Alzheimer's disease was the sixth leading cause of death for older Americans and accounted for one third of all deaths among older adults in 2013 (Alzheimer's Association, 2014). For individuals with advanced Alzheimer's disease or other types of dementia, it is their designated health care proxies (most often family members) who are responsible for communicating written or verbal preferences, as well as making important end-of-life (EOL) care decisions on their behalf (e.g., enrollment into hospice programs, initiation or withdrawal of life support in consultation with health care providers) (Givens, Kiely, Carey, & Mitchell, 2009; Mitchell et al., 2012; Silveira, Kim, & Langa, 2010).
Extant past studies reported that proxies, while confronted with the need to make complex EOL decisions, were uncertain about their roles and responsibilities, uninformed about treatment options, and unsupported by professionals in their decision making (Gessert, Forbes, & Bern-Klug, 2000; Swigart, Lidz, Butterworth, & Arnold, 1996). Moreover, proxies reported generalized stress; conflicts with preferences of the patient, other family members, and health care providers; and experienced delay in decision making (Breen, Abernethy, Abbott, & Tulsky, 2001; Kramer, Boelk, & Auer, 2006). A more recent study of nursing home residents with advanced dementia (in Boston) reported that health care proxies reported high levels of overall decision satisfaction in care decisions made, with a mean total satisfaction score of 78 of 100 (Givens et al., 2009). However, their level of satisfaction with the process of decision making was significantly lower than satisfaction with the decisions because of dissatisfaction with the level of perceived involvement by nursing home primary care providers. Other studies have shown that families often believe that support from health care providers is inadequate and experience conflict with providers (Biola et al., 2007; Fassier & Azoulay, 2010; Wendler & Rid, 2011).
In efforts to assist proxies in decision making, advance care planning (ACP) has been used and received the most attention (Ditto & Hawkins, 2005; Kwak, Allen, & Haley, 2011). ACP is defined as a process of communication about the decisions of current and future medical care in the event that an individual with dementia cannot make decisions on his/her own. During the ACP process, individuals are encouraged to complete advance directives. The most common form of advance directive is a living will (i.e., a written document that specifies preferences for different life support treatments). Individuals are also encouraged to designate a health care proxy (e.g., durable power of attorney for health care) to make medical decisions on their behalf. For individuals with dementia, advance directive completion rates range between 60% and 70% in community and nursing home care settings (Lingler et al., 2008; Pasman et al., 2004).
Review of the literature on ACP and proxy decision making suggests that evidence for effectiveness of ACP on improving proxies' knowledge or directing care consistent with patient preferences is yet to be established (Ditto & Hawkins, 2005; White & Arnold, 2011). For example, Shalowitz, Garrett-Mayer, and Wendler (2006) reported that in approximately one third of cases included in their review of 16 studies (a majority of which focused on patient and proxy samples of individuals without dementia), proxies inaccurately predicted patient preferences. Concerning proxy decision making for individuals with dementia, even fewer quantitative studies systematically examine the role of ACP. A review by Dening, Jones, and Sampson (2011) on ACP among individuals with dementia identified only three studies (one qualitative and two small mixed-methods studies [Forbes, Bern-Klug, & Gessert, 2000; Hirschman, Kapo, & Karlawish, 2006, 2008]) that specifically examined the relationship between ACP and proxy decision making by families. These studies explored multiple factors considered by proxies in making health care decisions, but not necessarily the role of ACP. Although a randomized trial of a professionally facilitated ACP intervention reported increased chances of patients' preferences being followed and reduced psychological distress among family members (Detering, Hancock, Reade, & Silvester, 2010), the intervention focused on discussion between cognitively competent patients at the terminal stage of the illness and their families.
Advance Care Planning and Decisional Conflicts Among Proxies
ACP may be effective in reducing decisional conflicts and thereby effective in helping families make informed decisions. However, more systematic, theoretically guided research is needed to examine the unique effect of ACP on the experience of proxy decision making. The current study aimed to examine the role of ACP in predicting proxies' decisional conflicts about EOL choices for individuals with dementia. The theoretical framework guiding the study is the Ottawa Decision Support Framework (ODSF), which is based on concepts from psychology, decision analysis and decisional conflict studies, social support, and economic concepts of expectations and values (O'Connor & Jacobsen, 2007). The ODSF is a broad framework that identifies modifiable factors that can be targeted for specific decision–support interventions to reduce decisional conflicts (O'Connor & Jacobsen, 2007). Decisional conflict is a concept broadly defined as uncertainty or difficulty in identifying the best alternative (Janis & Mann, 1977) and is manifested by verbalized uncertainty about choices, vacillation between choices, verbalized distress or physical signs of distress, and delayed decision making (O'Connor & Jacobsen, 2007). Decisional conflict has been a key area of assessment in evaluating the quality of a decision about care (i.e., whether the decision is informed and consistent with values).
