Israeli Nurses’ Palliative Care Knowledge, Attitudes, Behaviors, and Practices

Palliative care has a beneficial effect on survival, quality of life (QOL), and quality of death (Dahlin, 2015). It is designed to improve the QOL of patients and their families who are confronting life-threatening illness, doing so by preventing and relieving suffering through early identification, assessment, and treatment of pain and other issues, including those of a physical, psychosocial, and spiritual nature (Sepúlveda et al., 2002; World Health Organization, n.d.).

The philosophy of palliative care is relevant to the intensive care unit (ICU), and critical care organizations have encouraged its implementation. However, there has been some resistance to the philosophy and practice of palliative care because of the underlying culture of the ICU (i.e., to save life at any cost), the sense of failure when moving from curative treatments to palliative care (Mosenthal et al., 2012), and various cultural and religious attitudes toward death and dying (Loike et al., 2010). Palliative care has been closely connected to oncology practice (Saunders, 2001), so oncology nurses are likely more familiar with palliative care compared to nurses in the ICU, where the integration of palliative care is a more recent phenomenon. The objective of this study was to understand whether Israeli oncology nurses and ICU nurses perceive themselves as having similar end-of-life (EOL) palliative care knowledge, attitudes, and behaviors.

Theoretical Framework

The current study is based on Bandura’s (1977, 1982) theory of self-efficacy (Treece et al., 2006). According to this theory, if a person feels confident and capable of performing a specific behavior, called an efficacy expectation, then it is more likely that the individual will be motivated to perform that behavior. The theory of self-efficacy has been used in interventions designed to change clinician behavior (Cabana et al., 2002; Treece et al., 2006). Cabana et al. (2002) grouped efficacy expectations into knowledge, attitudes, and behaviors categories that were used to explain behavioral change (Treece et al., 2006).

Using Bandura’s theory to guide this research, knowledge, attitudes, and behaviors were applied to palliative care practices in the ICU and the oncology setting. Knowledge is defined as a person being confident that they possess the information necessary to perform palliative care practices, such as symptom control and spiritual support (Montagnini et al., 2012). Attitudes are defined as personal evaluations of palliative care practices, such as feeling comfortable discussing advance care planning with patients and families (Montagnini et al., 2012) and the process of death and dying (Downey et al., 2010). Palliative care behaviors are defined as engaging in advance care planning with patients and families, discussing code status with patients and families, and withholding life support (Montagnini et al., 2012).

Literature Review

The current authors’ assumption was that oncology nurses would report higher scores when measuring palliative care knowledge, attitudes, and behaviors because palliative care is linked to the oncology setting. However, studies have investigated the level of palliative care in the ICU and in the oncology setting and found that there is a poor delivery of palliative care in both.

Palliative care has been shown to be lacking in the ICU. Pain and other symptoms of patients in the ICU are poorly managed (Maciasz et al., 2013), and families of patients in the ICU may experience psychological consequences, such as anxiety, depression, and post-traumatic stress disorder (PTSD) (Salins et al., 2016). Wright et al. (2010) conducted a longitudinal study of patients with advanced cancer and their caregivers to determine whether place of death is associated with patients’ QOL at EOL and psychiatric disorders in caregivers. Patients’ QOL was assessed by caregiver report. Findings from the study by Wright et al. (2010) showed that patients who died in an ICU or hospital experienced more physical and emotional distress and worse QOL at EOL than patients who died at home with hospice. Caregivers of patients who died in an ICU also were at a heightened risk for developing PTSD, compared to caregivers of patients who died at home with hospice; for caregivers, hospital deaths also were associated with an increased risk of prolonged grief disorder, compared to home hospice deaths.

Very little research has been done about palliative care knowledge or attitudes among oncology nurses. It may be that there is a blurring of roles or role overlap (Pavlish & Ceronsky, 2007) between palliative care nurses and oncology nurses. Although the oncology nursing literature has many studies concerned with increasing knowledge about palliative care, few have assessed oncology nurse attitudes or behaviors. Palliative care nurses do not always identify themselves as oncology nurses. In addition, palliative care is built into the main competencies for oncology nursing (Oncology Nursing Society, 2016), so it may be that few have specifically examined this aspect of oncology nursing because it is viewed as a part of the oncology nurse’s job. For example, in South Africa, Competencies for Oncology and Palliative Nursing addresses both topics (South African Nursing Council, n.d.). Although palliative care is viewed as an aspect of oncology nursing by oncology nurses, not all palliative care nurses come from an oncology background (Pavlish & Ceronsky, 2007).

