Palliative Care in the Outpatient Setting A Comparative Effectiveness Report Final Report April 27, 2016 Completed by: Institute for Clinical and Economic Review ©Institute for Clinical and Economic Review, 2016 AUTHORS: Courtney Cunningham, MPH Program Director Karin Travers, DSc Research Director Rick Chapman, PhD, MS Director of Health Economics Anne Loos, MA Research Associate Erin Lawler, MA Program and Communications Associate Shanshan Liu, MS, MPH Research Associate Steven D. Pearson, MD, MSc President Daniel A. Ollendorf, PhD Chief Scientific Officer DATE OF PUBLICATION: April 27, 2016 ICER would like to thank Rebecca Kirch, JD, Consultant to the Center to Advance Palliative Care for her peer review of the draft report. ©Institute for Clinical and Economic Review, 2016 Page i About ICER The Institute for Clinical and Economic Review (ICER) is an independent non-profit research organization that evaluates medical evidence and convenes public deliberative bodies to help stakeholders interpret and apply evidence to improve patient outcomes and control costs. ICER receives funding from government grants, non-profit foundations, health plans, provider groups, and health industry manufacturers. Through all its work, ICER seeks to help create a future in which collaborative efforts to move evidence into action provide the foundation for a more effective, efficient, and just health care system. More information about ICER is available at http://www.icer- review.org About New England CEPAC The New England Comparative Effectiveness Public Advisory Council (New England CEPAC) – a core program of ICER – provides a public venue in which the evidence on the effectiveness and value of health care services can be discussed with the input of all stakeholders. New England CEPAC seeks to help patients, clinicians, insurers, and policymakers interpret and use evidence to improve the quality and value of health care. The New England CEPAC is an independent committee of medical evidence experts from across New England, with a mix of practicing clinicians, methodologists, and leaders in patient engagement and advocacy. All Council members meet strict conflict of interest guidelines and are convened to discuss the evidence summarized in ICER reports and vote on the comparative clinical effectiveness and value of medical interventions. More information about New England CEPAC is available at http://icer-review.org/programs/new-england-cepac/. ©Institute for Clinical and Economic Review, 2016 Page ii Table of Contents Executive Summary ............................................................................................................................ ES1 1. Background ........................................................................................................................................ 1 1.1 Introduction ................................................................................................................................. 1 1.2 Scope ............................................................................................................................................ 2 2. The Topic in Context .......................................................................................................................... 6 2.1 History and Evolution of Palliative Care Services......................................................................... 6 2.2 Structure and Settings of Palliative Care ..................................................................................... 8 2.3 Barriers and Opportunities ........................................................................................................ 12 2.4 Existing Approaches to Palliative Care Outside of the Hospital Setting .................................... 20 3. Summary of Coverage Policies ......................................................................................................... 24 4. Comparative Clinical Effectiveness .................................................................................................. 27 4.1 Overview .................................................................................................................................... 27 4.2 Methods ..................................................................................................................................... 28 4.3 Results ........................................................................................................................................ 32 4.4 Elements for Successful Palliative Care Programs ..................................................................... 48 5. Comparative Value ........................................................................................................................... 53 5.1 Overview .................................................................................................................................... 53 5.2 Prior Published Evidence on Costs and Cost-Effectiveness of Outpatient Palliative Care Programs .......................................................................................................................................... 53 5.3 Potential Budget Impact of Outpatient Palliative Care Programs ............................................. 56 5.4 Summary and Comment ............................................................................................................ 59 6. Voting Results .................................................................................................................................. 61 6.1 About the New England CEPAC Process .................................................................................... 61 6.2 Clinical Effectiveness Voting Results .......................................................................................... 65 6.3 Care Value Voting Results .......................................................................................................... 65 6.4 Provisional Health System Value Voting Results ....................................................................... 66 7. Recommendations to Guide Policy and Practice ............................................................................. 67 For Payers......................................................................................................................................... 67 For Providers .................................................................................................................................... 69 For Patients and Families ................................................................................................................. 70 ©Institute for Clinical and Economic Review, 2016 Page iii For Policymakers .............................................................................................................................. 70 Future Research Needs .................................................................................................................... 71 References ........................................................................................................................................... 73 A1. Palliative Care Compared to Hospice ............................................................................................ 74 A2. Definitions of Palliative Care.......................................................................................................... 75 A3. Medicare Care Choices Model ....................................................................................................... 76 A4. Search Strategies ........................................................................................................................... 78 A5. Clinical Guidelines .......................................................................................................................... 83 A6. Previous Systematic Reviews ......................................................................................................... 86 A7. Ongoing Studies ............................................................................................................................. 90 A8. Evidence Tables ............................................................................................................................. 95 A9. Key Informant Interviews ............................................................................................................ 120 A10. Policy Roundtable Participants .................................................................................................. 121 ©Institute for Clinical and Economic Review, 2016 Page iv List of Abbreviations Used in this Report ABMS American Board of Medical Specialties ACA Affordable Care Act ACGME Accreditation Council for Graduate Medical Education ACHPN Advanced Certified Hospice and Palliative Nurse ACHP-SW The Advanced Certified Hospice and Palliative Social Worker AIM Advance Illness Management APN Advanced Practice Nurse ATT Average Treatment Effect on Treated Patients CAPC Center to Advance Palliative Care CARES-MIS Cancer Rehabilitation Evaluation System Medical Interaction Subscale CCA Commonwealth Care Alliance CCCC Coalition for Compassionate Care of California CES-D Center for Epidemiologic Studies Depression Scale CG Caregivers CHF Chronic Heart Failure CHPN Certified Hospice and Palliative Nurse CHPPN Certified Hospice and Palliative Pediatric Nurse CHPVA Certified Hospice and Palliative Care Administrator CMS Centers for Medicare and Medicaid Services COPD Chronic Obstructive Pulmonary Disease CPLC Certified Nurse in Perinatal Loss Care DHMC Dartmouth Hitchcock Medical ED Emergency Department ESAS Edmonton Symptom Assessment Scale FACIT-Pal Functional Assessment of Chronic Illness Therapy- Palliative Care FACIT-Sp Functional Assessment of Chronic Illness Therapy- Spiritual Well-Being FACT-G Functional Assessment of Cancer Therapy- General FACT-L Functional Assessment of Cancer Therapy- Lung FAMCARE-p16 The Family Satisfaction with End-of-Life Care (16-item measure of patient satisfaction) FFS Fee For Service GME Graduate Medical Education HADS-A Hospital Anxiety and Depression Scale of anxiety HADS-D Hospital Anxiety and Depression Scale of depression HCFA US Health Care Financing Administration HMP Hospice and Palliative Medicine HRQoL Health-Related Quality of Life HRSA Health Resources and Services Administration HRU Health Care Resource Utilization IAHPC International Association for Hospice and Palliative Care IPAL-OP Improving Outpatient Palliative Care Project LASA Linear Analog Scales of Assessment LCS Lung Cancer Subscale LOS Length of Stay NR Not reported NS Not significant MBCB Montgomery- Borgatta CG Burden ©Institute for Clinical and Economic Review, 2016 Page v MGH Massachusetts General Hospital mITT modified Intent-To-Treat MS Multiple Sclerosis MSW Master of Social Work NHPCO National Hospice and Palliative Care Organization NIH National Institutes of Health NSCLC Non-Small Cell Lung Cancer PCLC Palliative Care Leadership Center PC Palliative care PCP Primary Care Physician PHQ-9 Patient Health Questionnaire POM Profile of Mood States PPCHETA Palliative Care and Hospice Education and Training Act PPS Palliative Performance Scale PROMIS Patient Reported Outcomes Measurement Information System PRISMA Preferred Reporting Items for Systematic Reviews QoL Quality of Life QUAL-E Quality of Life at the End of Life RCT Randomized Control Trial TOI The Trial Outcome Index USPSTF U.S. Preventive Services Task Force ©Institute for Clinical and Economic Review, 2016 Page vi Executive Summary Background Palliative care is a management approach that provides symptom relief and comfort care to patients with serious or life-threatening illnesses, with the goal of improving quality of life (QoL) for both patients and their families.1 Unlike hospice care, which is typically restricted to individuals with a prognosis of survival of six months or less, palliative care can begin at diagnosis and is often provided along with treatment aimed at prolonging life, such as chemotherapy or radiation therapy for cancer.2 One of the primary objectives of palliative care is to help patients prioritize their goals of care, and it may include conversations around advance care planning (e.g., a “living will”) depending the anticipated disease trajectory.3 Topic in Context In summarizing the contextual considerations for outpatient palliative care, we seek to describe the following:  The history and evolution of outpatient palliative care in the U.S.  A description of settings in which outpatient palliative care is provided  An overview of barriers and opportunities impacting the provision of palliative care as described by national and regional experts and the policy literature  A snapshot of selected programs to shed light on existing approaches to outpatient palliative care History Palliative care in the U.S. grew largely out of the hospice movement. As life expectancy has increased and the number of individuals with chronic, advanced, or serious illness continues to rise, so has the need for care that is targeted towards improving quality of life (QoL) through relief of physical and psychological symptoms associated with life-threatening illnesses. Palliative care was initially and continues to be offered predominately in the inpatient setting. However, due to the growing recognition of perceived barriers related to access, many feel that the next step in the evolution of palliative care is expansion of services into the outpatient setting. Individual health systems have begun to identify ways to expand access through programs focused on providing palliative services beyond the hospital setting in outpatient clinics, offices, and community- and home-based environments. This movement is intended to meet the needs of the growing number of patients who would benefit from palliative care services throughout the course of illness and care continuum.4 Figure ES1, taken from Clinical Practice Guidelines for Quality Palliative Care, outlines the progression of palliative care.1 ©Institute for Clinical and Economic Review, 2016 Page ES1 Figure ES1: Palliative Care’s Place in the Course of Illness Graphic from the National Consensus Project for Quality Palliative Care For the purpose of our evidence review, palliative care was defined by services that incorporate elements directed at both physical and psychosocial patient care, as both are considered essential and definitional components of palliative interventions.5,6 Given the variability in health care systems and delivery worldwide, we have focused our evidence review on publications from Canada and the U.S. only. Settings of Care Palliative care services are administered in a number of settings, including hospitals, outpatient clinics, skilled nursing facilities, and in the home.4 Because of the growing demand for palliative care, due primarily to the increasing number of patients living with chronic or serious illness, the field has evolved to include a number of approaches to maximize delivery of these services. Given that these populations are spending increasingly more time outside of the hospital, our report focuses on services provided in the outpatient setting, which has been identified by many in the field as the next frontier in palliative medicine.7 Structure of Care While the structure of how palliative care services are provided varies across population types and settings, two dominant models have emerged. For the purposes of this report, these models are defined as generalist palliative care and specialist palliative care. Generalist Palliative Care is care provided by professionals who have some clinical experience and basic training in palliative care concepts but whose primary specialty is not palliative care. These professionals may include primary care physicians, generalists, oncologists, nurse practitioners, and professionals in other disciplines. These providers conduct needs assessments, educate patients about their disease, and provide basic symptom management and psychosocial support. Generalist palliative care (also sometimes called primary palliative care) can be provided by individual providers alone or in conjunction with a multi-disciplinary team (e.g., social workers, chaplains, ©Institute for Clinical and Economic Review, 2016 Page ES2 etc.); the main differentiating feature from specialty palliative care is the absence of a clinician with advanced training/certification in hospice and palliative care.8 Specialist Palliative Care involves hospice and palliative medicine (HPM) board certified clinicians and advanced practice nurses certified in hospice and palliative nursing. Specialist palliative care providers are those who have extensive training and experience in palliative care and focus on those more complex aspects of disease management, including controlling refractory physical and emotional symptoms and worsening depression or anxiety; assisting with conflict resolution; identifying and communicating patients’ goals and preferences for care; and counseling around issues of end of life planning and bereavement.8 Barriers and Opportunities There are a number of barriers impeding the widespread use of outpatient palliative care. It is important to recognize that the landscape for palliative care is constantly evolving; therefore, our summary of barriers and opportunities should be viewed as a “snapshot” of the status at the time of the report’s publication. Through our evaluation we identified barriers that affect access to palliative care, including:  Insufficient workforce capacity and a need for additional training and education opportunities for palliative care providers  A lack of awareness of palliative care services among non-specialty providers  Geographic and socioeconomic access disparities  Limited knowledge about palliative care services among patients and families  Limited availability of robust data on the impact of palliative care  Scant funding opportunities for palliative care research  Complex billing processes and inadequate reimbursement for palliative care services We also identified a number of opportunities that may contribute to the ongoing evolution of outpatient palliative care. In our overview we discuss opportunities related to:  New payment methodology  Use of electronic medical records to encourage care coordination  Patient and provider education  Resources for existing providers and health systems  Legislation aimed at eliminating barriers to providing palliative care All of these barriers and opportunities are discussed in comprehensive detail in the full report. ©Institute for Clinical and Economic Review, 2016 Page ES3 Existing Models for Outpatient Palliative Care There are a number of ways in which palliative care programs function in the outpatient setting. Several existing programs provide exemplars and context for how palliative care is organized and coordinated through health systems. We found that most outpatient programs are operated by hospital systems, though some models involve an integrated approach through an accountable care organization (ACO) framework. Several examples of programs available in New England are provided in the full report. Comparative Clinical Effectiveness of Outpatient Palliative Care Programs Our review of published evidence examined higher quality comparative studies of palliative care interventions delivered outside the hospital. Interventions of interest were those implemented in outpatient settings that incorporate elements directed at both physical and psychosocial patient care, as both are considered essential and definitional components of palliative interventions.5,6 We recognized that such studies vary substantially in terms of their entry criteria and description of the interventions. In order to inform our analysis of the comparative clinical effectiveness of outpatient palliative care, we defined several characteristics of the interventions as of a priori interest including:  the timing of the intervention (whether it was specified as early [as defined in relation to time of diagnosis or minimum period of time prior to death] or not);  the target population of the intervention (patients with cancer only or patients with mixed diagnoses, given the interest in expanding palliative care to patients with other progressive diseases);9 and  the type of service provided in the intervention (i.e., whether the level of provider training was characterized as specialist [with input from care providers described as having a palliative care specialty or certification] or generalist [with input from experienced care providers without reference to a specialty or certification]). Our review focused on specific clinical benefits reported in terms of patient QoL, mood, symptom burden, patient satisfaction, health care resource utilization (HRU), psychosocial outcomes, caregiver impact, and survival, as an outcome that might indirectly result from other intervention outcomes. ©Institute for Clinical and Economic Review, 2016 Page ES4 Overall Evidence The body of evidence derived from higher quality comparative studies of outpatient palliative care interventions in the U.S. and Canada is somewhat constrained and variable with respect to intervention target and structure. We identified 14 good and fair quality studies describing a variety of outpatient palliative care interventions. Specialist interventions were described by 10 studies,10-22 while generalist interventions were described by four.23-26 We note, however, that most of the interventions described as “generalist” appear to be multi-disciplinary, follow core principles of palliative care, and are delivered by experienced personnel, making the distinction between these approaches and those led by certified professionals somewhat artificial. Four studies10,11,21,22 evaluated the effect of palliative interventions described as early while 10 studies12,13,17-20,23-26 did not specify early initiation of the intervention. Five studies10,11,21,22,25 evaluated outpatient palliative care’s effect on patients with cancer while nine studies12,13,17-20,23,24,26 did so in populations of patients with mixed diagnoses. Although we sought to identify any potential harms associated with palliative care, the studies included in our review did not report any adverse events connected with the interventions; therefore, our evaluation of net benefit is based solely on the noted benefits. Overall, the evidence describing outpatient palliative care’s benefit is stronger for QoL, resource utilization outcomes, patient satisfaction, and mood outcomes, with more limited evidence suggesting benefits on survival, symptom burden, psychosocial, and caregiver outcomes. These observations are consistent with the fact that outpatient palliative care programs are designed to increase patient social support, patient self-advocacy, and coordinated medical care; while palliative care is not focused on improved survival as an indicator of effectiveness, the survival benefit may be mediated by the other more directly influenced outcomes of interest (e.g., mood, QoL). Table ES1 captures the strength of evidence around the relative effect of outpatient palliative care compared to usual care for the key outcomes of interest, with evidence around the more granular aspect of the intervention type (specialist vs. generalist). We present the overall numbers of studies that reported on each outcome by type of intervention. ©Institute for Clinical and Economic Review, 2016 Page ES5 Table ES1. Strength of Evidence for Outpatient Palliative Care Compared to Usual Care Outcome Specialized care Generalist care Overall QoL 5 studies 2 studies 7 studies Resource Use 8* studies 3 studies 11 studies Survival 3* studies 1 study 4 studies Symptom Burden 5 studies 2 studies 7 studies Patient Satisfaction 3 studies 1 study 4 studies Psychosocial and Spiritual 1 study 2 studies 3 studies Mood 4 studies 2 studies 6 studies Caregiver Outcomes 1 study 0 studies 1 study * Bakitas 2015 was not included in these results because the primary comparison for these outcomes was early vs. delayed palliative care. Color Code: High Moderate Low No evidence Many studies lacked detailed descriptions of the interventions, precluding analysis of the effect of specific elements previously identified by professional bodies as being essential to palliative care. In an effort to define potentially contrasting characteristics identified within the interventions included in our review, we noted that there were interventions which could be classified as specialist or generalist levels of palliative care, based on the definitions offered by Quill et al.8 Such specialist interventions reported a higher level of benefit on QoL and resource utilization than did the generalist studies; however, the findings are hampered by the small number of studies and by the temporal bias given the specialty’s introduction in 2006. Notably, our inclusion of the 2015 Bakitas study11 on early versus delayed palliative care was primarily based on the outcomes reported at three months, which was the point at which the standard care group began receiving the intervention. Results for specific outcomes can be found in the sections that follow. Quality of Life We identified three studies10,21,22 of good quality and four studies11,18,24,25 of fair quality that reported on patient QoL. The evidence base suggests that palliative care either improved patient QoL or had comparable improvements in QoL relative to usual care. Both study population (cancer vs. mixed) and timing of palliative care may be drivers of the variable findings on QoL. ©Institute for Clinical and Economic Review, 2016 Page ES6 Mood Outcomes We identified a total of six studies evaluating mood outcomes, specifically anxiety and depression, using a number of validated instruments, including two good quality10,21 and four fair quality11,18,24,25 RCTs. Of these six studies, four10,21,24,25 found statistically significant results in favor of the intervention, though not all studies evaluated both outcomes. No clear relationship based on target population, timing, or type of palliative care intervention emerged as potentially influencing outcomes. Symptom Burden The same seven RCTs reporting QoL outcomes also reported on symptom burden, and four of seven found no statistically significant differences in any symptoms between groups. Three studies showed reductions in some disease-specific symptoms (measured on the lung cancer subscale [LCS] of the FACT-L scale)21 and physical symptoms (measured on the Linear Analog Scales of Assessment (LASA)-physical well-being scale14,25 or the University of California, San Diego Shortness of Breath Questionnaire24). As with mood outcomes, we found no clear evidence of correlation with the study population, timing or type of palliative care intervention for symptom burden outcomes. Patient Satisfaction The available evidence suggests that palliative care improves patient satisfaction relative to usual care. One good quality prospective cohort study13 and three RCTs, including one of good quality22 and two of fair quality,12,24 reported on patient satisfaction; three of these studies found higher satisfaction or a greater increase in satisfaction in the intervention group compared to controls. Another RCT in patients with mixed diagnoses found no difference in satisfaction between groups, likely due to small sample size.24 We found no consistent evidence of differences according to study population, timing or type of palliative care intervention. Survival Our review identified four good quality studies evaluating the hypothesis that outpatient palliative care may affect patient survival, of which three were RCTs,10,12,21 and one a propensity-matched cohort study.26 The evidence base suggests that palliative care provided in the outpatient setting does not negatively affect patient survival,10,12,26 and may in fact result in increased survival in populations with advanced cancer,21 particularly when adjusting for potential confounders of the relationship between the intervention and survival (e.g., The Eastern Cooperative Oncology Group [ECOG] performance status score, or survival for less than one year).10,21 Both target population (cancer) and timing of palliative care (early) may be drivers of variable findings on survival; however, these factors are inextricably linked in the limited evidence base describing survival, with both studies in cancer patients also describing their interventions as early.10,21 In addition, an RCT ©Institute for Clinical and Economic Review, 2016 Page ES7 that directly compared early versus delayed palliative care in cancer patients showed a significant improvement in one-year survival rate in the early group (63% vs 48%, p=0.04).11 Resource Utilization Among the 11 studies providing resource utilization data, there was a generally observed benefit of outpatient palliative care on overall resource utilization, with eight studies reporting improvement in at least one utilization outcome, measured variously as an increase in hospice utilization or death at home, or a decrease in utilization of hospital or acute care services.10,12,13,16-20,23,24,26 Benefits were noted more frequently in cohort studies than RCTs, possibly as a result of larger sample size and longer follow-up in the former. There was a suggestion of reduced acute care resource utilization among the five studies evaluating outpatient palliative care interventions targeted to populations with mixed diagnoses.12,13,18,20,24 Three studies evaluated the impact of outpatient palliative care on the location of death, and suggest that outpatient palliative care results in a larger proportion of patients dying at home, or a smaller proportion dying in the hospital.12,16,20 Additionally, five studies evaluated hospice utilization and suggested favorably higher hospice utilization associated with palliative care interventions.12,16,17,19,23 Correlations with study population, timing, and type of palliative care intervention are inconclusive given the limited number of studies. Psychosocial and Spiritual Outcomes We identified three RCTs of fair quality evaluating psychosocial outcomes for patients, including spiritual well-being and advance care planning.18,24,25 One RCT showed a statistically significant improvement in spiritual well-being, as measured on the Spiritual Well-Being Scale, and a higher proportion of patients completing funeral arrangements after a year of follow-up; however, between-group comparisons for consideration of durable power of attorney or plans for disposition of possessions after death were not significant.24 Another study also reported a statistically significant benefit for overall spiritual well-being on the Functional Assessment of Chronic Illness Therapy (FACIT) tool in favor of the intervention after one month; however, this effect did not persist at weeks eight or 27.25 Caregiver Outcomes We identified only one good quality study reporting outcomes for caregivers, including depression, QoL, and caregiver burden.15 This study found that caregivers of patients receiving early palliative care had greater reductions in depressive symptoms than those receiving usual care after three months of follow-up. After 36 months, stress burden was also significantly better for those receiving palliative care at enrollment compared to those who received the intervention three months later.15 ©Institute for Clinical and Economic Review, 2016 Page ES8 Summary and Comment Despite the limitations of the published evidence derived from comparative studies, it is possible to use the ICER rating system to generate an estimate of the magnitude of the difference between outpatient palliative care and usual care. In summary, it is our judgement that there is moderate certainty that outpatient palliative care confers a comparable or better net health benefit relative to usual care. Additionally, given the lack of harms associated with such interventions, we have high certainty that the net health benefit is at least comparable. This yields an overall ICER Evidence Rating of C+: Comparable or Better for outpatient palliative care. Components of Outpatient