Approximately half of all United States (U.S.) deaths occur in a hospital with more than twenty percent of all U.S. health care dollars being spent on patients in the last year of their lives (Halpern, 2012). The emotional and financial cost of caring for terminally ill loved ones can be overwhelming, with 10-16% of families having to take out a loan or mortgage, spend all their savings, or obtain another job to cover medical costs. Seventy seven percent of caregivers report missing work due to their caregiving duties (Donley & Danis, 2011). More often, patients and loved ones express the desire to have involvement in decision making related to end of life care and to choose where and how to die, rather than just hope for a peaceful death. Organizations like Aging with Dignity, the Conversation Project, and many others encourage patients, family members, and health care providers to sit down and discuss end of life decisions in advance, before patients are unable to express their wishes for a peaceful death (Aging with Dignity”, 2012). This project explores end of life discussions, identifies a PICOT question, and describes the framework guiding the project. Also included is a review of the pertinent literature and appraisal of the evidence. Finally, it discusses implications for practice and develops an initial plan of action for implementation of advance care planning. END OF LIFE DISCUSSIONS IN THE COMMUNITY 6 Prevalence of Problem Even though medical technology has advanced dramatically since the 1970s, health care providers’ willingness or ability to discuss the course of disease and its trajectory has not advanced (Lokker, 2012). If the patient is aware of the progression of disease and the prognosis or potential outcomes, this knowledge may allow health care providers to help patients evaluate their goals and avoid poorly considered decisions (Lokker, 2012). If patients are only approached near the time of their death to discuss their end of life wishes, then opportunities are lost to help patients be more fully involved in developing advance directives and choosing where and under what circumstances they wish to die. The One Slide is a quick outline of the most important questions that health care providers and patients need to discuss, and that family members need to ask of their loved ones. It provides a minimal script that broaches the subject to get patients, families, and health care providers talking about end of life care, (Appendix A, The One Slide). If at least these questions are asked, then a further discussion can be planned later to get more information or allow a patient to confer with their family. The impact of not having communicated or not developed Advance Care Plan is enormous for the patient as well as the family. The economic impact can be substantial, with one study noting that 10-16% of families with a terminally ill family member have to take out a loan or mortgage, spend their savings, or get another job to cover costs of medical care. Patients can feel guilty about the financial toll their illness places on their loved ones and can feel helpless about the financial fallout that might occur after their death (Donley & Danis, 2011). Emotionally, the impact is more profound. Without Advance Care Planning, patients may die, not at home with their family and friends to comfort them, but in an END OF LIFE DISCUSSIONS IN THE COMMUNITY 7 ambulance, the emergency room (ER), or in the Intensive Care Unit (ICU). Patients and families still pursue unrealistic treatments, desperately hoping for a cure, with the eventual death leaving surviving family members with guilt and regret (Murray & Jennings, 2005). Equally disconcerting, even with all the treatments and medications that are available to alleviate pain and suffering, is that many people still die in severe pain due to lack of provider training, unnecessary regulations that impede practice, and financial barriers that limit access to hospice and palliative care during the course of the illness and at the end of life (Murray & Jennings, 2005). The SUPPORT Trial A landmark two year observational and two year controlled clinical trial published in 1995 sought to improve end of life decision making and decrease frequency of a painful and prolonged dying process. The Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT) (JAMA, 1995) followed 9105 adults with one or more of 9 (nine) life-threatening illnesses with a 6 month mortality rate of 47%. Trained nurses met with patients, families, physicians, and hospital staff to elicit care preferences. The phase I observational trial noted deficiencies in communication, frequency of aggressive care, and characteristics of hospital death. Only 47% of physicians knew when patients wanted to avoid forego cardiopulmonary resuscitation (CPR), 47% of Do Not Resuscitate (DNR) orders were written within 2 (two) days of death, 38% of patients who died spent at least 10 days in intensive care, but more importantly, during the phase II intervention, there was no improvement in communication or promotion of a peaceful death with only 37% of control patients and 40% of intervention patients discussing CPR preferences (JAMA, 