Sepsis assessment and management in critically Ill adults: A systematic review

Background Early assessment and management of patients with sepsis can significantly reduce its high mortality rates and improve patient outcomes and quality of life. Objectives The purposes of this review are to: (1) explore nurses’ knowledge, attitude, practice, and perceived barriers and facilitators related to early recognition and management of sepsis, (2) explore different interventions directed at nurses to improve sepsis management. Methods A systematic review method according to the PRISMA guidelines was used. An electronic search was conducted in March 2021 on several databases using combinations of keywords. Two researchers independently selected and screened the articles according to the eligibility criteria. Results Nurses reported an adequate of knowledge in certain areas of sepsis assessment and management in critically ill adult patients. Also, nurses’ attitudes toward sepsis assessment and management were positive in general, but they reported some misconceptions regarding antibiotic use for patients with sepsis, and that sepsis was inevitable for critically ill adult patients. Furthermore, nurses reported they either were not well-prepared or confident enough to effectively recognize and promptly manage sepsis. Also, there are different kinds of nurses’ perceived barriers and facilitators related to sepsis assessment and management: nurse, patient, physician, and system-related. There are different interventions directed at nurses to help in improving nurses’ knowledge, attitudes, and practice of sepsis assessment and management. These interventions include education sessions, simulation, decision support or screening tools for sepsis, and evidence-based treatment protocols/guidelines. Discussion Our findings could help hospital managers in developing continuous education and staff development training programs on assessing and managing sepsis in critical care patients. Conclusion Nurses have poor to good knowledge, practices, and attitudes toward sepsis as well as report many barriers related to sepsis management in adult critically ill patients. Despite all education interventions, no study has collectively targeted critical care nurses’ knowledge, attitudes, and practice of sepsis management.


Introduction
Sepsis is a global health problem that increases morbidity and mortality rates worldwide and which is one of the most common complications documented in intensive care units (ICUs) [1]. About 48.9 million cases of sepsis and 11 million sepsis-related deaths were documented in 2017 worldwide [2]. Sepsis is an emergency condition leading to several life-threatening complications, such as septic shock and multiple organ dysfunction and failure [3]. Sepsis has negative physiological, psychological, and economic consequences. Untreated sepsis can lead to septic shock; multiple organ failure, such as acute renal failure [4]; respiratory distress syndrome [5]; cardiac arrhythmia (e.g. Atrial Fibrillation) [6]; and disseminated intravascular coagulation (DIC) [7]. Also, sepsis is associated with anxiety, depression, and post-traumatic stress disorder [8]. As for the financial burden of sepsis on the healthcare system, the cost of healthcare services and supplies for ICU critical care patients with sepsis is high [1]. In 2017, the estimated annual cost of sepsis in the United States (US) was over $24 billion [2].
Previous studies have shown that among nurses, misunderstanding and misinterpretation of the early clinical manifestations of sepsis, poor knowledge, attitudes, and practices related to sepsis, and inadequate training might lead to delayed assessment and management of sepsis [9][10][11]. Moreover, the limited numbers of specific and sensitive assessment tools and standard protocols for the early identification and assessment of sepsis in critical care patients leads to delayed management, therefore increasing sepsis-related mortality rates [10].
Critical care nurses, as frontline providers of patient care, play a vital role in the decisionmaking process for the early identification and prompt management of sepsis [11]. Therefore, improving nurses' knowledge, attitudes, and practices related to the early identification and management of sepsis is associated with improved patient outcomes [12,13]. To date, there remains a wide gap between the findings of previous research and sepsis-related clinical practice in critical care units (CCUs). Furthermore, there is no evidence in the nursing literature regarding nurses' knowledge, attitudes, and practices related to the early identification and management of sepsis in adult critical care patients and the association of these factors with patient health outcomes. Therefore, summarizing and synthesizing the existing research on sepsis assessment and management among adult critical care patients is needed to guide future directions of sepsis-related clinical practice and research. Accordingly, this review aims to identify nurses' knowledge, and attitudes, practices related to the early identification and management of sepsis in adult critical care patients.

Materials and methods
The present review used a systematic review design guided by structured questions constructed after reviewing the nursing literature relevant to sepsis assessment and management in adult critical care patients. The authors (MR, DB, AH) carefully reviewed and evaluated the selected articles and synthesized and analyzed their findings to reach a consensus. This review was guided by the following questions: (a) what are nurses' knowledge, attitudes, and practices related to sepsis assessment and management in adult critical care patients?, (b) what are the perceived facilitators of and barriers to the early identification and effective management of sepsis in adult critical care units?, and (c) what are the interventions directed at improving nurses' sepsis assessment and management?

