A quantitative, descriptive cross-sectional design was used to conduct the study.

The data were collected in CCUs at hospitals in Sana’a, the capital city of Yemen. Hospitals in Yemen are categorized as governmental, military/police, teaching, or private. According to this categorization, a convenient sample was obtained from 8 randomly selected hospitals (two hospitals from each category with at least a 10-bed capacity CCU). Based on an estimated population size of 1200 CCNs, a power level of 0.95, a margin of error of 0.05, and a response rate of 50%, a sample size of 292 CCNs were considered sufficient. All the CCNs with at least 6 months experience in the CCU, who provided bedside care and EN to critically ill patients, and who agreed to participate in the study were included regardless of gender or qualification level. Nurses with less than 6 months of clinical experience, not involved in providing EN support, or unwilling to participate were excluded. All the types of CCUs were included. In total, 517 questionnaires were distributed in the eight hospitals; 308 were completed, giving a response rate of 59.6%. Sixteen questionnaires were incomplete, and they were excluded from the analysis.

The data were collected using a self-administered Nurses’ Perceptions of EN questionnaire developed by Persenius et al. [16]. After obtaining permission, the authors modified the questionnaire, and the final questionnaire contained two parts with 50 items. The first part contained 19 items about the participants and workplace characteristics. The second part measured the CCNs’ perceptions of EN and contained five subscales with 31 items. The first, second, and third subscales contained five questions each that assessed the nurses’ perceptions of the responsibility, knowledge, and support from the documentation system regarding EN using the nursing process framework. The fourth subscale assessed the nurses’ perceptions of the sources of knowledge regarding EN (7 sources). The fifth subscale assessed the nurses’ perceptions of their educational needs regarding EN (5 items) and the format of the required education (4 items). The items in the five subscales used a five-point Likert scale ranging from (1) “very small extent” to (5) “very great extent.” The mean score of each item was calculated out of 5; a score < 3 was considered low, 3–3.9 moderate, and 4–5 high [8]. The final questionnaire was revalidated by three experts in critical care nursing, translated into Arabic, translated back to English, and piloted. The Cronbach’s alpha coefficients for the five subscales of the second part (second to sixth) were 0.61, 0.83, 0.86, 0.88, and 0.84, respectively.

The Research Ethical Review Committee approved the study at Al-Razi University (No. 022/FOMS/2020). After obtaining consent from the hospitals, the researchers met the head/charge nurses of the participating CCUs and explained the objectives and procedures. The head/charge nurses explained the study and distributed the information sheet and questionnaire to the nurses. Participation was voluntary, anonymous, and confidential. Informed consent was obtained from the nurses after they received the information sheet and an adequate explanation. The head/charge nurse in each unit received the completed questionnaires confidentially. The data were collected between March and June 2021.

The Statistical Package for the Social Sciences (SPSS) version 16 (IBM, Armonk, NY, USA) was used for the data analysis. Descriptive analysis was performed with the demographic data and the nurses’ knowledge of the responsibility for NS. The mean scores of each subscale were obtained by dividing the sum of the raw scores by the number of items in the subscale. An analysis of variance (ANOVA) and an independent t-test was used to compare the mean scores of the perceptions of responsibility, knowledge, documentation system support, sources of knowledge, and educational needs. If the ANOVA test was significant, a post hoc test was used. P < 0.05 was considered statistically significant.

Table 1 presents the participants’ demographic and workplace characteristics. Most of the respondents were male (60.6%), single (52.1%), aged 30 years or less (65.1%), had an associate degree in nursing (57.5%), and 5 years or less experience in the nursing profession and critical care nursing (64.4% and 73.3%, respectively). About half of the respondents did not receive any in-service education/training about EN.

Table 2 describes the characteristics of the workplace. More than a third (38%) of the respondents worked in general CCUs, 22.9% in cardiac CCUs, and 17% in Neuro CCUs. The highest proportion (39%) worked in governmental hospitals, and 24% in military/police hospitals. Only 38% of the nurses reported that their CCU had guidelines or protocols for EN, 34.3% that a nurse or a dietitian was responsible for NS issues in their CCU, 29.4% that they had a nutritional support team (NST) in their units, and 35.6% that there was an NST at their hospital. Most nurses indicated that physicians prescribed the EN amount, type, rate, and probiotics.

