Abstract
Background
Delirium is a condition characterized by an acute and transient disturbance in attention, cognition, and consciousness. It is increasingly prevalent at the end of life in patients with cancer. While non-pharmacological nursing interventions are essential for delirium prevention, their effectiveness in terminally ill patients with cancer remains unclear.
Objectives
This study examined the nursing support practices employed to prevent delirium in terminally ill patients with cancer in palliative care units (PCUs) in Japan.
Methods
This cross-sectional study administered an online survey to 2448 nurses from 162 institutions with PCUs in Japan.
Results
Regarding nursing practices with patients with a prognosis of months and weeks, multi-component intervention, family support, and dehydration prevention were practiced frequently, whereas bright light therapy was practiced less frequently. The specialist team approach was practiced by approximately 50% of participants. Regarding multicomponent intervention in cases with a prognosis of months and weeks, adjustments in the sleep environment and pain management were practiced by more than 90% of the participants, whereas early exercise, vision or hearing assistance, and patient education on delirium were implemented by less than 50%. The management of intravenous injection routes and catheters was implemented by participants approximately 20% more frequently in cases with a prognosis of weeks than those with a prognosis of months.
Conclusion
The primary methods of nursing support for preventing delirium in patients with terminal cancer were multicomponent intervention, family support, and dehydration prevention. Further research is necessary to develop and implement appropriate nursing support strategies.
Keywords
Introduction
Delirium, characterized by an acute and transient disturbance in attention, cognition, and consciousness, 1 is the most common neuropsychiatric syndrome in patients with advanced cancer.2,3 It occurs in 9-57% of patients seen by inpatient palliative care teams, 6-74% of those admitted to palliative care units, and 42-88% of patients before death. 4 Delirium is categorized into hyperactive, hypoactive, and mixed types, 1 with hypoactive delirium being the most frequent at the end of life (EOL).4-7
Although pharmacological therapy is the main management strategy for delirium at the end of life, a combination of non-pharmacological therapies is recommended.2,3 Non-pharmacological therapies target factors that contribute to the onset, worsening, or persistence of delirium, including physical, psychological, and environmental factors, such as cognitive, physical, and sensory support. These therapies are effective in preventing the onset of reversible delirium. 8 At the EOL, reversible delirium has been reported in 27-49% of cases.9,10 However, although the preventive effects of non-pharmacological therapies have been suggested, sufficient evidence remains lacking. 11
Nurses provide nursing support, primarily in the form of non-pharmacological therapies, to relieve terminally ill patients’ suffering. Nurses play an important role in the prevention of delirium onset.12-15 Wang et al. in their systematic review and meta-analysis identified nurse-led non-pharmacological interventions—such as multicomponent approaches including sleep, sensory support, dehydration and disorientation prevention, light therapy, and rehabilitation. 12 These interventions significantly reduced the incidence of delirium and in-hospital mortality compared to usual care. However, the practical application in terminal cancer care remains unclear. Therefore, this study explored the role of palliative care nurses in preventing delirium in patients with cancer with prognoses of months and weeks in Japan.
Methods
Study Design and Setting
This study, part of the Evidence–Practice GAP Study, is a cross-sectional anonymous survey of registered nurses aimed at clarifying nursing practices related to cancer-related symptoms16,17 in terminal patients with cancer and the caregiving burden on their families 18 (UMIN000052329). This survey targeted 389 palliative care units (PCUs) in Japan, as of September 14, 2023, within medical institutions that met the Ministry of Health, Labour, and Welfare’s standards and submitted the necessary notifications. PCUs were selected because they are the primary providers of specialized palliative care in Japan. This is because specialized home care services remain underdeveloped, although government is working to enhance their efficiency through legal, reforms and collaborative projects. 19
In a previous study 20 targeting hospice and PCU nurses, the facility consent rate was 50%, and the response rate was 80%. However, owing to the ongoing strain on medical settings from the COVID-19 pandemic, a decrease in both rates was anticipated. Therefore, all registered PCUs at the time of the survey were included as target facilities.
Participants and Study Procedure
The inclusion criteria were as follows: (1) nurses providing direct care in PCUs and (2) nurses who consented to participate in this study.
