptive and inferential statistics, as these were calculated to obtain the measures of central tendency (i.e., mean, median, and dispersion). The chi-square test of independent variance was used to determine if there were significant relationships among the responses from the participants. A chi-square p value less than .005 was considered statistically significant. Cramer's V was calculated to produce a statistically significant result with a value greater than .01.
Validity and Reliability
The pretest and posttest questionnaires were evaluated by the statistician, and the pretest and posttest questionnaires and educational intervention were reviewed by five experts, including an intensivist, a critical care nursing educator at the local university, a nurse manager at a private college, a unit manager at a private hospital, and a unit manager at a public hospital. The reviewers considered the cost efficiency, user friendliness, and availability of the equipment used to implement the educational intervention based on the SSC guidelines as well as the educational intervention's relevancy in terms of context, scope of practice of the nurses, face validity, and internal and external validity. To ensure reliability, the pretest and posttest questionnaires and the educational intervention were piloted in ICU 3 on 14 participants. After the pilot study, five questions in Section B (Statements B4.3 and B4.4) and Section C (Statements C1.6, C1.8, and C1.12) were revised. The results of the pilot study were not included in the main study.
Study Permission and Ethical Clearance
The relevant ethical clearances were obtained from the university (Ref. No. H17-HEA-NUR-021) as well as the Eastern Cape Department of Health (Ref. No. EC_201801_005). Furthermore, hospital managers and unit managers in the selected ICUs granted permission for data collection. Written consent was obtained from all of the nurses who agreed to participate. After the conclusion of data collection, the control group (ICU 5) was given a copy of the PowerPoint presentation and the printed materials based on the 2016 SSC guidelines.
Results
In the pretest phase, questionnaires were distributed to 127 participants. One hundred twenty-one questionnaires were returned, with four discarded because of incompleteness, giving a response rate of 92%. In the posttest phase, questionnaires were distributed to 117 participants. One hundred one questionnaires were returned, with seven discarded because of incompleteness, giving a response rate of 80%.
Demographic Data
Demographic data for the group of participants at enrollment are presented in Table 1.
A predominant number of participants were female, with most between the ages of 40 and 59 years. Almost all of the participants were permanently employed and involved with patient care. More than half of the participants had more than 10 years of experience in the ICU. Nearly 40% held a 4-year diploma in nursing science, and more than half had received specialized critical care training.
Knowledge Related to Sepsis in Mechanically Ventilated Adult Patients in Intensive Care Units
Descriptive statistics for the items on knowledge related to sepsis in mechanically ventilated adult patients in ICUs are presented in Table 2.
Most of the items concerning participants' knowledge related to sepsis in mechanically ventilated adult patients in ICUs in both the pretest and posttest questionnaires were correctly answered by participants. Moreover, the rate of correctness of answers, with the exception of answers to Items B4.3 and B4.5, had improved on the posttest questionnaire. There was a medium significant difference on Item B4.3 at posttest, [chi]2 (df = 4, n = 94) = 17.25, p = .002, V = 0.30 (medium).
Practices Related to Sepsis in Mechanically Ventilated Adult Patients in Intensive Care Units
Descriptive statistics for the answers to questions on the practices related to sepsis in mechanically ventilated adult patients in ICUs are shown in Table 3.
Statements C1.1-C1.12
The findings revealed that most participants gave correct answers for most items with an improvement in practices for the posttest, with the exception of Statements C1.2, C1.7, C1.8, and C1.12. Medium statistical differences were only noted for Statements C1.2 for both pretest, [chi]2(df = 8, n = 117) = 25.74, p = .001, V = 0.33 (medium), and posttest, [chi]2(df = 8, n = 94) = 18.55, p = .017, V = 0.31 (medium); C1.8 for the pretest only, [chi]2(df = 8, n = 117) = 16.20, p = .040, V = 0.26 (medium); and C1.12 for both pretest, [chi]2(df = 8, n = 117) = 20.09, p = .010, V = 0.29 (medium), and posttest, [chi]2(df = 8, n = 94) = 17.39, p = .026, V = 0.30 (medium).
Comparison Within the Three Groups During Pretest and Posttest
The results of the comparisons within Intervention Group 1, Intervention Group 2, and the control group in terms of the responses to questions on the knowledge and practices related to sepsis in mechanically ventilated adult patients in ICUs are presented in Table 4.
The knowledge score was higher in the pretest (57.72) than the posttest (54.61) in Intervention Group 1, although the difference was not significant. A significant increase in knowledge score was seen in the scores for both Intervention Group 2 (62.18 vs. 53.28) and the control group (56.72 vs. 70.05). With regard to practice scores, similar results were noted, but with a significance only for the control group (p = .002).
Discussion
This study was designed to assess the effect of an educational intervention on nurses' knowledge and practices of sepsis in mechanically ventilated patients in adult public ICUs. The results study show that an educational intervention based on the SSC guidelines, including a 20-minute PowerPoint presentation and handing out of printed materials, may have some effect on sepsis knowledge related to mechanical ventilation of adult patients. Including monitoring visits as part of the educational intervention did not show a positive effect in terms of increasing knowledge or practice scores. Studies have been done to investigate the effect of educational interventions on healthcare professionals' knowledge in various healthcare contexts (Abu Farha et al., 2018; Melo E Lima et al., 2018; Patel et al., 2015). However, studies including an educational intervention on sepsis related to mechanical ventilation in the adult public ICUs are scarce. Therefore, further research is required to expand knowledge in the field of nursing education with respect to sepsis related to mechanically ventilated patients in the ICU context. This study is believed to be the first of its kind to be undertaken in South Africa.