In general, the nurses perceived that they had a low level of responsibility, a low level of knowledge, and a moderate level of support from the documentation system regarding EN care. The nurses’ perception of their knowledge had the lowest mean score (2.76 ± 0.89), and the perceptions related to documentation support the highest (3.15 ± 1.06). Of the items related to the level of knowledge, the highest mean score was for “planning and implementing interventions.” Knowledge regarding “prevention of complications” was second, and the lowest mean score was for “evaluating the outcome.” The nurses reported the highest mean score for responsibility and documentation support for the “prevention of complications” and “planning and implementing interventions” were second. The lowest responsibility and documentation support was “evaluating outcome” (Table 3).

The nurses in the teaching hospitals perceived a significantly lower responsibility for EN than those in the private hospitals; however, the nurses in all types of hospitals experienced a low level of responsibility. The nurses who received in-service education about EN and working in CCUs with policy or NST perceived having a significantly higher level of knowledge regarding EN than those who did not receive in-service education and nurses working in CCUs without policy or NST, though all had a low level of knowledge. The nurses working in the Neuro CCU felt they had significantly higher support from documentation than those in the general CCU. Similarly, the nurses with associate degrees felt they had significantly higher support from documentation regarding EN compared to those with bachelor’s degrees (Table 4).

Scientific workshops and conferences scored highest as the source of knowledge related to EN. All other sources had similar scores. The nurses from the private hospitals scored higher in educational courses in the nursing school than those from the military/police hospitals (p = 0.045). Scientific workshops and conferences received higher scores from the nurses of the military/police hospitals than those from the teaching hospitals (p = 0.039). In terms of their educational needs, the outcome evaluation scored the highest, followed by goal setting and nutritional assessment. The nurses reported that their educational needs could be met mostly through ward rounds, lectures, and presentations at team meetings. There were no significant differences between the nurses in the four types of the hospital regarding their educational needs and the format (Table 5).

Most nurses reported that there were no guidelines in place or were unaware of the availability of guidelines in their units. The internal conflict in Yemen affected the dissemination and implementation of EN guidelines in CCUs. In contrast to our results, Fulbrook et al. [26] and Roynette et al. [27] found that 75.7% (280/370 units) of CCUs in 20 European countries had guidelines or protocols for EN. However, the results of the current study are consistent with the findings of Darawad et al. [8], who reported that most Jordanian nurses reported no guidelines. The hospital administration must adopt and implement current evidence-based guidelines for EN in CCUs. The orientation of new nursing staff and periodical in-service education are essential for providing EN care consistently [8].

The lack of staff or team responsible for NS at the unit and hospital levels is consistent with the literature [16, 26, 27]. Establishing such a team will positively affect patient outcomes, nurses’ knowledge, and institutional costs [8, 16]. In the current study, physicians were primarily responsible for prescribing the EN type, rate, amount, and probiotics, similar to the literature [8, 11, 16]. CCNs administer the prescribed type, amount, rate, and probiotics of EN, but they need to understand the reasons for and effects of the prescribed EN to achieve the desired outcomes.

The CCNs perceived having a low responsibility and low knowledge regarding EN care, and their responsibility score was lower than that reported in Sweden [16] and Jordan [8], congruent with the studies that reported a lack of knowledge in nurses and healthcare workers [4, 7-9, 17-22]. Interestingly, the Yemeni CCNs perceived receiving reasonable support from documentation systems. This finding is comparable with the results of the Jordanian nurses [8] and inconsistent with the findings of the Swedish nurses [16].

Inadequate knowledge remains a barrier to implementing guidelines and providing adequate EN [4, 7]. Not using research regarding EN may be one factor contributing to the nurses’ perceptions of having a low level of knowledge [7, 8]. The main general knowledge sources were in-service education, guidelines and protocols, scientific journal articles, and the Internet [7, 8, 25]. Unfortunately, these sources are unavailable in most Yemeni hospitals due to the unstable situation. The high level of knowledge found by Morphet et al. [13] in CCNs could be attributed to the qualifications of the respondents, who were all members of the Australian College of Critical Care Nurses with postgraduate critical care qualifications. In the current study, most nurses have newly graduated with associate degrees.

In relation to the nursing processes, the CCNs indicated higher levels of responsibility, knowledge, and documentation support for the middle stage of the nursing process, compared with the earlier stages of assessment and the last stage of outcome evaluation. The CCNs’ autonomy and knowledge were prominent in performing actions rather than assessment and decision-making regarding EN. This finding is in accordance with Kim and Chang [7], Persenius et al. [16], Al Kalaldeh et al. [25], and Fletcher and Carey [12]. Globally, nursing roles in CCUs have changed and expanded in terms of increasing autonomy with greater responsibility [16], but the physician’s traditional dominant role over the other healthcare team members’ roles still exists in CCUs in Yemen. This situation may prevent nurses from adequately assessing, managing, and evaluating their patients’ nutritional status [13, 8].