The exclusion criteria were: (1) being a manager of the PCU (e.g., a head nurse); (2) nurses from other wards who were being trained in the PCU; and (3) nurses from other hospitals who were being trained in the PCU.
Nursing administrators of all eligible study facilities were sent a research explanatory document, a consent form for facility participation, and a return envelope to obtain their agreement. Facilities that did not respond were contacted once more to confirm their willingness to participate. For consenting facilities, we first inquired about the number of nurses working in their palliative care units. Research explanation documents, corresponding to the number of nurses, were then mailed for distribution among their nursing staff. Only nurses who agreed to participate completed the online questionnaire. The data for this survey were collected from October 1, 2023, to March 31, 2024. The nurses completed an anonymous Internet survey using the LimeSurvey Cloud offered by LimeSurvey GmbH (cf. https://www.limesurvey.org/ja).
Measurements
The measurements were based on the results of a scoping review 21 and a Delphi study. 22 First, the scoping review identified nursing support for non-pharmacological therapy to prevent delirium in patients with cancer and their families. A total of 794 articles were screened, and 10 met the inclusion criteria. The thematic analysis categorized nursing support into four components: (1) multicomponent interventions23-28: after conducting delirium screening (or assessment), nurses manage distressing symptoms, prevent dehydration and constipation, adjust the sleep environment, review high-risk medications for delirium, manage urinary catheters and intravenous (IV) lines, and promote exercises to encourage daytime wakefulness, all in collaboration with a multidisciplinary team and tailored to the patient’s condition; (2) bright light therapy 29 : nurses expose patients at high risk of delirium to light with an intensity of 5000 lx in the morning for a period of 2-5 d, according to their condition, using specialized equipment; (3) specialist team approaches 30 : the treatment and care for delirium are provided through collaboration with a team of doctors and nurses specializing in geriatrics and psychiatry, providing comprehensive support and advice on the diagnosis, treatment, and care for delirium; and (4) family support31,32: Nurses educate families of patients at high risk of delirium on its symptoms and preventive measures, using verbal explanations or pamphlets. Second, we evaluated the applicability of nursing support for delirium in patients with cancer with a prognosis of months and weeks, as well as their families. Third, a preliminary survey of nine experienced PCU nurses was conducted to identify nursing support for dehydration prevention, in addition to the four categories of nursing support.
Nursing support for patients with cancer, with a prognosis of weeks or months, was assessed using a five-point Likert scale (1 = rarely; 2 = seldom; 3 = sometimes; 4 = frequently/often; 5 = very frequently/very often). The patient’s prognosis was based on the nurses’ experience, and prognosis prediction tools were not required. In this study, a prognosis of weeks was defined as 1-3 weeks, and a prognosis of months was defined as 4 weeks or more. The survey addressed each type of nursing support twice: once for a prognosis in weeks and once for a prognosis in months. For example, regarding dehydration prevention support to reduce the risk of delirium, we inquired about nurses experience in using shaved ice or ice chips for patients with difficulty in ingesting fluids. Nurses were asked to report the extent of this support for patients with both month- and week-based prognosis.
For multicomponent intervention, various types of support, such as orientation and cognitive stimulation, as well as vision and hearing correction, are assumed. Therefore, we collected data on nursing practices following Hosie et al.’s5 review (Figures 2 and 3). Among nurses who responded “sometimes,” “frequently/often,” or “very frequently/very often” to multicomponent interventions, the various types of these interventions were assessed based on whether they were practiced or not practiced in patients with cancer with monthly and weekly prognoses.
Demographic data, such as sex, age, years of nursing experience, years of PCU experience, educational background, and qualifications, were collected from the participants.
Statistical Analyses
Descriptive statistics were used to analyze the demographic information and nursing practice frequencies. EZR, a component of the R software (Saitama Medical Center, Jichi Medical University, Saitama, Japan), was used for statistical analysis. 33
Research Ethics
Approval was obtained from the Clinical Research Ethics Review Committee of the Mie University Hospital (U2023-011). All participants provided their consent to participate in the study.
