ADEPP Tool for End-Stage Dementia EOL Care Planning

Background and Hospice Utilization in End-Stage Dementia

The U.S. Centers for Medicare and Medicaid Services (CMS) has been tracking hospice patients by diagnosis for the past decade, and the data collected reveals significant improvements in the number of days dementia patients spend in hospice. Between 1998 and 2009, average length of stay (LOS) for patients with a diagnosis of Alzheimer's disease increased from 67 to 106 days, while patients with a diagnosis of non-Alzheimer's dementia increased from 57 to 92 days.¹ This seems like good news, given the findings of a limited study which revealed that the benefits of hospice services noticed by family members peaked around 90 days.² The national data therefore suggests dementia patients have succeeded in reaching a reasonable average LOS.

Not all locations within the United States appear to be benefiting equally, however. In Lee County, Florida, for example, the average hospice LOS for all patients, regardless of diagnosis, was just 10 days.³ This statistic is troubling in part because the percentage of residents 65 years of age and older within this community is double the national average.^4,5 These statistics would also suggest that twice as many Lee County residents suffer from dementia compared to the rest of the nation. The prevalence of dementia for this age group is unknown, but estimates suggest that 5 million Americans 65 years of age and older suffered from Alzheimer's disease in 2014,⁶ the most common cause of dementia.⁷ The number suffering from non-Alzheimer's dementia probably represents another 3% of the population,⁸ for a total of approximately 14% for this age group. Applying this estimate to the residents of Lee County would suggest that almost 24,000 residents 65 years of age and older⁴ suffer from some form of dementia.

In 2012, the sole hospice provider in Lee County, Hope Hospice and Community Services, provided hospice services to 4,705 Lee County residents.³ The State of Florida estimated that approximately 13% of hospice patients would have had a primary diagnosis of dementia, which translates into an estimated 612 residents. With an average life expectancy of 6 years following diagnosis,⁷ the number of Lee County residents 65 years of age and older who were suffering from dementia and had less than 6 months to live would be approximately 2,000. Based on these calculations, approximately 1,400 residents — or 70% of all residents with end-stage dementia — probably never receive hospice services, let alone receive them in a timely manner. In addition, it is important to note that only about half of all Americans estimated to suffer from dementia ever receive a formal diagnosis.⁹ Clearly, hospice is being underutilized by dementia patients in Lee County, and when services are obtained the estimated average LOS is only 10 days.

The underutilization of hospice by dementia patients has been attributed to a number of variables. According to a survey of 800 U.S. hospice program directors, the primary barriers to hospice utilization were provider and family ignorance of hospice services,¹⁰ a conclusion confirmed by a number of other studies.^11–13 Primary caregivers residing in Connecticut almost uniformly reported provider indifference to an expressed need for end-of-life (EOL) planning and palliative/hospice services.¹³ Whether perceived provider indifference was due to ignorance¹⁰ or the lack of Medicare reimbursement for clinician-initiated EOL planning¹⁴ is unknown, but it seems likely that both could be contributors.

Other barriers include prognostic accuracy and provider discomfort with terminating curative treatments;^15,16 however, determining an accurate prognosis for dementia patients can be challenging.^17,18 The difficulty of determining an accurate prognosis is exacerbated by the Medicare definition of "terminally ill" as having less than 6 months to live,¹⁹ thereby making this task even more difficult. In a recent survey, nearly all physicians caring for patients with life-limiting illnesses reported feeling comfortable discussing advance care planning (ACP) with patients and family members, but did so only 43% of the time.²⁰ In addition, less than 30% believed advance care wishes were ever honored. Physicians are also less likely to discharge dementia patients from inpatient psychiatric care to nursing homes (NHs) offering hospice services,²¹ and NH organizational structure and care culture can play a role in determining whether hospice services are offered and utilized.^22–25 Based on these findings, there are significant provider and organizational barriers to timely hospice referrals for patients suffering from end-stage dementia.