Two studies were identified that focused on palliative care delivered by oncology nurses. In the first study, Pavlish and Ceronsky (2007) used focus groups to examine 33 oncology nurses’ perceptions about palliative care, finding that there was agreement on nurses’ descriptions about the nature of palliative care but divergence related to how palliative care intersects with other areas of practice. In addition, Pavlish and Ceronsky (2007) observed that differing perspectives about the intersections of palliative care with oncology nursing, such as the belief that oncology nurses were providing palliative care services themselves, may inhibit timely consultation with specialty palliative care teams.

In the second study, Harden et al. (2017) aimed to improve oncology nurses’ palliative care knowledge, attitudes, and behaviors using a palliative care nursing education program. The Scale of EOL Care in the ICU (EOLC-ICU), created by Montagnini et al. (2012), was employed; results showed that there was a statistically significant increase in knowledge, attitudes, and behaviors after the educational intervention.

In Israel, the concept of palliative care is most often linked to EOL care (Coffey et al., 2016). The Israel Cancer Association is involved in implementing and upgrading palliative care services in Israel to include a more holistic approach (Kislev et al., 2013). Palliative care education for Israeli oncology nurses started in the early 1980s and included the establishment of a professional organization (Ami & Yaffe, 2015). However, despite advances in nursing education about palliative care and the provision of palliative care services by nurses, Coffey et al. (2016) reported that palliative care has not changed in Israel; it is significantly less developed than palliative care in other countries, like the United States, and is still focused on EOL care. A study of 105 Israeli oncology and palliative care nurses by Feder et al. (2018) found that 42% of the sample reported that their workplace provided little or no palliative care education or resources to nurses.

Israel has a unique history related to palliative care because its legal system is based on Jewish religious and secular law. In 2005, Israel’s Knesset (the legislative branch of the Israeli government) passed the Dying Patient Law, which determines who is considered a patient with a terminal disease and under what conditions palliative medical care is required (Ministry of Health, 2005). The law includes clear definitions of the roles of nurse, physician, patient, and family member in deciding on the type of care provided, including decision making regarding goals of care. In 2009, the Israeli Ministry of Health (2009b) published an executive order mandating that all inpatient and outpatient healthcare institutions provide palliative care services in all settings. However, only recently have institutions begun to comply with this order. In addition, in an attempt to ensure that the executive order was feasible, the Ministry of Health (2009a) legalized the first advanced nursing practice role in the area of palliative care.

Globally, implementing palliative care into the ICU has been evolving to include additional training of ICU physicians and nurses, a focus on patient need for palliative care services rather than outcome, a move away from the idea that palliative care means EOL care, and more interest in developing expertise in palliative care even in the ICU (Mathews & Nelson, 2017). ICU nurses who are trained to save a life at any cost are now being challenged by the idea that if patients cannot be cured, they need supportive care, even in the ICU (Katz, 2018). About 10%–20% of all patients admitted to the ICU die in the ICU (Teno et al., 2013). Most of these deaths do not come as a surprise to patients, families, or caregivers, and ICU clinicians have learned to change the goal of care from cure to comfort care (Ely et al., 2019). Therefore, implementation of palliative care into the ICU is of utmost importance in Israel, and there is an international movement to recognize that palliative care is appropriate to the ICU.

Methods

This comparative cross-sectional study involved completion of an online survey by 126 Israeli nurses (79 oncology nurses and 47 ICU nurses) who belonged to a professional nursing organization in Israel, either the Israel Association of Cardiology and Critical Care Nurses or the Israeli Oncology Nurses Organization. An invitation to participate in the study that contained a link to the online survey was sent to members by the two professional nursing organizations. Inclusion criteria were willingness to complete the survey and membership in one of the two professional nursing organizations. Data were collected after institutional approval by the Hadassah Medical Center Helsinki Ethics Committee in Jerusalem, Israel. Responses were returned to the authors without any identifying information. Consent was implied with return of the survey. The data were collected in a file compatible with IBM SPSS Statistics, then transferred into an IBM SPSS Statistics, version 25.0, file.