Palliative Care Programs Associated with Treatment Success Although the interventions reported in the available literature were highly heterogeneous, we were able to identify several commonly described components of palliative care interventions in our evidence base, which we evaluated for their potential association with successful outcomes. Given that one of the primary goals of palliative care is to improve a patient’s QoL,27 we defined treatment success as a statistically significant effect on any QoL measure in favor of the palliative care intervention. Of the seven studies we identified as evaluating QoL, four10,21,22,25 were considered successful and three11,18,24 were not. We found that two21,22 of the four successful studies provided at least monthly in-person medical appointments with physicians or nurses present as part of the package of palliative services, while the studies not showing a QoL benefit followed up through telephone or home visits from a volunteer patient advocate on the palliative care team. Other elements evaluated (multidisciplinary care team, specialist care, patient and family education, advance care planning) did not appear in substantially different proportions of successful and unsuccessful studies. Comparative Value of Outpatient Palliative Care We reviewed the published literature for analyses that have examined the economics of palliative care programs, including studies of the costs that are potentially offset through the use of such programs (e.g., reduced end-of-life medical costs). We also explored the potential health system budgetary impact of outpatient palliative care programs over a near-term time horizon, utilizing published information on program costs and cost offsets, as well as the potential population eligible for such services. Our findings and analyses suggest that outpatient palliative care services appear to be cost-effective or even cost-saving for the health care system, by providing more care in home and outpatient settings rather than inpatient. Expanding the use of these palliative care programs to larger ©Institute for Clinical and Economic Review, 2016 Page ES9 proportions of eligible patients has the potential to substantially reduce the costs of health care for patients with advanced illness. Costs and Cost-Effectiveness Most recent studies that have compared direct health care costs for palliative care to usual care have found such programs to be cost-saving. Many of the costs of intensive inpatient admissions can be avoided, as more patients are able to be cared for at home or in an outpatient setting. A recent annual review by Hughes et al.28 of the growth of palliative care (both inpatient and outpatient) in the U.S. reports that several studies of outpatient programs found cost savings through reduced hospital admission rates and movement of patients from high-cost settings such as hospitals to lower-cost settings such as home health care. In addition, three specific studies found lower costs (ranging from about $6,000 to $8,000 per patient) with outpatient palliative care compared to usual care in patients with mixed diagnoses (cancer and other serious illnesses). Importantly, however, these studies do not provide clear indication of the start-up and implementation costs of outpatient programs, so a full assessment of return on investment could not be performed. Cost Burden and Unpaid Caregiving While many economic analyses of palliative care take a payer or health system perspective, some have found that at least some of the acute services that outpatient and home-based palliative care avoids by reducing hospitalizations and emergency department (ED) visits may actually be provided in the home setting, often by informal, unpaid caregivers. In some settings, it has been estimated that one-half to three-quarters of the costs of care in the last six to twelve months of life are attributable to family caregiving, suggesting a substantial economic burden on these individuals. Unfortunately, none of the comparative studies in our evidence base have attempted to formally assess this shift in costs in order to estimate the incremental effects attributable to palliative care. Potential Budget Impact Model To inform possible cost savings at the health system level, we also estimated the potential budgetary impact of outpatient palliative care programs among candidate populations for such treatment in the U.S., based on estimates from the literature. We combined estimates of the mean cost per patient with estimates of the population potentially eligible for outpatient palliative care programs and different assumed levels of uptake of such programs. We used the estimated annual savings per patient of $11,508 (after updating to 2015 U.S. dollars) to estimate the total impact on the U.S. health care budget of implementing outpatient palliative care at different levels. Results suggest that, with immediate implementation for 10% of the eligible population, an estimated 78,665 individuals would receive palliative care each year. After one year of treatment, with net annual savings of $11,508 per patient, one-year budget impact is estimated ©Institute for Clinical and Economic Review, 2016 Page ES10 to constitute savings of approximately $905.3 million. Over the entire five-year time horizon, we estimate that a cumulative total of 393,325 patients would be enrolled in palliative care, with a total potential savings of approximately $4.5 billion. Assuming that larger proportions of eligible patients could be enrolled in palliative care resulted in even greater savings. Alternatively, assuming that capacity to deliver palliative care has to be built up over time (i.e., 2% of eligible patients each year through year five) leads to fewer eligible patients being enrolled and results in fewer savings than with immediate implementation; our five-year savings estimate in this scenario is $2.7 billion. Full results at greater assumed uptake percentages (i.e., 25% and 50%) can be found in Table ES2 as well as in the full report. If outpatient palliative care was targeted to only patients with cancer, 74% of these estimated savings would be realized. Table ES2. Estimated Total Potential Five-Year Budget Impact (BI) of Outpatient Palliative Care: Immediate Implementation for Fractions of the Eligible U.S. Population (N=786,628) Immediate Implementation Gradual Implementation Percent Cumulative Cumulative Total Cumulative Number Cumulative Total BI Enrolled Number Treated BI (billions) Treated (billions) 10% 393,325 -$4.53 235,995 -$2.72 25% 938,275 -$11.32 589,965 -$6.79 50% 1,966,575 -$22.63 1,179,945 -$13.58 Finally, we also estimated the budget impact of outpatient palliative care programs for a hypothetical health plan population of 1 million members. Enrolling as few as 10% of patients expected to die of cancer, CHF, and COPD in outpatient palliative care (n=248) would result in cost savings of approximately $2.8 million per year, or $0.24 per member per month. For benchmarking purposes, this is comparable to payments the state of Colorado makes to accountable care organizations for well-child visits. Summary and Comment Most studies that attempted to compare the cost to the health care system of providing outpatient or home-based palliative care to that of providing usual care found that such palliative care decreased overall costs, mainly through avoided hospitalizations and ED visits near the end of life. However, it was often unclear whether these evaluations incorporated the start-up costs of developing an outpatient palliative care program. While the literature seems to support cost savings associated with outpatient palliative care for the health care system, other studies have looked at the economic impacts of these programs from a societal perspective. These studies point out that, while outpatient and home-based palliative care may succeed in avoiding unnecessary or unwanted hospitalizations and ED visits, some of the care that would be provided in those settings may now be provided in the home, often by informal, unpaid caregivers. Unfortunately, we did not identify any comparative studies that attempted to estimate the effects of outpatient palliative care on both medical and caregiver costs relative to usual care. ©Institute for Clinical and Economic Review, 2016 Page ES11 We also used the health care costs for palliative care and usual care from an RCT conducted in 2002-2004 to estimate the impact on the U.S. health care budget of palliative care for the treatment of cancer, COPD, and CHF patients near the end of life. The potential money saved from the health care budget was substantial, and scaled up as more patients were assumed to enroll in palliative care. Even assuming a 2% uptake rate per year (i.e., 10% enrollment after five years) would result in estimated cumulative savings of about $2.7 billion after 5 years. ©Institute for Clinical and Economic Review, 2016 Page ES12 1. Background 1.1 Introduction The vast majority of health care in the United States is curative in nature; health care professionals are trained to execute scientifically-based practices that are primarily intended to diagnose, treat, and prevent disease and do not incorporate an equivalent focus on quality of life (QoL) needs and values of patients and their loved ones. As the life expectancy and treatment objectives for patients with complex and chronic illnesses have evolved, clinicians are now faced with the challenge of meeting the entire breadth of patient and caregiver needs, of which curative or life-prolonging therapies may only be a part of the solution. Palliative care is specialized medical care that is focused on improving QoL through relief of pain, symptoms, and distress of serious illness.29 Palliative care is typically provided by an interdisciplinary team of doctors, nurses, and other providers who lead a team or collaborate with a patient’s long-term care provider to administer an additional layer of support for the patient and family.30 Ideally, palliative care is initiated at diagnosis and is provided concomitantly with curative therapy.2 The National Consensus Project for Quality Palliative Care emphasizes that palliative care is best achieved through a coordinated partnership with the treating provider from diagnosis through end stages of illness, across the continuum of care, and in a variety of care settings and living situations. Figure 1, taken from Clinical Practice Guidelines for Quality Palliative Care,1 outlines the progression of palliative care. Beginning at diagnosis, care focuses predominantly on curative treatments, with palliative care services provided as needed for such elements as pain and symptom management, or advance care planning. As a serious illness progresses, more emphasis is placed on palliative services while the emphasis on curative care decreases, ultimately culminating in hospice care when curative therapies no longer serve to benefit the patient. Figure 1: Palliative Care’s Place in the Course of Illness Graphic from the National Consensus Project for Quality Palliative Care ©Institute for Clinical and Economic Review, 2016 Page 1 Additional definitions described by national and international palliative care organizations can be found in Appendix 2. To date, palliative care has predominantly been offered to patients within the inpatient hospital setting due in part to the structures within the fee-for-service (FFS) delivery system. FFS is a payment system in which providers receive reimbursement for each service that they provide, such as a test, office visit, or procedure.31 In the inpatient setting, palliative care has demonstrated success in improving patient satisfaction and QoL, among other factors.32 Subject matter experts mentioned that hospitals were an early focus because they serve as a good entry point; at some point in the course of any serious illness there is likely to be a hospital admission, giving the palliative care team the opportunity to offer services. A number of clinical trials have demonstrated that palliative care brings substantial benefits to patients and caregivers.28 However, due to barriers surrounding access to care, the evolving needs of a growing patient population, and evidence supporting the use of palliative care outside of the hospital setting, many in the field suggest that hospital-based inpatient palliative care has reached a tipping point. As the aging and chronically or seriously ill population grows, palliative care will likely continue to evolve.4 In response to the existing need and lack of access, many health systems are developing new programs to expand access to palliative care beyond the hospital setting to outpatient clinic, office, and community- and home-based care. By providing palliative care in outpatient settings, services may be better suited to meet patient needs throughout the disease trajectory and continuum of care.4 If outpatient-based palliative care is to become the next frontier in palliative medicine,7 decision makers will likely need to consider a number of barriers that affect the widespread use of palliative care in outpatient settings. Barriers include a limited workforce, a lack of awareness or reluctance towards services among both patients and physicians, and issues of access related to geographic location or socioeconomic status. For an in-depth description of barriers please see Section 2.3. As stakeholders consider ways to expand access, they may benefit from a deeper understanding of the comparative clinical effectiveness and economic impact of outpatient palliative care services and the elements related to successful outcomes. 1.2 Scope Our review of the evidence examined studies of palliative care delivered outside the hospital, including outpatient clinic, office, and community- and home-based settings. Palliative care was defined based on services that incorporate elements directed at both physical and psychosocial patient care, as both are considered essential and definitional components of palliative interventions.5,6 We recognize that studies vary substantially in terms of their entry criteria and definitions of the interventions, and as such have attempted to identify the components of ©Institute for Clinical and Economic Review, 2016 Page 2 palliative care programs most closely associated with treatment success. Given the variability in health care systems and delivery worldwide, we have focused our review on publications from Canada and the U.S. Evidence on intervention effectiveness was limited to good- or fair-quality randomized clinical trials or comparative cohort studies of any duration. For more information on methods used to ascertain study quality, please see Section 4.2. The scope for this review is described below using the PICOTS (Population, Intervention, Comparators, Outcomes, Timing, and Settings) framework. Population The populations of interest include all adults (>17 years old) with serious or life-threatening illnesses, including but not limited to advanced cancer, chronic obstructive pulmonary disease (COPD), and congestive heart failure (CHF). Interventions Interventions of interest include those palliative care services provided either independently of, or in conjunction with, curative or life-prolonging treatment for serious and life-threatening illnesses. Studies which describe “hospice,” “end-of-life care,” “supportive care,” “comfort care,” or any such variation were considered alongside studies with interventions specifically described as “palliative.” We included all studies regardless of their definition of the interventions. However, we have also attempted to identify components of care across studies, with the goal of identifying those that may influence the direction of effect. We also included studies describing “early” palliative care (as defined by initiation at or around the time of diagnosis) as compared to “late” interventions (e.g., initiated at or near the end of life) or no palliative services. Comparators Primary comparators of interest are current models of usual care, which may take multiple forms. This is alternatively referred to as either “usual” or “standard” care. Outcomes Outcomes of interest included the impact of palliative care on:  Survival and mortality  Health resource utilization, including location of death  Symptom severity, including pain, breathlessness, fatigue, and nausea  Patient, caregiver, and/or family psychological distress (e.g., stress, anxiety, depression) ©Institute for Clinical and Economic Review, 2016 Page 3  Ratings of patient, caregiver, family, and provider satisfaction  Health-related QoL  Economic outcomes, including program start-up costs, costs per patient, potential cost offsets, and measures of cost-effectiveness The review included evidence drawn from all measures of these outcomes as reported in the evidence base, but focused on those that are drawn from validated instruments. Timing Evidence on palliative care provided outside the hospital or nursing care facility setting was gathered from the year 2000 through November 2015, reflecting the timeline and evolution of palliative care in these settings. Evidence on intervention effectiveness was limited to higher quality randomized clinical trials or comparative cohort studies, of any duration. Settings While study participants could be identified in multiple settings, we focused on palliative care programs delivered outside the hospital, including outpatient or home-based settings. Included interventions were those in which services were clearly described as being offered in outpatient settings, and not in settings such as residential nursing or hospice facilities.6 In addition to conducting a review of available literature and analyzing the comparative value of palliative care, ICER staff conducted semi-structured interviews with national and regional experts in an effort to gain their perspectives on practice and delivery system innovations, barriers to change, and opportunities for improving palliative care in the outpatient setting. These key informants included experts from academic institutions, palliative care programs, hospitals, patient advocacy organizations, and health plans. A full methodology and list of organizations represented in interviews are available in Appendix 9. The report also provides an overview of legal, regulatory, and financial landscape factors related to palliative care in the U.S. ©Institute for Clinical and Economic Review, 2016 Page 4 The report attempts to answer some of the key issues confronting patients, provider organizations, payers, and other policymakers and includes the following: An overview of the contextual factors impacting the delivery of outpatient palliative 1 care services, including a summary of clinical guidelines and payer reimbursement policies relevant to palliative care. 2 An evaluation of the evidence on the comparative clinical effectiveness of outpatient palliative care services relative to usual care or alternative approaches. 3 An assessment of the costs, cost-effectiveness, and potential budget impact of outpatient palliative care. ©Institute for Clinical and Economic Review, 2016 Page 5 2. The Topic in Context Extending the duration of human life ranks as one of modern medicine’s most significant achievements. In the U.S., average life expectancy has increased from approximately 71 years for persons born in 1970 to 79 years for those born in 2011,33 which represents more than 25% of all gains in life expectancy since 1900. However, for many, living longer also means living with a chronic, advanced, or serious illness. Over 90 million Americans live with at least one chronic illness, and seven out of ten Americans die from chronic disease.34 Within the Medicare population, nine of ten deaths are associated with at least one of nine chronic illnesses: CHF, chronic lung disease, cancer, coronary artery disease, renal failure, peripheral vascular disease, diabetes, chronic liver disease, and dementia. Evidence suggests that these patients often receive low-quality care as indicated by fragmentation and/or overuse of services, high frequency of medical errors, and poor QoL.32 As the population living with chronic, debilitating, and life-threatening illness continues to grow, the need for palliative care has become more acute. 2.1 History and Evolution of Palliative Care Services In the U.S., palliative care came to existence in part due to the hospice movement which began after Florence Wald, former Dean of Yale University’s School of Nursing, established the Connecticut Hospice in 1974.35 The formalization of hospice began when the U.S. Health Care Financing Administration (HCFA; the precursor agency to the Centers for Medicare and Medicaid Services) initiated the hospice demonstration program in 1979. A Medicare benefit for hospice was introduced by statute shortly thereafter in 1982.35 The Medicare hospice benefit is highly standardized by the Medicare Conditions of Participation, and eligibility requirements include a prognosis of living less than six months.32 Patients must provide documentation from two physicians indicating a life expectancy of six months or less and agree to forgo insurance coverage for life-prolonging and curative treatments.36 The term palliative care originated in Montreal, Canada in 1974 when Dr. Balfour Mount, a physician at McGill University, began using the term in an effort to engage patients and families experiencing serious or life-limiting disease in holistic care practices without using the word hospice. Both hospice and palliative care services were aimed at addressing physical, psychological, social, or spiritual distress with the intent of improving QoL for patients and families. In the U.S., palliative care differs from hospice in that services are accessible to patients without the restriction of a six-month life expectancy. Outside of the U.S., this distinction is less prominent, and palliative care and hospice are relatively synonymous terms.32 Additional detail on the distinctions between palliative care and hospice is available in Appendix 1. ©Institute for Clinical and Economic Review, 2016 Page 6 Modern palliative care took shape in the U.S. during the 1990s predominantly within academic medical centers in response to the recognition that patients with a variety of serious illnesses could benefit from the core components of hospice care, regardless of their life expectancy. Many cite the 1997 Institute of Medicine (IOM) report titled Approaching Death: Improving Care at the End of Life that shed light upon deficiencies in end-of-life care in the U.S. as having influenced this recognition.37 More recently, palliative care has expanded beyond this setting due to a number of factors, including:35  Monetary support from philanthropic foundations to improve care for the seriously ill including an effort on behalf of the Robert Wood Johnson Foundation and George Soros’ Open Society Institute to fund efforts aimed at expanding the use of palliative care in mainstream medicine;37  An increase in the number of clinical demonstration projects, reviews, and randomized controlled trials documenting major inadequacies in care for the seriously ill and demonstrating the benefits of palliative medicine;  The 2002 establishment of a hospice and palliative care nursing certification by the American Board of Nursing Specialties;  The development and release of the 2004 Clinical Practice Guidelines for Quality Palliative Care which specifically expanded the focus of palliative medicine beyond care for those at the end of life;  The 2006 publication of a National Framework and Preferred Practices for Palliative and Hospice Care Quality by the National Quality Forum which established quality guidelines for standardized palliative care;  The 2006 recognition of the subspecialty of Hospice and Palliative Medicine by American Board of Medical Specialties (ABMS) and the Accreditation Council for Graduate Medical Education followed by first examination for the Hospice and Palliative Medicine (HPM) specialty in 2008; and  The 2011 formation of the palliative care certification for hospital programs by the Joint Commission. The predominant model of palliative care in the U.S. is the hospital consultation model. This model typically involves a team of physicians, nurses, social workers, and other support staff who assist primary care providers and specialists in the hospital.32 Palliative care is most commonly seen in the oncology setting. Almost all National Cancer Institute designated cancer centers have palliative care services available, with 98% having a palliative care program, and 59% having an outpatient palliative care clinic.28 The widespread use of palliative care in oncology is likely due in part to the nature of the disease and the need for emphasis on symptom relief. The significant side effects of many treatments for cancer make it necessary to provide both curative care in tandem with palliative services such as pain and symptom management. Even in settings and specialties where palliative care is accepted there remain challenges to access and appropriate use. There are a number of studies that describe limitations to access, even in ©Institute for Clinical and Economic Review, 2016 Page 7 oncology.38 Some stakeholders interviewed noted that while palliative care services are prevalent in oncology settings, services or consults are often initiated late. In order to extend the potential benefits of palliative care and address barriers to access, decision-makers may turn to expanding palliative care services outside the hospital. 2.2 Structure and Settings of Palliative Care During interviews, key informants emphasized that palliative care is expanding into outpatient settings. Stakeholders stressed that outpatient palliative care provides physicians with an opportunity to truly influence decision-making about the patient’s care and prevent patients from needing to access the hospital in the first place. Palliative care in the outpatient setting can follow a number of different models. This section outlines common elements and models of outpatient care. Program Structure Palliative care programs aim to address physical and psychological symptoms, spiritual needs, treatment goals, and end-of-life preferences.2 Palliative care can be provided concurrently with, or independently of, curative or life-prolonging care at any stage of a serious illness. These services are administered in a number of settings, including hospitals, outpatient clinics, skilled nursing facilities, and in the home.4 Because of the growing demand for palliative care, due primarily to the increase in the number of patients living with chronic and serious illness, the field has evolved to include a number of approaches to maximize delivery of these services. While the structure of how these services are provided varies across population type and setting, two dominant models have emerged and are described in detail below. For the purposes of this report, we have defined these models as generalist palliative care and specialist palliative care. Generalist Palliative Care Palliative care is often provided by professionals that have some clinical experience and basic training in palliative care concepts but whose primary specialty is not palliative care, including primary care physicians, generalists, oncologists, nurse practitioners, and professionals in other disciplines. These providers conduct needs assessments, educate patients about their disease, and provide basic symptom management and psychosocial support. Generalist palliative care (also sometimes called primary palliative care) can be provided by individual providers alone or in conjunction with a multi-disciplinary team (e.g., social workers, chaplains, etc.); the main differentiating feature from specialty palliative care is the absence of a clinician with advanced training/certification in hospice and palliative care.8 ©Institute for Clinical and Economic Review, 2016 Page 8 Specialist Palliative Care Specialist palliative care teams can include clinicians who are board certified in hospice and palliative medicine (HPM), as well as advanced practice nurses certified in hospice and palliative nursing. Specialist palliative care providers are those who have extensive training and experience in palliative care, and focus on more complex aspects of disease management, including controlling refractory physical and emotional symptoms and worsening depression or anxiety; assisting with conflict resolution; identifying and communicating patients’ goals and preferences for care; and counseling around issues of end of life planning and bereavement.8 HPM Certification The HPM certification is a medical subspecialty that was approved by the American Board of Medical Specialties (ABMS) in 2006. The first HPM exam was administered in 2008, and from 2008 to 2012 there was a "grandfathering period," where physicians with extensive experience in the field were able to achieve certification through direct pathway without having to complete a 12- month fellowship. As of 2012, any provider seeking an HPM subspecialty certification must complete a 12-month HPM fellowship through an accredited program. Once a fellowship has been completed, the provider is able to take the certification exam.39 There are fewer than 200 fellowships available per year.32 More information can be found at: http://aahpm.org/hpm/number-certified Certification of Nurses and Team Members The Hospice and Palliative Credentialing Center (HPCC), formerly the National Board for Certification of Hospice Nurses, was incorporated in 1993 with the goal of initiating a certification process for hospice nurses. The first certification exam was administered in 1994. To become certified, applicants must provide verification of licensure and meet practice hour requirements. A passing score on the HPCC exam provides certification for a period of four years. Certifications are available for members of the hospice and palliative nursing care team, administrators, and professionals in perinatal loss care and include: • Advanced Certified Hospice and Palliative Nurse (ACHPN) • Certified Hospice and Palliative Nurse (CHPN) • Certified Hospice and Palliative Pediatric Nurse (CHPPN)  Certified Hospice and Palliative Licensed Nurse (CHPLN)  Certified Hospice and Palliative Nursing Assistant (CHPNA) • Certified Hospice and Palliative Care Administrator (CHPCA) • Certified in Perinatal Loss Care (CPLC) ©Institute for Clinical and Economic Review, 2016 Page 9 Additional information can be found at: http://hpcc.advancingexpertcare.org/competence/certifications-offered/ Advanced Certified Hospice and Palliative Social Worker (ACHP-SW) The National Association of Social Workers Specialty Certification program began in 2000 to address an increased need for specialization. The ACHP-SW was established 2008 in partnership with the National Hospice and Palliative Care Organization (NHPCO) for MSW-level certification. This certification is intended to improve specialized knowledge, skills, and abilities of professional social workers in hospice and palliative care settings. Additional information can be found at: http://www.socialworkers.org/credentials/credentials/achp.asp Chaplaincy Certification The Board of Chaplaincy Certification Inc. provides certification for professional chaplains interested in pursuing a specialization in palliative care. In addition, the College of Pastoral Supervision and Psychotherapy provides training and credentialing in hospice/palliative care for clergy, including chaplains, pastoral supervisors, and pastoral psychotherapists. Additional information can be found at: http://www.cpsp.org/certification Settings of Care As previously mentioned, palliative care services are administered in a number of settings, including hospitals, outpatient clinics, skilled nursing facilities, and in the home.4 Given that populations living with serious or life-threatening illnesses are spending increasingly more time outside the hospital, our report focuses on services provided in the outpatient setting.7 Table 1 outlines different models of outpatient palliative care as informed by data from the Center to Advance Palliative Care’s (CAPC’s) Improving Outpatient Palliative Care Project (IPAL-OP) as well as our evidence review. ©Institute for Clinical and Economic Review, 2016 Page 10 Table 1. Models of Outpatient Palliative Care40 Location Description Financing Clinic Based Models Outpatient Stand-Alone Clinic These are independently functioning, specialty clinics where patients receive palliative care Palliative care clinic responsible for all services. costs. Outpatient Co-Located Clinic These clinics operate in a shared space with other medical services. Referrals to palliative Costs may be shared between the care services may come from physicians within the clinic in other specialties, or from other palliative care clinic and the host clinic. practices entirely. Outpatient Embedded These services share similar characteristics with outpatient co-located clinics. Palliative care The host clinic is typically responsible for providers share space and work closely with other providers. There may be protocols all costs. defining how palliative care functions in tandem with other medical services. Community Based Models Home Visits Palliative care consultants visit patients in their homes. This model is generally well-suited Billing can vary greatly depending on the for patients who have complex needs that would require more time than is possible in an services provided and the structure of office setting, patients for whom travel to an office appointment would be difficult, or for the home care program. practices with insufficient office space. Facility-Based Visits Palliative care consultants visit patients in a residential care facility and provide palliative Consultants bill for consultation services to complement other medical services being administered by the facility’s providers. services. Information adapted from https://central.capc.org/eco_download.php?id=1130 ©Institute for Clinical and Economic Review, 2016 Page 11 2.3 Barriers and Opportunities There are a number of barriers impeding the widespread use of outpatient palliative care. This section provides an overview of the regulatory, financial, and administrative contexts affecting how palliative care is delivered in the U.S. It is important to recognize that the landscape for palliative care is constantly evolving; therefore, this section should be considered a “snapshot” of the status at the time of the report’s publication. Barriers Workforce: Capacity, Education, and Training A primary barrier to the expansion of palliative care identified in our interviews with key stakeholders was a shortage of adequately trained providers. As of 2014, the American Academy of Hospice and Palliative Medicine (AAHMP) reported that 7,000 physicians in the U.S. are HPM certified.39 Data from a study examining 30-day readmissions among seriously ill older adults indicated the availability of only one HPM physician for every 20,000 older adults with life-limiting illness, and one HPM physician for every 11,000 Medicare deaths.28 There is an estimated shortfall of 6,000-18,000 palliative care physicians in the U.S. and an approximately equal deficit in the availability of advanced practice nurses.41 This shortage is exacerbated by the limited availability of specialty training opportunities for Masters-prepared nurses, as well as for physicians.32 Key informants pointed to education and training requirements as a key barrier to expansion of palliative care programs. New physicians may be interested in pursuing fellowships in HPM; however, the regulations may deter mid-career professionals from pursuing HPM certification, as they will likely be required to pause their career to complete fellowship training. The limited number of fellowship opportunities, paired with the lack of a training exemption for physicians already working in the palliative care field, contribute to a shortage of HPM-certified providers. Awareness of Palliative Care Among Non-Specialist Providers Key informants also identified attitudinal barriers and a lack of awareness of palliative care among non-specialist providers. For many physicians, the perceived dichotomy between curative treatment and palliative care likely affects their willingness or ability to refer patients to palliative care programs. Some physicians may not recognize the distinction between palliative care and hospice care, and may be reluctant to approach either due to the perception that beginning palliative care would interfere with the opportunity to offer the patient curative treatment. Stakeholders identified the perception that referral to palliative care indicates that a patient or practitioner has “given up hope” as a common misconception and a major barrier for consults. Many components of palliative care can be offered by a primary care physician, generalist, or ©Institute for Clinical and Economic Review, 2016 Page 12 disease specific-provider; however, many such providers either lack the training to discuss and provide palliative care services, or are not aware of services offered within their medical system to which they could refer patients. Geographic and Socioeconomic Access Disparity In the outpatient setting, there are limited data related to access. However, data from CAPC indicate that regions that lack access to inpatient palliative care tend to:42  have fewer academically-affiliated, nonprofit hospitals with formal teaching programs;  be located in the southern U.S.; and  have fewer faith-based health care organizations. While less documented than other health disparities, there appear to be issues associated with access to symptom management and communication services among African American and Hispanic patients as compared to white patients.43 Factors affecting access to palliative care are not well-understood, but may include:  gaps in palliative care knowledge among minority older adults;  spiritual and religious beliefs that conflict with perceived palliative care practices;  mistrust of the health care system due to past experience;43 and  cultural barriers including absence of minority staff, interpreters, and outreach/education to diverse communities. Patient and Family Knowledge of Palliative Care Patients, families, and caregivers may not understand what palliative care entails and do not necessarily know at what point to request services. Palliative care is a relatively new concept to many patients and to some is still synonymous with end-of-life or hospice care. Nearly 90% of adults surveyed as part of a recent study reported having either limited or no knowledge of palliative care. However, upon learning what palliative care entails, more than 90% of the respondents stated that they would want palliative care for themselves or a family member and that it should be universally available.2 Availability of Data While many specialty organizations advocate for the use of palliative care, the availability of data from high powered, well-designed RCTs is generally lacking.28 Our review of the published evidence derived from higher-quality comparative studies supports the observation that such data are lacking, with only 14 relevant studies identified. Stakeholders referenced the need for additional research demonstrating that palliative care effectively improves patient experience, reduces resource utilization, and is not a high-risk ©Institute for Clinical and Economic Review, 2016 Page 13 investment. Our evidence review was able to highlight the relative impact of outpatient palliative care programs on particular outcomes of interest; however, we were not able to substantively identify particular components of such inherently multidisciplinary programs that may be able to best predict success. Such studies are difficult given the significant heterogeneity of palliative care interventions, as well as patient populations, across different settings (e.g., rural vs. urban) and patient populations (e.g., advanced cancer vs. CHF). Our finding that at least one monthly in-person clinical interaction was associated with significant improvements in QoL measures suggests that additional studies might quantify this element of outpatient palliative care programs, and better characterize its relationship to success (see Section 4 for details). Appendix 7 of this report describes comparative studies of outpatient palliative care currently ongoing in the US and Canada, some of which may be able to provide information lacking in the current evidence base. While many health systems have developed programs to address the needs of complex populations through pilot demonstrations, systems level interventions, and value-based reimbursement models, there exists a need for additional outcomes data and cost-effectiveness analyses.44 A number of the stakeholders that we interviewed felt that the available data indicate that outpatient palliative care decreases costs by reducing hospital readmissions. However, they stressed that health systems are looking for more robust data demonstrating that outpatient palliative care is both cost-effective and improves patient-centered outcomes such as QoL. Stakeholders also highlighted a need for studies to address symptom relief, specifically referencing the need for more explicit guidelines on the prescription of opioids to treat chronic pain in patients with life-threatening illnesses. Funding While there has been an increase in the number of NIH-funded investigations and grants in the past decade, only 1% of the current NIH budget is dedicated to palliative care research. Institutional programs have been created in an effort to strengthen available evidence, but the field continues to require additional extramural funding.28 Stakeholders noted that when starting a new palliative care program, financial support on the local level, from hospitals or within provider organizations, and executive level buy-in were critical to success. Billing for Services Based on conversations with stakeholders and a review of available resources we found that billing for outpatient palliative care services covers a portion of direct staff costs but does not typically support all services provided. For example, a survey of 12 outpatient palliative care programs conducted by the Palliative Care Leadership Center (PCLC) Initiative of the Center to Advance Palliative Care found that billing supports 50% of outpatient palliative care services. Reimbursement for services is impacted by a number of factors including:45 ©Institute for Clinical and Economic Review, 2016 Page 14  Documentation and billing processes  Palliative care team provider composition (impacts who can bill for direct costs)  Contracts with payers and type of payer (Commercial or Medicare/Medicaid)  Effort towards direct patient care vs. non-billable efforts Many of the barriers listed in this section may be alleviated through increasing awareness of and access to palliative care services in the outpatient setting. We have identified a number of opportunities that may help to foster the continued development of outpatient palliative care programs, as listed in the sections that follow. Opportunities New Payment Methodology Through stakeholder interviews and our review of the cost literature, we found that many consider integrated delivery systems operating under capitated or risk-bearing managed care payments to be best suited to execute outpatient palliative care programs. Kaiser Permanente and the Veterans Administration are examples of integrated systems that operate in multiple states and have initiated palliative care models designed to improve access to outpatient palliative care through care coordination efforts.29 Integrated delivery systems are uniquely positioned to provide outpatient palliative care because of the established systems-level structures that facilitate sharing of information, communication, and finances that allow for coordination of care across settings. The willingness and capacity of individual health systems to engage in care models and payment methodology outside of the fee-for-service realm will also have an impact on the continued expansion of outpatient palliative care. While the integrated delivery system has been identified by many as a promising avenue for palliative care, these systems are currently available to a minority of patients, and access varies geographically. Access to palliative care could be further improved regardless of delivery system structure through policy initiatives that encourage the use of palliative care by providing additional reimbursement mechanisms for services.29 For example, reimbursement for use of telehealth and mobile technologies may improve access. A recent example of a policy intended to change reimbursement mechanisms in order to support palliative care comes from the Centers for Medicare and Medicaid Services (CMS). In January 2016, CMS introduced the Medicare Care Choices Model, intended to allow Medicare patients to access palliative care services from specific hospice providers without giving up curative treatment. This program is intended to address barriers to access by extending home-based palliative care services to as many as 150,000 Medicare beneficiaries.46 The model is also predicted to allow greater access to supportive care services provided by hospice; improve QoL and satisfaction for patient and families; and play a role in informing new payment systems for the Medicare and Medicaid programs. For a detailed list of programs in New England please see Appendix 3. ©Institute for Clinical and Economic Review, 2016 Page 15 Use of Electronic Medical Records to Encourage Care Coordination Several stakeholders emphasized the need for a mechanism to appropriately identify those who will benefit from palliative care services. One way to increase access to palliative care is through an EMR system-based checklist similar to that used in hospitals for vital signs by nurses. By screening for certain factors related to functional impairment and entering assessments in a shared records system, providers across the continuum of care can more accurately recognize a patient’s need for palliative care. Suggested factors for screening include:  Frailty  Cognitive impairment  Number of chronic conditions  Family caregiver burden  Distress Provider Education Continued investment in palliative care education may help to address the existing shortage in the palliative care workforce. Stakeholders stated that medical schools and residency programs may want to consider expanding existing training options for palliative care. Currently program curricula provide training-related core competencies of palliative care including communication, pain and symptom management, and psychosocial assessment, but many feel that there exists a need to expand the offerings.2 The clinical experts and stakeholders interviewed recommend a number of options to help build the existing workforce including:32  Distributing currently unused graduate medical education (GME) slots to accredited palliative medicine fellowship training programs in order to address the shortage of fellowship opportunities;  Providing Health Resources and Services Administration (HRSA) Title VII–supported career development awards (loan forgiveness programs) to incentivize medical and nursing faculty to enter the field; and  Offering educational loan forgiveness for physicians and advance-practice nurses. It may also be important to facilitate a conceptual understanding of palliative care among mid- career professionals who do not plan to specialize in HPM and who may not have received extensive training in palliative care techniques during medical school or residency programs. Primary care and general practitioners often perform elements of palliative care as part of their practice but many lack formal training or capacity to address symptom management, spiritual needs, communication about care goals, and economic issues. Recently the IOM described a need for generalist-level palliative care training designed to enhance communication skills among ©Institute for Clinical and Economic Review, 2016 Page 16 medical students and practicing clinicians in order to increase provider competency in identifying patient and family preferences and care goals.47 Additionally, it is recommended that patients with more complex needs be referred by PCPs and other generalists to specialized palliative care teams. Providing continuing education training on elements of palliative care and encouraging referral to palliative care specialist teams when providers encounter these patients may help address the perceived barriers. A few of the stakeholders interviewed suggested that hospitals, nursing homes, and rehab facilities should conduct a palliative care needs assessment for all patients upon admission for outpatients living with serious or complex illnesses. However, some noted that the processes for identification and referral have not been worked out on a wide scale. Efforts could also focus on changing perceptions among providers that palliative care is appropriate only at the end of life. It may be necessary to emphasize that palliative care can be administered alongside curative treatment and is designed to ease symptoms and suffering throughout a serious illness, and should be considered at diagnosis. Resources for Existing Providers and Health Systems A number of agencies have developed materials intended to support practicing physicians and other providers, health systems, and communities expand and enrich the practice of palliative care.  CAPC has developed tools and best practices for the inpatient setting and now are focusing on developing the same training support for palliative care outside the hospital called Improving Outpatient Palliative Care (CAPC IPAL-OP). This program is intended to serve as a central repository for learning how to start an outpatient palliative care program and in turn improve outpatient palliative care practices. Membership is required for full access to resources, tools and references from health care systems at the forefront of developing innovative outpatient palliative care services. More information can be found at CAPC’s IPAL-OP homepage.  The Coalition for Compassionate Care of California (CCCC) offers a number of resources for providers and families including monthly palliative care webinars, custom tools, presentations, studies, and fact sheets about palliative care. These resources are designed to provide support to organizations and communities in an effort to expand palliative care across the continuum of care. More information can be found on CCCC’s palliative care site.  The IOM produced a report titled Dying in America which includes recommendations regarding the need for increased palliative care communication training for all practitioners. IOM recognized VITALtalk as having a strong program for promoting basic palliative care communication skills to prepare providers for difficult conversations surrounding palliative care. VITALtalk communication skill-building resources are available at http://vitaltalk.org/ ©Institute for Clinical and Economic Review, 2016 Page 17 Patient Education Patients may benefit from conversations with providers or a public awareness campaign designed to inform individuals and families about palliative care options. Federal and private sector investment in a major social marketing campaign could help address this issue.48 Key informants and clinical experts noted that early engagement with patients is crucial. This may be easier with some patient populations than others. One participant noted that the success of palliative care in oncology is partially due to the nature of the disease. When a patient is diagnosed with cancer, they know they are sick and are typically committed to a care plan. Patients with other chronic diseases such as diabetes and heart disease may potentially be less engaged in their treatment which can have a negative impact on the effectiveness of palliative care. It is worth noting that until the workforce is adequate to address need, patient education may promote awareness but may not increase access. Legislation Aimed at Improving the Palliative Care Landscape Recently policymakers at the state and federal level have proposed legislation intended to address a number of barriers to widespread use of palliative care. Examples include:  On the federal level, representatives recently introduced a bill titled The Palliative Care and Hospice Education and Training Act (PCHETA/HR 3119) which is intended to strengthen support for palliative care. The bill calls for increased investment in health care workforce training and public education, and for assistance from the National Institutes of Health to promote research.46 Many stakeholders interviewed highlighted the importance of continued support from government agencies in order to foster growth and development of new programs and research initiatives through grants and other funding mechanisms in order to create opportunities for future research.  In Massachusetts, state level palliative care legislation was recently enacted. Chapter 478 An Act to Improve Quality of Life by Expanding Access to Palliative Care authorized creation of a task force to advise the Commonwealth’s strategic initiatives for improving access to services. The act can be found at: https://malegislature.gov/Laws/SessionLaws/Acts/2014/Chapter478  In 2013, Rhode Island enacted House Bill 5204 An Act Relating to Health and Safety-State Palliative Care and Quality of Life Act intended to improve public awareness of and access to palliative care. The legislation calls for an Advisory Council-led Palliative Care needs assessment in Rhode Island and the creation of a Palliative Care Consumer and Professional Information Education Program. The long-term goal of this legislation is the creation of a system for identifying those who could benefit from palliative care, and promoting awareness of and appropriate access to palliative care services for those with serious illness.49 Stakeholders cited this legislation as the first of its kind in the nation. ©Institute for Clinical and Economic Review, 2016 Page 18 More information can be found at: http://webserver.rilin.state.ri.us/BillText13/HouseText13/H5204aa.htm  Additionally, a multi-stakeholder collaborative spearheaded by the American Cancer Society and its advocacy affiliate, the American Cancer Society Cancer Action Network (ACS CAN), launched a palliative care campaign that includes efforts aimed at increasing research grant support, improving availability of information, promotion and support of outreach efforts, and implementing palliative care legislation. The overall emphasis of this campaign is to treat an individual beyond the disease. It outlines three main objectives:50  Increase palliative care awareness, education, and research emphasis;  Boost workforce capacity, clinical communication skills, and palliative care training support; and  Pursue integration of palliative care services and quality standards in all care settings and associated payment reform promoting interdisciplinary care. ©Institute for Clinical and Economic Review, 2016 Page 19 2.4 Existing Approaches to Palliative Care Outside of the Hospital Setting There are a number of ways in which palliative care programs function in the outpatient setting. The evidence review in Section 4 explores the comparative effectiveness of different models of outpatient palliative care and seeks to identify the key program components that correspond to improved patient outcomes. In this section we aim to describe several existing programs to provide context for how palliative care is organized and coordinated through health systems. We found that most outpatient programs are operated by hospital systems. However, some models involve an integrated approach through an accountable care organization (ACO) framework. The following content comes from publicly available websites and links to specific sites are included at the beginning of each description. Hospital System Managed Models The Palliative Care Program, Dartmouth Hitchcock, New Hampshire and Vermont http://www.dartmouth-hitchcock.org/palliative_care/about_us.html Dartmouth Hitchcock Medical Center (DHMC) offers both inpatient and outpatient palliative care services to patients in the hospital or being treated for serious illnesses in clinics. When needed, the staff works with hospice team in the patient’s community to ensure an easy transition from the hospital to care at home or at an assisted living facility.  Staff: The Palliative Care Program involves care by a variety of providers. Physicians, nurse practitioners, and registered nurses manage the physical and emotional symptoms of serious illness. Social workers provide guidance and coordinate care, while other providers, including healing arts practitioners, spiritual care coordinators, and community volunteers provide additional supportive services.  Outpatient specific services: DHMC runs an outpatient palliative care clinic where patients can receive evaluations and interim care through regular appointments with the palliative care team. When possible, these appointments are scheduled in coordination with other medical visits to avoid repeat trips to the medical center. Home health services are available in conjunction with the VNA and Hospice of Vermont and New Hampshire. DHMC has 24/7 on call service, family education services, and bereavement support. Results from an RCT carried out in this setting are described in Section 4 of this report.10 ©Institute for Clinical and Economic Review, 2016 Page 20 Palliative Care Program, Massachusetts General Hospital, Massachusetts http://www.massgeneral.org/palliativecare/about/ Palliative care services at Massachusetts General Hospital (MGH) are offered to any patients diagnosed with a life-limiting or chronic condition at various stages of illness. Services focus on pain and symptom management, as well as on supporting patients and families in defining goals of care. As much as possible, palliative services are coordinated with other specialists providing the patient’s medical care. MGH has an inpatient and outpatient clinic, offers home visits for nearby patients, and refers to hospice care when needed.  Staff: MGH’s palliative care division includes 16 physicians, eight nurse practitioners, fellows, clinical social workers, and a triage nurse.  Outpatient Specific Services: Services provided in the outpatient setting typically focus on management of pain and symptoms, individual or family counseling, advance care planning, and arrangement of care through community resources. Results from an RCT evaluating early outpatient palliative care to patients newly diagnosed with non-small cell lung cancer (NSCLC) in this setting are described in Section 4.16,21 Hoag Memorial Hospital Presbyterian CARES/Palliative Care Program, Tustin, CA http://www.hoag.org/Specialties-Services/Other-Programs-Services/Palliative-Care/Services.aspx The CARES team at Hoag Memorial Hospital offers both inpatient and outpatient palliative care services to any patient with a chronic and progressive illness, including cancer, advanced heart or lung disease, kidney or liver disease, and dementia.  Staff: The CARES team includes a clinical nurse specialist, licensed clinical social worker, a palliative care certified physician, and a family medicine physician.  Outpatient Specific Services: The CARES team outpatient services including complex pain and symptom management; outpatient clinic visits; supportive counseling; coordination of care with community resources; emotional and spiritual support; and advance care planning. ©Institute for Clinical and Economic Review, 2016 Page 21 ACO Managed Models Kaiser Permanente, California, Colorado, Georgia and Hawaii http://www.growthhouse.org/palliative/ Kaiser Permanente is an integrated managed care consortium based in Oakland, California. Kaiser Permanente offers palliative care services in several states, including California, Colorado, Georgia, and Hawaii. The information below is based on a publicly available toolkit outlining the Kaiser Permanente (TCPC) in Downey, CA. TCPC offers both inpatient and outpatient services.  Staff: TCPC teams typically include a physician, a registered nurse skilled in management of pain and symptoms, and a social worker (licensed clinical social worker or Masters-level social worker).51 Based on patient need, additional providers such as home health aides, a chaplain, or physical therapists may be added to the team.  Outpatient services: Patients are able to receive home visits by all members of their care team for medical services, psychosocial support, and education. Care management services are offered to make sure all of the patient’s needs are met, and telephone support or after- hours visits are available 24/7. Patients and families are encouraged to engage in advance care planning discussions.  Financing: The program receives fixed monthly reimbursements from Kaiser Permanente for each enrollee. This reimbursement is a capitated rate intended to cover all inpatient and outpatient services.51 Commonwealth Care Alliance (CCA), Massachusetts http://www.commonwealthcarealliance.org/model-of-care CCA is a Massachusetts-based, nonprofit, fully integrated, prepaid health care delivery system. CCA serves Medicare and dual eligible seniors through its Senior Care Options plan, and individuals with disabilities through its Disability Care Program. There are more than 5,500 members enrolled in CCA, the majority of whom are eligible for nursing home placement. Stakeholders emphasized the complexity of the patient population; all patients have at least two chronic conditions.  Staff: Several care delivery programs at CCA rely on a multi‐disciplinary care teams managed by nurse practitioners (NP). In addition to the NP, the team includes social workers, behavioral health providers, community health workers, and other specialists.  