Eligibility criteria
The review questions were developed according to the PICOS (Participants, Interventions, Comparisons, Outcome, and Study Design) framework, as displayed in Table 1.
Inclusion criteria. The articles were retrieved and assessed independently by two researchers (MR, DB) according to the following inclusion criteria: (1) being written in English, (2) having an abstract and reference list, (3) having been published during the past 10 years, (4) focusing on critical care nurses as a target population, (5) examining knowledge, attitudes, and practices related to the assessment and management of sepsis, and (6) having been conducted in adult critical care units.
Exclusion criteria. Studies were excluded if they were (1) written in languages other than English, and (2) conducted in pediatric critical care units or non-ICU. Dissertations, reports, reviews, editorials, and brief communications were also excluded.
Search strategy. An electronic search of the databases CINAHL, MEDLINE/PubMed, EBSCO, Embase, Cochrane, Scopus, Web of Science, and Google Scholar was conducted using combinations of the following keywords: critical care, intensive care, critically ill, critical illness, knowledge, awareness, perception, understanding, attitudes, opinion, beliefs, thoughts, views, practice, skills, strategies, approaches, barriers, obstacles, challenges, difficulties, issues, problems, limitations, facilitators, motivators, enablers, sepsis, septic, septic shock, and

Item Description
Participants patients aged 19 years and older who were admitted to critical care settings with sepsis, septic shock, or septicemia Intervention Training/educational interventions (e.g., regular lectures, simulations, algorithms, decision support tools, and sepsis protocol) Comparison No restriction was applied on the number or type of comparison group as the impact of the intervention could be determined. Comparison groups could include no intervention, standard protocol, and other types of intervention which was educational

Outcome
The primary outcomes of interest in this review were the effective assessment and prompt management of sepsis and nurses' knowledge, attitudes, practice, perceived barriers, and enablers related to sepsis assessment and management. sepsis assessment and management could be assessed using either patient or nurse objective measures. Sepsis assessment and management were quantified as mean times required for sepsis recognition and treatment initiation, sepsis protocol adherence, and decline in mortality rate in-hospital sepsis-related complications. nurses' knowledge, attitudes, and practice related to sepsis could be assessed using either nurse-reported tools or performancebased tests, while nurses' perceived barriers and enablers could be assessed using nurse-reported tools.
septicemia. The search terms used in this review were described in S1 File. The search was initially conducted in March 2021, and a search re-run was conducted in April 2022. The search was conducted in the selected databases from inception to 4/2022. The initial search, using the keywords independently, resulted in 1579 articles, and after using the keyword combinations, this number was reduced to 241 articles. Then, after applying the inclusion and exclusion criteria, the number of articles was reduced to 92. A manual search of the reference lists of the 92 articles was carried out to identify any relevant publications not identified through the search. The researcher (MR) used the function "cited by" on Google Scholar to explore these publications in more depth. The researchers (MR, DB) then screened the identified citations of these publications, applying the eligibility criteria. In case of discrepancies, the researchers (MR, DB) discussed their conflicting points of view until a consensus was reached. Then, after careful reading of the article abstracts, 61 irrelevant articles were excluded, and a total of 31 articles were included in this review.

Data extraction
The following data were extracted from each of the selected studies: (1) the general features of the article, including the authors and publication year; (2) the characteristics of the study setting (e.g., single vs. multisite); (3) the sociodemographic and clinical characteristics of the target population, including mean age, and medical diagnosis (e.g., sepsis, septic shock, and SIRS); (4) the name of the sepsis protocol used, if any; (5) the characteristics of the study methodology (e.g., sample size and measurements); (7) the main significant findings of the study; and (8) the study strengths and limitations. All extracted data were summarized in an evidence-based table (Table 2). Data extraction was performed by two researchers (MR, DB). An expert third researcher (AH) was consulted to reach a consensus between the two researchers throughout the process of data extraction.

Ethical considerations
There was no need to obtain ethical approval to conduct this systematic review since no human subjects were involved.

