An integrative review design was the most suitable method to explore and produce a new understanding from various types of literature (experimental, non-experimental, and theoretical) to enhance understanding of the phenomenon under investigation (i.e., EOLC). This method also facilitated nursing science by informing further research, care provision, and policy-making. It also highlights strengths, weaknesses, limitations, and gaps in knowledge, and supports what is already known about theories relevant to our topic [24]. Therefore, this design helps meet this review’s purposes.

The search process involved four phases which were developed by the first author (YR) and validated by two expert authors (MCC and KLA) as follows: (1) identifying the problems related to the research question, (2) conducting a systematic literature search, (3) screening the articles to develop themes, and (4) performing critical analysis to develop the themes.

From October 30, 2023, to November 10, 2023, electronic literature searches were conducted using major databases such as MEDLINE, Cochrane, CINAHL, EBSCO, and ScienceDirect.

Search methods were defined using the MeSH (Medical Subject Headings) descriptors of the keywords “end-of-life care,” “barriers,” and “critical care nurses.” Additionally, the reference lists of all identified articles were manually searched for additional studies. The operators used in this search included “AND” and “OR,” as well as the truncation tools of each database. A refined search was performed with terms such as “critical care nurses’ perceptions” OR “opinions” AND “quality end-of-life care” OR “quality of death and dying.” Subsequently, terms like “barriers” OR “obstacles” OR “challenges” AND “quality end-of-life care” OR “quality of death and dying” were employed. Finally, the descriptors “critical care nurses’ perceptions,” “barriers,” and “quality end-of-life care” were used (Fig. 1).

PRISMA search flow diagram

The inclusion criteria for this search to select relevant articles were as follows: (1) Full-text articles, (2) Papers published in the English language from 2010 to 2023, and (3) Articles that specifically describe the barriers perceived by critical care nurses that affect the quality of end-of-life care.

Intervention studies and studies that describe barriers to providing quality end-of-life care from other perspectives, such as physicians and patients’ families were excluded. For the studies who included nurses and other health care workers within the context of critical care, the researchers included the results that relevant to nurses and excluded the others.

The data extraction and analysis were carried out to collect and consolidate the data from the selected studies into a standard format relevant to the research field. The extracted data included specific descriptions of the settings, populations, study methods, and outcome measures (Table 31). Two authors (YSR and KLA) independently extracted the data and reached an agreement after discussion with the third author (MCC).

Following the review process, the authors made the final decision on studies that met the study criteria. Out of a total of 103 articles, 9 duplicates were removed. The abstracts of the remaining 94 articles were initially found to be somewhat relevant to the research topic. However, after examining the articles in terms of research methodology and results, 36 articles that matched the selection criteria for this study were ultimately chosen. The full text of the 36 articles was reexamined based on the title first for suitability. Subsequently, the abstracts of the studies were reviewed, leading to the exclusion of 23 articles for various reasons, leaving 13 studies for further consideration in this study. However, two articles were disqualified as they did not contain a specific research methodology or reviewed literature papers; they relied solely on theoretical information. This step resulted in the inclusion of 11 research articles in this integrative review of the literature (Table 1).

To ensure the methodology’s quality and avoid bias in the design, highly credible and respected search engines were adopted to select peer-reviewed studies according to the inclusion criteria in this review. The articles chosen in this review were categorized into two sections based on study design and research methodology: quantitative and qualitative studies. These were evaluated manually and independently for each study, with any disagreements resolved by two experts (KLA, Professor, and MCC, Associate Professor) who have experience in research methodology, using the Mixed Methods Appraisal Tool (MMAT) version 2018 [25]. This tool includes specific criteria for evaluating the quality of quantitative, qualitative, and mixed-method studies. The MMAT consists of a checklist of five research components for each type of study with a rating scale including “Yes,” “No,” and “Can’t tell.” The overall results suggest that the evidence quality across the ten studies was high (Table 2).

Thematic analysis in this review involves a systematic process of coding and theme development, using both inductive and deductive approaches. This method ensures a comprehensive synthesis of diverse data sources, providing valuable insights into the research topic [24, 26]. Thematic analysis was employed for all studies to investigate the subject of interest. The coding for the themes in this review followed the six recommended phases: Familiarizing with the data; making initial codes; searching for themes; reviewing themes and making a thematic plan; defining and naming themes; generating the final picture of the report [24]. The coding was conducted by the primary author (YSR) and confirmed by the three secondary authors (LH, SM, and LY). Any discrepancies were discussed and resolved through consensus.

The search process yielded a total of 103 articles. All articles resulting from the search process were independently reviewed by all authors in this study for the research process, purpose, methodology, tools, main findings, recommendations, and limitations.

Eight cross-sectional descriptive studies and three qualitative studies were selected, which were conducted in the following countries: two from the USA [27, 28] and a single study from each of the following countries: Saudi Arabia [22], Jordan [29], Egypt [12], Malaysia [13], Scotland [30], Poland [31], Hong Kong [32], South Africa [33], and China [34].

In this comprehensive analysis of 11 studies, a diverse range of methodologies and findings were examined across different countries and healthcare settings. The studies included a mix of quantitative and qualitative approaches, with sample sizes varying from small convenience samples to larger cohorts. Key barriers to providing End of Life Care (EOLC) were identified, such as challenges in communication with families, lack of support from managers, and insufficient training in EOLC. The studies highlighted the importance of addressing these barriers to improve the quality of care provided by nurses in critical care settings. Notably, demographic characteristics and their impact on EOLC provision were not consistently addressed across the studies, indicating a potential area for further research and exploration in this field (Table 31).

The thematic analysis of included studies revealed several key themes and sub-themes related to barriers in End of Life Care (EOLC). These themes encompassed various aspects, including challenges related to patients and their families, healthcare institutions and the environment, as well as barriers specific to nurses. Communication and collaboration between patients, nurses, and families included issues such as seeking updates about patient status, misunderstandings about life-saving measures, misunderstanding poor prognosis, troubled family dynamics, and conflicts within families regarding life support decisions [22, 34]. Additionally, barriers related to Institution Policy and procedures highlighted concerns such as insufficient standard procedures, communication challenges in decision-making, inadequate ICU design, inappropriate staffing policies, and deficiencies in rooms, supplies, and noise control. Furthermore, barriers associated with nurses encompassed their emotional experiences and socio-demographic characteristics [12] (Table 4).