According to the ODSF, modifiable factors are referred to as decisional needs and include knowledge, value, and support, which are hypothesized to contribute to decisional conflict. In the context of proxy decision making, knowledge domain refers to having knowledge about risks and benefits of treatment options available to individuals with dementia. Value domain refers to the proxy's understanding of values considered important by the individual with dementia. Support is defined as the perceived support from family and health care providers for proxies in decision making for individuals with dementia. The more uncertain the proxy feels about these three domains, the more uncertainty and decisional conflict he/she is expected to experience. These three decisional needs areas may be targeted for specific decision–support interventions (e.g., deficits in knowledge can be addressed with education and unclear values with focused counseling to help clarify and prioritize goals).
The main hypothesis of the current study was that if individuals with dementia completed advance directives and proxies understood their preferences for EOL treatments, proxies would (a) then be more informed about their values, (b) expect to receive more support from families and health care providers in EOL decision making, and (c) report less uncertainty in decision making. Based on the ODSF, it was expected that the more knowledgeable the proxy was about dementia progression, the more informed and clear the proxy would be about EOL decisions. It was further hypothesized that the relationship quality between the proxy and other family members and physicians would influence how supported the proxy felt about receiving support in making EOL decisions. To test these hypotheses, a cross-sectional survey was conducted with a sample of designated health care proxies of individuals with cognitive impairment, Alzheimer's disease, or other types of dementia.
Participants and Procedures
The study sample consisted of spouse/partner and adult-child caregivers of individuals with cognitive problems or a diagnosis of Alzheimer's disease or related dementia. Caregivers were recruited from the League of Experienced Family Caregivers (LEFC). The LEFC is a registry of family caregivers who volunteer to participate in surveys about caregiving experiences conducted by the University of Wisconsin–Milwaukee. At the time of the current study, the LEFC registry consisted of 393 caregivers. Of these, 39 could not be reached. Research assistants contacted and explained the purpose and details of the study to 354 caregivers. Of these, four refused to complete the survey and 87 were ineligible to participate because they were no longer caregiving (n = 12), their care receiver was deceased (n = 70), or for some other reason (n = 5). Once agreed, 263 caregivers participated in the study by completing a mailed (89.4%), online (9.9%), or telephone (0.7%) survey (based on their preference). The final study sample consisted of 141 designated health care proxies caring for a spouse/partner or parent with cognitive impairment. The study was approved by the University of Wisconsin–Milwaukee Institutional Review Board.
Demographic Characteristics. Proxies answered questions about their relatives' demographic information, living arrangement, cognitive and functional status, and health status (i.e., number of falls, pneumonia incidences, and hospitalizations in the past 6 months). Proxies also reported their relationship to their relative (i.e., spouse/partner or adult-child) and their own demographic information.
Advance Care Planning. Proxies reported if their relative had (a) completed a living will or durable power of attorney for health care, (b) shared a copy of the completed advance directive with a physician, and (c) communicated to the proxy his/her preferences regarding three life support treatments, including cardiopulmonary resuscitation (CPR), a ventilator, and a feeding tube.
Knowledge About Dementia Trajectory. Proxies were asked seven questions about the trajectory of advanced dementia, of which four came from the Alzheimer's Disease Knowledge Scale (Carpenter, Balsis, Otilingam, Hanson, & Gatz, 2009). Proxy knowledge scores were calculated by summing the number of correct responses to the seven knowledge items and dividing by the total number of items, with the score ranging from 0 to 1. Example items included: “Eventually, an individual with Alzheimer's disease will need 24-hour supervision”; “Alzheimer's disease cannot be cured”; and “An individual with Alzheimer's disease becomes increasingly likely to fall as the disease gets worse.”
Family Support and Relationship. To assess the proxy's perceived level of support and relationship quality with other family members, 14 items were used from the Family Assessment Device Scale (Epstein, Baldwin, & Bishop, 1983), which is a 6-point Likert-type scale with response options ranging from 0 (strongly agree) to 5 (strongly disagree). A higher score indicated higher support and better relationship quality in the proxy's family.
Trust of Physicians. Eleven items from the Trust in Physician Scale, developed by Anderson and Dedrick (1990), assessed proxies' trust of the physicians of the individual with dementia. Response options ranged from 0 (strongly agree) to 5 (strongly disagree). A higher score indicated a higher level of trust of physicians.