Quality assessment and data synthesis
A quality assessment of the selected studies was performed independently by two researchers based on the guidelines of Melnyk and Fineout-Overholt [14]. Disagreements between the two researchers (MR, DB) were identified and resolved through a detailed discussion held during a face-to-face meeting. For complicated cases, the researchers (MR, DB) requested a second opinion from a third researcher (AH). According to the guidelines of Melnyk and Fineout-Overholt [14], twelve of the studies were at level 3 in terms of quality, four studies at level 5, and nine studies at level 6. A qualitative synthesis was performed to synthesize the findings of the reviewed studies. The following steps were applied throughout the process of data synthesis: 1. The data in the selected studies were assessed, evaluated, contrasted, compared, and summarized in a table (Table 2). This data included the design, purpose, sample, main findings, strengths/limitations, and level of evidence for each of the studies.
2. The similarities and differences between the main findings of the selected studies were highlighted.
3. The strengths and limitations of the reviewed studies were discussed.

Description of the selected studies
Most of the reviewed studies were conducted in Western countries [9,11,12], with only one study conducted in Eastern countries [1], and two in Middle-Eastern countries [15,16]. The detailed geographical distribution of the studies and other characteristics are described in Table 2.

Nurses' knowledge, attitudes, and practices
Nine of the selected studies assessed nurses' knowledge and attitudes related to sepsis assessment and management in critically ill adult patients [1,9,12,15,[17][18][19][20][21] ( Table 3). Nucera et al. [18] found that ICU nurses had poor attitudes towards blood culture collection techniques and timing and poor levels of knowledge related to the early identification, diagnosis, and management of sepsis. For example, the majority of nurses reported that there is no need to sterilize the tops of culture bottles, and there is no specific time for specimen collection [18]. However, the participating nurses reported good levels of knowledge related to blood culture   procedures and the risk factors for sepsis. Similarly, R. J. Roberts et al. [19] found the participating nurses to have good knowledge of septic shock and good attitudes toward the initiation of antibiotics for critically ill adult patients with sepsis. Only two studies assessed nurses' To examine the difference in mean times required for sepsis recognition and treatment initiation between nurses exposed to 2 different monitor displays in response to simulated case scenarios of sepsis practices related to sepsis assessment and management [15,19]. For example, in the study of R. J. Roberts et al. [19], 40% of the nurse participants reported that they were aware of the importance of initiating antibiotics and IV fluid within one hour of septic shock recognition [20]. Also, Yousefi et al. [15] found the participating nurses to have good practices related to sepsis assessment and management.

Barriers to and facilitators of sepsis assessment and management
The reviewed studies identified three types of barriers to the early identification and management of sepsis, namely patient-, nurse-, and system-related barriers (Table 4). Meanwhile, only nurse-and system-related facilitators were reported in the reviewed studies. The mostreported barriers and facilitators were system-related. The reported barriers included (a) the lack of written sepsis treatment protocols or guidelines adopted as hospital policy [22,23]; (b) the complexity and atypical presentation of the early symptoms of sepsis [19]; (c) nurses' poor level of education and clinical experience [1,12]; (d) the lack of sepsis educational programs or training workshops for nurses [22,23]; (e) the high comorbid burden among patients with sepsis, which complicates the critical thinking process of sepsis management [19]; (f) nurses' deficits in knowledge related to sepsis treatment protocols and guidelines [22][23][24]; (g) the lack of mentorship programs in which junior nurses' actions/activities are strictly supervised by experienced nurses [17,23]; (h) heavy workloads or high patient-nurse ratios [22]; (i) the shortage of well-trained and experienced physicians, particularly in EDs [19,22,23]; (j) the lack of awareness related to antibiotic use for patients with sepsis [19,22]; (k) the lack of IV access and unavailability of ICU beds [25]; (l) the non-use of drug combinations for the treatment of sepsis [22,26,27], and (m) poor teamwork and communication skills among healthcare professionals [22,26]. Only three facilitators of sepsis assessment and management were identified in the reviewed studies. These facilitators were (1) nurses' improved confidence in caring for patients with sepsis, (2) increased consistency in sepsis treatment, and (3) positive enforcement of successful stories of sepsis management [22,27].

Measurement tools of sepsis-related knowledge, attitudes, and practices
One of the reviewed studies used a Knowledge, Attitudes, and Practice (KAP) questionnaire developed according to the Surviving Sepsis Campaign (SSC) guidelines [15] to measure nurses' knowledge, attitudes, and practices related to sepsis assessment and management. Meanwhile, eight studies [1,9,12,[17][18][19][20][21] used self-developed questionnaires based on the literature and SSC guidelines and validated by expert panels. Details of these measurement tools and their psychometric properties are summarized in Table 5.