Outpatient Services: CCA runs a statewide program that provides care in the home and community settings utilizing a consultative model.52 In this model, CCA initiates care with a formal palliative care consultation after which care is typically managed by a registered nurse, nurse practitioner, or a physician assistant case manager. ©Institute for Clinical and Economic Review, 2016 Page 22  Financing: CCA has risk‐adjusted capitated contracts with Massachusetts Medicaid and CMS.53 It also has a contract with Neighborhood Health Plan to provide services through its Disability Care Program. ©Institute for Clinical and Economic Review, 2016 Page 23 3. Summary of Coverage Policies While many payers, both public and private, have coverage policies specific to hospice care, few offer publicly available policies specific to palliative care services offered in tandem with curative care. Medicare covers many services essential to provision of palliative care under parts A and B but does not have an explicit palliative care benefit. Similarly, New England state Medicaid programs do not have publicly available documentation outlining services covered as palliative care. Some private insurers offer programs that support the use of palliative care services. Since payer policies are continually evolving and information is not always publicly available, this section is not intended to be a complete picture of private and public payer efforts related to palliative care. While specific coverage policies vary by payer, the Center to Advance Palliative Care (CAPC) highlights a number of variables that affect the proportion of costs that are billable to an insurer. These variables include:  Quality of the documentation provided by clinicians, and the quality of the billing process  Mix of providers on the team  Contracts with payers  Amount of time the team spends on direct patient care as opposed to other activities that are related to patient care but may not be billable CAPC offers tools and documents to assist palliative care teams in improving billing practices. Their resources fall into several key areas of consideration, including:  Utilizing clinicians who are credentialed in HPM  Strategies to increase consistency and efficiency of documentation  Review of options for billing based on the complexity of services instead of billing for time  Billing extended time codes  Priorities for billing reports  Using data to improve team performance Medicare A white paper authored by the Home Care and Hospice Financial Managers Association and the National Association for Home Care and Hospice summarizes Medicare billing rules for palliative care services. The white paper is available in full here, and is briefly summarized below. Medicare does not have a designated palliative care benefit but covers many services related to palliative care under Parts A and B. Under Part A, many home health agencies that provide palliative care can bill for services such as nursing, social work, and spiritual care. Home health services are ©Institute for Clinical and Economic Review, 2016 Page 24 covered under the Medicare Home Health Prospective Payment system. Certified Medicare Home Health Agencies may contract with hospice providers to purchase pain control services from the provider for patients who have been diagnosed with a terminal illness but have opted to continue curative treatments, making them ineligible for the Medicare hospice benefit. Many services related to palliative care are also covered under Medicare Part B. Physicians use a hospice palliative care code when billing for palliative care services. The code does not provide increased reimbursement rates, but does provide information for collection of data. Under Part B, physicians, nurse practitioners, and licensed clinical social workers are able to bill for services. In many cases, the patient is responsible for a 20% coinsurance. Medicaid Of the six New England states, none offer publicly available coverage policies specific to palliative care. This does not mean that services related to the provision of palliative care are not covered, but they are not explicitly classified as palliative care services in any publicly available coverage policies. Private Insurers As with public payers, while many private insurers have coverage policies directly related to the provision of hospice benefits, few insurers have publicly available coverage policies specific to palliative care. Many services related to palliative care may be covered but are not explicitly identified as palliative care services. Some insurers also offer case management and chronic care programs that may share some overlap with palliative care services, but are not classified as such. Two examples of these programs are described briefly below: Aetna Compassionate Care Program https://www.aetna.com/individuals-families/member-rights-resources/compassionate-care- program/compassionate-care-description.html The Aetna Compassionate Care Program offers supportive services for patients with a serious illness as well as for their family members. The Compassionate Care Program offers an extended hospice benefit, meaning that members are able to access services typically reserved for hospice care during the last six months of life throughout the last 12 months of life. In addition, these benefits can be offered in tandem with curative services, whereas typical hospice benefits require that patients forgo any and all curative treatments. In addition to accessing services earlier, patients and families also have telephonic access to nurse case managers skilled in discussing the physical, emotional, spiritual, and cultural needs of patients facing serious illness. These case managers can help address issues with coverage, connect patients ©Institute for Clinical and Economic Review, 2016 Page 25 to appropriate resources, assist doctors and caregivers in managing pain or symptoms, aid in coordination among providers, and facilitate advance care planning. Aetna reports an average reduction of about $12,000 in health care spending for each member enrolled in the program. In Aetna’s Medicare Advantage members, there has been an 82% reduction in the number of days spent in the hospital for acute care, an 86% reduction in days spend in intensive care units, and a 78% reduction in ER use. Blue Cross Blue Shield of Massachusetts Blue Cross Blue Shield of Massachusetts’ Complete Care for Advanced Illness program was launched in January 2016 with the goal of improving quality of life for individuals with advanced illness. Under this program, BCBSMA will reimburse both medical providers as well as behavioral health providers for having conversations with patients about planning for advanced illness and end of life care. This program will also provide more education and training support to help patients, families, and clinicians manage a progressive illness. BCBSMA will also extend this program to help patients with advanced illness receive high-quality palliative care in their homes. Read BCBSMA’s recent press release on this program here. https://www.bluecrossma.com/visitor/newsroom/press-releases/2015/2015-12-28.html ©Institute for Clinical and Economic Review, 2016 Page 26 4. Comparative Clinical Effectiveness 4.1 Overview Our review of published evidence examined higher quality comparative studies of palliative care interventions delivered outside the hospital. Interventions of interest were those implemented in outpatient settings that incorporate elements directed at both physical and psychosocial patient care, as both are considered essential and definitional components of palliative interventions.5,6 We recognized that such studies vary substantially in terms of their entry criteria and description of the interventions. In order to inform our analysis of the comparative clinical effectiveness of outpatient palliative care, we defined several characteristics of the interventions as of a priori interest:  the timing of the intervention (whether it was specified as early [as defined in relation to time of diagnosis or minimum period of time prior to death] or not);  the target population of the intervention (patients with cancer only or patients with mixed diagnoses, given the interest in expanding palliative care to patients with other progressive diseases);9 and  the type of service provided in the intervention (i.e., whether the level of provider training was characterized as specialist [with input from care providers described as having a palliative care specialty or certification] or generalist [with input from experienced care providers without reference to a specialty or certification]). A potential hurdle to drawing conclusions across studies is variability in the design of the outpatient palliative care interventions. As such, we investigated the potential contrast in outcomes between studies in which the intervention was described as being delivered by a care team incorporating individuals with higher specialist training (i.e., a specialist approach), or one made up of individuals with basic clinical experience in delivering palliative or end-of-life care and/or some level of in- service training but not described as receiving a specialty certification (i.e., a generalist approach). We also recognized that some of the literature lacked detailed descriptions of the interventions, including the type of training and education the palliative care providers received, making it difficult to ascertain whether the care team involved a specialized palliative care provider. Additionally, the HPM certification (as described in Section 2) was not available until 2008. Despite these limitations, we tried to determine if outcomes may have been influenced by the type of service (i.e., specialist vs. generalist) patients received in the palliative care intervention groups. The comparator treatment of interest was usual care (also referred to variously in the literature as standard care), which was typically defined with very little detail beyond that it incorporated standard access to clinical services provided at the study site. Notably, our inclusion of the 2015 Bakitas study11 on early versus delayed palliative care was primarily based on the outcomes reported at three months, which was the point at which the standard care group began receiving ©Institute for Clinical and Economic Review, 2016 Page 27 the intervention; comparisons between groups beyond this point were considered for our subgroup analysis of the timing of the intervention. Our review focused on specific clinical benefits reported in terms of patient QoL, mood, symptom burden, patient satisfaction, health care resource utilization (HRU), psychosocial and spiritual outcomes, caregiver impact, and survival. 4.2 Methods Data Sources and Searches Procedures for the systematic literature review assessing the evidence available for outpatient palliative care interventions followed established best methods used in systematic review research.54 We conducted the review in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.55 Our search was conducted among MEDLINE-, CINAHL-, PsychInfo-, and Cochrane-indexed articles, and the timeframe for our search spanned the period from January 2000 through November 2015. We limited each search to studies of human subjects, and excluded articles indexed as guidelines, letters, editorials, narrative reviews, case reports, conference abstracts, or news items. To supplement the above searches and ensure optimal and complete literature retrieval, we performed a manual check of the references of recent relevant reviews and meta-analyses. Further details on the search strategy are available in Appendix 4. Our search was limited to studies conducted in the U.S. and Canada as we felt that differences in the use of palliative care and health-system interactions in other countries would limit generalizability to the U.S. system. Study Selection We screened the studies identified through our searches, first examining the titles and abstracts for clear exclusion criteria, and subsequently screening the full text of the remaining publications for presence of specific criteria. Trained investigators conducted screening according to defined inclusion and exclusion criteria, with any questions regarding their application being resolved in discussion with a second or third investigator. We did not exclude any study at abstract level screening due to insufficient information. For example, an abstract that did not describe reporting any outcome of interest was accepted for further review in full text if no exclusion criteria were present. We retrieved the citations that were accepted during abstract-level screening for full text appraisal. Two investigators reviewed full papers, and provided justification for exclusion of each excluded study. Inclusion and exclusion criteria are described below: Inclusion Criteria:  Any palliative care intervention outside hospital in outpatient clinic, home or community setting, in conjunction with active treatment or not  RCT or higher quality observational studies (case-control or cohort studies) ©Institute for Clinical and Economic Review, 2016 Page 28  Adult population 18 years or older with serious or life-threatening illnesses  Conducted in the U.S. or Canada  Evaluated at least one of the following outcomes of interest: o Survival or mortality, including location of death o Health resource utilization o Symptom severity, including pain, breathlessness, fatigue, and nausea o Patient, caregiver, and/or family psychological distress (e.g., stress, anxiety, depression) o Ratings of patient, caregiver, family, and provider satisfaction o Health-related QoL o Economic outcomes, including program start-up costs, costs per patient, potential cost offsets, and measures of cost-effectiveness Exclusion Criteria:  Studies outside of the U.S. or Canada  Studies conducted in children or mixed population  Studies conducted in a population residing in a hospital, nursing facility, or hospice  Evaluates a drug or procedure trial that is not part of a larger palliative care intervention  Study design has no comparison group (usual care as control)  No outcomes of interest Data Abstraction and Quality Assessment Summary tables capturing evidence abstracted from the reviewed studies are available in Appendix 8. We abstracted outcome data for each treatment group, and used criteria modified slightly from those published by the U.S. Preventive Services Task Force (USPSTF) to assess the quality of RCTs and comparative cohort studies, using the categories “good,” “fair,” or “poor” as described below:56 Good: Meets all criteria: Comparable groups were assembled initially and maintained throughout the study (follow-up at least 80 percent); reliable and valid measurement instruments were used and applied equally to the groups; interventions were spelled out clearly; all important outcomes are considered; and appropriate attention paid to confounders in analysis. In addition, for RCTs, intention to treat analysis was used. Fair: Studies were graded "fair" if any or all of the following problems occurred, without the fatal flaws noted in the "poor" category: Generally comparable groups are assembled initially but some question remains whether some (although not major) differences occurred with follow-up; measurement instruments were acceptable (although not the best) and generally applied equally; some but not all important outcomes were considered; and some but not all potential confounders were addressed. At least modified intention to treat (mITT) analysis was done for RCTs. ©Institute for Clinical and Economic Review, 2016 Page 29 Poor: Studies were graded "poor" if any of the following fatal flaws existed: Groups assembled initially were not close to being comparable or maintained throughout the study; unreliable or invalid measurement instruments were used or not applied equally among groups (including not masking outcome assessment); and key confounders were given little or no attention. For RCTs, intention to treat analysis was lacking. Several RCTs evaluated the treatment groups based on a modified intent-to-treat (mITT) analysis in which those subjects who did not receive the intervention after randomization were excluded due to death or worsening illness. This approach is generally accepted as the primary evaluation method in palliative care given the progression of life-limiting diseases in the study populations.57 We categorized those studies as “fair” quality rather than “poor” quality which would otherwise be employed on the basis of not utilizing a strict ITT analysis. Assessment of Level of Certainty in Evidence (ICER EBM Matrix) We used the ICER Evidence Rating Matrix (see Figure 2) to evaluate the evidence for a variety of outcomes. The evidence rating reflects a joint judgment of two critical components: a) The magnitude of the difference between a therapeutic agent and its comparator in “net health benefit” – the balance between clinical benefits and risks and/or adverse effects AND b) The level of certainty in the best point estimate of net health benefit.58 ©Institute for Clinical and Economic Review, 2016 Page 30 Figure 2. ICER Evidence Rating Matrix ©Institute for Clinical and Economic Review, 2016 Page 31 4.3 Results Study Selection and Patient Population The literature search for comparative studies of outpatient palliative care identified 3,191 potentially relevant references (see Figure 3 on the following page), of which 1710-26 publications (describing 14 studies) met our inclusion criteria and focused on good or fair quality evidence. An additional six studies of poor quality were also identified, but are not described in detail below. We abstracted evidence from all studies included in this review. Details around the populations, interventions, and outcomes are summarized in Appendix 8. Assessment of Publication Bias Scanning of the clinicaltrials.gov site to identify additional studies completed more than two years ago that would have met our inclusion criteria but have not been published revealed only one study that was completed in 2009 (NCT00648609). In this small comparative study (n=23), the investigators randomly assigned COPD patients 60 years or older to home-based self-management services provided by an interdisciplinary palliative care team or usual care. During the one-year follow-up period, health service utilization, symptom scores, and patient QoL were evaluated; no results were provided. ©Institute for Clinical and Economic Review, 2016 Page 32 Figure 3. PRISMA Diagram 3,191 potentially relevant references screened 2,800 citations excluded Population: 101 Intervention: 545 Comparator: 264 Outcomes: N/A Source: 789 391 references for full text Setting: 599 review Duplicates: 502 369 citations excluded (not conducted in North America, not a multidisciplinary intervention, comparisons of 23 publications individual components, 14 RCTs no access) 2 secondary analyses 7 non-randomized comparative studies Overall Evidence Quality As noted earlier, we used modified criteria from USPSTF to rate the quality of the comparative studies. Based on these criteria, we considered seven studies (three RCTs and four cohort studies) to be of good quality, with study arms comparable at baseline, valid instruments to evaluate outcomes, and limited differential attrition occurring during the outcomes assessment.10,13,17,20-22,26 Also using these criteria, seven studies (five RCTs and two cohort studies) were categorized as fair quality,11,12,18,19,23-25 most often on the basis of the use of a modified rather than full intent-to-treat analysis of randomized treatment arms. We also abstracted data from six studies deemed to be of poor quality59-64 but do not describe them in detail in this review. We have included mention of these studies only when important for additional context around our findings. Studies were categorized as poor quality primarily because of substantial attrition without adjustment of data, or due to substantial and unaccounted for differences in the proportion of patients lost to follow-up in each treatment arm. ©Institute for Clinical and Economic Review, 2016 Page 33 The overall dearth of RCTs and higher quality comparative studies in this area is perhaps unsurprising, as it is very difficult to implement a well-designed comparative study given ethical concerns and challenges with standardizing outpatient palliative care regimens across health care systems. Most studies described as fair compared patient groups with disparate demographic or clinical characteristics. Those described as poor did not present enough information to make this determination or did not sufficiently attempt to control for variables potentially confounding the relationship of the interventions to outcomes (in some cases reporting imbalances in factors such as disease mix or age of patients at baseline, and in some instances not reporting such baseline characteristics). Interventions Among the 14 higher quality original studies identified, the palliative care interventions were highly heterogeneous. As described in the methods, we recognized that such studies vary substantially in terms of described interventions, and we defined several characteristics as of a priori interest: intervention timing (early or not early); target population (cancer or mixed diagnoses); and type of service (specialist or generalist). Specialist interventions were described by 10 studies,10-22 while generalist interventions were described by four.23-26 We note, however, that most of the interventions described as “generalist” appear to be multi-disciplinary, follow core principles of palliative care, and are delivered by experienced personnel, making the distinction between these approaches and those led by certified professionals somewhat artificial. Four studies10,11,21,22 evaluated the effect of palliative interventions described as early while 10 studies12,13,17-20,23-26 did not specify early initiation of the intervention. Five studies10,11,21,22,25 evaluated outpatient palliative care’s effect on patients with cancer while nine studies12,13,17-20,23,24,26 did so in populations of patients with mixed diagnoses. Key Studies Earlier sections of this report describe several model programs for delivery of outpatient palliative care. Below we describe several of the key studies arising out of those programs. Bakitas 200910 This RCT was performed at Dartmouth-Hitchcock Medical Center, whose palliative care services are described earlier in this report (Section 2.4, Existing Approaches to Palliative Care). This publication described the implementation of the ENABLE II program, a specialized telehealth palliative care program designed to provide care to patients with advanced cancer living in rural areas. The outpatient palliative care intervention was based on a case management approach, with four educational sessions conducted by an advanced practice nurse (APN) with specialty training in palliative care; patients also received at least monthly follow up by telephone until death or study ©Institute for Clinical and Economic Review, 2016 Page 34 end. Additionally, patients and caregivers in the intervention group were invited to monthly telephone-based shared appointments. This RCT was designed to evaluate the effect of this outpatient palliative care program on QoL, symptoms, mood, and resource utilization. Investigators randomized 322 patients enrolled from 2003 to 2008 to either the intervention (n=161) or to usual care (n=161). The authors state that patients in the control arm were allowed unrestricted use of all oncology and supportive services at the study sites: the Norris Cotton Cancer Center/Dartmouth College and the Veterans Administration Medical Center, White River Junction, VT. Temel 201021 This RCT was performed at Massachusetts General Hospital, whose palliative care services are described earlier in this report (Section 2.4, Existing Approaches to Palliative Care); this study evaluated the impact of a specialized outpatient palliative care program on patients newly diagnosed with metastatic non-small cell lung cancer (NSCLC). Patients randomly assigned to the intervention met with either a board certified palliative care physician or APN within three weeks of enrollment. Meetings continued at least monthly thereafter until death, with additional visits scheduled at the discretion of the patient or provider. Investigators randomized 151 ambulatory patients to the above-described intervention (n=77) or to standard care (n=71) within eight weeks of diagnosis; the control group was described as routine oncologic care. Outcomes evaluated included QoL and mood at 12 weeks, as well as end-of life-care as documented in electronic medical records. Brumley, 200712 This RCT was performed at two health maintenance organizations participating in Kaiser Permanente’s home-based palliative care program in Hawaii and Colorado. Investigators randomized homebound terminally ill patients to the palliative care intervention (n=145) or to usual care, for which no additional description was provided (n=152). The specialized in-home palliative care plus usual care intervention was delivered to patients with mixed diagnoses by an interdisciplinary team who coordinated care from a variety of health care providers, including palliative care specialists and the patients’ primary care physician; the focus was on measuring patient satisfaction, resource utilization, and location of death. The control arm received usual care, described only as “standard care to meet the needs of the patients following Medicare guidelines for home health care criteria.” The study population had primary diagnoses of cancer (47%), CHF (33%), and COPD (21%). Rabow 200424 This RCT evaluated the effect of providing an outpatient palliative care intervention based on a case management approach which integrated care provided by family caregivers and primary care physicians for patients with either cancer, advanced COPD, or advanced CHF. The intervention aimed to improve QoL and advance care planning, reduce health care utilization, and support caregivers through resources provided by a multidisciplinary team. The authors randomized ©Institute for Clinical and Economic Review, 2016 Page 35 90 patients with a life expectancy between one and five years to either outpatient palliative care or usual care, for which no additional description was provided. Seow 201420 This comparative cohort study was carried out in Ontario, Canada, where several independently developed community-based, specialist palliative care teams operate; these teams served patients with mixed diagnoses (80% had cancer) from different catchment areas but had the same core team members within the same health financing system. In this study, 3,109 patients receiving specialist palliative care were matched by propensity score to the same number of patients receiving usual care. The purpose of this retrospective study was to capture the impact of a specialist palliative care team on emergency department (ED) visits and hospital admissions in the last two weeks of life, as well as the number of patients dying in the hospital. Strength of Evidence In an effort to define potentially contrasting characteristics of the outpatient palliative care interventions described in the available literature, we noted that there were interventions which could be classified as “specialist” or “generalist” levels of palliative care. These were defined according to the education level and training of palliative care providers as described in Section 2.2 of this report. Table 2 represents the strength of evidence around the relative effect of outpatient palliative care compared to usual care for the key outcomes of interest, with evidence around the more granular aspect of the intervention type (specialist vs. generalist). We present the total numbers of studies that reported on each outcome in the “overall” column, and also by type of intervention in separate columns. ©Institute for Clinical and Economic Review, 2016 Page 36 Table 2. Strength of Evidence for Outpatient Palliative Care Compared to Usual Care Outcome Specialized care Generalist care Overall QoL 5 studies 2 studies 7 studies Resource Use 8* studies 3 studies 11 studies Survival 3* studies 1 study 4 studies Symptom Burden 5 studies 2 studies 7 studies Patient Satisfaction 3 studies 1 study 4 studies Psychosocial and Spiritual 1 study 2 studies 3 studies Mood 4 studies 2 studies 6 studies Caregiver Outcomes 1 study 0 studies 1 study * Bakitas 2015 was not included in these results because the primary comparison for these outcomes was early vs. delayed palliative care. Color Code: High Moderate Low No evidence Specifically, high strength of evidence is defined as consistent reporting of statistically significant results in favor of the intervention in at least four studies. Moderate strength of evidence is suggested by at least half of the studies consistently reporting statistically significant results in the evidence base, and consistent findings of benefit in at least three studies. Low strength of evidence is suggested by a statistically significant benefit observed in fewer than three studies. No evidence is suggested where there are either no studies evaluating the effect of a particular type of intervention on the outcome of interest, or where there is no evidence of statistically significant benefit. While we found no strong evidence of the benefit of specialist-led outpatient palliative care for any of the outcomes evaluated, we did find moderate evidence for such benefit on QoL, resource utilization and patient satisfaction. Weaker evidence (low level) also suggests a benefit on survival, symptom burden, mood, and caregiver outcomes. There is no evidence for benefit on psychosocial outcomes as represented by the one study of a specialist intervention evaluating this outcome. Findings for those three studies evaluating generalist palliative care were mixed, likely due to the constrained evidence base. Overall, the evidence describing outpatient palliative care’s benefit is stronger for QoL, resource utilization outcomes, patient satisfaction, and mood outcomes, with weaker evidence suggesting benefits on survival, symptom burden, psychosocial, and caregiver outcomes. These observations are consistent with the fact that outpatient palliative care programs are designed to increase ©Institute for Clinical and Economic Review, 2016 Page 37 patient social support, patient self-advocacy, and coordinated medical care; while palliative care is not focused on improved survival as an indicator of effectiveness, the survival benefit may be mediated by the other more directly influenced outcomes of interest (e.g., mood, QoL). Results for specific outcomes can be found in the sections that follow. QoL We identified three studies10,21,22 of good quality and four studies11,18,24,25 of fair quality that reported on patient QoL. The evidence base suggests that palliative care either improved patient QoL or had comparable effects relative to usual care. Of note, no clinically important differences were identified for any of the QoL instruments used in the studies, nor were any responder analyses undertaken in the literature reviewed. Four10,21,22,25 of the seven studies that evaluated palliative care interventions, all of which were conducted in cancer patients only, showed positive results for QoL outcomes. In the previously- described ENABLE II study, QoL on the FACIT-Pal instrument was statistically significantly improved at 13 months of follow-up in the early (initiated within 8-12 weeks after diagnosis) palliative care group (mean difference, 4.6; p=0.02).10 QoL was also improved in Temel’s study of early palliative care (initiated within eight weeks after diagnosis) as measured by the Functional Assessment of Cancer Therapy- Lung (FACT-L) at 12 weeks (98.0 vs. 91.5 for usual care; p=0.03).21 Zimmermann evaluated early initiation of a palliative care consultation and follow-up in an oncology clinic provided by a palliative care physician and nurse in 461 patients with an estimated survival of 6-24 months. At four months, the overall QoL measured by FACIT-Sp (a combination of FACT-General and Sp-12 spiritual well-being) had a greater increase in the intervention group versus usual care (adjusted mean difference, 4.34; 95% CI, 0.70 to 7.98; p=0.02).22 Finally, Rummans et al. evaluated the feasibility and effectiveness of a four-week, structured, multidisciplinary intervention with a physical therapy component targeted to maintain the overall QoL in 103 patients with advanced cancer undergoing radiation therapy.25 At week four, the Spitzer QoL Uniscale in the intervention group increased by three points from baseline, compared to a nine-point decline in the control group (p=0.009); however, these differences did not persist at week eight or 27. In contrast to the studies above, our review identified three studies11,18,24 suggesting that outpatient palliative care’s effects on QoL were comparable to usual-care approaches (i.e., both groups saw similar improvements in QoL); these comparable results may be explained by lower statistical power in these studies given small sample sizes (n<100 in two studies), shorter durations of follow- up (three months in ENABLE III vs. 13 months in ENABLE II), or the use of instruments that may not be generalizable to patients with indications other than cancer. ©Institute for Clinical and Economic Review, 2016 Page 38 Target Population, Timing, and Type of Service Both study population and timing of palliative care may be drivers of the variable findings on QoL. In general, the effect of palliative care on QoL was more often evaluated in cancer patients than other patient populations (five cancer-only studies vs. two studies in mixed populations). Timing of care was directly addressed in one study11 that compared early versus delayed palliative care, while three compared early palliative care to usual care. It was difficult to separate the role of timing of intervention from target population because all four early studies were conducted in cancer patients. Early palliative care in cancer patients consistently yielded positive results compared to usual care. Specifically, palliative care for cancer patients yielded statistically significantly beneficial QoL effects (four out of five studies), including all three studies that compared early interventions to usual care. However, early care showed no advantage over delayed care in cancer patients when compared