1995). The in-hospital death characterized in the SUPPORT trial demonstrated uncontrolled pain, prolonged suffering, and caregiver END OF LIFE DISCUSSIONS IN THE COMMUNITY 8 distress. Significant confusion and moderate anxiety were present in 25% of patients near the end of their life (Collins, Parks, & Winter, 2006). Data from SUPPORT also reveals that surrogates who are asked to demonstrate “substituted judgement” to interpret how the patient, if able to understand circumstances, would make treatment decisions, was poor, with surrogate understanding of patient preferences only slightly better than chance (Collins, et al 2006). Later studies looking at surrogate inaccuracy demonstrated that access to a patient’s Advance Directives did not improve proxy accuracy. Other reasons for the discordance is that surrogates routinely underestimate patient preferences for less aggressive end of life care, their own values impact their decisions, and depression and anxiety are common in surrogates and may affect decision making accuracy (Collins, et al 2006). Even if patients have completed an Advance Directive, and it is available to the health care provider, and the surrogate decision makers are aware of what it states, many times it is not respected. Several barriers affect health care providers’ willingness to discuss end of life care: lack of provider education, inadequate methods to provide information, lack of protocols for communication, reluctance to discuss death, difficulty discussing the unknown, projecting undesirable outcomes, asking patients to make decisions about an uncertain “future self”, lack of shared understanding of values among the patient, family, and health care provider (Waldrop, 2012). A multi-center trial with 53 patients and 43 physicians demonstrated that patients wanted providers to be honest about end of life issues and wanted their physician to play an integral role in discussions and expected outcomes. They also wanted information earlier in the course of their disease (Pfeiffer, Sidorov, Smith, Boero, Evans, Settle, et al, 1994). END OF LIFE DISCUSSIONS IN THE COMMUNITY 9 Unfortunately, even if an Advanced Directive is available, it typically is not used since health care providers are unaware of its existence, or it addresses a hypothetical medical catastrophe that may not be relevant to the patient’s particular situation, or it may speak to interventions like CPR that might help a healthy individual who has a sudden accident or heart attack, but would be of limited benefit to an elderly patient with advanced illness or frailty (Tuohey & Hodges, 2011). Many web sites and organizations have been developed over the last several years to encourage discussion with patients and decision makers about what a patient’s end of life wishes are and how to discuss these with family and other loved ones (See Appendix B, Websites to help end of life discussions). Significance of Problem The Patient Self Determination Act (PSDA), signed in 1991, was a federal mandate allowing patients to accept or refuse treatment, and be given information about ADs on admission to a health care facility. Efforts to promote use of Advance Directives have experienced only limited success, with only 5-15% of patients having a completed Advance Directives (Waldrop, 2012). Advance Care planning is a process of review and discussion among patients, family members, and healthcare providers that focuses on goals of care rather than specific treatments. Advance Care Planning should be initiated by the health care provider well before the patient is terminally ill or develops diminishing cognitive abilities (Waldrop, 2012). Forms that may be used to assist with Advance Care Planning include Living Wills: documents which explicitly state patient preferences, usually regarding life sustaining treatments such as renal dialysis, feeding tubes, antibiotics, mechanical ventilation, and pain control; Durable Power of Attorney for HealthCare (DPOAHC): a documents which names END OF LIFE DISCUSSIONS IN THE COMMUNITY 10 a surrogate decision maker in the event the patient cannot speak for him or herself. Another document developed at the University of Oregon twenty years ago with is gaining broader support in several states, is the Physician Orders for Life Sustaining Care (POLST/MOLST). This is a medical order developed with the patient and health care provider, which follows patients throughout the health system and translates the patient or surrogate decision maker’s precise wishes for end of life care into specific medical orders that pertain to the patients care. The goal of this form is to ensure shared and informed medical decisions. This document is implemented in part or whole in 30 states, with programs being developed in 26 more, including Ohio (Bomba, Kemp, & Black, 2012). Do Not Resuscitate (DNR): a document which states that if a person’s heart or breathing ceases, that only comfort will be provided and attempts to restart the heart or breathing via CPR will not occur. (Dea Moore, 2007). Finally, Allow Natural Death (AND) was developed in 2000 by Reverend Chuck Meyer as an alternative to a traditional DNR, which to some people implies that no