The pretest-to-posttest increases in knowledge scores were relatively high for all three intervention groups but increased most significantly in Intervention Group 2 and the control group. High overall knowledge scores may be attributed to educational level (Khalil et al., 2019). For example, 61% of the participants were ICU trained, with approximately 40% of the ICU trained nurses having completed their training recently. This may have been associated with the generally high knowledge scores as well as the knowledge improvement in the posttest, which is consistent with Elkalmi et al.'s (2014) study on motivation and obstacles for adverse drug reaction among healthcare professionals, which found a significant association between improvement in knowledge and the educational level of participants.
For Intervention Group 2, the increase in knowledge scores was attributed to the educational intervention, which included a PowerPoint presentation and printed materials (Abu Farha et al., 2018; Tahoon et al., 2020). Prior studies have reported a similar effect for educational interventions using PowerPoint presentations, videos, WhatsApp group sharing, guidelines, and clinical scenarios on knowledge improvement outcomes (Abu Farha et al., 2018; Prins & Human, 2019; Tahoon et al., 2020). In this study, for the control group, the significant improvement in knowledge scores may be attributed to participants' involvement in the study having enhanced their knowledge about sepsis.
The practice scores, which were generally higher than the knowledge scores in the pretest phase, also were higher at posttest for both Intervention Group 2 and, significantly, the control group. However, these increases were not as significant as the increases in knowledge scores. This finding is congruent with a similar study by Melo E Lima et al. (2018). Furthermore, an increase in knowledge scores may also translate into increased scores for practices at posttest.
The finding of this study regarding the increase in practice scores for Intervention Group 2, although not significant, is similar to that of other studies that found an improvement in health professionals' practice scores, especially in relation to the effectiveness of an educational intervention (Bisallah et al., 2018; Kissoon, 2014; Varallo et al., 2017). Therefore, it is recommended that training using up-to-date clinical practice guidelines such as SSC guidelines be incorporated into the daily routine of critical care nurses and that quality systems be implemented to ensure these guidelines are adhered to and are enforced for all nurses in ICUs. In addition, comparisons of the two intervention groups against the control group indicate a small, statistically significant improvement in knowledge and practices among nurses. However, when these results were analyzed separately, comparing knowledge and practice scores, practice scores were higher than knowledge scores on the pretest yet did not increase as significantly as knowledge scores. Furthermore, most of the participants gave correct answers during pretest and posttest questionnaires. Many factors may have contributed to the reported findings, including differences in the control group versus the intervention groups in terms of, for example, age, level of experience, level of motivation, and bed occupancy rate in the ICU. Finally, a traditional approach of some nurses in the management of patients, including using their intuition, and information from their peers or colleagues, may have contributed to Intervention Group 1's performance (Jordan et al., 2016).
It is recommended that an observational study be conducted to describe the practices of nurses in relation to their knowledge of sepsis in mechanically ventilated adult patients in ICUs to better elucidate the findings highlighted in this study.
Furthermore, more studies on factors influencing the increase or decline in knowledge and practice scores should be conducted, especially in resource-constrained settings such as the South African public health system.
Limitations
This study was affected by several limitations. First, the study was conducted in the ICUs of public hospitals in Eastern Cape Province only. Thus, the results should not be generalized to private hospitals or to public hospitals in other provinces.
Second, convenience sampling was used in this study to enroll as many participants as possible. However, the anonymous nature of the questionnaire made it difficult to ensure that all of the participants completed the questionnaires at both time points.
Third, the researcher used a self-administered structured questionnaire to obtain responses regarding the practices related to sepsis treatment. This did not allow the truthfulness of reporting to be verified. This limitation would be addressed in an observational study format. Furthermore, the educational intervention could include a demonstration to enhance practices regarding sepsis related to mechanical ventilation in adult patients in ICUs.
Fourth, the posttest questionnaire was done 3 months after the start of the educational intervention. Although time frames of 3 months or less have been used in other educational intervention studies using a theory-based approach (Arrogi et al., 2017; Bosch et al., 2019), this time frame may be overly short for real uptake and usage by participants to take place. Furthermore, the methods (e.g., a 20-minute PowerPoint presentation) used in the educational intervention may not have had the same effect on all of the participants, and confounding variables such as employment of new staff, shortage of staff, and in-service education as well as uncontrolled demographic differences between the control and intervention groups may have influenced the implementation of the educational intervention and its effect on knowledge and practice scores. Finally, the study only targeted nurses working in ICUs. As nurses do not decide on the treatment plan alone, the guidelines should be adapted for and introduced to the entire team, including other stakeholders such as physicians and technicians. Therefore, it is recommended that both the educational intervention and the questionnaire be adapted and further tested with a more diverse sample. The educational intervention should use alternative methods of education, which may be implemented over a longer period to evaluate its impact on outcomes such as length of stay and patient prognosis.
Conclusions
The findings of this study indicate that the availability of SSC guidelines has a small but significant and positive effect on sepsis knowledge related to mechanical ventilation. However, including monitoring visits in the educational intervention had no effect on related knowledge or practices. Thus, the educational intervention used in this study should be further adapted and explored to improve efficacy.
The results of this study may be used by both hospital managers and nurses to improve adherence to best practices with regard to sepsis in adult mechanically ventilated patients in public ICUs. Furthermore, the results may be used in the training of intensive care nurses. Finally, more research is required, especially using more-diverse samples, to explore the factors that contribute to improving knowledge and practices and the differing effects that educational interventions using different teaching and implementation strategies have in the context of the management of sepsis in adult mechanically ventilated patients in public ICUs.
Acknowledgments
The authors would like to thank Vicki Igglesden for editing the article and the respondents for participating in the study.
Author Contributions
Study conception and design: EZH, WTHB, PJ
Data collection: EZH, BO
Data analysis and interpretation: EZH
Drafting of the article: EZH, WTHB, PJ
Critical revision of the article: WTHB, PJ
References