The nurses in all types of hospitals indicated experiencing a low level of responsibility, low level of knowledge, and moderate documentation support regarding EN. However, the nurses in the private hospitals had the highest scores for responsibility and documentation support, and their self-perceived responsibility was significantly higher than those in teaching hospitals. This result supports the finding of Darawad et al. [8] that nurses in a private hospital had significantly higher responsibilities and support from the documentation system. In the CCUs in Yemen, EN policies are unavailable, updated, or followed, or the nurses may be unaware. However, policy availability with a clear description of the responsibilities and well-defined monitored and controlled systems in the private sector with specific functions for each healthcare member would ensure that nurses know their roles and responsibilities [8, 25].

The nurses who received EN-related in-service education and those working in CCUs with policy or NST reported significantly higher levels of knowledge than those without in-service education, policy, and NST groups. This finding is logical and supports the positive effect of in-service education on the EN level of knowledge, as also reported by Kim and Chang [7]. The presence of NST and policy/protocol regarding EN in the unit positively affects the nurses’ knowledge [8, 16]. Previous studies reported that NST and colleagues were the main sources of information regarding the EN [7, 16, 22, 25]. Similarly, the procedure manual and policy/guidelines/protocol regarding EN were among the main resources of EN information [6, 13, 22, 25].

The nurses with associate degrees or who worked in the Neuro CCU had significantly higher perceptions of support from the documentation system than those with a bachelor’s degree or nurses working in a general CCU. It appears that the nurses with bachelor’s degrees misunderstood the concept of support from documentation and that the documentation systems in specialized CCUs, including a Neuro CCU, were clear and provided moderate support for the nurses in their NS activities. Other possible reasons could be related to the heavier workload in general CCUs and the greater availability of clear policy documentation and procedure manuals in the specialized CCUs compared with the general CCUs [8].

In the current study, age, gender, nursing career experience, CCU experience, and job position had no significant effect on the nurses’ perceptions of responsibility, knowledge, or documentation support. This finding concurs with Morphet et al. [13], reporting that the level of qualification or the job classification had no significant effect on the self-rated EN level of knowledge.

Surprisingly, the perceived main sources of the nurses’ knowledge were scientific workshops and conferences, followed by postgraduate specialized education courses. Persenius et al. [16], Kim and Chang et al. [7], and Metin and Pars [22] found that consulting colleagues was the main source of knowledge for Swedish, Korean, and Turkish nurses. However, Darawad et al. [8] and Al-Ghabeesh et al. [28] reported the main sources of knowledge as education and the Internet. Morphet et al. [13] found that dietitians and protocols were the main sources of Australian CCNs information about EN. The low scores for undergraduate education could be because, in many countries, including Yemen, nutrition courses in nursing schools are provided by other schools or programs, such as medical or agriculture schools [8]. In-service educational programs in most Yemeni hospitals have been suspended or terminated, and the nurses worked with the knowledge they acquired during their undergraduate education. Surprisingly, Morphet et al. [13] found that even postgraduate studies in critical care nursing provided limited education related to nutrition, which was inadequate for informing EN decisions. In addition, the in-service education provided by Australian hospitals was limited.

The nurses’ perceptions of their educational needs confirmed their perceptions of their level of EN knowledge. The educational needs regarding the outcome evaluation scored the highest, followed by goal setting and nutritional assessment. The nurses reported that their educational needs could be met through ward rounds, lectures and presentations at team meetings. In-service education and liaising with colleagues were reported as the preferred format for meeting their educational needs regarding EN [8, 29].

These findings highlight that CCUs in Yemen urgently need to create or adopt EN evidence-based guidelines, protocols, and policies and establish multidisciplinary NST with clear roles and responsibilities for each member. Healthcare facilities should provide newly hired nurses with in-service education and training related to EN guidelines, policies, and practices. Similarly, nursing schools and faculty members should include nutritional courses, update the curricula, and integrate evidence-based guidelines into their courses. Increasing the availability of knowledge sources, such as textbooks, scientific journal articles, and international evidence-based guidelines, with periodical refresher in-service programs, is highly recommended. Documentation systems need to be examined and modified to provide better support for nurses. Further studies are recommended to examine the nurses’ practices and compliance with guidelines in relation to EN.

This study’s limitation was using a convenience sample that offered a low-bias control. Inclusion and exclusion criteria were formulated to decrease possible bias associated with the sampling technique, and the nurses in all the hospitals were approached during the three working shifts of the day.