The overall goal of hospice care is to eliminate the unmet care needs of patients and family members as death approaches. Accordingly, researchers at the UCLA Medical Center implemented an End-of-Life Symptom Management Order (EMSO) protocol to help determine whether patients with life-limiting illnesses were eligible to receive a continuous opiate infusion for effective pain management.²⁶ Certain criteria needed to be met, including an active DNR, documented discussions of care goals with patient and family members, and a consensus reached between clinicians, patient, and family members about transitioning to comfort-oriented care. Clinicians also considered the extent of limitations for activities of daily living (ADL) and the disease and caregiver burden. Importantly, a post-implementation study revealed the EMSO protocol significantly reduced the prevalence of unmet care needs among dying patients.²⁶

The success of the EMSO protocol raises the question of whether a similar intervention could improve hospice utilization by end-stage dementia patients in Lee County, Florida, thereby reducing the prevalence of unmet care needs among this patient group. Towards this goal, an Advanced Dementia EOL Planning and Prognosis (ADEPP) tool was developed to educate clinicians about best-practice guidelines for ACP and to provide guidance for making an accurate prognosis (Appendix A).

Study Methods and Design

Study Design

The study design was primarily descriptive and was based on data collected from a questionnaire and interviews. The questionnaire was essentially identical to that developed and utilized by Snyder and colleagues,²⁰ except that the demographic question about provider academic degree (question 23) was expanded to include nurse practitioner (NP). The first step was to ask participating providers to complete the questionnaire and return it to the researchers prior to reviewing the ADEPP tool (Appendix A). The second step involved providers reviewing the ADEPP tool and using it in their clinical practice if they wished. The perceived acceptance and utility of the tool was then evaluated using a set of interview questions (Appendix B) communicated by phone.

Subjects

Project participants were physicians and nurse practitioners (NPs) practicing in Lee County, Florida. Provider names and contact information were obtained from the Lee County Medical Society, via internet searches, and personal contacts. Letters of inquiry were sent to clinicians selected by the following criteria: (1) providers licensed in the State of Florida, (2) residing in Lee County, Florida, and (3) providing care for patients suffering from advanced dementia. There were no exclusion criteria. No attempt was made to select for a specific demographic defined by age, gender, race, or ethnicity.

Tools

The questionnaire,²⁰ modified slightly to include the category of NP for question 23, was used to evaluate physician attitudes and practices before exposing participants to the ADEPP tool. The reliability of the survey was sufficient (Cronbach's alpha = .68) and internal validity was verified using a number of different questions about the same topic.²⁰ The questionnaire was designed to collect quantitative and qualitative data, in addition to demographic data. The set of interview questions (Appendix B) was designed to evaluate the acceptance and clinical utility of the ADEPP tool (Appendix A).

Intervention and Data Collection

A sample of 100 dementia care providers received by mail a single envelope containing a letter of inquiry, a consent form, a copy of the Primary Care Physician Questionnaire,²⁰ and a separate envelope containing the ADEPP tool. The letter of inquiry explained the purpose of the quality improvement project, while the consent form asked providers to voluntarily participate in the survey and whether they would be willing to be interviewed. Participants were asked to complete the consent form and questionnaire before returning both in the provided envelope. Participants then reviewed the ADEPP tool and decided whether to apply the information it contained to their practice. Participants who expressed an interest in being interviewed were contacted by phone and answers were recorded with pen and paper. The participation rate obtained by Snyder and colleagues²⁰ was 42%, but the questionnaire return rate in the current study was only 21%.

Analysis and Evaluation Plan

The primary themes used to evaluate the questionnaire data, which were based on the National Consensus Project,²⁷ were the following:

1. Clinician comfort with discussing EOL care plans with patient and family
2. Knowledge and attitudes about palliative and hospice services
3. Timeline of EOL planning, palliative care initiation, and referral to hospice
4. Perceived barriers to palliative or hospice care referrals

Since this was primarily a descriptive study, quantitative analysis was limited to the difference in the recommended and observed timing of ACP, palliative care, and hospice services discussion initiations during the disease process. Each variable was compared using the Wilcoxon rank-sum test. The interview data were analyzed for evidence of increased clinician comfort with EOL planning, knowledge about EOL planning, awareness and implementation of recommended timelines for palliative and hospice care referrals, utilization of palliative and hospice services, and/or diagnostic accuracy.