Results
Participants’ Demographic Characteristics (n = 539).
Abbreviation: SD, standard deviation.
In nursing practice with patients with cancer with a prognosis of months and weeks, multi-component intervention, family support, and dehydration prevention were practiced more frequently, while bright light therapy was practiced less frequently; additionally, the specialist team approach was practiced by approximately 50% of participants (Figure 1). Nursing practice for preventing delirium in patients with cancer based on monthly and weekly prognoses (n = 539).
In the nursing practice of multicomponent interventions for a prognosis of months and weeks (Figures 2 and 3), adjustment of the sleep environment and pain management were practiced by more than 90% of participants; by contrast, early exercise, vision or hearing assistance, and patient education on delirium were implemented by less than 50% of the participants. The management of IV routes and catheters was implemented approximately 20% more frequently in cases with a prognosis of weeks than in those with a prognosis of months. The frequency of nursing practice showed a similar trend; however, the percentages of respondents who answered that they practice “very frequently/very often” were higher for Hearing assistance, Dehydration compensation, Oxygen inhalation, Patient education on delirium, and Family education on delirium than for “sometimes.” Nursing practice for a multicomponent intervention to prevent delirium in patients with cancer based on monthly prognoses. Nursing practice for a multicomponent intervention to prevent delirium in patients with cancer based on weekly prognoses.

Discussion
This study identified potential trends in the actual state of nursing support for preventing delirium in patients with cancer and with a prognosis of months or weeks, as provided by PCU nurses in Japan who agreed to participate in this study. Multicomponent intervention, family support, and dehydration prevention were implemented frequently in patients with a prognosis of months and weeks. In terms of multicomponent interventions, including dehydration prevention, we assume that these interventions can be implemented regardless of the prognosis because it is non-pharmacological, relatively non-invasive, and easy to implement.34,35 Regarding family support, family plays a significant role in the EOL care by influencing decision-making, communication styles, and prioritization of values and beliefs.36,37 We assume that education on delirium, including knowledge of delirium and how to respond to it, was provided, as was psychological support for anxiety and worry about delirium symptoms.38,39 Regarding bright light therapy and the specialist team approach, we assume that these supports depend on the resources of the facility because of the need for specialized equipment and the enrollment of specialists. 22
In multicomponent intervention, the content of EOL interventions has been reported.24,27 However, there is a lack of articles identifying which interventions are practiced by nurses in EOL care and to what extent. In this study, almost all PCU nurses implemented sleep environment adjustment and pain management as multicomponent interventions for patients with a prognosis of months and weeks. Pain is one of the most frequent symptoms at the EOL 40 and promotes delirium.2-5,41 Given that opioids used for pain management also pose a risk of delirium, assessing pain, including the treatment status, is crucial.42,43 However, in patients with dementia or other cognitive decline, if the presence, location, or extent of pain is unclear, pain should be identified through comprehensive assessment, including examination findings, facial expressions, and behavior during treatment. 44 Additionally, pain should be assessed over time rather than at a single point. Therefore, it is important to gather data from family members familiar with patient’s behavior, healthcare professionals who often interact with the patient, and nursing assistants to ensure a thorough evaluation. Adjusting the sleep environment is crucial for managing delirium, as it supports regular sleep-wake cycles vital for cognitive function and overall health, by creating a calm, comfortable environment that reduces stress and anxiety. 45 Key components include ensuring a quiet room, comfortable bedding, and stable room temperature.
Exercises, vision or hearing assistance, and education on delirium for patients with cancer with a prognosis of months or weeks were performed by less than half of the PCU nurses. In terms of exercises, they may not have been actively implementing them depending on the patient’s medical condition and status. Rehabilitation at the end of life is for the prevention of disuse syndrome as well as improved quality of life. 46 Therefore, it can be performed in bed (if it is difficult for the patient to leave the bed) or with family caregivers without any physical burden. The use of hearing aids and glasses is recommended for hearing and visual impairments, especially for patients with dementia, 44 while education on delirium prevention for patients and their families is important. 3 It is not clear whether the PCU nurses in this study had fewer opportunities to care for patients who used hearing aids or glasses, or whether they asked family members to bring these aids home beforehand to avoid the risk of damage in the event of delirium. Additionally, details regarding awareness of delirium education for patients and their families and the timing of explanations were not included in this survey; thus, we believe that further research is needed in this regard.