Instruments

A personal and work characteristics questionnaire included the following:

• Age 

•  Gender

•  Marital status

•  Religion, religiosity, and national origin (three measures of ethnicity in Israel)

•  Number of years as an ICU nurse or an oncology nurse

•  Number of years at current institution and in current unit

•  Current role

•  Whether additional education following a basic certification course had been completed and in what area

•  Type of nursing education

•  Previous exposure to palliative care (personal experience with family member or friend receiving hospice or palliative care, professional education in palliative care, professional experience using palliative care)

The 28-item EOLC-ICU is based on that created by Montagnini et al. (2012) to assess how prepared and comfortable nurses feel about their ability to provide palliative care, as measured by the categories of knowledge, attitudes, and behaviors. The EOLC-ICU employs a five-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). The scale was adapted for use with oncology nurses by replacing the words “intensive care” with “oncology.” Montagnini et al. (2012) reported adequate reliability of the EOLC-ICU and evaluated its validity with content validity. For the current study, Cronbach alphas were 0.89, 0.66, 0.88, and 0.89 for knowledge, attitudes, behaviors, and total score, respectively. The EOLC-ICU was forward-translated into Hebrew by native Israeli speakers and back-translated by native English speakers using the Brislin (1970) method.

Quality of death and dying (DeCato et al., 2013) was evaluated with a single item: “Rate the care patients on your unit received in the last several days of their lives while in the ICU from all of the doctors and other healthcare providers combined.” Participants could rate the item on a scale ranging from 0 (worst health care possible) to 10 (best health care possible). This item was found to have construct validity correlating with other measures of quality of EOL care (Curtis et al., 2002; Glavan et al., 2008).

Data Analysis

Descriptive statistics, including frequencies, means and standard deviations, and medians, were calculated for all items on the questionnaire as well as for the categories of knowledge, attitudes, and behaviors. Responses of ICU nurses and oncology nurses were compared based on the level of measurement of the dependent variable, with the chi-square test used for nominal data, the Mann–Whitney U test used for ordinal data, and independent t tests used for interval-level data.

Results

Most participants were female (n = 115), married (n = 99), and Jewish (n = 114) (see Table 1). No statistically significant differences were found among the demographic variables except for gender, where the ICU nurse sample had five male nurses and the oncology nurse sample had one male nurse (chi-square[1] = 5.69, p = 0.017). There were more Jewish nurses overall, compared to two Muslim and two Christian nurses. Level of religiosity was mostly secular (n = 74), followed by traditional (n = 24) and Orthodox (n = 17); nine participants preferred not to answer. A statistically significant difference was observed in participants’ number of years of experience in the oncology setting compared to the ICU (oncology: mean = 12 years, SD = 8; ICU: mean = 18 years, SD = 10; t[3.24] = 107; p = 0.002).

Most oncology nurses reported having taken courses in palliative care previously (n = 44) compared to ICU nurses (n = 8) (c2[1] = 18.87, p = 0.0001) (see Table 2). In addition, most oncology nurses (n = 46) had known someone who had received palliative care compared to ICU nurses (n = 11) (chi-square[1] = 14.56, p = 0.0001). No other work-related variables had a statistically significant difference among the oncology nurses and the ICU nurses.

Mean item scores for knowledge, attitudes, and behaviors related to palliative care delivery, as well as those for quality of death and dying and comfort discussing code status, are listed in Table 3. Higher scores indicate greater perceived competency regarding EOL care. Mean total scores for knowledge were 45.1 (SD = 7.9) and 41.7 (SD = 10.6) out of a possible 60 for the oncology nurses and the ICU nurses, respectively; these scores were not statistically significant.

Secondary analyses between knowledge scores and other variables were computed. There was a statistically significant difference in the level of self-perceived knowledge of those who had taken a palliative care course compared to those who had not (course: mean = 4, SD = 0.6; no course: mean = 3.58, SD = 0.79; t[117] = 3.134; p = 0.002). The difference in knowledge scores between those who had taken a palliative care course versus those who had not was found only among oncology nurses (course: mean = 4, SD = 0.62; no course: mean = 3.6, SD = 0.59; t[74] = 2.237; p = 0.028). In addition, there was a significant difference in knowledge levels between those who had had a personal experience with palliative care and those who had not (personal experience: mean = 3.9, SD = 0.68; no personal experience: mean = 3.6, SD = 0.77; t[119] = 2.004; p = 0.047).

The mean total attitude score for oncology nurses was 19.6 (SD = 4.3) and 19.1 (SD = 5) for ICU nurses out of a possible 25; these scores did not demonstrate a statistically significant difference. Nurses who had had a personal experience with palliative care had statistically significant more positive attitudes than nurses who had not (personal experience: mean = 4.1, SD = 0.77; no personal experience: mean = 3.8, SD = 0.8; t[119] = 2.073; p = 0.04). A statistically significant difference was observed among oncology nurses who had taken a palliative care course versus those who had not (course: mean = 4.2, SD = 0.71; no course: mean = 3.9, SD = 0.84; t[74] = 2.01; p = 0.048).