Questionnaire Results

Of the 100 questionnaires mailed to providers in Lee County, Florida, only 21 were completed and returned (Table 1). Among the participants who returned questionnaires, 8 were medical doctors (MDs), 4 were osteopathic doctors (DOs), and 9 were nurse practitioners (NPs). The vast majority — 86% — practiced in family medicine rather than internal medicine, and only 14% had a sub-specialty (2 geriatric and 1 family). Over 95% had at least 6 years of experience practicing medicine, and nearly a quarter had over 20 years of experience. The age range was between 30 and 80 years, while the mean and median ages were almost identical at approximately 57 years. The gender mix was balanced, with 52% women, and the vast majority reported being White (81%). Sixty-seven percent reported having a personal advance directive (AD), while 81% had helped family members and friends with ACP in the past. Fifty-seven percent had both a personal AD and had helped family and friends with ACP, while only 2 participants had neither.

The percent of patients per survey participant with progressive, chronic, life-limiting diseases was 50% (median) and 53.4% (mean), respectively (Table 2). By comparison, the percent of terminally ill patients being treated by each participant was 5% (median) and 32% (mean), respectively, which indicates that most participants were treating only a few terminally ill patients, while a small number of participants were treating a large number of these patients. Participants also reported discussing ADs with 33.4% of patients with a progressive, chronic, life-limiting condition and 43% of terminally ill patients. The mean number of palliative and hospice referrals made by participants within the past six months was 7.6 and 4.9, respectively. This data suggests only a few patients may be benefitting from palliative and hospice services among the many who may qualify.

When participants were queried about perceived barriers to palliative and hospice referrals, 43% responded in the affirmative (Table 3). Participants who noted perceived barriers on the questionnaire attributed resistance to the wishes of patients and family members, while none mentioned system-, organizational-, or provider-associated barriers. In response to a related question about who should be primarily responsible for initiating ACP discussions, 38% believed primary care providers should be responsible, and another 38% felt this duty should be shouldered by palliative care consultants (Table 3). One participant believed attorneys should be responsible, and another assigned this responsibility to specialists, while 3 participants felt that patients should be primarily responsible for initiating ACP discussions. Clearly, there is a lack of consensus about who should be responsible for initiating a discussion about ACP, although there was a slight bias in favor of specialists and hospitalists (Fig. 1).

When participants were queried about how well they understood palliative and hospice care, less than half agreed that palliative care was equivalent to hospice care (Fig. 1). Most participants, however, understood that palliative care is part of hospice care but not the reverse. When queried about whether palliative care should be used only when patients have less than six months to live, most disagreed. Most also understood that the sole purpose of palliative care was not the creation of an AD.

Participants were more evenly divided when queried about the ability of palliative and hospice care specialists to grant ACP wishes to patients and family members, or whether ACP is too time-consuming and too upsetting to patients and families (Fig. 1). This result may indicate a significant level of ambivalence among providers about the clinical value of ACP, an interpretation supported by the perceived low clinical utility of ACP documentation (Fig. 1). Provider comfort concerning ACP consults or when asked to determine a prognosis were also low, which is consistent with a general theme that ACP can be intrusive, unwelcome, and of little clinical value.

Providers were asked to recommend the optimum time to initiate ACP discussions with patients and family members during the disease process (Fig. 2). The data revealed wide disagreement, ranging from a peak at disease onset to 3 months before death. The recommended start of discussions concerning palliative care ranged from disease onset to 1 week before death, and those concerning hospice ranged from disease onset to the moment of death. When queried about the actual or observed timing of ACP, palliative, and hospice discussion initiations, most providers reported later initiation times. The difference between the recommended and observed timing for initiating these three discussions was statistically significant using the Wilcoxon rank-sum test (ACP, P = .0003; palliative care, P = .0006; hospice services, P = .0003).

References

1. U.S. Centers for Medicare and Medicaid Services. Average length of stay by year, by diagnosis, for diagnoses in the top 20 in 2009. Accessed March 15, 2015.

2. Lopez-Acevedo M, Lowery WJ, Lowery AW, Lee PS, Havrilesky LJ. Palliative and hospice care in gynecologic cancer: a review. Gynecol Oncol. 2013;131(1):215–221.