care whatsoever will be provided if the patient is dying. To some people, DNR orders are considered “passive orders…, which are a crisis response and not a plan to help a patient die comfortably” (Schlairet & Cohen, 2013). Allow Natural Death orders allow non- interference with a natural process but interventions to be provided or withheld are clearly discussed and documented, (Schlairet & Cohen, 2013). By using Allow Natural Death orders, health care providers can direct patients and surrogate decision makers in more detail to gain an understanding of natural death to help selectively refuse treatments that might prolong life and promote a peaceful death (Schlairet & Cohen, 2013). The benefit to a written health care directive is the patient’s ability to provide clear documentation of his/her intentions (Ward, 2009). The purpose of the discussion however, END OF LIFE DISCUSSIONS IN THE COMMUNITY 11 should not be simply to get forms signed, but to discuss with patients and families their desires and goals regarding treatment, including benefits and risks of treatment, issues related to medically futile care, and withdrawing or withholding treatment (Dea Moore, 2007). Purpose and Goal of Project Purpose Statement The purpose of this Evidence Based Practice (EBP) project was to determine if having a Advance Care Plan discussion with patients and surrogate decision makers using the Respecting Choices PC-Advance Care Planning format increases the likelihood that a patient is able to articulate his or her desires for end of life care to the decision maker and then to have them documented so that the decision maker and provider can understand and advocate for the patient’s wishes regarding treatment and care. PICOT Question (P) Population of interest- patients 50+ years of age in a community living facility (independent or assisted living), with at least one comorbidity and their surrogate decision makers (I)- Intervention of interest- discussion and education of end of life care with patient and decision maker using the Gunderson Lutheran Respecting Choices Patient Centered Advance Care Planning interview (PC-ACP). (C)- Comparison group- patients who do not have an end of life discussion or an Advance Care Plan (O)- Outcome- documentation of end of life discussion and preparing an Advance Directive (living will- durable power of health care, or DNR), modifications in answers END OF LIFE DISCUSSIONS IN THE COMMUNITY 12 regarding end of life preferences by the subject after Respecting Choices PC-ACP interview compared to prior to interview and change in answers for comfort with role of surrogate decision maker after the Respecting Choices PC-ACP interview (T)- Time- 60 days Theoretical Frameworks Glaser and Strauss- Levels of Awareness In 1966, Glaser & Strauss discussed how patients in hospital settings became aware of their impending death and how clinicians tended to avoid discussing it with patients, hoping the patient would figure it out on his or her own. They developed four “awareness contexts” based on the awareness of the health care team, patient and family awareness of the patient’s status, and their recognition of each other’s awareness (Glaser & Strauss, 1966). Closed awareness: the patient is unaware of his terminal state but everyone else is aware. This may be due to the family and health care team belief that the patient might not be able to cope with full disclosure of the severity of the illness. Suspected awareness: The patient suspects what everyone else knows; this may be triggered if the health care team and the family have inconsistent communication and behavior. Mutual pretense awareness: Everyone knows the patient is dying but pretends otherwise. Open awareness: Everyone knows the patient is dying and acknowledges that fact (Glaser & Strauss, 1966). If a person and his family are in closed awareness, then it is unlikely that any discussions about end of life care will occur. Patients who are aware of the imminence of death are more likely to die peacefully and more likely to feel like life was worth living (See Appendix C, Levels of Awareness). END OF LIFE DISCUSSIONS IN THE COMMUNITY 13 This model is crucial to Advance Care Planning and this project. If the patient and surrogate decision makers are not at the same level of awareness, then end of life discussions will not occur and the surrogate decision makers cannot advocate for the patient. Many times, the patient is aware of his increasing debility and severity of illness, but the surrogate decision maker is not, leading to lack of communication about the benefit or futility of continuing medical treatments related to the patient’s quality of life goals. Trajectories of Illness One hundred years ago, most deaths were sudden, usually caused by accidents, infection, or complications from childbirth. In developed society, deaths are usually attributable to serious progressive illnesses (cardiovascular or pulmonary disease, or cancer), which limits a person’s usual activities increasingly until death occurs. (Murray, Kendal, Boyd, & Sheikh, 2005). Three distinct illness trajectories have been developed, which are crucial to helping a patient and family prepare for