Proxies' End-of-Life Care Choices. Proxies were first presented with a hypothetical situation that was characterized by conditions likely to occur at the end stage of dementia. They were then asked to choose a goal of care and a set of treatment options if their relative were in the hypothetical situation. The initial scenario, EOL care goals, and treatment choices were adapted from Volandes et al. (2009), pilot tested with two caregivers, and reviewed by a neurologist and palliative care specialist for applicability for individuals with dementia. The EOL scenario included common end-stage dementia symptoms, such as inability to move around independently, recognize a close family member, answer questions, swallow, and feed oneself, and indicated that the probable life expectancy was less than 1 year. Three options for goal of care were: (a) life-prolonging care (i.e., emphasis on extending life), (b) limited care (i.e., focus on maintaining current physical and mental functions, but not necessarily extending life), and (c) comfort-only care (i.e., focus on maximizing comfort and controlling pain, but not on improving functioning or extending life). In addition, proxies were asked to choose among three sets of treatment options: (a) perform everything available, such as life support treatments and any other procedures available; (b) perform everything except for CPR or procedures used in the intensive care unit (ICU); and (c) perform only procedures for pain and symptom control, excluding hospitalization, CPR, feeding tubes, ventilators, or other ICU procedures.
Decisional Conflict. Proxies' decisional conflict was measured by the Decisional Conflict Scale (DCS; O'Connor, 1995), a 5-point Likert-type scale with response options ranging from 0 (strongly agree) to 4 (strongly disagree). The DCS measures uncertainty in making a decision and evaluates factors that contribute to uncertainty, such as feeling uninformed, unclear about personal values, and unsupported in making a decision. Twelve items from four subscales (i.e., informed, value clarity, support, and uncertainty) of the original DCS were used to measure decisional conflict. To obtain a score for each subscale, items were summed and averaged. Higher scores indicated greater uncertainty.
The informed subscale measured whether the proxy was informed about risks and benefits of given treatment options (3 items, alpha = 0.95). The value clarity subscale measured if the proxy had a clear understanding of values regarding the treatment choice, such as which benefits or risks mattered most to the individual with dementia and which was more important to him/her (3 items, alpha = 0.93). The support subscale measured if the proxy expected to receive support from family members and health care providers when making EOL decisions for the individual with dementia (3 items, alpha = 0.90). The uncertainty subscale measured the overall level of uncertainty about the treatment choice that the proxy made for the individual with dementia in the given EOL scenario (3 items, alpha = 0.91).
Descriptive statistics were used to characterize sociodemographic, functional, and ACP characteristics of individuals with dementia and proxies, as well as proxies' choices for EOL care goals and treatments for their relative in the hypothetical scenario, and scores on decisional conflict subscales. A series of multiple regression analyses was conducted to examine the association between ACP, proxies' knowledge of Alzheimer's disease or other dementia, family conflicts, trust of physicians, and the four domains of decisional conflict about EOL decision making. In addition to these key variables hypothesized to be key predictors, several demographic and background characteristics of individuals with dementia and proxies were also tested for their correlation with decisional conflict, including the functional and clinical status of individuals with dementia and relationship between the individual with dementia and proxy. Only predictor variables achieving statistical (p < 0.05) and clinical (r > 0.2) significance with decisional conflict subscales were entered in multiple regression analysis. All analyses were performed with IBM SPSS 21.0.
Participant Background Characteristics
Background and ACP characteristics of individuals with dementia are described in Table 1. The mean age of the individuals with dementia was 76.2 (SD = 11.8 years). The majority of individuals with dementia were female (53%), lived with the proxy (60%), designated their spouse/partner as their proxies (58%), and had Alzheimer's or dementia diagnosed (54%), and 46% had suspected or diagnosed memory problems. Approximately all proxies were White (97%) and most were female (72%). The average Alzheimer's/dementia knowledge score was 0.44 (SD = 0.19). Mean scores on the Family Assessment Device Scale and Trust in Physician scales were 42.8 (SD = 17.2) and 37.89 (SD = 10.98), respectively.