Interventions directed at improving nurses' sepsis assessment and management
Educational programs. Only four of the selected studies examined the impact of educational programs on nurses' knowledge, attitudes, and practices related to sepsis management and found significant improvements in nurses' posttest scores (Table 6) [11,15,28,29]. For example, Drahnak's study [28] implemented an educational program developed by the authors and integrated with patients' health electronic records (HER) and found significant improvements in nurses' post-test nursing knowledge scores. Another educational program developed by the authors was implemented to improve ICU nurses' knowledge, attitudes, and practices related to sepsis and found a significant improvement in posttest scores among the intervention group [15]. Another study was designed to examine the effectiveness of the Taming Sepsis Educational Program1 (TSEP™) in improving nurses' knowledge of sepsis [11]. A 15-minute structured educational session was developed to decrease the mean time needed to order a sepsis order set for critically ill patients through improving ER nurses' knowledge about SSC guidelines and found that the mean time was reduced by 33 minutes among the intervention group [29].
Simulation. Only two studies examined the effect of using simulation in improving the early recognition and prompt treatment of sepsis by critical care nurses (Table 6) [30,31]. Vanderzwan et al. [30] assessed the effect of a medium-fidelity simulation incorporated into a multimodel nursing pedagogy on nurses' knowledge of sepsis and showed significant improvements in six of the nine questionnaire items. While Giuliano et al. examined the difference in mean times required for sepsis recognition and treatment initiation between nurses exposed to two different monitor displays in response to simulated case scenarios of sepsis and showed a significant reduction in the mean times required for sepsis recognition and treatment initiation by those nurses who were exposed to enhanced bedside monitor (EBM) display [31].
Decision support tools. Four of the selected studies examined the effectiveness of decision support tools, adapted based on the SSC guidelines and the "sepsis alert protocol", on the early identification and management of sepsis and confirmed the effectiveness of these tools (Table 7) [32][33][34][35]. The decision support tools used in three of the studies guided the nurses throughout their decision-making processes to reach effective assessment, high quality and timely management of sepsis, and, in turn, optimal patient outcomes [32,33,35]. However, no significant differences in the time of blood culture collection and antibiotic administration were reported between the intervention and control groups in the study of Delawder et al. [34].
Sepsis protocols. Eight of the selected studies examined the effectiveness of sepsis protocols [24,[36][37][38] and sepsis screening tools [16,[39][40][41] for the early assessment and management of sepsis ( Table 7). All of these articles revealed that the implementation of sepsis screening tools or protocols based on the SSC guidelines leads to the early identification and timely management of sepsis, as well as the improvement in nurses' compliance to the SSC guidelines for the detection and management of sepsis. For example, in one study, patients who received Early Goal-Directed Therapy (EGDT) had a lower mortality rate as compared to patients who received usual care [16]. The sepsis screening tools and guidelines were also tested to examine their impact on some patient outcomes, and variabilities were identified. For example, the use of the Modified Early Warning Score (MEW-S) tool revealed no significant improvement in patient mortality rate [41]. In contrast, mortality rates were decreased by using the Nurse Driven Sepsis Protocol (NDS) [40], Quality Improvement (QI) initiative [38], and a computerized protocol [37]. In addition, nurses in the computerized protocol group had better compliance with the SSC guidelines than did nurses in the paper-based group [37]. One of the selected studies compared between a paper-based sepsis protocol and a computer-based protocol and found that antibiotic administration, blood cultures, and lactate level checks were conducted more often and sooner by nurses in the computerized protocol group [37].
Sepsis alert (Binary alarm system) The tool consists of two steps. The first step is the detection of actual or potential sepsis, and the second is screening & stratification conducted within 15 minutes 89% of septic patients were detected by the alert system, & screening and stratification was completed for 75% of the cases within an hour from notification. The tool sensitivity was 94%.
Long et al.
User interface alert User interface alert was designed for medical systems to a provide computer support system for decision-making related to sepsis  The readmission rate was reduced from 36.28% to 25% 8 weeks after the NDS protocol, and compliance to the sepsis intervention protocol improved but with no effect on mortality rate.

PLOS ONE
Two of the selected studies used the EGDT as a screening tool for sepsis and found no significant differences in times of diagnosis, blood culture collection, or lactate measurements between the control and intervention groups [16,24]. However, significant differences were found in the time of antibiotic administration in the study of Oliver et al. [24]. Although Elkhuri et al. [16] revealed no significant differences in the time of antibiotic administration, the mortality rate among patients in the intervention group declined significantly.