end of life planning (see Appendix A). Progressive Chronic Illness: usually cancer, involved progressive predictable decline over weeks, months, years, with weight loss and increasing debility occurring in the last months before death. This generally provides time to anticipate palliative needs and provide end of life planning. Organ Failure: This trajectory covers patients with cardiovascular, pulmonary, or other chronic organ failure; this is characterized by long periods of chronic illness, with occasional, acute exacerbations requiring intensive treatment with possible hospitalization. Each exacerbation may result in death; if the patient survives the episode, there is a progressive deterioration overall, but the timing of death is not clear. There may be periods of acute deterioration, some recovery, and sudden (perhaps unexpected) death. Finally, there is Progressive Disability: This is characterized by either END OF LIFE DISCUSSIONS IN THE COMMUNITY 14 brain failure related to dementia, or generalized fragility, where patients loses weight and functional capacity, then finally succumbs to a relatively minor event that would not affect a health person, but in conjunction with declining abilities may prove fatal to this patient. Death then occurs unexpectedly after a sudden illness or acute event. Trajectories of illness are important to Advance Care Planning and this project since if a patient and Surrogate decision maker understand the disease process and likely course of illness, they are more able to anticipate end of life needs and possible time periods to consider, as well as what treatment options would help or hinder the patient’s quality of life goals (See Appendix C, Trajectories of Illness). The Stetler Model The Stetler model will be used to guide this project and has five steps, preparation, validation, comparative evaluation/decision making, and translation/application. This model defines evidence as “information or facts that are systematically obtained” and this evidence can come from different sources and can vary in the degree it is systematically obtained (Ciliska, 2011). The Stetler model, (See Appendix D), is known as a practitioner oriented model because of its focus on critical thinking and ability to be used by an individual practitioner (Ciliska, 2011). END OF LIFE DISCUSSIONS IN THE COMMUNITY 15 Steps to using the Stetler model are: 1) Preparation: define and affirm a priority need- search, sort and select sources of research and supplemental evidence, consider influential factors, affirm nature, degree, and priority of problem, and define purpose and outcomes per issue. 2) Validation: assess evidence by critiquing each study with a utilization focus in mind. Perform utilization focused critique and synopsis per source, synthesize findings, and evaluate common aspects per criteria. 3) Comparative evaluation/decision making: making decisions about use after synthesizing the data using a set of utilization criteria then deciding if and what to use based on the identified need. (Fit and qualifiers/ feasibility/substantiating evidence/current practice). State decisions re: use of findings per strength of overall evidence. 4) Translation/application: converting the findings into a type of change to be made, planning application, putting plan into action, and then enhancing adoption and implementation with evidence based change plan. Confirm type, level and method per details/qualifiers. 5) Evaluation: evaluation of plan in terms of degree of implementation and whether goals for using evidence were met. Evaluate dynamically, identify goals for each “use,” obtain evidence re: the implementation, approach, targeted practice or system changes, and the result/outcomes (Melnyk & Fineout-Overholt, 2011). This model was applicable for the project since it is designed for a sole practitioner using critical thinking. This model can be used in a practice setting and allows for variation END OF LIFE DISCUSSIONS IN THE COMMUNITY 16 in the context of an individual patient’s preferences and circumstances (Ciliska, 2011), which are necessary when discussing end of life decisions with patients and loved ones. Review of Literature Search Strategies Databases searched included OhioLink, Google Scholar, Ebsco, Proquest, and EJC. These databases were utilized because they represent a wide range of healthcare related journals. The literature search began in February, 2013 and was ongoing during the project. Key words for search included “End of Life” and “Primary care” and “Primary care “ and “discussions, ”Do Not Resuscitate,” “advance directives,” “patient preferences”,, “protocols”, “code status,” “advance care planning”, and “palliative care,” POLST, trajectories of illness, trajectories of dying, ALLOW Natural Death, end of life guidelines, protocols and palliative care, Medical Directives, and health care directives. Critical Appraisal and Evaluation of the Evidence For this project, articles were reviewed for the following: time relevance (since 2000), primary care focus, tools to improve end of life discussion, and research articles or protocols and guidelines to implement end of life discussions in primary care. Research