Introduction
Advance care planning – definition and benefits
Promoting autonomy and improving end-of-life care are core public health issues (Rao et al., Reference Rao, Anderson and Smith2002). Advance care planning (ACP) is recognized as a mechanism to attain these goals (Institute of Medicine, 2014). ACP is broadly defined as a process by which adult persons understand and share their personal values, life goals, and preferences for future medical care (Sudore et al., Reference Sudore, Lum and You2017a). While originally focused on the completion of written, legal documents such as advance directives and durable power of attorney, today, ACP is conceptualized as a social process including formal and informal communication of end-of-life treatment wishes and values (Ryan et al., Reference Ryan, M-Amen and McKeown2017).
An increasing body of knowledge shows that ACP is associated with positive outcomes for both patients and family members, such as better end-of-life quality, increased satisfaction with regard to treatment and care, reduced family members’ burden and anxiety, and reduced hospitalization (Brinkman-Stoppelenburg et al., Reference Brinkman-Stoppelenburg, Rietjens and van der Heide2014). Yet, its implementation, in general, remains a challenging and complex task (Dixon et al., Reference Dixon, Karagiannidou and Knapp2018)
Factors associated with the completion of ACP
Surveys assessing ACP among the general population have identified several individual-level factors as barriers for the completion of ACP. These included sociodemographic characteristics such as younger age, lower educational level, and belonging to a minority group (Lovell & Yates, Reference Lovell and Yates2014; van der Steen et al., Reference Van der Steen, Radbruch and Hertogh2014; Lund et al., Reference Lund, Richardson and May2015; Ng et al., Reference Ng, Kuah and Loo2017; Clark, et al., Reference Clark, Person and Gosline2018; Howard et al., Reference Howard, Day and Bernard2018), as well as negative attitudes and poor knowledge of ACP (Ng et al., Reference Ng, Kuah and Loo2017). However, these factors had only a limited contribution to the understanding of ACP completion (Khosla et al., Reference Khosla, Curl and Washington2016). Obviously, additional factors have to be explored to understand this complex behavior better. In the present study, we examined the contribution of health beliefs on Alzheimer's disease as one of these factors.
ACP and Alzheimer's disease
Alzheimer's disease is characterized by the development of progressive cognitive impairments that ultimately interfere with the performance of daily social and occupational activities. Alzheimer's disease is the most common type of dementia and dominates public perceptions (Rizzi et al., Reference Rizzi, Rosset and Roriz-Cruz2014). ACP is of special importance in the area of Alzheimer's disease, as persons with the disease will ultimately not have the capacity to express their care preferences and values. Therefore, initiating ACP is strongly recommended in the case of Alzheimer's disease, as a means for supporting autonomy and expressing care preferences and values, as well as for improving end-of-life outcomes (Cotter et al., Reference Cotter, Spriggs and Razzak2018; Dixon et al., Reference Dixon, Karagiannidou and Knapp2018). Laypersons’ beliefs on Alzheimer's disease may be one of the factors involved in attaining this goal.
Health beliefs regarding Alzheimer's disease and the completion of ACP
Perceived susceptibility (defined as the perception of the likelihood of developing a disease), worry (defined as the subjective concerns and fear about developing a disease), and subjective knowledge of a disease are central components of cognitive models for encouraging involvement in health behaviors (Conner & Norman, Reference Conner and Norman2005). As ACP could be conceptualized as health behavior (Fried et al., Reference Fried, Bullock and Iannone2009; Ernecoff et al., Reference Ernecoff, Keane and Albert2016; Ng et al., Reference Ng, Kuah and Loo2017), we hypothesized that health beliefs regarding Alzheimer's disease would significantly contribute to the understanding of laypersons’ attitudes toward ACP.
ACP in Israel
In Israel, there are 2 ways of formally undertaking ACP for end-of-life care: advance directives and power of attorney. The Ministry of Health established these forms within the framework of the “Dying Patient Law,” enacted in 2005. According to this law, a patient is defined as terminally ill if s/he is suffering from an incurable disease and has a life expectancy of up to 6 months. While attention to palliative care, in general, and ACP, in particular, has developed over the last years in Israel (Bar-Sela et al., Reference Bar-Sela, Mitnik and Zalman2017; Bentur & Sternberg, Reference Bentur and Sternberg2017; Feder et al., Reference Feder, Collett and Conley2018), still less than 1% of the adult population has formally completed advance directives or signed a durable power of attorney (Shvartzman et al., Reference Shvartzman, Reuven and Halperin2015).
Thus, the current study aimed to assess attitudes and completion of ACP and their association with sociodemographic and Alzheimer's disease health beliefs among a sample of the adult population in Israel.