3. Bureau of Planning & Evaluation, Department of Elder Affairs, State of Florida. 2013 Report: Hospice Demographics and Outcome Measures. http://elderaffairs.state.fl.us/doea/Evaluation/2013%20Hospice%20Report%20FINAL.pdf. Accessed March 15, 2015.

4. U.S. Census Bureau. State & County QuickFacts: Lee County, Florida. http://quickfacts.census.gov/qfd/states/12/12071.html. Accessed March 16, 2015.

5. U.S. Census Bureau. Table 7. Resident population by sex and age: 1980 to 2010. Accessed March 15, 2015.

6. Hebert LE, Weuve J, Scherr PA, Evans DA. Alzheimer disease in the United States (2010–2050) estimated using the 2010 Census. Neurology. 2013;80:1778–1783.

7. Alzheimer's Association. Alzheimer's Association Report: 2014 Alzheimer's Disease Facts and Figures. Alzheimers Dement. 2014;10(2):e47–e92.

8. Plassman BL, Langa KM, Fisher GG, et al. Prevalence of dementia in the United States: the Aging, Demographics, and Memory Study. Neuroepidemiology. 2007;29(1–2):125–132.

9. Boustani M, Peterson B, Hanson L, Harris R, Lohr KN. Screening for dementia in primary care: a summary of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med. 2003;138(11):927–937.

10. Ryan T, Ingleton C. Most hospices and palliative care programmes in the U.S.A. serve people with dementia; lack of awareness, need for respite care and reimbursement policies are the main barriers to providing this care. Evid-Based Nurs. 2011;14(2):40–41.

11. van der Steen JT, van Soest-Poortvliet MC, Hallie-Heierman M, et al. Factors associated with initiation of advance care planning in dementia: a systematic review. J Alzheimers Dis. 2014;40(3):743–757.

12. Jenkins TM, Chapman KL, Ritchie CS, et al. Barriers to hospice care in Alabama: provider-based perceptions. Am J Hosp Palliat Care. 2011;28(3):153–160.

13. Lewis LF. Caregivers' experiences seeking hospice care for loved ones with dementia. Qual Health Res. 2014;24(9):1221–1231.

14. Advisory Board Company. Daily Briefing: Medicare may reimburse docs for end-of-life care. Accessed March 15, 2015.

15. Teno JM, Casarett D, Spence C, Conner S. Is it "too late" or is it? Bereaved family member perceptions of hospice referral when their family member was on hospice for seven days or less. J Pain Symptom Manage. 2012;43(4):732–738.

16. Lau DT, Masin-Peters J, Berdes C, Ong M. Perceived barriers that impede provider relations and medication delivery: hospice providers' experiences in nursing homes and private homes. J Palliat Med. 2010;13(3):305–310.

17. Mitchell SL, Miller SC, Teno JM, Davis RB, Shaffer ML. The Advanced Dementia Prognostic Tool: a risk score to estimate survival in nursing home residents with advanced dementia. J Pain Symptom Manage. 2010;40(5):639–651.

18. Mitchell SL, Miller SC, Teno JM, Kiely DK, Davis RB, Shaffer ML. Prediction of 6-month survival of nursing home residents with advanced dementia using ADEPT vs. hospice eligibility guidelines. JAMA. 2010;304(17):1929–1935.

19. U.S. Centers for Medicare and Medicaid Services. F tag 309 — Quality of Care — Advance Copy. Accessed March 16, 2015.

20. Snyder S, Hazelett S, Allen K, Radwany S. Physician knowledge, attitude, and experience with advance care planning, palliative care, and hospice: results of a primary care survey. Am J Hosp Palliat Care. 2013;30(5):419–424.

21. Epstein-Lubow G, Fulton AT, Marino LJ, Teno J. Hospice referral after inpatient psychiatric treatment of individuals with advanced dementia from a nursing home [published online ahead of print May 5, 2014]. Am J Hosp Palliat Med. 2014. doi:10.1177/1049909114531160.

22. Resnick HE, Foster GL, Hickman SE. Nursing home participation in end-of-life programs: United States, 2004. Am J Hosp Palliat Care. 2009;26(5):345–360.