Demographics and Characteristics of Individuals With Dementia and Proxies (N = 141)
Advance Care Planning and Life Support Preferences of Individuals With Dementia and Proxies' End-of-Life Care Choices
Table 2 displays the ACP and preferences for life support of the individuals with dementia, and proxies' goal of care, treatment options, and DCS subscale scores. The majority of individuals with dementia (88%) completed a living will and shared a copy with the proxy. However, only 38% shared the copy with a physician. Most proxies (78%) indicated that they knew their relative's preference for life support. The majority of proxies reported that the individual with dementia did not want life support to be provided (54% did not want CPR, 63% did not want a feeding tube, and 69% did not want a ventilator). More than 20% of proxies reported not knowing the preference for at least one of the three life support options (22%, 26%, and 23% did not know CPR, feeding tube, and ventilator preferences, respectively).
Advance Care Planning for Individuals With Dementia and Proxies' End-of-Life Decision Making (N = 141)
Sixty-eight percent of proxies chose the comfort-only care goal, followed by 26% of proxies who chose limited care. More than 70% chose the pain and symptom control-only option and 15% chose to perform everything but CPR and ICU treatments. Only 5% chose the life-prolonging goal of care and 6% chose the option to perform everything available.
The Role of Advance Care Planning on Proxies' Decisional Conflict
Summary of the final multiple regression analysis results are presented in Table 3 (higher scores on each subscale represent higher levels of uncertainty in the given area). Pearson's correlation coefficients for the following individual with dementia and proxy background variables did not reach statistical or clinical significance and thus were not included in the regression models: functional or clinical characteristics of the individuals with dementia (e.g., activities of daily living, confirmed diagnosis of dementia, number of falls, pneumonia incidences, or hospitalizations in the past 6 months), proxy's relationship to the individual with dementia, and number of years caregiving.
Summary of Regression Analyses for Variables Predicting Decisional Conflict Subscale Scores (N = 141)
Because the majority of individuals with dementia completed advance directives, the association between completing advance directives and decisional conflict could not be tested. Instead, whether the proxy's knowledge of preference for EOL treatments could be associated with subscale scores was examined. Contrary to the authors' expectation, it was found that proxies who did not know the preferences for any of the life support treatments (compared to those who knew preferences for some or all of the treatments) were significantly likely to report greater clarity in their understanding of the values (beta = −0.19, p < 0.05). At the same time, if the individual with dementia did not want any of the life support treatments, proxies were likely to anticipate less support (beta = 0.23, p < 0.01) and report greater uncertainty (beta = 0.21, p < 0.01). However, in a separate analysis (data not shown), the preference to have all three treatments was not associated with any domains of decisional conflict.
Concerning the predictors of proxy knowledge, family support, and trust of physician, findings were as expected. Higher scores on the knowledge of Alzheimer's disease/dementia trajectory were associated with being more informed (beta = −0.21, p < 0.05), clearer in understanding values (beta = −0.18, p < 0.05), and less uncertain (beta = −0.17, p < 0.05). Family support (beta = −0.34, p < 0.001) and trust of physicians (beta = −0.37, p < 0.001) were also significant predictors of support subscale scores. Higher family support and trust of physician scores were associated with higher levels of support that proxies anticipated receiving in EOL decision making. The trust of physicians was also associated with informed (beta = −0.46, p < 0.001), value clarity (beta = −0.51, p < 0.001), and uncertainty (beta = −0.53, p < 0.001) subscale scores. The more the proxy trusted the physician, the more informed about values and less uncertain proxies were about their decisions.
For a growing number of individuals with dementia, family members are responsible for communicating their preferences and making important EOL care decisions in consultation with health care providers (Givens et al., 2009; Mitchell et al., 2012). The current study aimed to systematically examine the role of ACP, proxy knowledge, and family and health care provider support in proxy decision making of EOL care for individuals with dementia. Four sets of findings provide implications for practice and future research with individuals with dementia and their proxies in making EOL decisions.
First, proxies reported 60% to 70% higher rates of advance directive completion by their relatives than those reported in past studies (Lingler et al., 2008). At the same time, less than 40% of individuals with dementia shared advance directives with physicians and more than 20% did not clearly communicate their preferences for life support treatments to their proxies. Hence, it is unclear if the ACP process has involved insufficient details regarding treatment preferences and a lack of conversations facilitated by health care professionals, which have been identified as contributors to inaccurate prediction of patient preferences and uncertainty by proxies (Ditto & Hawkins, 2005; Hines et al., 2001).
Second, an unexpected finding was that if proxies did not know preferences for any of the treatments, they reported having a clearer understanding of the values of the individual with dementia regarding EOL choices than if they knew all or some of the treatment preferences. Although this finding appears counterintuitive, it is possible that proxies feel more confident in their ability to evaluate treatment options when they have leeway to interpret and translate broad values or preferences for specific decisions about EOL care (Ditto & Hawkins, 2005; Hawkins, Ditto, Danks, & Smucker, 2005). Proxies may also feel more confident in evaluating treatment options against values without specific, detailed instructions because they are less constrained in projecting their own preferences onto making choices for their relative (Fagerlin, Ditto, Danks, Houts, & Smucker, 2001).