The mean total behavior score was 26.8 (SD = 13) for oncology nurses and 27.2 (SD = 9.7) for ICU nurses out of a possible 55; these scores did not demonstrate a statistically significant difference. In addition, there was no statistically significant difference related to higher scores for behavior between nurses who had had a personal palliative care experience and those who had not. However, a statistically significant difference was observed in behavior scores among nurses who had taken a palliative care course compared to those who had not (course: mean = 3.7, SD = 1.33; no course: mean = 3, SD = 0.88; t[84.528] = 3.182; p = 0.002).

Oncology nurses scored higher than ICU nurses on the following attitude items:

•  Feeling comfortable discussing code status (oncology nurses: mean = 3.88, SD = 1.3; ICU nurses: mean = 3.22, SD = 1.53; t[2.56] = 124; p = 0.016)

•  Behaviors such as consulting palliative care experts (oncology nurses: mean = 3.39, SD = 1.6; ICU nurses: mean = 2.3, SD = 1.38; t[3.75] = 124; p = 0.000)

•  Notifying patients about a change in clinician (oncology nurses: mean = 3.55, SD = 1.8; ICU nurses: mean = 2.74, SD = 1.6; t[2.54] = 124; p = 0.012)

The mean quality of death and dying score for oncology nurses was 7.66 (SD = 1.8) and 6.85 (SD = 2.1) for ICU nurses out of a possible 10, with higher scores indicating better perceptions of health care. This difference was statistically significant (t[2.27] = 120, p = 0.025). There was a statistically significant correlation between a nurse’s personal experience with palliative care and patient quality of death and dying (r = –0.19, p < 0.05).

Discussion

The oncology nurses and ICU nurses in this study showed moderate levels of self-perceived knowledge and attitudes toward palliative care. However, the level of their self-reported behaviors was low. Although oncology nurses tended to score slightly higher than ICU nurses in terms of self-reported knowledge, attitudes, and behaviors, these differences were not found to be significant.

Studies have examined how palliative care nursing practice has been affected by confidence and knowledge (Harden et al., 2017), personal fears and acceptance of death (Braun et al., 2010), level of religiousness (Braun et al., 2010), preparedness through education (Loike et al., 2010), and nurses’ autonomy (Razban et al., 2015). Multiple studies have looked at nurses’ attitudes, beliefs, and confidence levels to understand how nurses respond in clinical settings (Harden et al., 2017; Hou et al., 2013). A study by Braun et al. (2010) focused on 147 Israeli oncology nurses’ attitudes toward death and caring for dying patients. Results showed that attitudes of acceptance of death were significantly associated with the degree of nurses’ religiosity; attitude is closely tied to a religious worldview (Braun et al., 2010).

The knowledge, attitudes, and behaviors findings from the current study were lower than those from a study by Montagnini et al. (2012) with a U.S. sample. The current authors do not know why there is such a disparity in findings and hypothesize that cultural differences between U.S. and Israeli nurses may play a role. According to a study by Feder et al. (2018), Israeli oncology and palliative care nurses perceived a lack of palliative education and resources for nurses. A study by Razban et al. (2015) showed that Iranian nurses from a variety of units, including critical care and oncology, scored low on knowledge about palliative care.

Practitioners from the Middle East have tried to approach the issue of palliative care through the Middle East Cancer Consortium. Each year, conferences are offered related to palliative care and ways to improve care. Noting the desire for palliative care training in the Middle East, Silbermann et al. (2015) reported that in 15 Middle Eastern countries, including Israel, 86% of healthcare professionals responded that they wished to learn more about palliative care, and 90% believed that their clinical staff would like to learn more about the topic. Additional research is needed regarding how to make this happen.

A study by Hou et al. (2013) involving 311 RNs at a university hospital in China reported that the ICU nurses (n = 55) had higher total scores related to palliative care knowledge and attitudes than the oncology nurses (n = 40). However, results from the current study revealed that oncology nurses had higher scores than ICU nurses; this was not statistically significant. Culture as well as characteristics of the healthcare system might be factors associated with these results.