23. Miller SC. A model for successful nursing home-hospice partnerships. J Palliat Med. 2010;13(5):525–533.

24. Zheng NT, Mukamel DB, Caprio T, Cai S, Temkin-Greener H. Racial disparities in in-hospital death and hospice use among nursing home residents at the end of life. Med Care. 2011;49(11):992–998.

25. Zheng NT, Mukamel DB, Caprio TV, Temkin-Greener H. Hospice utilization in nursing homes: association with facility end-of-life care practices. Gerontologist. 2013;53(5):817–827.

26. Walling AM, Ettner SL, Barry T, Yamamoto MC, Wenger NS. Missed opportunities: use of an End-of-Life Symptom Management Order Protocol among inpatients dying expected deaths. J Palliat Med. 2011;14(4):407–412.

27. National Consensus Project. Clinical Practice Guidelines for Quality Palliative Care, Second Edition. Pittsburgh, PA: National Consensus Project; 2009. http://www.nationalconsensusproject.org/guideline.pdf. Accessed March 17, 2015.

Tables

Table 1. Clinician Demographics

Degree
MD: 8 (38%)
DO: 4 (19%)
NP: 9 (43%)

Type of Practice
Family Medicine: 18 (86%)
Internal Medicine: 3 (14%)

Years of Experience
0–5 yrs: 1 (4.8%)
6–10 yrs: 4 (19.0%)
11–15 yrs: 5 (23.8%)
16–20 yrs: 6 (28.6%)
>20 yrs: 5 (23.8%)

Sub-specialty
Yes: 3 (14%)
No: 18 (86%)

Age
Min/Max: 30/80 | Median: 58 | Mean: 56.8

Gender
Female: 11 (52%)
Male: 10 (48%)

Race/Ethnicity
African-American: 1 (4.76%)
Hispanic: 1 (4.76%)
Asian-American: 1 (4.76%)
Caucasian/White: 17 (81%)
Other: 1 (4.76%)

Clinician Advance Directive (AD)
Yes: 14 (67%)
No: 7 (33%)

Clinician personal ACP experience
Yes: 17 (81%)
No: 4 (19%)

Both clinician AD and personal ACP experience
Yes: 12 (57%)
No: 2 (9.5%)

Note: AD, advance directive; ACP, advance care planning.

Table 2. Patient Demographics

Percent with progressive, chronic, life-limiting disease
Min/Max: 1/100 | Median: 50 | Mean: 53.4

Percent discussed AD (among patients with progressive, chronic, life-limiting disease)
Min/Max: 0/100 | Median: 20 | Mean: 33.4

Percent terminally ill
Min/Max: 0/100 | Median: 5 | Mean: 32

Percent discussed AD (among terminally ill patients)
Min/Max: 0/100 | Median: 50 | Mean: 43

Number of palliative care referrals past 6 months
Min/Max: 0/40 | Median: 3 | Mean: 7.6

Number of hospice referrals past 6 months
Min/Max: 0/30 | Median: 2 | Mean: 4.9

Note: AD, advance directive.

Table 3. Barriers to Referrals

Are there factors limiting referrals?
Yes: 9 (43%)
No: 12 (57%)

Primary responsibility for initiating ACP discussions
Attorneys: 1 (5%)
Primary care physicians: 8 (38%)
Social workers: 0 (0%)
Palliative care (PC) consultants: 8 (38%)
Patients: 3 (14%)
Specialists: 1 (5%)

Note: ACP, advance care planning; PC, palliative care.

Supplemental Table 1. Interviewed Clinician Demographics

Degree
MD: 5 (56%)
DO: 1 (11%)
NP: 3 (33%)

Type of Practice
Family Medicine: 6 (67%)
Internal Medicine: 3 (33%)

Years of Experience
0–5 yrs: 0 (0%)
6–10 yrs: 2 (22%)
11–15 yrs: 1 (11%)
16–20 yrs: 3 (33%)
>20 yrs: 2 (22%)

Sub-specialty
Yes: 3 (33%)
No: 6 (67%)

Age
Min/Max: 34/80 | Median: 62 | Mean: 59.2

Gender
Female: 5 (56%)
Male: 4 (44%)

Race/Ethnicity
African-American: 0 (0%)
Hispanic: 0 (0%)
Asian-American: 0 (0%)
Caucasian/White: 8 (89%)
Other: 1 (11%)

Clinician AD
Yes: 7 (78%)
No: 2 (22%)

Clinician personal ACP experience
Yes: 7 (78%)
No: 2 (22%)

Both clinician AD and personal ACP experience
Yes: 5 (56%)
No: 0 (0%)

Note: AD, advance directive; ACP, advance care planning.