Third, proxies reported greater uncertainty about their choices and were less likely to expect to receive support from family and health care professionals if the individual with dementia wanted the least aggressive care (i.e., they were against all three life support treatments) as opposed to more aggressive care. It may be that proxies anticipated disagreements from other family members or health care providers if their decision was to restrict care rather than provide all care possible, especially when preferences were not clearly communicated to other family members or physicians. Past studies documented that families are more likely to choose overtreatments for patients in EOL situations due to the desire and obligation to do everything to avoid death (Ayalon, Bachner, Dwolatzky, & Heinik, 2012).
Fourth, it was found that proxies' decision-making experiences were associated with their knowledge of dementia trajectory, family support, and trust of physicians. These findings support the increasing consensus that proxy decision making is a complex, interactive process. The more knowledgeable the proxy was about dementia trajectory, the more informed and less uncertain he/she felt about decision making. Unlike cancer, dementia often presents with a slow, gradual decline in function and health, making it particularly challenging to identify the EOL phase. A study of nursing home residents at the EOL phase found that proxies' understanding of probable complications with advanced dementia and recognition of poor prognosis among residents predicted significantly lower chances of receiving burdensome intervention (Mitchell et al., 2009).
In past studies, lack of support from family members for proxies' specific decision-making roles as well as problematic relations with health care providers were found to lead to decisional conflict (Breen et al., 2001). Similarly, proxies who appraised their family to be supportive were more likely to anticipate receiving support from other family members in decision making. Trust of physicians was associated with all domains of decisional conflict. Higher levels of trust were associated with the proxy being more informed about values and feeling supported and less uncertain about his/her decision. It is likely that the proxy's ability to process information, deliberate decisions, and engage in shared decision making with health care teams may be impaired if he/she perceives the source (i.e., health care providers) of information critical for EOL decision making to be untrustworthy (Fassier & Azoulay, 2010; Forbes et al., 2000; Wendler & Rid, 2011).
The current findings underscore the complexity of the interactive nature of proxy decision making and the need for developing interventions that facilitate communication among individuals with dementia, proxies, and health care providers. A review of the state of advanced dementia research (Mitchell et al., 2012) identified two key factors associated with better EOL outcomes: (a) the presence of advance directives and (b) counseling of proxies in making EOL decisions. The current findings suggest that interventions involving ACP should consider providing education to help individuals with dementia and proxies in earlier disease trajectory gain more accurate and realistic understandings of the end stage of dementia, and ensure communicating preferences with direct health care providers. Such education may need to be delivered to individuals with dementia and their families earlier in the trajectory so that preferences for EOL treatments can be discussed and documented with knowledge of a more realistic context.
Moreover, facilitating detailed discussion regarding how much leeway proxies should have in actual EOL decision making may also help them cope with uncertainty (Hawkins et al., 2005). In addition, such interventions should address how to seek and handle input from other family members and health care providers. Decision coaching, which involves individualized, nondirective support provided by a health care professional (e.g., nurse, social worker) to assist family members in shared decision making with their health care team, may be a useful strategy (Murray, Miller, Fiset, O'Connor, & Jacobsen, 2004; Stacey et al., 2008). Decision coaching is similar to professionally facilitated ACP interventions, such as the Respecting Choices program (Detering et al., 2010), but may be specifically targeted to individuals with dementia and their proxies. In their role as decision coaches, health care providers could address proxies' needs for support in the process of translating values into specific treatment choices, as well as reduce barriers to shared decision making (e.g., distrust of health care providers) by mediating information exchange and deliberation between the proxy and health care team.
In considering these findings, a few study limitations should be noted. The cross-sectional design, with a convenience sample of proxies who were predominantly White, limited the generalizability of these findings. ACP and proxy preferences may also vary by ethnicity. In addition, the decision conflict reported is only a proxy for decisional conflict in actual decision-making situations. Although special efforts were made to develop the decision scenario as close as possible to a real EOL care, decision-making context, the experience in real situations may be different from the responses obtained using the vignette. However, this problem is not unique to the current study because the current practice of ACP across health care settings is based on patient preferences for future hypothetical illness states. Nevertheless, more prospective studies are needed. Another limitation is that proxies' specific knowledge of the treatments they were choosing (e.g., potential benefits and risks of these treatments) was not tested, which may need to be examined further in future studies as well as incorporated in future interventions. Lastly, preferences were obtained from proxies because the authors did not have access to advance directive documents. Future studies should further assess directly the role of ACP and actual EOL decision-making situations by using prospective, controlled designs.