Bandura’s theory of self-efficacy was used to guide this research, and it appears that the perceived knowledge and behaviors scores of the oncology nurses and the ICU nurses in this sample point to a need to improve the knowledge of both groups. Although the Israeli Oncology Nurses Organization and the Israel Association of Cardiology and Critical Care Nurses provide palliative care education to their members, results of this study indicate that more education is warranted.

On a positive note, Israeli ICU nurses’ levels of knowledge, attitudes, and behaviors are similar to those of the Israeli oncology nurses. Regarding the differences between oncology nurses and ICU nurses, it appears that the oncology nurses in this study had more personal experiences with palliative care involving loved ones and had taken courses on palliative care. It might be that people who took courses on palliative care were motivated after having had a personal experience with death. This is consistent with Bandura’s (1982) theory, which suggests that personal experience may be a motivating factor for self-efficacy. If a person believes that their experience has helped them, they may have more self-confidence (Bandura, 1982). Oncology nurses who had more education on and experience with palliative care had higher knowledge and attitudes scores compared to the ICU nurses, but the ICU nurses scored higher on behaviors than the oncology nurses; however, this was not statistically significant. Future research should explore this finding. According to Bandura’s theory, experience should indicate higher self-efficacy, but the results of this study did not support that.

Palliative care principles are embedded into nursing education (Kazanowski & Sheldon, 2014); nurses are taught to listen to and advocate for patients, promote comfort, and alleviate suffering (Dahlin, 2015). Although all nurses should have palliative care training, the current authors assumed that palliative care practices would be better among oncology nurses. However, this was not observed in the current study. Research exploring the basis of how nurses view their own palliative care behaviors may provide insight into how to motivate the use of palliative care in practice, even in countries still trying to influence change. The findings of this study add to the body of knowledge regarding palliative care and nursing care in the Middle East, including Israel, and illustrate that worldwide cultural issues still exist and may be barriers to adequate palliative care, as compared to Western countries (Bentur et al., 2012, Silbermann et al., 2015). In addition, with many nurses seeking global health experiences, an awareness of palliative care delivery in different parts of the world must occur.

Limitations

There were several limitations to this study. This online study had a poor response rate; therefore, there may be a bias among those who responded, with the study populations not being well represented. It is possible that reported scores are higher in the current study than what they would be for the study population otherwise; study participants may be more interested in or believe more in palliative care than those who did not participate. Regarding religion and religiousness, there was little variability in the data because most participants were Jewish and either secular or traditional.

Implications for Nursing

Based on results of the current study, oncology nurses and ICU nurses in Israel are in need of advanced palliative care education. Policies need to be amended to encourage increased palliative care behaviors across all units. Additional research should be conducted related to how to better introduce and implement palliative care in the Middle East. It is interesting that oncology nurses and ICU nurses reported diverse experiences, with oncology nurses having had more personal experience with palliative care and even having taken more courses. Little exists in the literature that compares characteristics of nurses who gravitate toward a particular specialty; however, the current study shows that even in a very small country, there are differences between subspecialties of nursing; personal experience may also play a role in individuals’ competencies. In addition, although oncology nurses had higher knowledge and attitudes scores, this did not reflect in their behaviors.

Conclusion

The results of this study point to a greater relationship between nurses’ personal experience and palliative care self-efficacy rather than self-efficacy driven by knowledge. In this instance, attitudes preclude behaviors. Nursing education for building competence about palliative care may be rooted in strengthening personal experience and providing experiential-type education. It is important to remember that although oncology nurses reported more education and experiences in palliative care in the current study, they did not demonstrate more palliative care behaviors.

The authors gratefully acknowledge the work of summer research intern Emma Finn.

About the Author(s)

Caryn Scheinberg-Andrews, PhD, FNP-BC, is a senior faculty member in the Henrietta Szold Hadassah School of Nursing at Hebrew University in Ein Kerem, Jerusalem, Israel, and an associate professor of practice in the School of Health Science at Simmons University in Boston, MA; and Freda DeKeyser Ganz, PhD, RN, is dean in the Faculty of Life and Health Sciences at Jerusalem College of Technology and a professor emeritus in the Henrietta Szold Hadassah School of Nursing at Hebrew University. No financial relationships to disclose. Both authors contributed to the conceptualization and design, completed the data collection, provided analysis and statistical support, and contributed to the manuscript preparation. Scheinberg-Andrews can be reached at candrews@hadassah.org.il, with copy to ONFEditor@ons.org. (Submitted July 2019. Accepted October 16, 2019.)