Supplemental Table 2. Interviewee Patient Demographics

Percent with progressive, chronic, life-limiting disease
Min/Max: 1/100 | Median: 50 | Mean: 45

Percent discussed AD (among patients with progressive, chronic, life-limiting disease)
Min/Max: 0/95 | Median: 1 | Mean: 19

Percent terminally ill
Min/Max: 0/100 | Median: 3 | Mean: 24

Percent discussed AD (among terminally ill patients)
Min/Max: 0/100 | Median: 5 | Mean: 31

Number of palliative care referrals past 6 months
Min/Max: 0/40 | Median: 3 | Mean: 7.8

Number of hospice referrals past 6 months
Min/Max: 0/30 | Median: 2 | Mean: 4.6

Note: AD, advance directive.

Supplemental Table 3. Barriers to Referrals (Interviewees)

Are there factors limiting referrals?
Yes: 4 (44%)
No: 5 (55%)

Primary responsibility for initiating ACP discussions
Attorneys: 1 (11%)
Primary care physicians: 2 (22%)
Social workers: 0 (0%)
Palliative care (PC) consultants: 4 (44%)
Patients: 1 (11%)
Specialists: 0 (0%)

Note: ACP, advance care planning.

Appendix A: Advanced Dementia End-of-Life Planning and Prognosis (ADEPP) Tool

Best Practice Recommendations

Palliative and hospice care focus on providing patient- and family-centered care, thereby improving quality of life through the relief of suffering and attention to the social, economic, and spiritual needs of patients and families. The key elements of palliative care, according to the National Consensus Project (2009), are the following:

1. Patients with life-threatening conditions, illnesses, or injuries
2. Patient- and family-centered care, with family defined by the patient or surrogate
3. Timely provision of palliative or hospice services, which typically begins when a life-threatening condition or illness is first diagnosed
4. Comprehensive care to address the physical, psychological, social, and spiritual needs of the patient and family. Assessments, diagnosis, planning, interventions, monitoring, and follow-up services should be provided in a formal clinical setting and conducted at regular intervals.
5. Skilled, multidisciplinary care for dementia patients at all stages of disease progression in order to address the physical, psychological, social, and spiritual needs of the patient and family. The professionals on the care team can range from gerontologists to psychiatrists and from social workers to clergy.
6. Effective relief of suffering, especially pain relief
7. Effective communications, consisting of information sharing, active listening, consensus treatment goals and preferences, and medical decision-making support, for both patient and families
8. Skilled determination of prognosis and assessment of family needs during end-of-life
9. Continuity of care, from home to emergency room and from the intensive care unit to hospice setting
10. Access to care regardless of disease, disease severity, disabilities, race, ethnicity, gender, sexual orientation, and financial resources
11. Delivery of high-quality care, as defined by the following:
   a. Timely provision of services, 24 hours a day and 7 days a week
   b. Patient-centered
   c. Achieves desirable outcomes
   d. Accessible and equitable
   e. Evidence- and knowledge-based
   f. Efficient delivery of care

Please see National Consensus Project (2009) and van der Steen et al. (2014a) for more detailed information.

End-of-Life Planning

The Centers for Medicare and Medicaid Services (CMS) (2012) have identified the goals that will help a dying patient experience a "good death." These are:

1. Patient autonomy
2. Effective pain management
3. Trusted clinicians
4. Courteous and caring staff
5. No unnecessary, intrusive treatments
6. Patient/family understanding of the progression of illness
7. Cleanliness
8. Making connections with family and friends
9. Minimal social isolation
10. Dignity

The American Nurses Association (2010) has outlined the ethical and clinical goals of EOL planning:

1. EOL planning should not be a single meeting between clinicians and patients, but a series of meetings between a multidisciplinary care team, the patient, and family members, in order to adapt the care plan to the patient's changing health and prognosis.
2. Nurses should try to understand the patient's and family's fears, needs, and values while discussing care options and providing counseling and support. The primary skill needed is the ability to actively listen.
3. For life-limiting diseases like dementia, palliative care should be one of the first treatment options discussed.
4. In the presence of the care team, advance directives should be discussed next, preferably beginning in an outpatient setting. The care team will perform best when it includes palliative and hospice specialists, social workers, and spiritual care providers, in addition to clinicians. The patient, concerned family members, and surrogates should also be present.
5. Advance directives can include living wills, do-not-resuscitate orders, durable power of health care attorney, healthcare proxies, and Physician Order for Life-Sustaining Treatment (POLST). Use of these documents can vary from state to state, so the relevant agency or website should be consulted. In the State of Florida, the primary advance care directives are a living will and health care surrogate (Florida Agency for Health Care Administration, n.d.). Under state law, both directives must be honored by clinicians regardless of setting. The health care surrogacy can be included in a durable power of attorney. A pre-hospital do-not-resuscitate order is also honored in Florida, and forms are available from attorneys, physicians, ambulance services, and the website of the Department of Elder Affairs for the State of Florida (n.d.). The other advance directive forms can also be found on this website.

Ethical Guidance (ANA, 2010)

1. Any patient or authorized proxy has the ethical and legal right to refuse treatment.
2. A clinician may ethically and legally withhold or withdraw life-sustaining treatment when it is disproportionately burdensome or not beneficial.
3. A patient's or proxy's lack of consent to treat is equally binding.
4. Withholding or withdrawing treatment, however, does not nullify a provider's responsibility to provide high-quality care.
5. Assisted suicide and active euthanasia are illegal in Florida and most of the United States.
6. Additional detailed information can be obtained from the Alzheimer's Association (2007) and within the research articles authored by van der Steen et al. (2014a, 2014b).

Prognosis Determination for Patients with Dementia

ADEPT Instrument (Adapted from Mitchell et al., 2010a)

Recent nursing home admission (<90 days): 3.3
65–69 years of age: 1.0
70–74 years of age: 2.0
75–79 years of age: 3.0
80–84 years of age: 4.0
85–89 years of age: 5.0
90–94 years of age: 6.0
95–99 years of age: 7.0
100 or >100 years of age: 8.0
Male: 3.3
Dyspnea: 2.7
At least one pressure ulcer ≥ Stage 2: 2.2
Activities of Daily Living score = 28 (total dependence)*: 2.1
Bedfast most of the day: 2.1
Insufficient oral intake: 2.0
Bowel incontinence: 1.9
Body mass index <18.5 kg/m²: 1.8
Weight loss: 1.6
Congestive heart failure: 1.5

*See table below to calculate activities of daily living score.

Probability of Death Prediction Table (Adapted from Mitchell et al., 2010a)

ADEPT Risk Score 10: Probability of death within 6 months = 21% (confidence: 73%); within 12 months = 37% (confidence: 70%)
Risk Score 13: 34% / 52%
Risk Score 14: 40% / 57%
Risk Score 16: 52% / 67%
Risk Score 18: 64% / 76%
Risk Score 20: 73% / 84%
Risk Score 22: 83% / 90%

Activities of Daily Living Scoring (Adapted from Mitchell et al., 2010a)

For each activity, score as follows: Independent = 0, Supervision Needed = 1, Limited Assistance = 2, Extensive Assistance = 3, Total Dependence = 4.

Activities: Bed mobility, Dressing, Toileting, Transfer, Eating, Grooming, Locomotion.
Sum all column totals to obtain the Total ADL Score.

Appendix B: Semi-Structured Interview Questions

1. Did you find the ADEPP tool useful? Please explain.
2. Was there anything that you disliked about the tool? Please explain.
3. Was there anything that you liked about the tool? Please explain.
4. Would you use it in your practice?
5. Would you recommend the ADEPP tool to your colleagues?
6. What would you change about the tool?
7. Did you feel more comfortable discussing EOL planning with patients and family members?
8. Did your attitude towards hospice care change? Please explain.
9. In your opinion, what are the most significant barriers to hospice utilization?