Findings from the current study provide insight into the complex process of proxy decision making for individuals with dementia and potential factors that may influence the EOL decision-making process. Future intervention and research efforts should take into consideration preferences of individuals with dementia, proxies' knowledge of disease trajectory, family support, and relationship with health care providers to better support individuals with dementia and their families.
- Alzheimer's Association. (2014). 2014 Alzheimer's disease facts and figures. Retrieved from http://www.alz.org/downloads/facts_figures_2014.pdf
- Anderson, L.A. & Dedrick, R.F. (1990). Development of the Trust in Physician Scale: A measure to assess interpersonal trust in patient-physician relationships. Psychological Reports, 67, 1091–1100. doi:10.2466/PR0.67.8.1091-1100 [CrossRef]
- Ayalon, L., Bachner, Y.G., Dwolatzky, T. & Heinik, J. (2012). Preferences for end-of-life treatment: Concordance between older adults with dementia or mild cognitive impairment and their spouses. International Psychogeriatrics, 24, 1798–1804. doi:10.1017/S1041610212000877 [CrossRef]
- Biola, H., Sloane, P.D., Williams, C.S., Daaleman, T.P., Williams, S.W. & Zimmerman, S. (2007). Physician communication with family caregivers of long-term care residents at the end of life. Journal of the American Geriatrics Society, 55, 846–856. doi:10.1111/j.1532-5415.2007.01179.x [CrossRef]
- Breen, C.M., Abernethy, A.P., Abbott, K.H. & Tulsky, J.A. (2001). Conflict associated with decisions to limit life-sustaining treatment in intensive care units. Journal of General Internal Medicine, 16, 283–289. doi:10.1046/j.1525-1497.2001.00419.x [CrossRef]
- Carpenter, B.D., Balsis, S., Otilingam, P.G., Hanson, P.K. & Gatz, M. (2009). The Alzheimer's disease knowledge scale: Development and psychometric properties. The Gerontologist, 49, 236–247. doi:10.1093/geront/gnp023 [CrossRef]
- Dening, K.H., Jones, L. & Sampson, E.L. (2011). Advance care planning for people with dementia: A review. International Psychogeriatrics, 23, 1535. doi:10.1017/S1041610211001608 [CrossRef]
- Detering, K.M., Hancock, A.D., Reade, M.C. & Silvester, W. (2010). The impact of advance care planning on end of life care in elderly patients: Randomised controlled trial. British Medical Journal, 340, c1345. doi:10.1136/bmj.c1345 [CrossRef]
- Ditto, P.H. & Hawkins, N.A. (2005). Advance directives and cancer decision making near end of life. Health Psychology, 24(Suppl. 4), S63–S70. doi:10.1037/0278-6133.24.4.S63 [CrossRef]
- Epstein, N.B., Baldwin, L.M. & Bishop, D.S. (1983). The McMaster Family Assessment Device. Journal of Marital and Family Therapy, 9, 171–180. doi:10.1111/j.1752-0606.1983.tb01497.x [CrossRef]
- Fagerlin, A., Ditto, P.H., Danks, J.H., Houts, R.M. & Smucker, W.D. (2001). Projection in surrogate decisions about life-sustaining medical treatment. Health Psychology, 20, 166–175. doi:10.1037/0278-618.104.22.168 [CrossRef]
- Fassier, T. & Azoulay, E. (2010). Conflicts and communication gaps in the intensive care unit. Current Opinion in Critical Care, 16, 654–665. doi:10.1097/MCC.0b013e32834044f0 [CrossRef]
- Forbes, S., Bern-Klug, M. & Gessert, C. (2000). End-of-life decision making for nursing home residents with dementia. Journal of Nursing Scholarship, 32, 251–258. doi:10.1111/j.1547-5069.2000.00251.x [CrossRef]
- Gessert, C.E., Forbes, S. & Bern-Klug, M. (2000). Planning end-of-life care for patients with dementia: Roles of families and health professionals. Omega, 42, 273–291.