directly in one RCT11 of early palliative care versus a three-month delay in palliative care initiation, either before the delayed group started the intervention or during the one-year follow-up. The other two studies in mixed populations both reported non-significant results. There was no consistent evidence according to type of service (i.e., specialized versus generalist); however, this observation is not conclusive due to the limited number of studies evaluating generalist palliative care. Mood Outcomes We identified a total of six studies evaluating mood outcomes, including anxiety and depression: two good quality10,21 and four fair quality11,18,24,25 RCTs. Of these six studies, four10,21,24,25 found statistically significant results in favor of the intervention, though not all studies evaluated both outcomes. Only two10,21 of the five studies evaluating depression showed statistically significant improvements from palliative care interventions. Both Bakitas trials10,11 evaluated depression on the CES-D measurement tool but only one10 had a statistically significant finding in favor of the intervention based on a mixed model analyses for repeated measures (TE= -1.8, p=0.02). A longer duration of follow-up in the earlier study (13 months vs. three months in the 2015 study) may have influenced this outcome. However, in the Temel RCT,21 statistically significant findings were in favor of an early palliative care intervention based on the proportion of patients with elevated depression scores on the HADS-D (16% vs. 38% for usual care, p=0.01) and with symptoms of major depression on the PHQ-9 (4% vs. 17% for usual care, p=0.04) after only three months of follow-up. For studies evaluating anxiety, only two24,25 out of four studies found statistically significant between-group differences in favor of the intervention. The Rummans RCT25 found a statistically ©Institute for Clinical and Economic Review, 2016 Page 39 significant difference on the POMS tension/anxiety subscale in the intervention group after one month (p=0.042), but the authors did not report the mean scores associated with this outcome. Rabow et al.24 reported that anxiety was reduced in the intervention group but increased for those receiving usual care between six and 12 months of follow-up in an analysis controlling for anxiety at baseline and adjusting for group-by-time interaction (-1.5 vs. +0.4 for control, p=0.05). Target Population, Timing, and Type of Service No clear relationship emerged based on the three factors we identified as potentially influencing outcomes. Three10,21,25 of the four studies conducted in cancer patients indicated relative benefit of the intervention on mood outcomes, while one24 of the two studies with mixed populations favored the intervention. Of the six studies evaluating mood outcomes, three RCTs10,11,21 evaluated early palliative care and two RCTs10,21 found significant differences in favor of an early palliative care intervention. Similarly, two of the three studies not described as early reported statistically significant differences in favor of the intervention.24,25 Four of the six studies were considered specialist palliative care. Two of them found significant differences in favor of the intervention10,21, while both studies considered to be generalist care showed beneficial effects of the intervention.24,25 However, it is difficult to extrapolate any meaning from these findings due to the limited number of studies. Symptom Burden Most available studies found that palliative care had no impact on overall symptom burden compared to usual care, but some studies found beneficial effects on certain disease-specific symptoms and physical symptoms in general. The commonly-reported symptoms included pain, fatigue, drowsiness, nausea, anxiety, depression, appetite, dyspnea, and sleep quality, as measured by composite scores or frequency and intensity of specific symptoms. The same seven RCTs reporting QoL outcomes also reported on symptom burden, and four of seven found no statistically significant differences in any symptoms between groups. Possible explanations for lack of benefit in these studies included some symptom scales were not validated in populations receiving home- based palliative care in one study;18 a relatively low symptom severity in both groups at baseline in two studies;10,22 and measurement insensitivity as patients approach death, which was mentioned in one study.11 The other three studies, one of good quality21 and two of fair quality,24,25 showed reductions in some disease-specific symptoms and physical symptoms. A secondary analysis of the previously- described Rummans RCT found that physical symptoms, as measured by Linear Analog Scales of Assessment (LASA) physical well-being scale, were reduced in the intervention group while increased in the control group at week four (-10.0 vs. +0.4, p=0.022).14 However, additional ©Institute for Clinical and Economic Review, 2016 Page 40 subscales measuring the severity and frequency of pain, fatigue, activity, and symptom distress showed no differences. The Temel RCT found that the intervention group scored better at 12 weeks on the lung cancer subscale (LCS) of the FACT-L scale (21.0 vs. 19.3 for usual care, p=0.04), which evaluates seven symptoms specific to lung cancer.21 Finally, Rabow et al. reported that patients in the control group reported a significantly higher rate of dyspnea than the intervention group (OR=6.07, 95% CI, 1.04 to 35.56), as well as significantly higher degree of interference from dyspnea on daily activities (25.4 vs. 40.6 for controls, p=0.01) at 12 months based on the University of California, San Diego Shortness of Breath Questionnaire.24 Target Population, Timing, and Type of Service As with mood outcomes, we found no consistent evidence associated with the study population or timing of care for symptom burden outcomes. Results associated with symptom burden were similar in cancer-only studies and studies with mixed populations: two of five cancer-only studies and one of two studies with mixed populations reported positive effects. There also was no clear pattern in terms of timing of care: two of three studies that did not specify early initiation of palliative care reported positive effects, while one of three studies that compared early palliative care to usual care in cancer patients showed a benefit. The one RCT comparing early versus delayed palliative care showed no statistically significant difference between groups. In terms of type of service, both generalist interventions significantly improved symptoms compared to usual care, but only one of five specialist interventions had positive effects; this finding is not conclusive, however, due to the limited number of generalist care studies. Patient Satisfaction The available evidence consistently suggests that palliative care improves patient satisfaction relative to usual care. One good quality prospective cohort study13 and three RCTs, including one of good quality22 and two of fair quality,12,24 reported on patient satisfaction; three of these studies found higher satisfaction or a greater increase in satisfaction in the intervention group compared to controls. Another RCT in patients with mixed diagnoses found no difference in satisfaction between groups, likely due to small sample size.24 In a two-year study, Brumley et al. prospectively compared 161 patients enrolled in a home-based palliative program for end-of-life care and 139 who received usual care with a life expectancy of less than one year.13 Patient satisfaction was measured by the Reid-Gundlach Satisfaction with Services instrument. At 60 days, the satisfaction score of the intervention group was statistically significantly higher than baseline (t=-2.75, p=0.01) while the control group had no change (t=-0.5, p=0.6); statistical differences between groups were not evaluated, however. In a more recent RCT by the same authors, Brumley et al. evaluated the same intervention in 298 terminally ill patients using the Reid-Gundlach Satisfaction with Services instrument.12 In this study, the authors defined the rate of ©Institute for Clinical and Economic Review, 2016 Page 41 satisfaction as the proportion of patients reporting being “very satisfied” (score of 37 or above), and found increased rates in the intervention group at 30 days (OR=3.37 vs. controls; 95% CI, 1.42 to 8.10; p=0.006) and 90 days (OR=3.37 vs. controls; 95% CI, 0.65 to 4.96, p=0.03). Another study reported that patients receiving early palliative care had significantly improved satisfaction at four months as measured on the FAMCARE-P16 (+3.7 vs. -2.4 for control, p<0.0001); sensitivity analyses also revealed similar results.22 Target Population, Timing, and Type of Service It is difficult to compare patient satisfaction results according to study population, intervention timing, or type of service because only one study22 described an early palliative care intervention in cancer patients and only one24 is considered to be generalist palliative care. Survival Our review identified four good quality studies evaluating the hypothesis that outpatient palliative care may affect patient survival, of which three were RCTs10,12,21, and one a propensity score- matched cohort study.26 The evidence base suggests that palliative care provided in the outpatient setting does not negatively affect patient survival,12,26 and may in fact result in increased survival in populations with advanced cancer,10,21 particularly when adjusting for potential confounders of the relationship between the intervention and survival (e.g., The Eastern Cooperative Oncology Group [ECOG] performance status score, or survival for less than one year).10,21 Temel et al.21 reported that median survival was significantly improved in the early outpatient palliative group compared to the usual care group (11.6 vs. 8.9 months, p=0.02). In an adjusted regression analysis, the investigators estimated the hazard ratio for death in the usual care group to be 1.70 (95% CI, 1.14 to 2.54; p=0.01).21 Another RCT, conducted by Bakitas et al., also reported a longer overall median survival in the intervention group but these results were not statistically significant.10 As in the Temel paper, the authors present a Cox regression model adjusting for chemotherapy and site of care, which found a significantly reduced risk of death in the intervention group during the first year of follow-up (HR=0.67; 95% CI, 0.50 to 0.91; p=0.009).10 Two other studies identified in our review reported no association between outpatient palliative care and survival of patients with mixed diagnoses: one a retrospective comparative cohort study26 examining survival at 30 days, and another an RCT incorporating Kaplan-Meier survival analysis.12 Target Population, Timing, and Type of Service Target population and timing of palliative care are inextricably linked in the limited evidence base describing survival, with both studies in cancer patients also describing their interventions as early, making it difficult to identify clear patterns.10,21 However, an RCT that directly compared early versus delayed palliative care in cancer patients showed a significant improvement in one-year survival rate in the early group (63% vs 48%, p=0.04).11 Three of the four studies evaluating survival ©Institute for Clinical and Economic Review, 2016 Page 42 as an outcome of palliative care described specialist interventions, rendering it difficult to contrast these findings.10,12,21 Resource Utilization Among the 11 studies reviewed with resource utilization data, there was a generally observed benefit of outpatient palliative care on overall resource utilization, measured variously as an increase in hospice utilization or death at home, or a decrease in utilization of acute care services.10,12,13,16-20,23,24,26 Benefits were noted more predominantly among cohort studies than among RCTs, possibly a result of such data being more often collected in such study designs.16,17,24 We analyzed the effect of outpatient palliative care on specific health resource utilization outcomes as outlined below: death at home/outside the hospital (evaluated as one category of outcome), ED visits, hospital utilization, and hospice utilization. Among the 10 studies providing appropriately powered results, eight reported a statistically significant association of outpatient palliative care with at least one of these outcomes.12,13,16,17,19,20,23,26 Two of these studies reported no such associations with either hospital or ED visits.10,24 Death at Home/Outside the Hospital The available studies suggest that outpatient palliative care results in a larger proportion of patients dying at home, or a smaller proportion dying in the hospital. Three studies (two fair quality RCTs, one good quality cohort study) evaluated the impact of outpatient palliative care on the location of death.12,16,20 While presented here as a resource utilization outcome, the concept of death at home is highly nuanced and incorporates elements of patient desire and satisfaction. Brumley et al. reported the results of an RCT estimating that the proportion of patients who died at home over the period of follow-up was statistically significantly greater in the outpatient palliative care group than in the control group after adjusting for age, survival time, and medical conditions (69% vs. 50%; OR, 2.20; 95% CI, 1.3 to 3.7; p<0.001).12 An alternative to the analysis of death at home as an outcome was evaluated in a propensity-matched cohort study, which found that fewer patients receiving outpatient palliative care died in the hospital compared to those receiving usual care (16.2% vs. 28.6%; RR, 0.46; 95% CI, 0.40 to 0.52).20 In contrast to these findings, a secondary analysis of data derived from the Temel RCT21 yielded no significant difference in the proportion of patients dying at home between the groups receiving early palliative care and usual care.16 ED Visits Mixed results were noted regarding associations of outpatient palliative care with a reduction in ED visits. Our review identified seven good and fair quality studies evaluating the effect of outpatient palliative care on ED visits, of which five were RCTs.10,12,13,18,20,21,24 Three of these studies reported ©Institute for Clinical and Economic Review, 2016 Page 43 significant associations of outpatient palliative care with a reduction in ED visits,12,13,20 while four reported no significant effect.10,18,21,24 In three publications, outpatient palliative care was associated with fewer ED visits.12,13,20 A cohort study conducting multivariate analyses (MANCOVA) controlling for differences in severity of illness at baseline revealed that patients receiving outpatient palliative care had significantly fewer ED visits (0.93 vs. 2.3 visits; p<0.001).13 In a later RCT by the same author, investigators randomized patients to a palliative care program delivered by an interdisciplinary team (n=145), or to usual care (n=152); regression analysis controlling for survival time indicated that the intervention reduced the number of ED visits by 0.35 (r2=0.04, p=0.02) compared to the usual care group.12 Similarly, a propensity-matched cohort study estimated the relative risk (RR) of an ED visit in the last two weeks of life among the intervention group compared to the control to be 0.77 (95% CI, 0.69 to 0.86).20 Hospital Utilization Our review identified eight publications describing the effect of outpatient palliative care on hospital admissions and hospital days.10,12,13,18,20,21,24,26 Of the six studies, four RCTs reported no effect of outpatient palliative care on measures of hospital utilization10,18,21,24, while one RCT12 and three cohort studies reported a significant reduction in such measures13,20,26. This contrast by study design may be attributed to larger sample size and longer follow-up in these cohort studies. In a previously-described cohort study, Brumley et al. reported that patients treated with specialized outpatient palliative care had significantly fewer hospital visits than did those treated with usual care (2.36 vs. 9.35, p<0.001).13 In another RCT, Brumley et al. reported that patients treated with specialized outpatient palliative care were less likely to be admitted to hospital (36% vs. 59%, p<0.001) and stayed in hospital for fewer days (a reduction of 4.36 days, p<0.001) than did those treated with usual care.12 Another previously-described propensity-matched cohort study26 compared patients receiving home palliative care (n=392) to patients receiving usual home care (n=890). While 30-day mortality rates were similar in the two groups, the outpatient palliative care group had a significantly lower probability (9.1%) of a 30-day hospital readmission compared to usual care (17.2%), yielding an average treatment effect on treated patients of 8.3% (95% CI, 8.0 to 8.6). A third cohort study also reported a reduction in hospitalization in the last two weeks of life in patients with mixed diagnoses receiving specialized outpatient palliative care (RR, 0.68, 95% CI, 0.61 to 0.76).20 Hospice Utilization Our review identified seven studies evaluating the effect of outpatient palliative care on hospice utilization.10,12,16,17,19,21,23 Four of these, one secondary analysis16 of an RCT21 and three cohort studies,17,19,23 reported a significant association of outpatient palliative care on some element of hospice utilization. In contrast, three RCTs reported no such relationship for the proportion of ©Institute for Clinical and Economic Review, 2016 Page 44 patients enrolling,12 the mean duration of time spent,21 or number of patients being referred to hospice care.10 In a secondary analysis of data derived from an RCT,21 Greer et al. evaluated the effect of early outpatient palliative care on hospice care and found that while there was no significant difference in the rates of referral to hospice, significantly more patients receiving outpatient palliative care were enrolled in hospice at least week before death (60.0% vs. 33.3% for control, p=0.004).16 This study also found that in an analysis of only those patients who were referred to hospice, the mean duration of time spent in hospice care was statistically significant in favor of early palliative care (24.0 vs. 9.5 days, p=0.02). Similar findings were observed in Scheffey et al., who compared prior outpatient palliative care reported in a cohort of hospice patients diagnosed with mixed terminal illnesses with that reported by a matched group without such prior care.19 This study found that the outpatient palliative care group (n=342) had a significantly longer length of stay (LOS) in hospice than did the usual care group (n=1,368) (median LOS 24 vs. 15 days, p<0.001). A retrospective cohort study conducted in 435 patients with mixed indications and a life expectancy of less than six months evaluated 140 patients receiving outpatient palliative care and two different groups of patients receiving usual care (n=68 and n=227, respectively).23 In this study, Ciemins et al. reported that a higher proportion of patients receiving outpatient palliative care received hospice referrals (47.2%) compared to those receiving usual care (33.8%) (p=0.003); this finding was particularly pronounced among African American patients, of whom 47.0% receiving outpatient palliative care were referred to hospice compared to 18.8% receiving usual care (p=0.008). Another retrospective database study evaluating 149 deceased patients who had received home-based palliative care found that significantly more intervention patients enrolled in hospice compared to 537 propensity-matched comparison patients (70% vs. 25%, p<0.0001).17 The median time spent in hospice was also statistically significant in favor of the intervention group (34 vs. 9 days for control, p=0.0003). Target Population, Timing, and Type of Service Even though the evidence base suggests target population and timing may drive the variability of the observed outcomes, with one16 of two studies of early palliative care in cancer patients versus seven12,13,17,19,20,23,26 of eight studies of palliative care in patients with mixed diagnoses showing benefits on health resource utilization, the fact that these two factors are inextricably linked makes it difficult to distinguish their roles from each other. An additional RCT evaluating early versus delayed specialist palliative care in advanced cancer patients also evaluated the relative rates of hospital and intensive care unit days, as well as number of patients dying at home, but none of these outcomes were statistically significant.11 As to type of service, independent of target population and timing, all five studies describing specialist palliative care in patients with mixed diagnoses12,13,17,19,20 showed some improvement in ©Institute for Clinical and Economic Review, 2016 Page 45 health resource utilization, while two of three studies describing generalist palliative care in these patients reported no such association.23,26 However, these contrasts are inconclusive given the limited number of studies. Psychosocial and Spiritual Outcomes We identified three RCTs of fair quality evaluating psychosocial outcomes for patients, including spiritual well-being and advance care planning.18,24,25 Radwany et al. was a small pilot study not powered to detect differences in these outcomes, so those non-significant results are not described in further detail.18 The Rabow RCT showed a statistically significant improvement in spiritual well-being after a year of follow-up as measured on the Spiritual Well-Being Scale for both the mean effect over time (f=8.21, p=0.007) and group-by-time (f=4.24, p=0.05) interaction analyses in favor of the palliative care intervention.24 Advance care planning was also measured based on a validated questionnaire which asked if patients had completed or considered a durable power of attorney, funeral plans, and plans for disposition of possessions after death.24 Of the patients who had not completed funeral arrangements at baseline, 35% of intervention patients and 5% of control patient had done so by the end of the study (p=0.03), but between-group comparisons for consideration of durable power of attorney or plans for disposition of possessions after death were not significant. Although Rummans et al. also reported statistically significant effects for overall spiritual well-being on the FACIT tool in favor of the intervention after one month (92.9 vs. 83.9 for control; p=0.003), this effect did not persist at weeks eight or 27.25 Target Population, Timing, and Type of Service Due to the limited number of studies evaluating psychosocial outcomes, it is unclear whether the patient population in these studies (those with cancer vs. mixed indications) or timing (early vs. late) had any effect on these outcomes. Interestingly, the only study18 with a palliative care specialist was the only one that did not find statistically significant outcomes in favor of the intervention for spiritual well-being. As previously mentioned, however, this study was not powered to detect between-group differences. Caregiver Outcomes We identified only one good quality study15 reporting outcomes for caregivers, including depression, QoL, and caregiver burden. While this study did not find any differences between groups for QoL or caregiver burden, caregivers of patients receiving early palliative care had greater reductions in depressive symptoms than those receiving usual care based on the CES-D after three months (mean difference, -3.4; SE, 1.5; p=0.02). ©Institute for Clinical and Economic Review, 2016 Page 46 We identified three additional RCTs60-62 in which the intervention specifically targeted caregiver outcomes and appeared to favor the palliative care intervention with regards to lessening strain/burden, reducing stress, and improving QoL. However, these studies received a poor quality rating and were excluded from our narrative analysis; details of these studies can be found in the data abstraction tables (Appendix 8). Target Population, Timing, and Type of Service Given that only one higher quality study reported on caregiver outcomes, no comparisons of intervention characteristics across studies for the target population or service type can be made. However, Dionne-Odom et al.15 also evaluated caregivers who received palliative care at enrollment compared to a delayed group receiving the intervention three months later and found that for caregivers of those patients that died, a terminal decline analysis demonstrated that depression (CES-D; mean difference, -3.8; SE, 1.5; p=0.02) and stress burden (MBCB; mean difference, 1.1; SE, 0.4; p=0.01) were statistically significantly better in the early intervention at 36 weeks of follow-up; however, the findings were not significant for caregiver QoL, objective burden, or demand burden. Overall Summary: Net benefit and level of certainty for all comparisons The body of evidence derived from higher quality comparative studies of outpatient palliative care interventions in the U.S. and Canada is somewhat constrained and variable with respect to intervention target and structure. Some studies reported on interventions specifying early enrollment of patients into a palliative care program, while others made no such distinction; some studies evaluated interventions in patients with cancer, while others were implemented in patient populations with a variety of serious illnesses; and some studies described interventions in which at least one member of the core care team had a specialty in palliative care, while others described interventions implemented by care teams with experience in palliative care but no specialty or certification. It is also challenging to pool information across comparative studies because these studies examined distinct patient populations with different disease entities and variable severities of illness. Despite the limitations of the published evidence derived from comparative studies, it is possible to use the ICER rating system to generate an estimate of the “net health benefit” describing the balance between clinical benefits and risks and/or adverse effects for outpatient palliative care relative to usual care. Although we sought to identify any potential harms associated with palliative care, the studies included in our review did not report any adverse events connected with the interventions; therefore, our evaluation of net health benefit is based solely on the noted advantages. ©Institute for Clinical and Economic Review, 2016 Page 47 It is our judgement that there is moderate certainty that outpatient palliative care confers a comparable or better net health benefit relative to usual care. Additionally, given the lack of harms associated with such interventions, we have high certainty that the net health benefit is at least comparable. This yields an overall ICER Evidence Rating of C+: Comparable or Better for outpatient palliative care. 4.4 Elements for Successful Palliative Care Programs In the sections that follow we describe essential components of palliative care as defined by professional organizations as well as quality reporting standards promulgated by federal legislation. We then undertake our own analysis of the elements of outpatient palliative care programs described in the higher quality studies identified in our review of the evidence that appear to be correlated with successful treatment outcomes. ©Institute for Clinical and Economic Review, 2016 Page 48 Essential Elements of Palliative Care as Described by Professional Bodies The International Association for Hospice and Palliative Care lists the following essential practices in palliative care: Figure 4. IAHPC List of Essential Practices in Palliative Care.5 ©Institute for Clinical and Economic Review, 2016 Page 49 Measurement The Patient Protection and Affordable Care Act (ACA) mandated a quality reporting program for hospice programs; CMS is tasked with setting the exact quality measures that hospices must utilize. Unlike hospice, there are currently no external quality-reporting requirements for palliative care programs, which means that data measuring the quality of palliative care is limited.32 There are efforts under way to increase measurement of outcomes for quality improvement and public reporting of palliative care programs.32 The Center to Advance Palliative Care (CAPC) launched a voluntary program registry in 2009 for voluntary reporting on palliative care structure and process measures derived from the National Quality Forum Framework and Preferred Practices. Outcomes are available in various formats and can be found at https://registry.capc.org/cms/ The Joint Commission’s Advanced Certification Program for Palliative Care was launched in 2011 under the ACA to create a mechanism to recognize hospital inpatient programs that demonstrate exceptional patient and family-centered care and optimize QoL for adult and pediatric patients with serious illness. The incentives associated with this program are believed to have created a demand for the expansion of palliative care services in new care settings.2 While essential elements of palliative care have been outlined by various entities, the evidence linking specific practices to outcome measures is less clear. This may be in part due to the regulatory requirements related to collecting data on palliative care outcomes. Efforts to evaluate such linkages will further our understanding of how the essential elements of palliative care contribute to success or positive outcomes. Elements of Outpatient Palliative Care Aligned with Treatment Success Subsequent to our description of the comparative effectiveness of outpatient palliative care, we attempted to identify specific elements of a palliative care program that may be aligned with treatment success. Given that one of the primary goals of palliative care is to improve a patient’s QoL,27 we defined treatment success as a statistically significant effect on any QoL measure in favor of the palliative care intervention, and evaluated those higher quality original studies included in our sample that quantitatively assessed such outcomes regardless of the measurement tool used. We then compared the frequencies of various specific components of the interventions described in “successful” studies relative to those described in the “unsuccessful” studies (those in which the outcome was measured, but not found to be significantly affected by the intervention). The comparative studies we reviewed did not provide evidence to distinguish successful and unsuccessful studies on the basis of many of the essential elements of palliative care described above. However, we were able to identify several commonly described components of palliative care interventions described in our evidence base. Table 3 on the following page lists the components of palliative care we evaluated for their potential association with successful outcomes. Notably, all studies identified in our review that measured QoL as a primary or secondary ©Institute for Clinical and Economic Review, 2016 Page 50 outcome had detailed descriptions of the intervention, including the frequency and duration of contact with patients. Table 3. Components of Palliative Care Across Studies Evaluating QoL Bakitas, Rabow, Radwany, Bakitas, Rummans, Temel, Zimmermann, Author, Year 201511 200424 201418 200910 200625 201021 201422 Program Unsuccessful Studies Successful Studies Components Multidisciplinary N Y Y N Y N N Care Team Palliative Care Y N Y Y N Y Y Specialist Patient & Family Y Y Y Y Y N Y Education Advance Care Y Y Y Y N Y Y Planning Monthly In-person N N N N N Y Y Office Visit Of the seven studies we identified as evaluating QoL, four10,21,22,25 were considered successful and three11,18,24 were not. We found that half21,22 of the four successful studies provided at least monthly in-person medical appointments with physicians or nurses present as part of the package of palliative services, while the studies not showing a QoL benefit followed up through telephone or home visits from a volunteer patient advocate on the palliative care team. These in-person appointments gave patients and caregivers an opportunity to have in-depth discussion about issues related to their disease and treatment, including symptom management, insurance, and social services. These successful studies also described an “early” palliative care intervention for advanced cancer patients. The Rummans RCT,25 which included patients with cancer who were diagnosed in the last 12 months, was considered successful but did not specify monthly medical appointments as part of the palliative care intervention. Notably, patient QoL showed improvement at week four, but not at week eight and week 27. Thus, the timing of a palliative care intervention (i.e., at or around the time of diagnosis) and the length of follow-up may also be correlated with successful outcomes. While Bakitas et al. offered patients the opportunity to attend shared medical appointments, given the rural location of the majority of the participants only 18% were able to make these monthly visits.65 In addition to the structural components of palliative care, we also considered other characteristics of the interventions as potentially influencing success. For those palliative care interventions that were considered successful, the data are most robust for those studies evaluating early palliative ©Institute for Clinical and Economic Review, 2016 Page 51 care for oncology patients. Importantly, one of the reasons why the most recent study11 assessing early versus delayed palliative care did not show a benefit may have been due to control patients receiving palliative care earlier than anticipated, which may have inflated the benefit of the intervention in the delayed group. ©Institute for Clinical and Economic Review, 2016 Page 52 5. Comparative Value 5.1 Overview We reviewed the published literature for analyses that have examined the economics of palliative care programs. This included studies of the costs that are potentially offset through the use of such programs (e.g., reduced end-of-life medical costs). Below we summarize recent literature about the economic impact of palliative care programs and components, the strength and validity of that evidence, and where gaps in knowledge still exist. Unfortunately, data on costs for these programs was not generally reported in a way that would allow us to highlight those programs or program components that best correlated with cost-offsets or favorable cost-effectiveness. There was also a dearth of studies on the cost to initiate and operate outpatient palliative care programs or specific components of such programs. We also explore the potential health system budgetary impact of outpatient palliative care programs over a near-term time horizon, utilizing published information on program costs and cost offsets, as well as the potential population eligible for such services. 