was excluded if it was published prior to 2000 unless it was a seminal document or it was not primary care based. Research was excluded if it singularly focused on emotional aspects related to dying or palliative care. One study (Schwartz, Wheeler, Hammes, Basque, Edmunds et al (2002) was evaluated using the AGREE too (See Appendix E Literature Synthesis). Literature was searched from 2000-2014 looking at studies related to Advance Care Planning and Advance Directives. Terms used included “primary care” + “advance END OF LIFE DISCUSSIONS IN THE COMMUNITY 17 directives/”end of life”/”DNR”, POLST, Allow Natural Death (AND), /”protocols”/”treatment preferences”/”palliative care/medical directives/Advance care planning.” The most successful research was obtained through OHIOLINK, though Cochrane database and EvidenceUpdate also provided data. Because of the landmark nature of the SUPPORT trial, it was included even though it was outside the desired time frame. Articles Included: Any articles pertaining to completion of Advance Directives, barriers to completion, stability of end of life preferences over time, and emotional aspects to promote end of life discussions. Because discussing end of life care is so emotionally fraught, literature on factors to promote Advance Directives discussions was included as well as those focusing on Primary Care Provider’s role in promoting discussion. Also included is research on Advance Directives that discussed whether surrogate decision makers could understand an AD from a loved one. Literature which focused on the AD process (not just the form) was also included. Literature excluded: was written before 2000 unless seminal research (SUPPORT trial), theoretical articles that were not research based, articles linking AD with other topics such as organ donation, articles on palliative care that did not address AD completion, and any research solely discussing hospice, since hospice is a more type of end of life care, and not specifically related to completion of an Advance Directive. After evaluating the available research, it is evident that promoting discussions regarding Advance Directives and its preparation remains problematic. However, the Respecting Choices Advance Care Planning program developed at Gunderson Lutheran Medical Center is showing great promise. An early study by Schwartz, Wheeler, Hammes, Basque, Edmunds et al (2002) using this program illustrates positive results in 61 END OF LIFE DISCUSSIONS IN THE COMMUNITY 18 ambulatory geriatric patients with the majority of patients/surrogate decision makers showing congruence in their understanding of patient goals of END OF LIFE treatment. However, this study is limited by the small sample size. Two studies by Kirchhoff, Hammes, Kehl, Briggs, and Brown (2010, 2012) also have shown success in improving patient end of life care with regards to their desired wishes in surrogate decision makers (2010), and in patients with specific medical diagnoses both using 313 patient/decision maker pairs. More recently, a retrospective study in 2014 looking at the use of Respecting Choices among 722 hospitalized patients looked at Advance Directives in minorities compared to white patients did find significant improvement in the use of Advance Directives among minorities, (25.8 & to 38.4%), though not in white patients, (35.9 % to 42.9%), with high consistency for all orders (74%-96%) (Pecanac, Repenshek, Tannenbaum, and Hammes, (2014). Recommended Practice Change End of life discussions are not always effective despite the use of multiple approaches. While much research identifies the need for end of life discussions, and desire of patients to have the conversation while still healthy, patients and family members do not have the discussion and health care providers often do not initiate the conversation. Even if the discussion does occur, the family is often unable or unwilling to advocate the patient’s wishes to the health care providers, the patient is unable to advocate for himself, or the health care provider does not follow the advance directives. Rationale from Literature The Agency on Healthcare Research on Quality (AQHR) has published protocols and guidelines on AD in nursing homes, for disease management, and for patients with cancer, END OF LIFE DISCUSSIONS IN THE COMMUNITY 19 but many of the interventions used in research only modestly increased the number of advance care planning discussions or preparation of Advance Directives. Most evidence is level 3 or 4, and only 4 random controlled studies (RCT) were found in the literature, likely due to the sensitive nature of the topic. Three of the studies, (Schwartz, 2002, Kirchhoff et al, 2010 and 2012), did find significant improvement in Advance Directive preferences being documented and followed using a training program Patient Centered- Advance Care Planning from Gunderson Lutheran Health system. Most review of literature focused on emotional aspects of end of life care, barriers to discussions both for patients and providers, and how patients and providers felt about having or not having a discussion about end of life care. Studies by the Agency for Healthcare Research & Quality show that