- Givens, J.L., Kiely, D.K., Carey, K. & Mitchell, S.L. (2009). Healthcare proxies of nursing home residents with advanced dementia: Decisions they confront and their satisfaction with decision-making. Journal of the American Geriatrics Society, 57, 1149–1155. doi:10.1111/j.1532-5415.2009.02304.x [CrossRef]
- Hawkins, N.A., Ditto, P.H., Danks, J.H. & Smucker, W.D. (2005). Micromanaging death: Process preferences, values and goals in end-of-life medical decision making. The Gerontologist, 45, 107–117. doi:10.1093/geront/45.1.107 [CrossRef]
- Hines, S.C., Glover, J.J., Babrow, A.S., Holley, J.L., Badzek, L.A. & Moss, A.H. (2001). Improving advance care planning by accommodating family preferences. Journal of Palliative Medicine, 4, 481–489. doi:10.1089/109662101753381629 [CrossRef]
- Hirschman, K., Kapo, J. & Karlawish, J. (2006). Why doesn't a family member of a person with advanced dementia use a substituted judgment when making a decision for that person?American Journal of Geriatric Psychiatry, 14, 659–667. doi:10.1097/01.JGP.0000203179.94036.69 [CrossRef]
- Hirschman, K., Kapo, J. & Karlawish, J. (2008). Identifying the factors that facilitate or hinder advance planning by persons with dementia. Alzheimer Disease and Associated Disorders, 22, 293. doi:10.1097/WAD.0b013e318169d669 [CrossRef]
- Janis, I. & Mann, L. (1977). Decision making: A psychological analysis of conflict, choice, and commitment. New York, NY: Free Press.
- Kramer, B.J., Boelk, A.Z. & Auer, C. (2006). Family conflict at the end of life: Lessons learned in a model program for vulnerable older adults. Journal of Palliative Medicine, 9, 791–801. doi:10.1089/jpm.2006.9.791 [CrossRef]
- Kwak, J., Allen, J.Y. & Haley, W.E. (2011). Advance care planning and end-of-life decision making. In Dilworth-Anderson, P. & Palmer, M.H. (Eds.), Annual review of gerontology and geriatrics: Pathways through the transitions of care for older adults (Vol. 31, pp. 143–165). New York, NY: Springer.
- Lingler, J.H., Hirschman, K.B., Garand, L., Dew, M.A., Becker, J.T., Schulz, R. & DeKosky, S.T. (2008). Frequency and correlates of advance planning among cognitively impaired older adults. American Journal of Geriatric Psychiatry, 16, 643–649. doi:10.1097/JGP.0b013e31816b7324 [CrossRef]
- Mitchell, S.L., Black, B.S., Ersek, M., Hanson, L.C., Miller, S.C., Sachs, G.A. & Morrison, R.S. (2012). Advanced dementia: State of the art and priorities for the next decade. Annals of Internal Medicine, 156, 45–51. doi:10.7326/0003-4819-156-1-201201030-00008 [CrossRef]
- Mitchell, S.L., Teno, J.M., Kiely, D.K., Shaffer, M.L., Jones, R.N., Prigerson, H.G. & Hamel, M.B. (2009). The clinical course of advanced dementia. New England Journal of Medicine, 361, 1529–1538. doi:10.1056/NEJMoa0902234 [CrossRef]
- Murray, M.A., Miller, T., Fiset, V., O'Connor, A. & Jacobsen, M.J. (2004). Decision support: Helping patients and families to find a balance at the end of life. International Journal of Palliative Nursing, 10, 270–277. doi:10.12968/ijpn.2004.10.6.13268 [CrossRef]
- O'Connor, A.M. (1995). Validation of a decisional conflict scale. Medical Decision Making, 15, 25–30. doi:10.1177/0272989X9501500105 [CrossRef]
- O'Connor, A.M. & Jacobsen, M.J. (2007). Decisional conflict: Supporting people experiencing uncertainty about options affecting their health. Retrieved from http://homeless.ehclients.com/images/uploads/W-2_Ottawa_Decision_Making_tool--_Reading-1.pdf
- Pasman, H.W., Onwuteaka-Philipsen, B.D., Ooms, M.E., van Wigcheren, P.T., van der Wal, G. & Ribbe, M.W. (2004). Forgoing artificial nutrition and hydration in nursing home patients with dementia: Patients, decision making, and participants. Alzheimer Disease and Associated Disorders, 18, 154–162. doi:10.1097/01.wad.0000137522.69111.56 [CrossRef]