5.2 Prior Published Evidence on Costs and Cost-Effectiveness of Outpatient Palliative Care Programs Comparisons of Palliative Care to Usual Care Most recent studies that have compared direct health care costs for palliative care to usual care have found such programs to be cost-saving. Many of the costs of intensive inpatient admissions can be avoided, as more patients are able to be cared for at home or in an outpatient setting. A recent annual review by Hughes et al.28 of the growth of palliative care (both inpatient and outpatient) in the U.S. reports that several studies of outpatient programs found cost savings through reduced hospital admission rates and movement of patients from high-cost settings such as hospitals to lower- cost settings such as home health care. The three studies reviewed below found lower costs with outpatient palliative care compared to usual care in patients with mixed diagnoses (cancer and other serious illnesses); one specifically examined a palliative care program that included specialist (“board certified”) staff. However, while these studies compared ongoing costs of care, there is lack of evidence on the development and implementation costs of establishing outpatient palliative care programs. In addition, a limitation of many of these studies is that they are conducted on one particular type of program in one type of setting, limiting their generalizability. ©Institute for Clinical and Economic Review, 2016 Page 53 Brumley et al.12 evaluated costs in the Kaiser Permanente RCT that compared generalist in-home palliative care plus usual care to usual care alone in terminally ill patients with COPD, CHF, or cancer. In regression analyses controlling for days on service, age, severity of illness, and primary disease, patients receiving palliative care had significantly lower health care costs than did the patients receiving usual care (p=0.03), with adjusted mean cost of $12,670 for the palliative care group and $20,222 for usual care. This difference in costs was largely due to fewer ED visits (20% vs. 33% with ED visits, respectively; p=0.01) and hospitalizations (36% vs. 59%, respectively; p<0.001). Strengths of this study included its design as an RCT, and the relatively large sample for this type of study; limitations include being conducted only in one integrated health system and the use of proxy costs, which may limit generalizability to other settings. Prior to the 2007 study, a “comparison group study” in 1999-2000 by Brumley et al.13 compared end-of-life palliative care intervention to usual care, enrolling 558 patients with various life- threatening diseases (mainly cancer, COPD or CHF). Of the 300 patients who died during the study, 161 received the palliative care intervention and 139 usual care. The intervention was a generalist, home-based multidisciplinary approach to managing end-of-life care, including pain and symptom relief, patient education, and emotional and spiritual support. Brumley et al. found that the palliative care patients had 45% lower average costs than the usual care patients ($7,990 vs. $14,570, p<0.001), with fewer physician visits, ED visits, and days in hospital or skilled nursing facility. A more recent analysis by Lukas et al.66 examined hospital outcomes for a specialist, home-based (non-hospice) palliative medicine consulting service in 369 patients with “advanced complex illness” (life-limiting diagnosis with need for frequent or intense medical care, including cancer, chronic illness/end-stage organ failure, and frailty/dementia) in the mid-Atlantic region of the U.S. Retrospective chart reviews were used to compare hospital and ED utilization and costs for the 18 months prior to and after palliative care enrollment. In mixed models analyses of covariance with repeated measures on time (pre- versus post-palliative care enrollment), mean total costs for all hospitalizations decreased from $23,386 pre-enrollment to $16,467 per patient post-enrollment (p<0.001). Limitations of this study include that it is a pre–post design with no control group, and that hospitalization data was only available from one health care network, which may limit the generalizability of these results. Cost Burden and Unpaid Caregiving While many economic analyses of palliative care take a payer or health system perspective, some have examined home-based palliative and end-of-life care from a broader perspective, with an emphasis on assessing unpaid care. In general, these studies find that while outpatient and home- based palliative care may result in reduced costs for hospitalizations and ED visits, they may also result in increased services provided in the home setting, often by informal, unpaid caregivers. ©Institute for Clinical and Economic Review, 2016 Page 54 None of the comparative studies in our evidence base attempted to formally assess this shift in costs in order to estimate the incremental effects attributable to palliative care. In the U.S., Rhee et al.67 attempted to estimate the quantity and economic value of informal family caregiving for community-dwelling older persons in the last year of life, using data from the Health and Retirement Study (a biennial, nationally representative, longitudinal survey) for respondents who died between 2000 and 2002 (n=990). Total average hours of informal caregiving were estimated using ordinary least squares regression, and valued using the 2002 national average home aide wage ($9.16 per hour). Community-dwelling older persons received an adjusted average of 65.8 hours of informal caregiving per week. The annual replacement cost of informal care averaged $31,342, based on the average home aide wage. Replacing the informal care provided in the U.S. in the last year of life with home health aides was estimated to require approximately $1.4 billion in funds. Limitations of this study include the use of survey data, which may underestimate caregiving time, and the assumption that home health aide time can be substituted 1:1 for informal caregiving time. Finally, lost productivity and other burdens on caregivers were not included in these cost estimates. Several economic analyses from Canada examine home-based palliative care from a societal perspective. For example, Yu et al.68 compared the societal costs of home and hospital end-of-life care for a propensity score-stratified cohort of 186 palliative care cancer patients in Ontario, Canada. This hospital-based palliative care program provided “community and team-based multidisciplinary palliative care to patients at home,” with hospital care available when needed. Societal costs included health system costs, as well as patient out-of-pocket and informal caregiving costs. Mean total societal costs for end-of-life care for all patients was $34,198 (2012 Canadian dollars). Approximately 13% of this was for hospitalization, 33% for outpatient/home care, and 46% for unpaid caregiver time. Mean costs for the six months prior to death were not significantly different between home death and hospital death patients ($31,911 and $29,117, respectively). While hospitalization costs were lower for home death patients, outpatient and unpaid caregiving costs were higher, leading to small differences in overall costs. In another Canadian study, Chai et al.69 examined home-based palliative care in Toronto, including estimates for unpaid care. Caregivers of patients with malignant neoplasm (n=169) were interviewed at time of referral to palliative care and periodically thereafter until the patient’s death. Data were collected on palliative care resource use and costs, as well as unpaid caregiving time. Average monthly palliative care costs were $14,924 (2011 Canadian $) in the last year of life, 77% of which was unpaid caregiving costs, 21% public costs, and 2% out-of-pocket costs. The authors suggest that the burden of this unpaid cost to patients’ families could threaten the sustainability of home-based palliative care. Finally, Dumont et al.70 conducted a prospective cohort analysis of 248 Canadian palliative care patients and their caregivers. Data on health care resource utilization and payments were collected ©Institute for Clinical and Economic Review, 2016 Page 55 via biweekly interviews for up to six months or until the patient’s death. Their analysis found that the largest component of health care costs was for inpatient care, followed by home care and informal caregiving time. They estimated that the mean total cost per patient ($18,446, in 2005/2006 Canadian dollars) was comprised of 71% public payments, 27% family payments, and 2% non-profit organization costs. 5.3 Potential Budget Impact of Outpatient Palliative Care Programs We have also provided an estimate of the potential budgetary impact of outpatient palliative care programs among candidate populations for such treatment in the U.S. Our estimates are based on those found in the literature. We combined estimates of the mean cost per patient with estimates of the population potentially eligible for outpatient palliative care programs and different assumed levels of uptake of such programs. Potential Budget Impact Model: Methods Potential budgetary impact was defined as the total incremental cost of outpatient palliative care for the treated population, calculated as the incremental health care costs of palliative care treatment minus health care costs that would be incurred without palliative care (i.e., usual care). All costs were undiscounted and estimated over one- and five-year time horizons. The potential budget impact analysis included the entire candidate population for palliative care programs in the U.S., which was considered be comprised of those with cancer, COPD, or CHF with an expected life expectancy of less than 12 months. To estimate the size of the potential candidate population for palliative care, we used tabulations from the National Center for Health Statistics of the number of deaths by cause for 2013. We summed the total numbers of deaths (across all ages) for cancer (identified as “malignant neoplasms”, ICD-10 codes C00-C97), COPD (identified as “other chronic lower respiratory diseases”, ICD-10 codes J44 and J47) and CHF (identified as “heart failure”, ICD-10 code I50). This resulted in a candidate population size of approximately 787,000 individuals in the US, of which approximately 585,000 (74%) were cancer decedents. In estimating potential budget impact, we recognized that not all patients nearing the end of life will have access to outpatient palliative care. Therefore, our calculations assume that the utilization of such programs reaches only some fraction of eligible patients. To estimate the population size that would use outpatient palliative care, we assumed that some percentage of the eligible population would enter palliative care in each year and that this percentage would stay constant over time (years one through five). Our assumed percentages were 10%, 25% and 50% of eligible patients. ©Institute for Clinical and Economic Review, 2016 Page 56 In another scenario, we assumed that utilization of outpatient palliative care programs may increase over time, with accompanying increases in the percent of candidate patients being treated over the five-year time horizon of our analysis. In this scenario, uptake was assumed to occur in equal proportions across the five-year timeframe, with costs adjusted to account for the different number of patients treated with palliative care each year. For example, if 10% of patients were assumed to be enrolled at the end of five years, 2% of eligible patients were assumed to initiate therapy in the first year, 4% in the second, 6% in the third, and 8% in the fourth year, before finally growing to 10% in the final, fifth year. Note that patients only incur costs (or savings) in the year in which they enroll in palliative care, as we assume that all patients have less than 12-month life expectancy at time of enrollment. To estimate the cost per patient of outpatient palliative care compared to usual care, we used the results from the U.S.-based RCT reported by Brumley et al.,12 with adjusted mean costs of $12,670 per palliative care patient and $20,222 per usual care patient, using 2002 costs. We updated these costs to 2014 dollars, using the medical care component of the Consumer Price Index, to obtain mean costs of $19,308 for palliative care and $30,816 for usual care. This indicates a cost savings of $11,508 per patient enrolled in palliative care rather than receiving usual care. Because the published cost estimates used for this analysis indicated that outpatient palliative care is cost-saving compared to usual care, there was no need to compare our estimates to a budget impact threshold. Potential Budget Impact Model: Results We used the estimated savings per patient of $11,508 to estimate the total impact on the U.S. health care budget of implementing outpatient palliative care at different levels. Table 4 presents the potential budgetary impact of one year and five years of palliative care in the candidate population, assuming immediate implementation of outpatient palliative care programs for different fractions of the population. Results are presented for both one-year and five-year time horizons. Results from the potential budget impact model showed that, with immediate implementation for 10% of the eligible population, an estimated 78,665 individuals would receive palliative care each year. After one year of treatment, with net annual savings of $11,508 per patient, one-year budget impact is estimated to be savings of approximately $905.3 million. Over the entire five-year time horizon, we estimate that a cumulative total of 393,325 patients would be enrolled in palliative care. Across the full five-year time horizon, the total potential savings are approximately $4.5 billion. Assuming that larger proportions of eligible patients could be enrolled in palliative care resulted in even greater savings. At 25% of eligible patients treated in years one through five, approximately $2.3 billion are estimated to be saved per year, summing to $11.3 billion over five years. At 50% enrollment, estimated savings increase to $4.5 billion in year 1 and $22.6 billion over five years. ©Institute for Clinical and Economic Review, 2016 Page 57 We also estimated the potential savings from implementing palliative care in only the eligible patients with cancer (74% of the total eligible population). Annual budget impact ranged from saving approximately $673 million per year at 10% enrollment, to approximately $1.68 billion at 25% and $3.37 billion at 50% enrollment of cancer patients. Cumulative savings over five years would range from approximately $3.4 billion at 10% enrollment to approximately $16.8 billion at 50% enrollment. Table 4. Estimated Total Potential Budget Impact (BI) of Outpatient Palliative Care: Immediate Implementation for Fractions of the Eligible U.S. Population (N=786,628) Analytic Horizon = 1 Year Analytic Horizon = 5 Years Percent Enrolled Number Treated Total BI (billions) Cumulative Number Cumulative Total BI Treated (billions) 10% 78,665 -$0.91 393,325 -$4.53 25% 196,655 -$2.26 938,275 -$11.32 50% 393,315 -$4.53 1,966,575 -$22.63 The above scenario assumes that there would be full capacity to absorb 10% to 50% of eligible patients into palliative care programs beginning in year one. However, it may take some time to ramp up the implementation of outpatient palliative care programs. In separate analyses, we estimated the budget impact of such a ramp up in implementation, assuming a 20% increase in capacity per year. For example, 10% enrollment at five years would imply 2% enrollment in year one, 4% in year two, etc. Assuming that as capacity is ramped up, there will be fewer eligible patients to enroll in palliative care which will result in fewer savings than with immediate implementation. This is displayed in Table 5. With a linear increase to 10% of the eligible population enrolled in year five, an estimated 15,733 individuals would receive palliative care in year one, increasing to 78,665 patients enrolled in year five. After one year of treatment, budget impact is estimated to be a saving of approximately $181 million. Over the entire five-year time horizon, we estimate a cumulative total cost saved of approximately $2.7 billion. Assuming linear increases to reach 25% of eligible patients by year five, approximately $453 million are estimated to be saved per year and $6.8 billion over five years. Ramping up to 50% enrollment by year five would imply an estimated savings of $905 million in year 1 and $13.6 billion over five years. ©Institute for Clinical and Economic Review, 2016 Page 58 Table 5. Estimated Total Potential Budget Impact (BI) of Outpatient Palliative Care: Gradual Implementation for Fractions of the Eligible U.S. Population (N=786,628) Analytic Horizon = 1 Year Analytic Horizon = 5 Years Percent Enrolled Number Treated Total BI (billions) Cumulative Number Cumulative Total BI Treated (billions) 10% 15,733 -$0.181 235,995 -$2.72 25% 39,331 -$0.453 589,965 -$6.79 50% 78,663 -$0.905 1,179,945 -$13.58 Finally, we also estimated the budget impact of outpatient palliative care programs for a hypothetical commercial health plan population of one million members. We applied the proportion of the 2013 U.S. population with deaths from cancer, COPD and CHF (0.002475) to estimate the number of patients eligible for palliative care (2,475). Enrolling 10% of these patients in outpatient palliative care would result in an estimated savings of approximately $2.8 million per year, or approximately $0.24 on a per-member-per-month basis. For benchmarking purposes, this is comparable to payments the state of Colorado makes to accountable care organizations for well- child visits.71 5.4 Summary and Comment We found several studies that attempted to compare the cost to the health care system of providing outpatient or home-based palliative care to that of providing usual care. Most of these studies found that such palliative care decreased overall costs, mainly through avoided hospitalizations and ED visits near the end of life. However, it was often unclear whether these evaluations incorporated the start-up costs of developing an outpatient palliative care program. We were unable to identify any publications that clearly delineated the costs of implementation vs. ongoing costs of such programs. In addition, there was little detailed information on the costs of specific components of palliative care programs in the U.S., which makes comparisons of different programs difficult, given that programs often vary in the specific palliative services provided. While the literature seems to support cost savings associated with outpatient palliative care for the health care system, other studies have looked at the economic impacts of these programs from a societal perspective. These studies point out that, while outpatient and home-based palliative care may succeed in avoiding unnecessary or unwanted hospitalizations and ED visits, some of the care that would be provided in those settings may now be provided in the home, often by informal, unpaid caregivers. The economic value of that unpaid caregiving time (valued at prevailing wages for paid home aides) may replace much (if not all) of the cost saving from reduced inpatient care. For example, Yu et al.68 concluded that: “Higher hospitalization costs for hospital death patients were replaced by higher unpaid caregiver time and outpatient service costs for home death patients. Thus, from a societal ©Institute for Clinical and Economic Review, 2016 Page 59 cost perspective, alternative sites of death, while not associated with a significant change in total societal cost of end-of-life care, resulted in changes in the distribution of costs borne by different stakeholders.” There is a need for studies that compare the economic outcomes of different types of outpatient palliative care programs in similar settings to each other and to usual care. Especially helpful would be studies that are designed to evaluate the effects of individual components of such programs, so that services can be tailored to provide palliative care in the most cost-effective manner. To estimate the up-front investment required, studies of the cost to initiate and operate palliative care programs or specific components of such programs would also be useful. Finally, formal cost- effectiveness analyses of outpatient palliative care programs are needed to allow comparisons of the quality-adjusted life-years gained by such programs compared to usual care. We used the health care costs for palliative care and usual care from an RCT conducted in the U.S. in 2002-200412 to estimate the impact on the U.S. health care budget of palliative care for the treatment of cancer, COPD, and CHF patients near the end of life. The potential money saved from the health care budget was substantial, and scaled up as more patients were assumed to enroll in palliative care. Even assuming a 2% uptake rate per year (i.e., 10% enrollment after five years) would result in estimated cumulative savings of about $2.7 billion after 5 years. Our estimates of levels of outpatient palliative care uptake in the health care system by five years were based on arbitrary assumptions, so actual uptake may not reach these levels this quickly. In addition, the costs used in our analysis came from only one study, and so may not be representative of the costs for such programs in other setting in the U.S. Finally, our budget impact analysis did not include the costs of setting up and implementing new palliative care programs, or the economic value of unpaid caregiving time. In summary, outpatient palliative care services appear to be cost-effective or even cost-saving for the health care system, by providing more care in home and outpatient settings and less inpatient care. Expanding the use of these palliative care programs to larger proportions of eligible patients has the potential to substantially reduce the costs of health care for patients with advanced illness. It should be kept in mind, however, that some portion of these services may now be provided as unpaid caregiving rather than being avoided altogether. ©Institute for Clinical and Economic Review, 2016 Page 60 6. Voting Results 6.1 About the New England CEPAC Process During public meetings of the New England CEPAC, the Council deliberates and votes on key questions related to the systematic review of the clinical evidence, an economic analysis of the applications of the medical technologies or treatments under examination, and the supplementary information presented. Council members are selected for three year terms and are intentionally selected to represent a range of expertise and diverse perspectives. To maintain the objectivity of New England CEPAC and ground the conversation in the interpretation of the published evidence, members are not pre-selected based on the topic being addressed. Acknowledging that any judgment of evidence is strengthened by real-life clinical and patient perspectives, clinical representatives with expertise in the subject matter are recruited for each meeting topic and provide input to Council members before the meeting to help clarify their understanding of the interventions being analyzed in the evidence review. The same clinical experts serve as a resource to the Council during their deliberation, and they help form recommendations with the Council on ways the evidence can be applied to policy and practice. At each meeting, after the Council votes, a Policy Roundtable discussion is held with the Council, clinical experts, and representatives from provider groups, payers, and patient groups. This is intended to bring stakeholders into the discussion on how best to apply the evidence to guide patient education, clinical practice, and coverage and public policies. Participants on Policy Roundtables are selected for their expertise on the specific meeting topic, are different for each meeting, and do not vote on any questions. At the March 31, 2016 meeting, the Council discussed issues regarding the application of the available evidence to help patients, providers, and payers address the important questions related to the management of high cholesterol. Following an evidence presentation and public comments, the Council voted on key questions concerning the clinical effectiveness and value of palliative care services provided in the outpatient setting. These questions are developed by the ICER research team for each assessment, with input from the New England CEPAC Advisory Board to ensure that the questions are framed to address the issues that are most important in applying the evidence to support clinical practice and medical policy decisions. The voting results are presented in the section below, along with comments reflecting considerations mentioned by the Council members during the voting process. In its deliberations and voting related to value, the Council made use of a value assessment framework with four different components of care value, a concept which represents the long-term perspective, at the individual patient level, on patient benefits and the incremental costs to achieve those benefits. The four components of care value are comparative clinical effectiveness, ©Institute for Clinical and Economic Review, 2016 Page 61 incremental cost per outcomes achieved, additional benefits or disadvantages, and contextual considerations regarding the illness or therapy. Once they made an overall assessment of care value as low, intermediate, or high considering these four components, the New England CEPAC then explicitly considered the affordability of specialist palliative care in assessing provisional health system value as low, intermediate, or high (see Figure 8 and Figure 9, as well as the detailed explanation that follows). Figure 8. Care Value Framework There are four elements to consider when deliberating on care value: 1. Comparative clinical effectiveness is a judgment of the overall difference in clinical outcomes between two interventions (or between an intervention and placebo), tempered by the level of certainty possible given the strengths and weaknesses of the body of evidence. The Council uses the ICER Evidence Rating Matrix as its conceptual framework for considering comparative clinical effectiveness. 2. Incremental cost per outcomes achieved is the average per-patient incremental cost of one intervention compared to another to achieve a desired “health gain,” such as an additional stroke prevented, case of cancer diagnosed, or gain of a year of life. Alternative interventions are compared in terms of cost per unit of effectiveness, and the resulting comparison is presented as a ratio: a “cost per outcome achieved.” Relative certainty in the cost and outcome estimates continues to be a consideration. As a measure of incremental costs per outcomes achieved, ICER follows common academic and World Health Organization (WHO) standards by using cost per quality-adjusted life years (QALYs) and adopting thresholds at $100,000 per QALY and $150,000 per QALY as guides to reasonable ratios of incremental costs per outcomes achieved. 3. Other benefits or disadvantages refers to any significant benefits or disadvantages offered by the intervention to the individual patient, caregivers, the delivery system, other patients, or the public that would not have been considered as part of the evidence on comparative clinical effectiveness. Examples of additional benefits include mechanisms of treatment delivery that ©Institute for Clinical and Economic Review, 2016 Page 62 require many fewer visits to the clinician’s office, treatments that reduce disparities across various patient groups, and new potential mechanisms of action for treating clinical conditions that have demonstrated low rates of response to currently available therapies. Additional disadvantages could include increased burden of treatment on patients or their caregivers. For each intervention evaluated, it will be open to discussion whether additional benefits or disadvantages such as these are important enough to factor into the overall judgment of care value. There is no quantitative measure for additional benefits or disadvantages. 4. Contextual considerations include ethical, legal, or other issues (but not cost) that influence the relative priority of illnesses and interventions. Examples of contextual considerations include whether there are currently any existing treatments for the condition, whether the condition severely affects quality of life or not, and whether the condition affects priority populations. There is no quantitative measure for the role of contextual considerations in an overall judgment of care value. In assessing provisional health system value, the Council was asked to vote whether interventions represent a “high,” “intermediate,” or “low” value. Figure 9. Health System Value Framework 1. Potential Health System Budget Impact is the estimated net change in total health care costs over a 5-year time-frame. 2. Provisional “Health System Value” represents a judgment integrating consideration of the long- term care value of a new intervention with an analysis of its potential short-term budget impact if utilization is unmanaged. The Council votes reflect a judgement on the provisional health system value of an intervention. 3. Mechanisms to Maximize Health System Value is an action step, ideally supported by enhanced early dialogue among manufacturers, payers, and other stakeholders. 