patients are willing to discuss end of life care but most patients have not participated in those discussions with their health care providers. Patients are more likely to accept or refuse treatments based on how invasive the treatment might be or how long the treatment might last. They are more likely to accept CPR vs. long term ventilation if they are in their current state of health. If they are given a hypothetical situation involving a stroke, fewer patients want either CPR or long term ventilation (Advance Care Planning, Preferences for Care at End of Life, 2003). They recommend that hypothetical scenarios be used to help patients identify their opinions on certain treatment patterns (Advance Care Planning, 2003). The following steps are recommended to encourage discussion about Advance Directives: 1. Initiate a guided discussion: use hypothetical scenarios related to a patient’s particular situation and find out patient preferences under certain circumstances END OF LIFE DISCUSSIONS IN THE COMMUNITY 20 2. Introduce the subject of Advance Care Planning and offer information: encourage patient to fill out both a Living Will and Durable Power of Attorney for Health Care. 3. Prepare and complete Advance Care Plan documents: standard language is often not specific enough to be helpful at directing care. If a patient does not want a ventilator, the provider should clarify if that is true under all, or only some, circumstances. 4. Review the patient preferences regularly and update documentation: remind the patient that documents can be revised at any time and the provider should note what preferences change, and what stays the same 5. Apply the patient’s desires to actual circumstances: Patient’s treatment was considered generally consistent with their preferences if those were clearly stated in an Advance Directives and the provider was aware the preferences. Even if a decision is needed for an unanticipated situation, providers and surrogate decision makers can make an educated guess based on knowledge about the patient’s goals and thresholds for treatment (Advance Care Planning, 2003). Research has shown that patients are more likely to accept treatment for conditions they consider better than death, and refuse treatment for conditions they consider worse than death (Advance Care Planning, 2003). Options that are more invasive (mechanical ventilation or dialysis) may be less acceptable then less invasive treatments (CPR or antibiotics) (Advance Care Planning, 2003). Very little research has been done on how to promote end of life discussion or completion of Advance Directives in healthy adults who live in community living END OF LIFE DISCUSSIONS IN THE COMMUNITY 21 settings. Recent studies in 2010 and 2012, as well as an earlier study in 2002, have showed promise by presenting information more focused towards patients’ end of life wishes and goals rather than simply educating about the need to fill out specific forms. None of the studies however, were done in a community setting to evaluate if the Respecting Choices Advance Care Planning Facilitator training would enhance discussion and decision making. In addition, the studies were done in Wisconsin, which has a wider acceptance for Advance Directives due to the work of Gunderson Lutheran in promoting Advance Directives, and a smaller minority population. Of note, the Agency for Healthcare Research & Quality innovation webpage (2013) discusses the use of the Respecting Choices training program and its effect on Advance Care Planning, not only in Wisconsin, but in Minnesota and Australia. Implications for Practice The Agency for Healthcare Research & Quality recommends that patients be asked about end of life preferences at regularly scheduled outpatient visits and when a provider could ask him or herself if they would be surprised if a patient had died in the next year, or within 6 months (Advance Care Planning, 2003). Patients should be screened at their initial visit and if they say they have an Advance Directive, it should be obtained and documented in the patient chart. The health care provider should then ask about and document the presence and type of advance directive. The health care provider should then make every effort to schedule an appointment at a later date to discuss the Advance Directive in detail and document the findings. The non-professional staff can print out and give any desired forms to the patient and answer simple questions. If needed, the questions can be routed to the provider through a telephone consult so the question can END OF LIFE DISCUSSIONS IN THE COMMUNITY 22 be answered in a timely fashion. Existing patients should have their charts reviewed at least annually during their Medicare wellness visit or wellness visit by their private insurance carrier. Patients who only come in for ill visits should be asked about an Advance Directive and asked to provide one if available, or schedule an appointment to discuss. Methods Implementation Initially this project was set for a family practice setting in Tallmadge, Ohio, and for the year prior to the study, attempts were made to educate the patient population on the need for Advance Care Plan during