- Shalowitz, D.I., Garrett-Mayer, E. & Wendler, D. (2006). The accuracy of surrogate decision makers: A systematic review. Archives of Internal Medicine, 166, 493–497. doi:10.1001/archinte.166.5.493 [CrossRef]
- Silveira, M.J., Kim, S.Y. & Langa, K.M. (2010). Advance directives and outcomes of surrogate decision making before death. New England Journal of Medicine, 362, 1211–1218. doi:10.1056/NEJMsa0907901 [CrossRef]
- Stacey, D., Murray, M.A., Legare, F., Dunn, S., Menard, P. & O'Connor, A. (2008). Decision coaching to support shared decision making: A framework, evidence, and implications for nursing practice, education, and policy. Worldviews on Evidence-Based Nursing, 5, 25–35. doi:10.1111/j.1741-6787.2007.00108.x [CrossRef]
- Swigart, V., Lidz, C., Butterworth, V. & Arnold, R. (1996). Ethical issues. Letting go: Family willingness to forgo life support. Heart & Lung, 25, 483–494. doi:10.1016/S0147-9563(96)80051-3 [CrossRef]
- Volandes, A.E., Paasche-Orlow, M.K., Barry, M.J., Gillick, M.R., Minaker, K.L., Chang, Y. & Mitchell, S.L. (2009). Video decision support tool for advance care planning in dementia: Randomised controlled trial. British Medical Journal, 338, b2159. doi:10.1136/bmj.b2159 [CrossRef]
- Wendler, D. & Rid, A. (2011). Systematic review: The effect on surrogates of making treatment decisions for others. Annals of Internal Medicine, 154, 336–346. doi:10.7326/0003-4819-154-5-201103010-00008 [CrossRef]
- White, D.B. & Arnold, R.M. (2011). The evolution of advance directives. Journal of the American Medical Association, 306, 1485–1486. doi:10.1001/jama.2011.1430 [CrossRef]
Demographics and Characteristics of Individuals With Dementia and Proxies (N = 141)
|Demographic/Characteristic||Individuals With Dementia, n (%)||Proxies, n (%)|
|Age (years), mean (SD)||76.21 (11.78)||62.41 (10.62)|
| Female||75 (53.2)||102 (72.3)|
| White||136 (96.5)|
|Caring for spouse/partner||82 (58.2)|
|Living with individual with dementia||85 (60.3)|
|Diagnosis and functional status|
| Dementia diagnosis||76 (53.9)|
| In the past 6 months...|
| Fell one or more times||35 (24.8)|
| Had pneumonia one or more times||13 (9.2)|
| Hospitalized one or more times||36 (25.5)|
| Number of ADLs needing help||2.67 (2.48)|
| Number of IADLs needing help||7.98 (2.83)|
Advance Care Planning for Individuals With Dementia and Proxies' End-of-Life Decision Making (N = 141)a
|Advance care planning|
| Completed a living will||124 (87.9)|
| Shared a copy with proxy||124 (87.9)|
| Shared a copy with physician||54 (38.3)|
|Preferences for life support|
| Cardiopulmonary resuscitation (CPR)|
| Yes||32 (23.2)|
| No||75 (54.3)|
| Do not know||31 (22.5)|
| Feeding tube|
| Yes||14 (10.1)|
| No||88 (63.3)|
| Do not know||37 (26.6)|
| Yes||10 (7.1)|
| No||97 (69.3)|
| Do not know||33 (23.6)|
|Proxy's goal of care|
| Life-prolonging||7 (5)|
| Limited care||37 (26.4)|
| Comfort-only||96 (68.6)|
|Proxy's choice of treatments|
| Perform everything available||9 (6.5)|
| Perform everything except CPR/ICU||21 (15.1)|
| Perform only pain and symptom control||109 (74.8)|
|Proxy's decisional conflict scorec|
| Informed||2.62 (1.07)|
| Value clarity||2.71 (1.1)|
| Support||2.66 (1.1)|
| Uncertainty||2.5 (0.86)|
Summary of Regression Analyses for Variables Predicting Decisional Conflict Subscale Scoresa (N = 141)
|B||SE B||β||B||SE B||β||B||SE B||β||B||SE B||β|
|Individual with dementia does not want life support treatment||0.27||0.19||0.13||0.17||0.19||0.08||0.47||0.17||0.23**||0.43||0.18||0.21**|
|Proxy does not know life support preferences||−0.13||0.29||−0.04||−0.61||0.28||−0.19*||−0.48||0.26||−0.15||−0.26||0.27||−0.08|
|Proxy's dementia knowledge||−1.14||0.47||−0.21*||−1.03||0.46||−0.18*||−0.60||0.42||−0.11||−0.91||0.43||−0.17*|
|Trust of physician||−0.04||0.01||−0.46***||−0.05||0.01||−0.51***||−0.03||0.01||−0.37***||−0.05||0.01||−0.53***|