4. Achieved Health System Value is the real-world result of health care stakeholder efforts to maximize the value of a given intervention. Usually, the care value and the provisional health care system value of an intervention or approach to care will align, whether it is “high,” “intermediate,” or “low.” For example, a treatment that is ©Institute for Clinical and Economic Review, 2016 Page 63 judged to represent high care value from the perspective of per-patient costs and benefits will almost always represent a high health system value as well. But health system value also takes into consideration the short-term effects of the potential budget impact of a change in care across the entire population of patients. Rarely, when the additional per-patient costs for a new care option are multiplied by the number of potential patients treated, the short-term budget impact of a new intervention of intermediate or even high care value could be so substantial that the intervention would be “unaffordable” unless the health system severely restricts its use, delays or cancels other valuable care programs, or undermines access to affordable health insurance for all patients by sharply increasing health care premiums. Under these circumstances, unmanaged change to a new care option could cause significant harm across the entire health system, in the short-term possibly even outweighing the good provided by use of the new care option itself. Provisional health system value builds upon the judgment of care value by integrating consideration of the potential short-term budget impact of a new intervention, a figure highly dependent upon an estimation of the potential uptake of the new drug across the entire population. In the ICER framework, the theoretical basis for the budget impact threshold is based on societal willingness to pay. This foundation rests upon the assumption that society would prefer health care costs to grow at a rate that does not exceed growth in the overall national economy. ICER has used estimates based on data from the World Bank, the Centers for Medicare & Medicaid Services (CMS), and other public sources to calculate a budget impact threshold for individual new drugs or devices that would identify those whose potential budget impact would contribute significantly to excessive health care cost growth. It should be noted that if, after considering potential budget impact, a health intervention judged to have high care value receives a judgment of “low” provisional health system value from the Council, this does not imply that the health system should not adopt the intervention; rather, the vote indicates that policy makers should consider implementing mechanisms related to patient selection, step therapy, pricing, and/or financing to ensure that the short-term budget impact of a high care value intervention does not lead to more harm than good. New England CEPAC votes on provisional health system value will therefore serve an important function by highlighting situations when policymakers need to take action and work together to align care value with health system value. ©Institute for Clinical and Economic Review, 2016 Page 64 6.2 Clinical Effectiveness Voting Results 1. Is the evidence adequate to demonstrate that specialist palliative care delivered in the outpatient setting is superior to usual care for: a. Improving quality of life? Yes: 19 votes No: 0 votes b. Reducing hospitalization and ED use? Yes: 19 votes No: 0 votes 2. Is the evidence adequate to demonstrate that generalist palliative care delivered in the outpatient setting is superior to usual care for: a. Improving quality of life? Yes: 3 votes No: 16 votes b. Reducing hospitalization and ED use? Yes: 2 votes No: 17 votes Comments: Members of the New England CEPAC voting no emphasized that their votes should not be interpreted to mean that generalist palliative care is not effective; rather, the currently available evidence is not yet adequate to prove the effect, and further research is needed. Additionally, members also found the definition of generalist palliative care to be problematic, as it often is not adequately defined in many studies. 6.3 Care Value Voting Results 3. Given the available evidence, what is the care value of specialist palliative care in the outpatient setting vs. usual care in the outpatient setting? Low: 0 votes Intermediate: 4 votes High: 15 votes Comments: A majority of the Council found that based on the evidence that specialist palliative care is superior to usual care for both improving quality of life and reducing hospitalization and ED use, and that it offers a net cost savings, specialist palliative care services represent a high care value. ©Institute for Clinical and Economic Review, 2016 Page 65 6.4 Provisional Health System Value Voting Results 4. Given the available evidence, what is the overall provisional health system value of specialist palliative care? Low: 0 votes Intermediate: 2 votes High: 17 votes ©Institute for Clinical and Economic Review, 2016 Page 66 7. Recommendations to Guide Policy and Practice Prior to the New England CEPAC public meeting, ICER staff conducted semi-structured interviews with national and regional experts in New England and nationally to gain their perspectives on practice and delivery system innovations, barriers to change, and opportunities for improving how palliative care services in the outpatient setting are delivered. These key informants included experts from academic institutions, hospitals, patient advocacy organizations, and health plans (see Appendix 9 for the complete list). The results of these interviews and research were used to inform a moderated discussion between Council members and regional policy roundtable participants. Clinical experts, health insurers, state agency representatives, and a family caregiver discussed with Council members various policy options for improving access to and provision of palliative care services in the outpatient setting (see Appendix 10 for a list of policy roundtable participants). Combining the insights gained from the earlier policy expert interviews with the votes on the evidence by the New England CEPAC (see Section 6 for a description of the voting process and a summary of the votes) and the ensuing policy roundtable discussion at the meeting, the following recommendations are presented to guide the implementation of palliative care in the outpatient setting, and are organized by stakeholder audience. Best practices and other resources are described in complementary Action Guides. Because the discussion at the meeting reflected multiple perspectives and opinions, the recommendations should not be taken as representing the views of individual members of New England CEPAC, policy roundtable participants, or as a consensus view held by all participants. For Payers: 1. Design coverage policies that avoid creating artificial lines between palliative care and curative care. Policies that support access to high quality palliative care also allow for the continuation of curative treatment while palliative care services are being delivered. Roundtable participants and experts agreed that payer policies that support concurrent curative treatments with palliative care services not only improve patient access and outcomes, but also help dispel myths about the role of palliative care in serious illness. In effect, palliative care should be “need-based,” not “prognosis- based.” Further, policies that add to the confusion of palliative care with hospice care - for example, those that require a patient to be within some number of months of end of life - may restrict access for patients and create artificial barriers for the provision of care. ©Institute for Clinical and Economic Review, 2016 Page 67 2. Expand the Medicare Care Choices Model demonstration project that allows curative treatment to be provided concurrently with hospice care. While hospice and palliative care are separate but related domains, changes to the Medicare hospice program that remove barriers to receiving supportive care alongside curative care may help expand access to palliative care services in other settings. Through the Medicare Care Choices Model, Medicare is testing a program where curative treatment need not be stopped in order to receive the Medicare hospice benefit. Programs like this could help patients, providers and policymakers understand the benefit of a palliative care approach that incorporates both curative treatment and comfort care. 3. Implement reimbursement policies that adequately reimburse the provision of outpatient palliative care services. Several experts noted the difficulty in a fee-for-service model of adequately covering the expenses for the entire palliative care team, especially social workers and chaplains. Institutions often rely on supplemental grant or other operational funds to support team members. Global payments and other capitated arrangements that pay for value, not volume, may help alleviate some of the challenges with financially supporting outpatient palliative care teams, while further incentivizing interdisciplinary, patient-centered care. For example, the Medicare Care Choices model provides a per member per month payment of between $200-400 to support the provision of care. 4. Consider outpatient palliative care services a key component of high quality patient- centered medical homes (PCMH). Several experts and New England CEPAC members felt that palliative care in the outpatient setting was essentially equivalent to high quality interdisciplinary primary care. Experts commented that one model for the delivery of palliative care could involve training primary care teams in PCMHs on the core components of palliative care, while supporting clinicians in identifying complex patients in need of specialized palliative care. By incentivizing PCMHs to deliver supportive services to those with serious illness alongside curative treatments, e.g. through supplemental reimbursements tied to demonstrating the ability to deliver palliative care in the outpatient setting, payer policies could support improved access and outcomes for seriously ill patients. 5. Collaborate with policymakers to take on the challenge of designing quality measures that foster high quality care without incentivizing one-size-fits all approaches to palliative care. A partnership with palliative care experts is needed for payers and other stakeholders to determine the appropriate set of quality measures to ensure the delivery of high quality palliative care in the outpatient setting. Experts agreed that there is a balance to strike when implementing new quality guidelines – the metrics should be tied to improved patient outcomes while giving providers the flexibility to deliver palliative care to varied, complex patient populations. Policymakers and payers should avoid “one-size fits all” approaches to defining high quality palliative care. For example, The National Consensus Project for Quality Palliative Care offers quality guidelines that outline key domains associated with high-quality palliative care. ©Institute for Clinical and Economic Review, 2016 Page 68 6. Create preferred provider networks with adequate representation from palliative care specialists, and reduce financial barriers to patients for accessing care. While many experts expect that PCPs and other non-palliative care specialists can deliver high quality outpatient palliative care, complex patients may still benefit from palliative care delivered by a specialist-led team. In those cases, having a preferred provider network with adequate inclusion of palliative care specialists will increase patient access. Further, reduction or removal of co-pays or co-deductibles will ease potential financial hurdles for patients needing access to palliative care. 7. Reimburse for alternative delivery methods, including remote access to care, such as telehealth. Experts identified workforce shortages and geographic access challenges as key barriers to accessing palliative care. New technologies, such as telehealth/telemedicine, have the potential to meet the palliative needs of seriously ill patients, and payers could improve patient access and outcomes by providing reimbursement for these alternative care delivery options. For Providers: 1. Train non-palliative care specialists, especially PCPs and other clinicians, on the core principles of palliative care. Fully acknowledging the increasing burdens on PCPs and other clinicians delivering primary care (nurse practitioners, physician assistants, etc.), experts on the panel noted that for non-complex, seriously ill patients, many of the principles of palliative care mirror high quality, integrated primary care. Efforts to educate PCPs and other clinicians, such as nurse practitioners and physician assistants, on core principles of palliative care would increase patient access and could improve outcomes. The Center to Advance Palliative Care (CAPC) offers numerous training and support resources for clinicians, including CME courses, webinars, and discussion forums. 2. Build awareness with PCPs, oncologists, cardiologists and other specialists about the availability of palliative care services in the outpatient setting. Many experts agreed that one key barrier to patient access to palliative care was a general lack of awareness among the broader provider community. In addition to formal training of PCPs on palliative principles, roundtable participants also recommended continuing education, peer-to-peer mentoring, and increased education and training in medical schools and residency programs to foster the general awareness of and familiarity with outpatient palliative care services. A stronger understanding of palliative care, and of the differences between hospice and palliative care, may help providers to more effectively communicate its goals to patients who may benefit from services. Providers can emphasize the role that curative treatments continue to play while palliative care services are being delivered, while highlighting the psychosocial and quality of life benefits associated with the symptom relief and responsible pain management palliative care delivers. ©Institute for Clinical and Economic Review, 2016 Page 69 3. Involve patients and family caregivers in advanced care planning. The role of the family caregiver in improving patient outcomes in palliative care was emphasized repeatedly at the New England CEPAC meeting and in expert interviews. The evidence is clear on the benefits to patients who are involved in advanced care planning, but when family caregivers are included as the second locus of care, outcomes may improve for both the patient and the caregiver. When appropriate, providers should involve family caregivers and patients in care planning and care execution. Organizations such as The Conversation Project provide tips and tools for initiating advanced care planning conversations. 4. Be flexible in how palliative care is delivered. As new technologies arise to help reach patients in rural areas, those with limited mobility, or those with increasing frailty, providers should evaluate the utility of telehealth/telemedicine to improve patient outcomes. Experts also noted the important role community health workers, who often see patients in the home, are increasingly playing in delivering high quality palliative care. For Patients and Families: 1. Build awareness among patients to help dispel myths about palliative care. Many patients may feel that discussions of palliative care are analogous to discussions of “giving up.” Providers and patients must build mutual trust to dispel commonly-held myths about the goals of palliative care. In addition, patients and families should be aware that palliative care also focuses on the needs of caregivers and can provide an additional layer of support for those caring for individuals with serious illness. Patients and families should understand that while “all hospice is palliative care, not all palliative care is hospice.” The public has greater awareness of hospice but often associates hospice only with end-of-life care. Understanding the option to continue curative care while receiving outpatient palliative care services may help patients embrace the provision of palliative care. Resources from GetPalliativeCare.org, a consumer-focused website sponsored by CAPC, provide information and tools for patients and families to learn more about palliative care and decide whether they could benefit. For Policymakers: 2. Expand the number of specialist palliative care providers by adjusting the certification process. The palliative care workforce lacks the number of specialist palliative care providers needed to meet demand. Because the certification program for specialist palliative care only began in 2008, many professionals who have practiced specialist palliative care lack the certification, since mid-career professionals are less likely to enroll in a fellowship program. Accreditation bodies should consider standards for awarding the certification to those with demonstrated proficiency in the principles of palliative care without requiring the fellowship. In response to this shortage, some programs have ©Institute for Clinical and Economic Review, 2016 Page 70 begun to emerge that allow mid-career providers to remotely complete coursework towards a graduate degree in palliative care without needing to take significant time away from work. 3. Deploy a national educational campaign on the role of palliative care. A national organization or federal agency such as the Office of the Surgeon General could undertake a national education campaign to help both patients and providers understand the role of palliative care in the care continuum. The goals of the campaign may include raising awareness of palliative care, as well as changing public perceptions related to the relationship between palliative care and end-of-life. 4. Balance needs of patients in pain with the goal of reducing opioid addiction. Council members and policy experts on the roundtable acknowledged the tension that exists between attempts to limit outpatient opioid prescribing and the delivery of high quality, personalized palliative care, which often involves pain management. Policymakers should consider polices that balance the need to reduce addiction while offering symptom relief to those who need it. Future Research Needs: 1. Identify the components of palliative care that lead to success, in order to support efforts to implement effective palliative care programs. Palliative care researchers are to be commended for the use of randomized controlled trials to study outpatient palliative care. ICER’s research revealed areas where more RCTs and additional data could help support the uptake of outpatient palliative care. Specifically, policymakers and providers are looking for information to understand the key components of a palliative care program that are key to success, in addition to a better understanding of the start-up, implementation, and ongoing costs for outpatient palliative care programs. Further, the discipline needs additional clarity on outcomes from specialist versus generalist palliative care providers, and further definition of what generalist palliative care entails. Research that describes the training and certification of the leaders of the care team in detail will help policymakers understand the distinct and important roles the two types of palliative care teams can play for patients. 2. Encourage research that focuses on the impact of palliative care on families and caregivers, especially their psychosocial and economic outcomes. The policy roundtable discussion spent time identifying the need for additional data on the true economic costs and savings for palliative are programs for patients, families, and caregivers. While some studies have mentioned the possibility costs from reduced acute medical services may shift to family members providing unpaid care, few studies have examined the true economic impact of that ©Institute for Clinical and Economic Review, 2016 Page 71 shift. Further, because the family caregiver is an additional focus of palliative care, adequately documenting psychosocial outcomes for family caregivers is critically important. 3. Study outpatient palliative care in varied clinical settings and diverse patient populations – not just in academic medical centers. The policy roundtable and New England CEPAC members cautioned that much of the evidence base for outpatient palliative care was derived from large, academic medical centers, as well as large risk- bearing integrated provider groups, which may not represent the patient experience in other settings. For example, care delivered in community health centers, rural areas, and through smaller hospitals may look different than care delivered from large, academic medical centers. Research should focus on how palliative care delivered in rural or smaller community settings can address issues of cultural competency. Further, research should examine outcomes for patients from various racial and socioeconomic backgrounds. 4. Add to the limited but promising research base examining a potential survival benefit for patients receiving palliative care. While the limited evidence showing a survival benefit was focused on cancer populations, researchers should examine survival benefit in other disease areas, in larger numbers of patients, and for longer periods of follow-up to further elucidate the potential survival benefits of palliative care. Roundtable participants emphasized that quality of remaining life (rather the improvement in life expectancy) is the primary goal of palliative care, but New England CEPAC members stressed the importance of further examining survival benefits, as the outcomes seen in currently-available studies rival those of new targeted medications for cancer. 5. Allocate resources to help thought leaders track and analyze the current provision of palliative care to identify additional research needs. As mentioned earlier in the report, only 1% of the current NIH research budget is dedicated to the study of palliative care. Roundtable participants emphasized the need for more funding to further understand the physical, psychosocial, and economic impact, as well as the components of successful programs and the long-term value of palliative care. 6. Study barriers to implementing high quality palliative care. Roundtable participants and experts noted that resources, best practices, and guidelines exist for the provision of palliative care, but those standards and best practices are not always followed. Research is needed to identify the barriers to implementation in a variety of settings. ©Institute for Clinical and Economic Review, 2016 Page 72 References APPENDICES ©Institute for Clinical and Economic Review, 2016 Page 73 A1. Palliative Care as Compared to Hospice Palliative Care as Compared to Hospice (NEJM 2015) Palliative Care as Compared with Hospice Characteristic Palliative Care Hospice Model Team of interdisciplinary providers that Team of interdisciplinary providers including may include nurses, social workers, physicians, social workers, chaplains, and chaplains, and other support staff. Primary volunteers as directed by statute; primary goal goal is to improve outcomes related to QoL of care are improved QoL and relief of physical, emotional and spiritual suffering Eligibility Patients of any age, diagnosis, or state of Patients of all ages who have a prognosis of 6 illness. Patients may continue life months or less if the disease follows the usual prolonging or disease directed therapy course. Patients must forgo Medicare coverage for curative or disease directed treatments related to the terminal diagnosis Location Most common in hospital setting. Also Most common in the home based setting. Also available in hospital clinics, group practices, available in assisted living facilities, nursing home care programs, and nursing homes homes, residential hospice facilities, inpatient hospice units, or hospice-contracted inpatient beds Payment Physician and nurse practitioner fees are Medicare hospice benefit; standard hospice covered by Medicare part B for inpatient or benefit from commercial payers is usually outpatient care; hospital teams are modeled after Medicare; Medicaid, although included within Medicare Part A or coverage varies by state; medication costs are commercial insurance payments to included for illnesses related to terminal illness hospitals for care episodes; flexible bundled payments under Medicare Advantage, Managed Medicaid, ACOs, and other commercial payers Source: http://www.nejm.org/doi/full/10.1056/NEJMra1404684 ©Institute for Clinical and Economic Review, 2016 Page 74 A2. Definitions of Palliative Care The following definitions reflect the modern concept of palliative care: WHO Definition of Palliative Care4 An approach that improves the quality of life of patients and their families facing the problem associated with life-threatening illness, through the prevention and relief of suffering by means of early identification and impeccable assessment and treatment of pain and other problems, physical, psychosocial and spiritual. The Centers for Medicare and Medicaid Services (CMS) Definition of Palliative Care4 Patient and family-centered care that optimizes quality of life by anticipating, preventing, and treating suffering. Palliative care throughout the continuum of illness involves addressing physical, intellectual, emotional, social, and spiritual needs and facilitating patient autonomy, access to information, and choice. The Center to Advance Palliative Care Definition of Palliative Care4 Specialized medical care for people with serious illnesses…focused on providing patients with relief from the symptoms, pain and stress of a serious illness — whatever the diagnosis. The goal is to improve quality of life for both the patient and the family. Palliative care is provided by a team of doctors, nurses and other specialists who work together with a patient’s other doctors to provide an extra layer of support. It is appropriate at any age and at any stage in a serious illness and can be provided along with curative treatment. National Hospice and Palliative Care Organization uses the National Consensus Project's Definition of Palliative care1 Palliative care is patient and family-centered care that optimizes quality of life by anticipating, preventing, and treating suffering. Palliative care throughout the continuum of illness involves addressing physical, intellectual, emotional, social, and spiritual needs and to facilitate patient autonomy, access to information and choice. The following features characterize palliative care philosophy and delivery:  Care is provided and services are coordinated by an interdisciplinary team;  Patients, families, palliative and non-palliative health care providers collaborate and communicate about care needs;  Services are available concurrently with or independent of curative or life-prolonging care;  Patient and family hopes for peace and dignity are supported throughout the course of illness, during the dying process, and after death ©Institute for Clinical and Economic Review, 2016 Page 75 A3. Medicare Care Choices Model The Centers for Medicare and Medicaid describes the Medicare Care Choices Model as follows: Through the Medicare Care Choices Model, the Centers for Medicare & Medicaid Services (CMS) will provide a new option for Medicare beneficiaries to receive palliative care services from certain hospice providers while concurrently receiving services provided by their curative care providers. CMS will evaluate whether providing hospice services can improve the quality of life and care received by Medicare beneficiaries, increase patient satisfaction, and reduce Medicare expenditures. Under current payment rules, Medicare and dually eligible beneficiaries are required to forgo curative care in order to receive services under the Medicare or Medicaid Hospice Benefit. Fewer than half of eligible Medicare beneficiaries use hospice care and most only for a short period of time. The model is designed to:  Increase access to supportive care services provided by hospice;  Improve quality of life and patient/family satisfaction; Model Summary Stage: Announced Number of Participants: 141 Category: Initiatives to Accelerate the Development and Testing of New Payment and Service Delivery Models Authority: Section 3021 of the Affordable Care Act Participating Organizations: Vermont Model Name Organization Name Address City State Notes Medicare Care Addison County Home Health 254 Ethan Allen New Haven VT Not Choices Model & Hospice Inc. New Haven, VT Hwy Applicable Medicare Care Bayada Home Health Care, Inc. 316 Main St., Norwich VT Not Choices Model dba Bayada Hospice Norwich, P.O. Box 1590 Applicable VT Medicare Care Caledonia Home Health Care 161 Sherman Saint VT Not Choices Model Saint Johnsbury, VT Dr Johnsbury Applicable Medicare Care Central Vermont Home Health 600 Granger Rd Barre VT Not Choices Model and Hospice Barre, VT Applicable Medicare Care Franklin County Home Health 3 Home Health Saint Albans VT Not Choices Model & Hospice Saint Albans, VT Circle, Suite 1 Applicable ©Institute for Clinical and Economic Review, 2016 Page 76 Medicare Care Lamoille Home Health Agency 54 Farr Ave Morrisville VT Not Choices Model Inc dba Lamoille Home Health Applicable & Hospice Morrisville, VT Medicare Care Orleans/Essex VNA & Hospice 46 Lakemont Newport VT Not Choices Model Inc. Newport, VT Rd Applicable Medicare Care Southwestern Vermont 7 Albert Cree Rutland VT Not Choices Model Hospice Network dba Rutland Dr Applicable Area Visiting Nurse Association and Hospice Rutland, VT Medicare Care Visiting Nurse Association & 205 Billings White River VT Not Choices Model Hospice of Vermont and New Farm Rd Bldg 5 Junction Applicable Hampshire White River Junction, VT Medicare Care VNA of Chittenden & Grand 1110 Prim Rd. Colchester VT Not Choices Model Isle Counties Colchester, VT Applicable Maine Medicare Care Community Health and 42 Cedar St Bangor ME Not Choices Model Counseling Services Bangor, Applicable ME Massachusetts Medicare Care Care Dimensions, Inc. Danvers, 75 Sylvan St., Danvers MA Not Choices Model MA Suite B-102 Applicable Medicare Care HopeHealth, Inc. Hyannis, MA 765 Attucks Hyannis MA Not Choices Model Lane Applicable Medicare Care Notre Dame Health Care, Inc 555 Plantation Worcester MA Not Choices Model dba Notre Dame Hospice Ave Applicable Worcester, MA ©Institute for Clinical and Economic Review, 2016 Page 77 A4. Search Strategies Ovid – Medline 1996 to Present with Daily Update and Cochrane Central Register of Controlled Trials #1 Palliative Care/ or Terminal Care/ or Terminally Ill/ or Palliative Medicine/ #2 ((palliat* adj25 care) or (supportive care or comfort care) or (terminal* and (care or caring or ill*)) or ((advanced or end stage or terminal*) adj4 (disease* or illness* or cancer* or malignan*)) or (last year of life or LYOL or life* end or end of life)).ti,ab. #3 1 or 2 #4 Home Care Services/ or Mobile Health Units/ or Community Health Nursing/ or Outpatients/ or Ambulatory Care/ #5 ((home adj4 palliat*) or ((home* or in-home or domicile or outreach or residential or housing or posthospital or post-hospital or communit* or mobile or ambulatory or outpatient or door to door) adj25 (team* or center* or centre* or clinic treat* or care or interven* or therap* or management or model* or program* or service* or base* or nurs*))).ti,ab. #6 4 or 5 #7 (((comprehensive* or integrative or systematic*) adj3 (bibliographic* or review* or literature)) or (meta-analy* or metaanaly* or "research synthesis" or ((information or data) adj3 synthesis) or (data adj2 extract*))).ti,ab. or (cinahl or (cochrane adj3 trial*) or embase or medline or psyclit or (psycinfo not "psycinfo database") or pubmed or scopus or "sociological abstracts" or "web of science").ab. or ("cochrane database of systematic reviews" or evidence report technology assessment or evidence report technology assessment summary).jn. or Evidence Report: Technology Assessment*.jn. or ((review adj5 (rationale or evidence)).ti,ab. and review.pt.) or meta-analysis as topic/ or