Discussion
Most of the reviewed studies focused on assessing critical care nurses' knowledge, attitudes, and practices related to sepsis assessment and management, revealing poor levels of knowledge, moderate attitude levels, and good practices. Also, this review revealed that the three most common barriers to effective sepsis assessment and management were nursing staff shortages, delayed initiation of antibiotics, and poor teamwork skills. Meanwhile, the three most common facilitators of sepsis assessment and management were the presence of standard sepsis management protocols, professional training and staff development, and positive enforcement of successful stories of sepsis treatment. Moreover, this review reported on a wide variety of interventions directed at improving sepsis management among nurses, including educational sessions, simulations, screening or decision support tools, and intervention protocols. The impacts of these interventions on patient outcomes were also explored. The findings of our review are consistent with the findings of previous studies which have explored critical care nurses' knowledge related to sepsis assessment and management [42]. Also, recent studies conducted in different clinical settings support the findings of our review regarding nurses' knowledge of sepsis. For example, a recent study conducted in a medicalsurgical unit revealed that nurses had good knowledge of early sepsis identification in non-ICU adult patients [43]. The variations in nurses' levels of knowledge related to sepsis assessment were attributed to variations in educational level and work environment (i.e., ICU vs. non-ICU).
The evidence indicates that the successful treatment of critically ill patients with suspected or actual sepsis requires early identification or assessment [44,45]. Early assessment is a critical step for the initiation of antibiotics for patients with sepsis, leading to improved patient outcomes and a decline in mortality rates [44]. The current review also revealed the significant role of educational programs in improving nurses' knowledge, attitudes, and practices related to the early recognition and management of sepsis. These findings are in line with the findings of another study, which tested the impact of e-learning educational modules on pediatric nurses' retention of knowledge about sepsis [45]. The study revealed that the educational modules improved the nurses' knowledge acquisition and retention and clinical performance related to sepsis management [45]. The findings of our review related to sepsis screening and decision support tools are in congruence with the findings of a previous clinical trial which assessed the impact of a prompt telephone call from a microbiologist upon a positive blood culture test on sepsis management [46]. The study revealed that this screening tool contributed to the prompt diagnosis of sepsis and antibiotic administration, improved patient outcomes, and reduced healthcare costs [46]. The findings of our review related to the effectiveness of educational programs in improving the assessment and management of sepsis were consistent with the findings of a recent quasi-experimental study. The study found that incorporating sepsis-related case scenarios in ongoing educational and professional training programs improved nurses' self-efficacy and led to a prompt and accurate assessment of sepsis [47]. One of the interventions explored in this review was a simulation that facilitated decision-making related to sepsis management. The simulation was found to be effective in mimicking the real stories of patients with sepsis and proved to be a safe learning environment for inexperienced nurses before encountering real patients, increasing nurses' competency, self-confidence, and critical thinking skills [48]. Also, a recent study showed that the combination of different interventions aimed at targeting sepsis assessment and management, including educational programs and simulation, may lead to optimal nurse and patient outcomes [49].

Limitations
The present review has several limitations. There is limited variability in the findings of the reviewed studies in terms of the main variable, sepsis. Moreover, the review excluded studies written in languages other than English and conducted among populations other than critical care nurses. However, there may be studies written in other languages which may have significant findings not considered in this review. Further, only eight databases were used to search for articles related to the topic of interest, which may have limited the number of retrieved studies. Finally, due to the heterogeneity between the selected studies, a meta-analysis was not performed.

Relevance to clinical practice
Our findings could help hospital managers in developing continuous education and staff development training programs on assessing and managing sepsis for critical care patients. Establishing continuous education, workshops, professional developmental lectures focusing on sepsis assessment and management for critical care nurses, as well as training courses on how to use evidence-based sepsis protocol and decision support and screening tools for sepsis, especially for critical care patients are highly recommended. Also, our findings could be used to development of an evidence-based standard sepsis management protocol tailored to the unmet healthcare need of patients with sepsis.

Conclusion
To date, nurses remain to have poor to good knowledge of and attitudes towards sepsis and report many barriers related to the early recognition and management of sepsis in adult critically ill patients. The most-reported barriers were system-related, pertaining to the implementation of evidence-based sepsis treatment protocols or guidelines. Our review indicated that despite all educational interventions, no study has collectively targeted nurses' knowledge, attitudes, and practices related to the assessment and treatment of sepsis using a multicomponent interactive teaching method. Such a method would aim to guide nurses' decision-making and critical thinking step by step until a prompt and effective treatment of sepsis is delivered. Also, despite all available protocols and guidelines, no study has used a multicomponent intervention to improve health outcomes in adult critically ill patients. Future research should focus on sepsis-related nurse and patient outcomes using a multilevel approach, which may include the provision of ongoing education and professional training for nurses and the implementation of a multidisciplinary sepsis treatment protocol.