scheduled office visits. Patients were asked, “Who is the person who would be your medical decision maker if you could not speak for yourself?” or “have you discussed with them what you would want?” Patients were then offered information from the brochure “Conversations that Light the Way” (www.lucas-co-probate- ct.org/forms/MiscCourtForms/conversations.pdf) and the Ohio Living Will and Durable Power of Attorney-HealthCare if they were interested. Most patients admitted they had not considered an Advance Care Plan, though many were aware of its necessity. Numerous patients were grateful for the opportunity for assistance to have the discussion and document their wishes, or for the opportunity to discuss wishes of family members. However, several barriers affected the ability to proceed with this process. Elderly patients who came in for acute medical visits were not always interested in the discussion, and patients who came in for chronic medical visits already had numerous other issues to address. Family members of elderly patients were often suspicious as to why this question was being asked of their 98 year old mother when she just “came in for a cold.” Time END OF LIFE DISCUSSIONS IN THE COMMUNITY 23 constraints were also a limiting factor as these discussions tended to take more than the allotted time. Asking patients to come back to have this discussion was problematic as many patients were on fixed incomes and thus could not afford another copay, did not have transportation and had to rely on others, or simply were not concerned. Even more challenging was the lack of administrative support. Though other physicians in the practice agreed that it was a necessary discussion, they were not willing to address the sensitive topic of advance care planning to their patients. Another concern was that the practice owner did not feel patients would be receptive a formal interview discussing advance care planning and that the project may have a negative impact on the practice. Fortunately, National Church Residences, whom the researcher had used for other clinical projects, recognized the need for this project and the benefits of advance care planning as a critical aspect of providing holistic care for its’ residents, and graciously stepped forward to implement the project. Respecting Choices Program The Respecting Choices Patient Centered Advance Care Planning facilitator certification is a copyrighted training program developed by Gunderson Medical Foundation which includes several hours of online training and continued training onsite. The goal is to train interested persons in protocols to initiate discussions with patients and decision makers through a series of questions to help direct the process of Advance Care Plan by identifying decision makers, goals of treatment for the patient, and identifying any religious, personal, or cultural views that would affect treatment choices. The program acknowledges that the discussion is not a “static” one time discussion, but is fluid and needs to take place over time as the patient’s health changes. A key element is identifying the surrogate decision END OF LIFE DISCUSSIONS IN THE COMMUNITY 24 maker and his/her willingness to advocate for the patient should it become necessary. (First Steps ACP facilitator training manual). Because this Patient Centered-Advance Care Plan is a structured interview technique, facilitators are encouraged to adhere the wording of the script. If a facilitator does not know an answer about the client’s health care, likely course of disease, or possible outcomes they are instructed to help the client make a list of questions for his healthcare provider. The Health Care Directive The Health Care Directive (See Appendix L) was a tool initially developed by Linda Emanuel in 1989 (Emanuel & Emanuel, 1991), and modified in 2004, to help individuals develop a plan of care for which interventions (CPR, ventilator support, artificial nutrition, surgery, dialysis, blood transfusion or simple diagnostic testing) they would want if they had a medical emergency, and were unable to speak for themselves. These interventions were then graded on a scale of 1-3; with 1 being “comfort care only”; 2 being “attempt cure” and 3 being “prolong life”. There were six (6) scenarios listed: 1. Coma and no hope of recovery 2. Coma, with small chance recovery, moderate chance of some disability, and large chance no recovery 3. Non reversible brain injury and can’t communicate or care for self and terminally ill 4. Non reversible brain injury, can’t communicate or care for self and not terminally ill END OF LIFE DISCUSSIONS IN THE COMMUNITY 25 5. Incurable chronic condition with physical or mental suffering and will eventually cause death, and a life threatening but potentially reversible condition 6. My usual state of health with a life threatening but potentially reversible condition Agent Comfort Form The Agent Comfort Form (See Appendix M) was developed by Delta Quest and asked six (6) questions on how comfortable an Agent (the surrogate decision maker) felt being a decision maker for an individual (the Principal). The answers were then scored on a range of 0-4 with 0 being not at all, 1 being a little bit, 2 being somewhat, 3 being quite a bit, and 4 being very much. The questions were: 1. I feel comfortable being a decision maker 2. I am concerned I will not make decisions reflecting my Principal’s wishes 3. I believe I understand my Principals preferences 4. I am not confident I will be able to make important medical decisions in a crisis. 