Methods
Design and participants
This cross-sectional study was based on a national sample of the adult (18 years or older) non-institutionalized population in Israel. Phone calls were conducted during April and May 2017, using random digit dialing, with the adult having the most recent birthday designated as the respondent from all available respondents in each household. If that person was absent, then the next eligible adult was selected as the respondent. Of the 1,044 phone numbers called, 459 refused to participate (69 had problems understanding the topic, and the rest did not have time), rendering a response rate of 56%. Additionally, 71 participants who did not complete all parts of the questionnaire were not included in the statistical analyses. Thus, a total of 514 adults participated in the study. The sample size was estimated based on previous studies examining knowledge and attitudes regarding Alzheimer's disease (Werner et al., Reference Werner, Goldberg and Mandel2013).
Survey questionnaire and measures
The survey questionnaire was pretested on 20 potential participants. Small modifications were made in 2 of the questions for increasing clarity. The questionnaire assessed awareness, attitudes, and completion of ACP, as well as health beliefs on Alzheimer's disease and sociodemographic variables. Items assessing attitudes and the completion of ACP were asked after a short definition of advance directives and durable power of attorney was read to the participants.
Awareness of ACP
Similar to Teixeira et al. (Reference Teixeira, Hanvey and Tayler2015), participants were asked whether they had ever heard of the terms, advance directives and durable power of attorney. A negative answer was rated as 0, and a positive answer was 1. An overall index of subjective knowledge of ACP was calculated by summing both items.
Attitudes toward ACP
Were assessed using an adapted version of Nolan and Bruder's (Reference Nolan and Bruder1997) Attitudes toward Advance Directives Scale (ADAS), containing 10 items rated on a 4-point Likert-type scale, ranging from 1, strongly disagree, to 4, strongly agree. Good internal reliability scores were reported for the English version of the scale (Douglas & Brown, Reference Douglas and Brown2002; Lee & Park, Reference Lee and Park2014; Marshall et al., Reference Marshall, Avery and Weed2017). For this study, the scale was translated back and forth into Hebrew, and an overall index was calculated by summing the items. The internal reliability of the index was adequate (Cronbach's alpha = 0.71).
Completion of ACP
Similar to Teixeira et al. (Reference Teixeira, Hanvey and Tayler2015), 4 items were used to assess the completion of ACP. The first 2 assessed the formal completion of advanced directives and durable power of attorney. The other 2 assessed the informal aspects of ACP: conducting conversations on the topic with family members or friends and with the family physician. All items were rated 0 if the participant answered “no” and 1 if s/he answered “yes.” Two indices indicating the formal and informal completion of ACP were calculated by summing the items.
Health beliefs on Alzheimer's disease
Similar to previous studies (Werner et al., Reference Werner, Goldberg and Mandel2013), these included: subjective knowledge of Alzheimer's disease; perceived susceptibility, and worry about developing it; and familiarity with the disease.
Subjective knowledge of Alzheimer's disease was assessed with a single question: “How much do you know about Alzheimer's disease?” Answers were rated on a 5-point Likert-type scale, ranging from 1, not much at all, to 5, very much.
Perceived susceptibility was assessed with a single question: “How likely do you think it is that you will develop Alzheimer's disease?” Answers were rated on a 5-point Likert-type scale, ranging from 1, not at all likely, to 5, very likely.
Worry about developing Alzheimer's disease was assessed by a single question: “How much do you worry that you will develop Alzheimer's disease?” Answers were rated on a 5-point Likert-type scale, ranging from 1, not at all worried, to 5, very worried.
Familiarity was assessed by asking participants if they knew someone with Alzheimer's disease among their relatives or acquaintances.
Sociodemographic variables included gender (female or male), age, years of education, religion (Jewish or non-Jewish), religiosity (secular or religious), marital status (not married or married), perceived health status (excellent, good, fair, bad, and very bad), and perceived economic status (excellent, good, fair, bad, and very bad).
Statistical analysis
The data were coded, cleaned, and analyzed using SPSS version 22.0. Descriptive statistics were used to describe the sample and main variables. Pearson correlations were calculated to assess the relationships among the dependent and independent variables. The determinants of attitudes were assessed using ordinary least squares (OLS) hierarchical regression. We tested for multicollinearity, and the results indicated that it was not a concern in our model. The variance inflation factor (VIF) did not exceed 2.3. Only variables that were found to be significantly correlated with attitudes toward ACP were included in the regression analyses. In the first step of the regression, we included sociodemographic factors, and in the second step, we included health beliefs on Alzheimer's disease.