The management of IV routes and catheters was implemented approximately 20% more frequently for patients with a prognosis of weeks, compared with those with a prognosis of months. The frequency of delirium increases with the approach of death, heightening concerns about falls 47 and the removal of medical lines owing to agitation and excitement. Therefore, nurses must prioritize patient safety by minimizing the use of lines and tubes. To prevent falls, the lines and tubes must be organized. To reduce the likelihood of line and tube removal, they can creatively cover insertion sites with bandages, making them less noticeable to patients.
The multicomponent intervention comprises multiple components. In the Hospital Elder Life Program, recognized worldwide as a multicomponent intervention program, effectiveness in preventing delirium onset has been reported when nurses as well as volunteers, social workers, and other multidisciplinary professionals work together.48,49
Additionally, it has been reported that higher adherence rates to care protocols involving multicomponent intervention, such as stimulation of disorientation and cognitive functions, exercise, and pain management, are effective in preventing the onset of delirium. 50 Therefore, adherence to these protocols is crucial when collaborating with multidisciplinary professionals. However, depending on the resources of the facility (e.g., personnel, funds, equipment, and organization), it is difficult to implement a multicomponent intervention based on previous research.48,49 Harrison et al. focused on the workflow of nurses to address these issues and reported that a practical component was set up and implemented, resulting in a high compliance rate and suggesting a preventive effect against the onset of delirium. 51 In Japan, nurses are primarily responsible for most of the care to prevent delirium in multicomponent intervention, 23 although it is possible to get volunteers to help in PCUs. Therefore, it is necessary to select components that fit the current situation in PCUs and develop a multicomponent intervention program that is nurse-driven and compliant.
Strengths and Limitations
This study’s strengths include its comprehensive evaluation of nursing practice for the prevention of delirium in terminally ill patients with cancer with different prognoses, as well as a clarification of actual nursing practice in multicomponent interventions.
However, this study has certain limitations. First, we conducted a web-based survey to allow participants to respond via smartphones or computers, even outside working hours. However, the response rate was low, limiting the generalizability of our findings. Previous research has shown that web-only surveys typically have lower response rates and less representative samples than paper surveys when targeting the public52,53. Therefore, to increase the response rate, using paper surveys may be necessary, although the associated costs, such as postage and printing, must be carefully evaluated. Second, since this study employed a self-administered survey, it may be subject to self-report bias. For example, the cases recalled by PCU nurses may differ regarding the prognosis (of months or weeks) and the type of delirium (hyperactive, hypoactive, or mixed). Third, the survey inquired about the frequency of nursing support by prognosis; however, it did not directly ask about the reasons for non-implementation. Therefore, the factors hindering implementation must be clarified in future studies. Furthermore, even if respondents reported providing nursing support, it was unclear whether it was actually implemented, as this could not be verified. Additionally, it is unknown whether patients or their families recognized the support, despite nurses’ perceptions of its provision. Despite these limitations, this study provides valuable insights for future research and clinical practice.
Conclusions
The primary methods of nursing support aimed at preventing delirium in patients with terminal cancer were multicomponent intervention, family support, and dehydration prevention. In multicomponent interventions, adjustment of the sleep environment and pain management were practiced. Trends in nursing support were similar for patients with a prognosis of weeks and months. Further research is necessary to develop and implement appropriate nursing support strategies.
Footnotes
Authors’ Contributions
All authors (Y.K., K.N., K.K., M.K., M.M., Y.M., and J.K.) contributed to the preparation and drafting of the manuscript. Y.K. and K.N. conceived the research idea, after which discussions with other authors (K.K., M.K., M.M., Y.M., and J.K.) who contributed to the finalization of the research idea were held. All authors contributed to manuscript preparation and editing, and have read and approved the final version.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by JSPS KAKENHI [Grant Number JP21H03236]. The funders had no role in the study design, data collection and analysis, decision to publish, or manuscript preparation.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