5. I feel adequately informed about my Principal’s wishes. 6. The responsibility I have for my Principal’s health care decisions is overwhelming. Project Setting/Population The setting for this project was senior living communities in the Summit County area managed or owned by National Church Residences (NCR). Subjects were recruited by the Service Coordinators of the buildings using data from the Care Path documentation program used by National Church Residences. Residents had a Mini Mental Status Exam (MMSE) on file as these are done routinely on residents. The project population included any patient 50 and older who had Chronic Obstructive Pulmonary Disease (COPD), Atherosclerotic Heart END OF LIFE DISCUSSIONS IN THE COMMUNITY 26 Disease (ASHD)/Congestive Heart Failure (CHF), or dementia. Patients and surrogate decision makers were approached and enrolled into the study as a pair. Residents and surrogate decision makers signed consent. The Health Care Directive Form, (see Appendix 11), was given to the patient and an Agent Comfort Form was given to the surrogate decision maker (also referred to as the Agent), (see Appendix 12), before the intervention to see if the patient and decision maker could articulate what the patient (the Principal) would want in terms of end of life care and how the surrogate felts about his/her competency in fulfilling this role. A structured interview was performed and the tools were completed again, after the intervention. Five buildings were used to recruit subjects with a total of nearly 400 potential subjects available for the project. A total of 22 subjects and SDMs completed the intervention and questionnaires. Human Subject Protection Human rights were protected by selecting persons who had a Mini Mental Status Score (MMSE) score 23 and above, and who could understand the consent form, sign and participate without undue distress. The surrogate decision maker also signed a consent form. The project was reviewed and approved by the University of Toledo (UT) Institutional Review Board (IRB). Implementation Plan The service coordinators distributed information sheets that described the project and who was eligible to participate. Residents who were interested in participating contacted the service coordinators. In addition, informational meetings were held in five buildings in Summit, Portage, and Stark Counties to inform potential participants of the purpose or the proposed project. Service coordinators assisted to contact and set up appointments to recruit END OF LIFE DISCUSSIONS IN THE COMMUNITY 27 interested participants and surrogate decision makers. Service coordinators made sure there was a documented MMSE on file, and any resident who scored less than 23 was excluded from the study. The investigator contacted the residents to sign the consent and give the Health Care Directive and set up a time to consent the surrogate and give the ACF. The Service Coordinators assisted residents if they had questions about forms and helped collect the forms. The investigator interviewed the resident and surrogate decision maker using Respecting Choices protocol and had them complete a 2nd Health Care Directive and Agent Comfort Form. Barriers and Facilitators Initial barriers that required intervention included a lack of willing participants initially when they were not directly approached by the Service Coordinator and encouraged to participate in the study. Residents did not understand the project even after explanation; “I already have my will done”, or insisted that even though they had a decision maker, they would not have time to participate. Several residents reported they had no family or decision maker to help them. Retrieving completed paperwork was also problematic since SDMs were not always prompt in returning their forms. Another barrier was that the Health Care Directive was difficult at times for some residents to fill out, and the Agent Comfort Form had some confusing questions that required some time to answer. The time needed was overestimated and tended to be a barrier to potential subjects. Interviews were scheduled on weekends and evenings to accommodate schedules of decision makers. Service Coordinators who used a more direct approach of actively recruiting residents and explaining why the intervention was needed tended to be more successful at recruiting residents and decision makers. Cost remained an issue since there was no financial assistance and the tools from END OF LIFE DISCUSSIONS IN THE COMMUNITY 28 Gunderson were purchased from the Respecting Choices Web site. Facilitators were the Service Coordinators, who understood the need for this intervention, and the NCR Corporate office in Columbus was very supportive of the project (See Appendix F, Barriers and Facilitators). Implementation Process using Stetler Model Validation.