Ethical considerations
The study was approved by the Ethics Committee of the University of Haifa.
Results
Characteristics of the participants
As shown in Table 1, the majority of the participants were female and Jewish, with a mean age of 53 years (range 18–94) and an average education of 14 years (range 0–30). Most were married and reported their income and health status as excellent or good. Regarding Alzheimer's disease measures (Table 2), slightly over one-half of the participants reported knowing someone with the disease, and the reported levels of knowledge, susceptibility, and worry were moderate to low.
Table 1. Distribution of participants by sociodemographic characteristics (n = 514)
Awareness of ACP
Only 13.8% of the participants reported having heard the term advance directives, and 39.5% reported having heard the term durable power of attorney. Overall, 43% reported hearing at least one of the terms. The percentage of participants who reported having heard at least one of the items was higher among Jewish than non-Jewish participants (49.6% and 19.4%, respectively, X2(1) = 30.4, p < 0.001), and among those who knew a person with Alzheimer's disease than among those without such familiarity (49.1% and 36.5%, respectively, X2(1) = 8.2, p < 0.01).
Attitudes toward ACP
Overall, attitudes toward ACP were positive, with the participants scoring above the median in all the items assessing attitudes. The least positive attitude was reported for the item relating to physicians respecting the participant's concerns for end-of-life treatment decisions (Table 2).
Table 2. Percentages and means for study (n = 514)
^Reversed item
aPercentage of those who heard the term bPercentage of the entire sample
Factors associated with attitudes about ACP
The correlation matrix (Table 3) reveals that attitudes toward ACP were significantly correlated with 4 sociodemographic variables (gender, education, religion, and religiosity), with higher levels of education and lower levels of religiosity being associated with more positive attitudes toward ACP. Additionally, female and Jewish participants reported more positive attitudes toward ACP than male and non-Jewish ones. As for the associations with health beliefs on Alzheimer's disease, our results showed that higher levels of subjective knowledge of the disease, as well as higher perceptions of susceptibility, were significantly correlated with more positive attitudes.
Table 3. Correlations between attitudes and study variables (n = 514)
** p < 0.01 * p < 0.05
Table 4 presents the results of the 2-step hierarchical regression analysis. Our results showed that in the first equation depicting sociodemographic variables, only gender and religiosity were significantly and negatively associated with attitudes. In the second equation, which included health beliefs on Alzheimer's disease, we found that gender, religiosity, and subjective knowledge of Alzheimer's disease were significantly associated with the dependent variable. Although, overall, a very small share of the variance was explained by the variables examined, adding health beliefs on Alzheimer's disease increased the amount of the variance explained from 3% to 6%.
Table 4. Two-step hierarchical regression analyses for attitudes on ACP1 (n = 514)
1 Numbers in the table are standardized betas.
*p < 0.05 ** p < 0.01
Completion of ACP
Regarding ACP completion, only 11.4% (n = 8) of the participants who heard about advance directives reported completing it, and 8.9% (n = 18) of those who heard of a durable power of attorney has designated a decision maker. Regarding informal measures of ACP, 23% (n = 116) of all participants reported having conversations with family or friends about treatment preferences, and 8% (n = 40) reported having these conversations with their family physician. Attitudes toward ACP were significantly higher among participants who engaged in conversations than among those who did not.
Discussion
This study aimed to assess attitudes and completion of ACP in a national sample of the general public in Israel. Though there have been several previous studies examining ACP in national samples, they concentrated mostly on examining completion of ACP (Raijmakers et al., Reference Raijmakers, Rietjens and Kouwenhoven2013; de Vemlick et al., Reference De Vleminck, Pardon and Beernaert2014; Rao et al., Reference Rao, Anderson and Lin2014; Huang et al., Reference Huang, Neuhaus and Chiong2016; Clark et al., Reference Clark, Person and Gosline2018) and did not assess associations with health beliefs associated with Alzheimer's disease. This is an important contribution for several reasons. First, as already stated, given the characteristics of the disease, persons with Alzheimer's disease will not be able at some stage of the disease to express their values and preferences for end-of-life treatment. Second, as the number of persons with Alzheimer's disease increases worldwide (Alzheimer's Disease International, 2018), an increased number of laypersons will become exposed to persons with the disease, some of them as caregivers.
Our findings showed that approximately two-fifths of the participants were familiar with at least one of the terms examined: advance directives and durable power of attorney. This percentage is higher than the one reported in previous surveys conducted on the general population in Canada (Teixeira et al., Reference Teixeira, Hanvey and Tayler2015) and Singapore (Ng et al., Reference Ng, Kuah and Loo2017), as well as with older persons in China (Zhang et al., Reference Zhang, Ning and Zhu2015). However, it should be noted that in these survey participants were asked, in general, about ACP. Indeed, awareness rates were very similar in the 4 countries while comparing this identical item, suggesting that despite the priority given worldwide to the topic, laypersons are still quite unfamiliar with the ACP concept. This is worrisome especially as a clear understanding of the concept by the general public was defined as central to the efforts to increase ACP (Conroy et al., Reference Conroy, Fade and Fraser2009; Ke et al., Reference Ke, Huang and O'Connor2015).
Similar to previous studies (Hong et al., Reference Hong, Yi and Johnson2018), laypersons pertaining to the minority group in Israel were less aware of the concepts of advance directives and durable power of attorney. This finding stresses the need to take culturally sensitive measures to increase the exposure of minorities to the meaning and importance of these concepts. A first step might be, similar to successful efforts described in the literature (Fischer et al., Reference Fischer, Sauaia and Min2012; Sudore et al., Reference Sudore, Barnes and Le2016), to provide minority persons with accessible information in their mother tongue.
Despite the low levels of awareness found in our survey, if introduced to the definitions of advanced directives and durable power of attorney, participants reported moderately positive attitudes toward ACP. While similar findings were reported in a survey on ACP conducted among community-dwelling older adults in China (Zhang et al., Reference Zhang, Ning and Zhu2015), we believe these findings rely on the characteristics of the Israeli society. Israeli society is characterized by strong values of familism and family solidarity (Katz et al., Reference Katz, Lowenstein and Halperin2015). The fact that 6 out of the 10 items in the scale used to assess attitudes related to the impact of ACP on family members might explain the overall positive attitudes toward ACP found in our study.
Examination of the factors associated with attitudes toward ACP reveals several interesting findings. First, the variables examined in this study explained only 6% of the variance in the dependent variable. This reflects the complexity and sensitivity of the topic (Sudore et al., Reference Sudore, Schillinger and Knight2010; Lovell and Yates, Reference Lovell and Yates2014; Jimenez et al., Reference Jimenez, Tan and Virk2018) and the need to expand this line of research, qualitatively and quantitatively. Second, being female was significantly associated with more positive attitudes toward ACP. Previous studies examining gender differences in the process of ACP have also demonstrated that women reported higher levels of awareness of advanced directives (Dobbs, Park, Jang, & Meng, Reference Dobbs, Park and Jang2015) and were more likely to have an informal discussion on the topic than men (Carr & Khodyakov, Reference Carr and Khodyakov2007; Clark et al., Reference Clark, Person and Gosline2018). Women's increased involvement in caring for family members along the life course (Silverstein et al., Reference Silverstein, Gans and Yang2006) might explain these findings. Indeed, a recent study examining individual and health care factors associated with involvement in ACP in 11 developed countries found that greater informal caregiving was associated with higher involvement in ACP activities such as conversations with family and health care providers (Sable-Smith et al., Reference Sable-Smith, Arnett and Nowels2017). Alternatively, the association between female gender and attitudes on ACP might be the result of a general tendency among women to plan for the future. Indeed, a study assessing older persons’ advanced care planning showed that female gender was associated with greater planning behavior in a variety of areas (such as in the finance and socioenvironmental domains), not just in the health domain (Black et al., Reference Black, Reynolds and Osman2008).
Finally, similar to Garrido and colleagues (Reference Garrido, Idler and Leventhal2012), multivariate analyses showed that religiosity, but not religion, emerged as an important sociodemographic characteristic associated with lower levels of positive attitudes toward ACP. While religion refers to an affiliation to a specific faith group, religiosity refers to the basic spiritual and existential beliefs of a person (Beit-Hallahmi, Reference Beit-Hallahmi2015). As the basic beliefs of the main monotheistic religions (Judaism, Islam, and Christianity) toward death, dying, and end-of-life issues are very similar (Puchalski & O'Donnell, Reference Puchalski and O'Donnell2005), our findings are not surprising, although they stress the need to examine the association between these beliefs and ACP further.
However, the principal finding of this study is the prominence of subjective knowledge on Alzheimer's disease as a correlate of attitudes toward ACP beyond the contribution of sociodemographic measures. While a couple of studies have examined the associations between ACP and patients’ knowledge of their disease (see a systematic review on the topic by Ng et al., Reference Ng, Leung and Chan2015), to our knowledge, our study is the first to establish the contribution of subjective knowledge on Alzheimer's disease to attitudes toward ACP. This is surprising especially because studies conducted with professional caregivers found that patients’ knowledge on their disease plays a central role in their own end-of-life decision-making (Braun et al., Reference Braun, Ford and Beyth2010; Mousing et al., Reference Mousing, Timm and Lomborg2018). The association between knowledge of Alzheimer's disease and ACP attitudes is especially important in the area of Alzheimer's disease because of 2 reasons. First, as stated above, the disease is associated with a certainty of losing the capacity to express wishes and preferences as the disease progresses. Second, although initiating ACP as early as possible is strongly recommended in the case of Alzheimer's disease (Ryan et al., Reference Ryan, M-Amen and McKeown2017) and is even included as a priority in national dementia strategies in many countries (Fortinsky & Downs, Reference Fortinsky and Downs2014; Nakanishi et al., Reference Nakanishi, Nakashima and Shindo2015) including Israel (Brodsky, Bentur, Laron & Ben-Israel, Reference Brodsky, Bentur and Laron2013; Bentur & Sternberg, Reference Bentur and Sternberg2017), studies show that in practice it is hardly applied (see Dixon et al., Reference Dixon, Karagiannidou and Knapp2018). Our findings suggest that increasing laypersons’ knowledge of the disease will improve their attitudes toward ACP and may, ultimately, increase its practice, as we showed that persons who reported being involved in informal ACP had more positive attitudes toward the process.
The lack of associations between attitudes toward ACP and beliefs of susceptibility, worry, and familiarity was surprising. Indeed, these are important factors in cognitive models of health behavior in general (Ferrer & Klein, Reference Ferrer and Klein2015), as well as in a study assessing these beliefs in relation to ACP among persons aged 65 years and over and their caregivers (Fried et al., Reference Fried, Bullock and Iannone2009). As it has been demonstrated that the importance of risk perceptions in health behavior models increases with the salience of the threat (Ferrer & Klein, Reference Ferrer and Klein2015), the young age of our sample might explain the lack of association we found. The use of single questions to assess complex health beliefs might provide another explanation for our findings.
This study had a number of limitations. First, its cross-sectional design limits our capacity to demonstrate causal relationships. Second, we examined participants’ awareness only in relation to advanced directives and durable power of attorney and did not ask whether they have heard of ACP. Future studies should also examine terms such as “choosing life-saving treatments,” “choosing comfort measures,” or, even, “advance care planning” itself as part of ACP awareness. Third, as stated, single questions were used to assess health beliefs on Alzheimer's disease. Fourth, the question used to assess knowledge reflected perceived knowledge (i.e., an individual's subjective assessment of general knowledge) rather than their actual knowledge (i.e., the extent to which an individual can recognize general factual information on a disease). The contribution of the present study would have been expanded if additional measures of Alzheimer's disease knowledge were included. Another limitation is that the examination of knowledge was focused only on Alzheimer's disease. Future research examining knowledge of other diseases is required to determine the relative importance of knowledge of Alzheimer's disease. Finally, similar to other public surveys, we relied on the participants’ own reports, which might be affected by their honesty, their understanding of the questions, and social desirability bias. However, we hope the use of a pretested, anonymous questionnaire and the extensive experience of the interviewers minimized these biases.
Despite these limitations, our findings have substantial theoretical and practical significance. Theoretically, they contribute to the emerging conceptualization of ACP within the framework of public health and health promotion. This view calls for educating and informing the general public as early as possible, with special attention to younger persons (Seymour, Reference Seymour2018). Our study, however, found no age-related differences in laypersons’ attitudes toward ACP, suggesting that interventions aimed at increasing awareness of the topic might use similar messages to younger and older persons.
Moreover, the present study encourages the use of cognitive models of health behavior by assessing intra-personal beliefs and knowledge (Galloway, Reference Galloway2003) to understand ACP attitudes. It also suggests that expanding knowledge about Alzheimer's disease, an aim that has been defined as a priority by national strategies worldwide to decrease stigma (Alzheimer Disease International, 2018), might be important in the case of ACP as well.
Practically, our study supports recent efforts to develop and implement effective and culturally sensitive website interventions to increase laypersons’ understanding of ACP and facilitate their involvement in the process (Sudore et al., Reference Sudore, Boscardin and Feuz2017b; Jimenez et al., Reference Jimenez, Tan and Virk2018; Lum et al., Reference Lum, Barnes and Katen2018; Pereira-Salgado, et al., Reference Pereira-Salgado, Mader and O'Callaghan2018). Moreover, it suggests including a module on knowledge of Alzheimer's disease as one of the components of these interventions.
Disclosures
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Author ORCIDs
Perla Werner, 0000-0002-0432-290X
Funding
This work was supported by the Israel National Institute for Health Policy and Health Services Research (grant number: 2016/36/a).