Meta- Analysis.pt. #8 randomized controlled trial.pt. or controlled clinical trial.pt. or randomized.ab. or placebo.ab. or Clinical Trials as Topic/ or randomly.ab. or trial.ti. #9 exp Animals/ #10 Humans/ #11 9 not 10 #12 8 not 11 #13 (Costs.mp. and "Cost Analysis"/) or Models, Economic/ or cost-effectiveness.mp. or Cost-Benefit Analysis/ or cost utility.mp. or Quality of Life/ or qol.mp. or quality adjusted life year.mp. or Quality-Adjusted Life Years/ or qaly.mp. or Cost of Illness/ or burden of disease.mp. or caregiver burden.mp. or Health Expenditures/ or Health Care Costs/ or Cost Control/ or Direct Service Costs/ or Cost Sharing/ or cost analysis.mp. ©Institute for Clinical and Economic Review, 2016 Page 78 #14 Cohort studies/ or Longitudinal studies/ or Follow-up studies/ or Prospective studies/ or Retrospective studies/ or cohort.ti,ab. or longitudinal.ti,ab. or prospective.ti,ab. or retrospective.ti,ab. #15 Control Groups/ or (control* adj2 (clinical or group* or trial* or study or studies or design* or method*)).ti,ab. #16 Controlled before-after studies/ or (before adj4 after).ti,ab. #17 (quasi-experiment* or quasiexperiment* or quasi random* or quasirandom* or quasi control* or quasicontrol* or ((quasi* or experimental) adj3 (method* or study or studies or trial or design*))).ti,ab,hw. #18 Pilot projects/ or pilot.ti. or (pilot* adj3 (program* or project? or study or studies)).ab. #19 Comparative study.pt. or (comparative and (study or studies)).ti. #20 Intervention studies/ or intervention.hw. or (intervention? or multiintervention? or multi-intervention? or postintervention? or post-intervention? or preintervention? or pre-intervention?).ti,ab. #21 Evaluation studies.pt. or (evaluation and (study or studies)).ti. #22 or/15-21 #23 14 and 22 #24 Case-Control Studies/ or Control Groups/ or Matched-Pair Analysis/ or ((case* adj5 control*) or (case adj3 comparison*) or control group*).ti,ab. #25 7 or 12 or 13 or 23 or 24 #26 Pediatrics/ or Adolescents/ or Child/ or Infant/ or Infant, newborn/ #27 (child* or adolescent* or infant* or baby or babies or neonat* or juvenil* or pediatric* or paediatric* or young person* or young people or youth* or young adult* or matern*).ti. #28 26 or 27 #29 25 not 28 #30 29 not (guideline or practice guideline or letter or editorial or news or case reports or clinical conference or congresses).pt. #31 limit 30 to (english language and humans and yr="2000 -Current") #32 3 and 6 and 31 #33 nursing home.mp. #34 32 not 33 ©Institute for Clinical and Economic Review, 2016 Page 79 Ovid – PsycINFO #1 exp palliative care/ or exp terminal care/ or terminally ill patients/ #2 palliat*.mp. #3 (terminal* and (care or caring or ill*)).mp. #4 ((advanced or end stage or terminal*) adj4 (disease* or illness* or cancer* or malignan*)).mp. #5 (last year of life or LYOL or end of life).mp #6 or/1-5 #7 exp home care/ or exp home visiting programs/ or community services/ or outpatients/ or ambulatory care/ #8 (home adj4 (hospital or palliat*)).mp. #9 (homecare or home-care or homebased or home-based).mp. #10 ((home or in-home or domicile or outreach or residential or housing or posthospital or post-hospital or communit* or mobile or ambulatory or door to door) adj2 (team* or center* or centre* or treat* or care or interven* or therap* or management or model* or program* or service* or base* or nurs*)).ti,ab. #11 (outpatient* or ambulatory).ti,ab. #12 or/7-11 #13 ((case* adj5 control*) or (case adj3 comparison*) or case-comparison or control group*).ti,ab. #14 ((cohort or longitudinal or prospective or retrospective).ti,ab,id. or longitudinal study.md. or prospective study.md. or retrospective study.md.) #15 (((comprehensive* or integrative or systematic*) adj3 (bibliographic* or review* or literature)) or (meta-analy* or metaanaly* or "research synthesis" or ((information or data) adj3 synthesis) or (data adj2 extract*))).ti,ab,id. or ((review adj5 (rational or evidence)).ti,ab,id. and "Literature Review".md.) or (cinahl or (cochrane adj3 trial*) or embase or medline or psyclit or pubmed or scopus or "sociological abstracts" or "web of science").ab. or ("systematic review" or "meta analysis").md. #16 exp clinical trials/ or cross-over studies/ or random allocation/ or double-blind method/ or single-blind method/ #17 exp clinical trials/ or crossover design/ or random assignment/ #18 exp clinical trials/ or double blind method/ or random allocation/ #19 random$.mp. ©Institute for Clinical and Economic Review, 2016 Page 80 #20 (cross-over or cross? over or (clinical adj2 trial$) or single-blind$ or single? blind$ or double-blind or double? blind$ or triple-blind or triple? blind).tw. #21 random sampling/ #22 or/13-21 #23 6 and 12 and 22 #24 (child* or adolescent* or infant* or baby or babies or neonat* or juvenil* or pediatric* or paediatric* or young person* or young people or youth* or young adult* or matern*).ti. #25 23 not 24 #26 limit 25 to (human and english language and yr="2000 -Current") EBSCO – CINAHL #1 MH palliative care OR MH Terminal Care OR MH Terminally Ill Patients #2 ((palliat* n25 care) or (supportive care or comfort care) or (terminal* and (care or caring or ill*)) #3 TI advanced or end stage or terminal*) and (disease* or illness* or cancer* or malignan*)) or (last year of life or LYOL or life* end or end of life) #4 AB advanced or end stage or terminal*) and (disease* or illness* or cancer* or malignan*)) or (last year of life or LYOL or life* end or end of life). #5 or/1-4 #6 MH Home Health Aide Services or MH Mobile Health Units or MH Community Health Nursing or MH Outpatients or MH Ambulatory Care #7 TI ((home and palliat*) or ((home* or in-home or domicile or outreach or residential or housing or posthospital or post-hospital or communit* or mobile or ambulatory or outpatient or door to door) and (team* or center* or centre* or clinic treat* or care or interven* or therap* or management or model* or program* or service* or base* or nurs*))) #8 AB ((home and palliat*) or ((home* or in-home or domicile or outreach or residential or housing or posthospital or post-hospital or communit* or mobile or ambulatory or outpatient or door to door) and (team* or center* or centre* or clinic treat* or care or interven* or therap* or management or model* or program* or service* or base* or nurs*))) #9 or/6-8 #10 (TI (systematic* n3 review*)) or (AB (systematic* n3 review*)) or (TI (systematic* n3 bibliographic*)) or (AB (systematic* n3 bibliographic*)) or (TI (systematic* n3 literature)) ©Institute for Clinical and Economic Review, 2016 Page 81 or (AB (systematic* n3 literature)) or (TI (comprehensive* n3 literature)) or (AB (comprehensive* n3 literature)) or (TI (comprehensive* n3 bibliographic*)) or (AB (comprehensive* n3 bibliographic*)) or (TI (integrative n3 review)) or (AB (integrative n3 review)) or (JN “Cochrane Database of Systematic Reviews”) or (TI (information n2 synthesis)) or (TI (data n2 synthesis)) or (AB (information n2 synthesis)) or (AB (data n2 synthesis)) or (TI (data n2 extract*)) or (AB (data n2 extract*)) or (TI (medline or pubmed or psyclit or cinahl or (psycinfo not “psycinfo database”) or “web of science” or scopus or embase)) or (AB (medline or pubmed or psyclit or cinahl or (psycinfo not “psycinfo database”) or “web of science” or scopus or embase)) or (MH “Systematic Review”) or (MH “Meta Analysis”) or (TI (meta-analy* or metaanaly*)) or (AB (meta-analy* or metaanaly*)) #11 (MH "Case Control Studies+") or (MH "Control Group") or (MH "Matched-Pair Analysis") or (TI (case or cases) n5 TI (control or controls)) OR (AB (case or cases) n5 AB (control or controls)) OR (TI (case or cases) n3 TI (matched)) OR (AB (case or cases) n3 AB (matched)) OR TI (control group*) #12 MH Prospective studies OR MH case control studies OR MH Correlational studies OR MH Nonconcurrent prospective studies OR MH Cross sectional studies OR ( (cohort and (study or studies) ) OR ( (observational and (study or studies)) ) #13 MH Clinical Trials OR PT Clinical trial OR TX clinic* n1 trial OR ( TX ((singl* n1 blind*) or (singl* n1 mask*)) or TX ((doubl* n1 blind*) or (doubl* n1 mask*)) or TX ((tripl* n1 blind*) or (tripl* n1 mask*)) or TX ((trebl* n1 blind*) or (trebl* n1 mask*)) ) OR TX randomi* control* trial* OR MH Random Assignment OR TX random* allocat* OR TX placebo* OR MH Placebos OR MH Quantitative Studies OR TX allocat* random* #14 or/10-13 #15 5 and 9 and 14 (Limiters - Published Date: 20000101-20151130; Exclude MEDLINE records; Human; Age Groups: All Adult; Language: English) ©Institute for Clinical and Economic Review, 2016 Page 82 A5. Clinical Guidelines While we did not identify any clinical guidelines specific to outpatient palliative care, a number of professional societies provide guidelines on palliative care generally (in all settings). Guidelines outlined below were identified through a web search as well as through conversations with key informants. Disease specific guidelines related to provision of palliative were also identified and are listed at the end of this section. Please note, this list is not intended to be exhaustive. National Consensus Project (2013) http://www.nationalconsensusproject.org/Guidelines_Download2.aspx Guidelines written by the National Consensus Project for Quality Palliative Care include the consensus views of key professional organizations including:  American Academy of Hospice and Palliative Medicine  National Hospice and Palliative Care Organization  Center to Advance Palliative Care  National Association of Social Workers  Hospice and Palliative Nurses Association  National Palliative Care Research Center The guidelines separate recommendations into eight domains, each addressing a distinct aspect of palliative care. The eight domains include:  Domain 1: Structure and Process of Care  Domain 2: Physical Aspects of Care  Domain 3: Psychological and Psychiatric Aspects of Care  Domain 4: Social Aspects of Care  Domain 5: Spiritual, Religious, and Existential Aspects of Care  Domain 6: Cultural Aspects of Care  Domain 7: Care of the Patient at the End of Life  Domain 8: Ethical and Legal Aspects of Care After reviewing the features of each domain, we have combined domains that share similar recommended considerations related to aspects of care. The combined domains include Psychological, Social, Spiritual, Legal, Ethical, and Cultural Aspects of Care to summarize characteristics. Structure and Process of Care: This domain makes recommendations related to care teams, care plans, and quality assessment. According to the guidelines, an interdisciplinary team should assess each patient and develop a care plan based on the patient and family’s specific goals and values. The plan should evolve as patient needs or goals change. The physical environment of care should meet the preferences and needs of the patient and family as much as possible. Quality improvement should involve a documented, data-driven process focused on making care safer, ©Institute for Clinical and Economic Review, 2016 Page 83 ensuring that each person and family is engaged in care, promoting coordination, ensuring use of the most effective treatment practices, and increasing affordability. Physical Aspects of Care: Guidelines in this domain recommend that management of pain and other symptoms be based on the best available evidence and that assessment and management be delivered in accordance with the status of the disease. Treatment for physical symptoms should be based on the disease, prognosis, and patient’s functional limitations. Psychological, Social, Spiritual, Legal, Ethical, and Cultural Aspects of Care: Guidelines within these domains recommend that these factors are considered in developing and executing a care plan. End of Life: This domain related specifically to care at the end of life. Institute for Clinical Systems Improvement (2013) https://www.icsi.org/_asset/k056ab/PalliativeCare.pdf ICSI recommends that palliative care discussions begin soon after a patient’s diagnosis with a serious illness, especially in cases where there are complex needs. Primary care clinicians should have palliative care conversations with patients to encourage early planning. Providers should complete a systematic assessment of patients’ palliative care needs and document goals and advance directives. The assessment should consider physical as well as cultural, psychological, social, spiritual, financial, ethical, and legal factors, and palliative care should seek to manage all of these areas when possible. Patients and families should be engaged in decisions about care, including setting of realistic goals. Palliative care should be compatible with any other medical treatments being administered. American Academy of Family Physicians (2006) http://www.aafp.org/afp/2006/0315/p1104.html AAFP guidelines address recommendations for the care team as well as for the care plan. Care teams should be skilled in communication to establish the goals of the patient and family and should collaborate with professionals across all health care systems to ensure coordination. The team should include individuals with specialist-level skill in physical, social, psychological, spiritual, and legal aspects of medical care. Support and education should be available for the team, as well as the patient and family. Community-based health resources should be incorporated when appropriate to aid in continuity of care. A patient’s care plan should be based on assessment of the goals of the patient and family and should be regularly reviewed and adapted to changing needs and preferences. All treatment decisions should be made in line with the goals of the care plan, with a focus on management of pain and other symptoms, and should incorporate pharmacologic ©Institute for Clinical and Economic Review, 2016 Page 84 treatments as well as nonpharmacological and complementary therapies. When possible, the setting of care should be based on the preferences of the patient and family. The option of hospice referral should be introduced as the patient’s health declines, and end-of-life concerns, hopes, fears, and expectations should be discussed openly. Following death, bereavement support should be offered to the family for at least 12 months. Other Organizations A number of other organizations have developed guidelines for palliative care relevant to specific disease areas. These guidelines include (but are not limited to):  Joint Guidance by the American Society of Clinical Oncology and the American Academy of Hospice and Palliative Medicine  The National Comprehensive Cancer Network  The American Thoracic Society  The American College of Chest Physicians  The American Society for Radiation Oncology ©Institute for Clinical and Economic Review, 2016 Page 85 A6. Previous Systematic Reviews We identified nine systematic reviews which examined the effectiveness of outpatient or home- based palliative or end-of-life care on various patient and caregiver outcomes and utilization indicators. These reviews are summarized by population type in the sections below. We also looked for reviews conducted by other national and international HTA organizations, but did not find any that specifically evaluated palliative care in outpatient or home-based settings. Life-limiting and Advanced Illnesses Luckett 2013 Luckett et al. is a systematic review and meta-analysis comparing community specialist palliative care services (SPCSs) offering home nursing to no SPCSs in patients with life-limiting illnesses. Meta- analysis of all nine studies indicated a significantly higher rate of home death for SPCSs with home nursing (odds ratio 4.45, 95% CI 3.24-6.11; p<0.001); two high-quality RCTs found no effect, however. Among the two studies measuring symptom/QoL and five studies measuring cost, one study reported a significant effect on symptom control in favor of SPCSs and three studies reported significantly lower cost in SPCSs, while the others found no effect. Gomes 2013 A Cochrane systematic review of 23 studies (16 RCTs), including 37,561 participants and 4,042 family caregivers, examined the clinical effectiveness and cost-effectiveness of home palliative care services for adults with advanced illness and their caregivers. Meta-analysis showed a higher rate of dying at home in patients receiving home palliative care compared to usual care (OR 2.21, 95% CI 1.31 to 3.71; p=0.003 [7 trials; n=1222]). There was a small but statistically significant beneficial effect of home palliative care on reducing symptom burden. Evidence was conflicting on patients’ experience of pain, breathlessness and sleep disturbance, caregiver burden, and patient and caregiver satisfaction. Moderate evidence showed no statistically significant effect on nausea/vomiting, constipation, diarrhea, fatigue and appetite loss. Evidence on cost-effectiveness was inconclusive. Thomas 2006 Thomas and colleagues evaluated 23 RCTs about care at the end of life, one of the themes being the effect of providing palliative care through dedicated community teams on quality of life, management of symptoms, satisfaction of care, duration of palliation period, and place of death. A high risk of bias was detected in most of the included RCTs, and conflicting results were found for all outcomes of interest. Six studies found some improvement in ratings of quality of life and perceived management of symptoms, while three RCTs found no improvement in symptoms. In terms of ©Institute for Clinical and Economic Review, 2016 Page 86 satisfaction, one study reported higher patient satisfaction in the intervention patients than the controls, one found equal increases in both groups and two found no significant increase in the intervention group over time. Three studies found increased satisfaction of caregivers in the intervention group over time. In addition, one study found hospital-at-home (24-hour home care from nurses for final two weeks) did not increase the likelihood of home death and another study found survival time in the palliative care group was shorter but the difference was non-significant, but there was a significantly higher rate of home death (p=0.02) and lower rate of dying in nursing homes (p<0.01). Shepperd 2011 A systematic review included four RCTs about end-of life care at home evaluating death at home, functional status, psychological well-being or cognitive status, admission to hospital, and patient satisfaction. The authors found patients receiving home-based end of life care were statistically significantly more likely to die at home (RR 1.33, 95% CI 1.14 to 1.55, p=0.0002 [three trials; n=652]) than those receiving usual care. But there was no statistically significant difference for any other outcome of interest. HIV/AIDS Harding 2005 Harding and colleagues identified 17 studies evaluating 22 programs of home and inpatient palliative care on patient outcomes in patients with HIV/AIDS; the results were synthesized by evidence grade (grade 1-4: strong, fairly strong, weaker, and weak, and qualitative). Only those results for outcomes associated with home palliative care are summarized here. An RCT (n=57) provided strong evidence (grade 1) that integrated multi-professional case managed home care showed a possible advantage in quality of wellbeing and survival over usual home care, but this difference did not reach statistical significance. A small observational study (n=42) provided fairly strong evidence (grade 2) that patients accessing home palliative care in addition to hospital care were not statistically different in Quality of Wellbeing Scale but had lower hospital admissions and reduced length of hospital stays compared to patients attending hospital care only. They found mixed results from five studies providing weaker evidence (grade 3) in terms of symptom and pain control. Additionally, a cross-sectional study provided weak evidence that patients (n=52) accessing HIV home care reported higher satisfaction, more control over care and communication, and more frequent contact with staff while perceived health status remained unchanged. Qualitative data from interviews with patients receiving home care reported less disturbance of daily routine, fewer hospital visits, higher satisfaction, and better quality of care, but some reported anxiety and fear induced by home care and poor communication.) of cancer patients using home-based palliative care services ©Institute for Clinical and Economic Review, 2016 Page 87 Caregivers Harding 2003 The authors reviewed interventions for carers, including family members. They identified 22 relevant studies that included/targeted carers, of which six interventions were included for evaluation, including home care, respite care provision, social networks and activities, and individual- and group-based sessions. Rigorous evaluation of interventions was rare, with only two (quasi-) experimental evaluations identified in this review. An RCT found that one-to-one intervention had no impact on emotional, social or physical function, while another RCT evaluating the effect of a stress and activity management group found that attendees improved knowledge scores, achieved activity goals, coped better with medical situation, and were more satisfied with care. The authors called for development and evaluation of targeted interventions for palliative care carers. Candy 2011 A Cochrane systematic review evaluated the effect of supportive interventions to improve the psychological and physical health of informal caregivers of patients with terminal illnesses. The 11 RCTs included in this review involved 1,836 caregivers. There was low quality evidence that interventions directly supporting caregivers significantly reduced psychological distress in short term (SMD -0.15, 95% CI -0.28 to -0.02), but effects on coping skills and quality of life were not statistically significant. Evidence was less clear on the indirect interventions: two trials found that interventions that provided support to the patient reduced caregiver psychological distress but none of the assessments were statistically significant. Another trial found no difference between trial arms in the proportion of caregivers reporting good physical health. Not Specified Finlay 2002 In a systematic review of palliative care in the hospital, hospice, and at home, the authors identified 22 studies examining home care services, most of which were classified as grade III evidence indicating poor quality. In general, the outcomes were in favor of home care services, showing improved satisfaction and pain symptom control in home care services compared with conventional care. All studies that considered costs suggest lower cost for the home care group. Taken together, the authors suggested that home care is cost-effective. Davis 2015 A systematic review of outpatient and home palliative care for patients with serious illnesses including 15 RCTs of outpatient palliative care and 13 RCTs of palliative home care. While some trials demonstrated advantages of palliative care over usual care, including improved depressive ©Institute for Clinical and Economic Review, 2016 Page 88 symptoms and patient quality of life; reduced aggressive care at the end of life; increased advanced directives; reduced hospital LOS and hospitalization; improved caregiver burden and quality of life; reduced medical cost; and patient and family satisfaction. Thus, the authors concluded that evidence on the benefits of palliative care was mixed due to methodological issues and suggested that better designed and executed studies be conducted. ©Institute for Clinical and Economic Review, 2016 Page 89 A7. Ongoing Studies Title/ Trial Sponsor Study Comparators Patient Population Primary Outcomes Estimated Design Completion Date Does RCT Standard of care N=300 Changes in the September Outpatient Palliative Subjects Quality of Life 2018 Care Improve Age 40 years and older (QOL) at 6 months Patient-centered Interdisciplinary Inclusion Criteria: Outcomes in outpatient palliative Parkinson's Disease? care  Fluent in English Changes in caregiver  UK Brain Bank criteria for diagnosis of probable PD distress at 6 months  At high risk for poor outcomes as identified by the Palliative Care Needs Assessment Tool (PC-NAT) NCT02533921 Exclusion Criteria:  Immediate and urgent palliative care needs  Unable or unwilling to commit to study procedures including; a. randomization, b. study visits or c. the addition of a neurologist to their care team  Presence of additional chronic medical illnesses which may require palliative services  Already receiving palliative care and/or hospice services. Randomized Study of RCT Early palliative care and N=700 Change in FACT-G November Early Palliative standard oncology care scores from baseline to 2016 Care Integrated With Age 18 years and older 12 weeks Standard Oncology Care Versus Standard oncology care Standard Study Patient Participant Eligibility Requirements: Oncology Care Alone in Patients With  Documentation of Disease: Confirmed advanced lung cancer (NSCLC, Incurable Lung or small cell lung cancer, or mesothelioma) or non-colorectal GI cancer Non-Colorectal (esophageal, gastric, hepatic, biliary, or pancreatic) not being treated with curative intent. ©Institute for Clinical and Economic Review, 2016 Page 90 Gastrointestinal  Diagnosed with incurable disease within the previous 8 weeks. Malignancies  Age ≥ 18 years  ECOG Performance Status 0-2  Ability to read and respond to questions in English or able to complete questions with minimal assistance required from an NCT02349412 interpreter or family member.  Planning to receive all medical care for cancer at the enrolling institution.  Participants must be under the care of an oncologist, but their current plan may or may not include chemotherapy or other forms of tumor-directed therapies. Study Family Caregiver Participant Eligibility Requirements:  Relative or friend who is identified by the patient participant who plans to regularly accompany the patient to the majority of their clinic visits.  Family caregiver must live with the patient or have in-person contact with him or her at least twice per week.  Ability to read and respond to questions in English or able to complete questions with minimal assistance required from an interpreter or family member.  Age ≥ 18 years Note: An eligible patient may participate in this trial without an eligible family caregiver being registered. Clustered, RCT Active Comparator: N=476 Trajectory of change in March Randomized, Existing family caregiver quality 2017 Controlled Trial of Home Care Nursing Age 18 years and older of life while caring for the Home Care Nurse Practice a palliative patient at Inclusion Criteria: Carer Support Needs home, as measured by Assessment Practice the Quality of Life in For home care nurse participants: Tool With Family Life Threatening Illness Experimental: Practice Caregivers  registered or licensed home care nurse at one of the following - Family carer version Support Tool of Palliative Patients Vancouver Island Health Authority Home and Community Care (QOLLTI-F) total score Intervention at Home offices: Victoria, Esquimalt/Westshore, Peninsula, Royal Oak, at 8 to 96 weeks Duncan/Ladysmith, Oceanside, Comox Valley, Campbell River  practiced in home care for > 6 months  work a minimum of 10 shifts per month  communicate well in English ©Institute for Clinical and Economic Review, 2016 Page 91 NCT02261935 For family caregiver participants: Family caregiver grief symptoms after  currently a family caregiver of palliative patient at home patient death, as  patient and family caregiver must be 18 years of age or older measured by the Texas  must communicate well in English Revised Inventory of Exclusion Criteria: Grief (TRIG) subscale scores 12 weeks after For family caregiver participants patient death  diagnosed cognitive impairment  identified safety risks in the home care environment Collaborative Care to RCT Collaborative Care to N=312 Difference in Kansas December Alleviate Symptoms Alleviate Symptoms City Cardiomyopathy 2016 and Adjust to Illness and Adjust to Illness Age 18 years and older Questionnaire (KCCQ) in Chronic Heart (CASA) overall score at 6 Failure (CASA) Trial months Inclusion Criteria: Usual care NCT01739686  Age 18 years of age or older  Able to read and understand English  Consistent access to a telephone  Patients have a primary care or other provider who is willing to facilitate intervention medical recommendations  A diagnosis of heart failure with at least one of the following: [hospitalization primarily for heart failure in the year prior (including current); taking at least 20 mg oral furosemide (or equivalent) daily in a single or divided dose; Brain natriuretic peptide(BNP) ≥ 100 or N- terminal prohormone of brain natriuretic peptide(NT-proBNP) ≥ 500; EF≤40%]  Report a low health status (KCCQ-SF≤70)  Bothered by at least one target symptom: [Pain; Depression; Fatigue; Breathlessness] Exclusion Criteria:  Previous diagnosis of dementia ©Institute for Clinical and Economic Review, 2016 Page 92  Active substance abuse or dependence, defined by either a diagnosis of abuse or dependence or an AUDIT-C ≥ 8, or self- reported substance abuse in the past 3 months  Comorbid metastatic cancer  Nursing home resident  Heart Transplant recipient  LVAD recipient Randomized Study of RCT Early palliative care N=350 Change in Functional April 2018 Early Assessment of Cancer Age 18 years and older Therapy (Quality of life Palliative measure) at 12 weeks Care Integrated With Standard of care Standard Inclusion Criteria: Oncology Care Versus Standard  Confirmed metastatic lung cancer (NSCLC, small cell lung cancer, Oncology Care Alone and mesothelioma)or non-colorectal GI cancer (esophageal, gastric in Patients With and hepatobiliary) not being treated with curative intent Advanced Lung and  Informed of metastatic disease within the previous 8 weeks Non-colorectal  No prior therapy for metastatic disease Gastrointestinal  Able to read questions in English or willing to complete Malignancies questionnaires with the assistance of an interpreter  Relative or friend of patient who will likely accompany the patient to clinic visits Exclusion Criteria: NCT01401907  Significant psychiatric or other co-morbid disease SmartCare: RCT CAU+(Enhanced Care as N=420 Change in depression June 2017 Innovations in Usual) from baseline at 4 Caregiving Age 21 years and older months on the Interventions shortened CES-D Inclusion Criteria: CAU+ and SmartCare (online symptom Care recipient: NCT02058745 management intervention for  Over 21 years of age.  Newly (within 1 month) diagnosed with a PMBT (tumor verified via caregivers) pathology report to be a glioblastoma multiforme, anaplastic ©Institute for Clinical and Economic Review, 2016 Page 93 astrocytoma, anaplastic oligodendroglioma, anaplastic oligoastrocytoma, medulloblastoma, or anaplastic ependymoma). Enhanced Care as Usual Caregiver: and SmartCare and Beating Blue (online  Primary nonprofessional, non-paid caregiver, as identified by the therapy for depressive care recipient.  Over 21 years of age with telephone access. symptom)  Reads-speaks English  Obtains a score of >6 on the shortened CES-D.  Caregivers may or may not be receiving pharmacotherapy for depressive symptoms Exclusion Criteria: Caregiver:  Currently considers self to be a primary caregiver for anyone else other than children  Currently receiving any type of formal counselling for depressive symptoms ©Institute for Clinical and Economic Review, 2016 Page 94 A8. Evidence Tables Author & Study Study Design Interventions Mean/Median Inclusion and Patient Outcomes Year Quality (sample size) Duration of Exclusion Characteristics Follow-up Criteria Aiken Poor RCT PC group (100) NR Patients ≥18 Age: 68.5 PC group, Patient 2006 -MCO medical years with COPD % female: 64% QoL: treatment for Outcomes or CHF (class 3b SF36, physical function (based on 100pt scale)-Slope disease plus PC presented or 4) with an Indications: over 9 months including: every 12 estimated 2-year COPD, CHF CHF: 0.18 -Symptom self- weeks up to 36 life expectancy COPD: 1.00 management and weeks and marked SF36, general health (based on 100pt scale)- education limitation of Slope over 9 months -Advance care physical CHF: 0.16 planning functioning and COPD: 0.54 -Interdisciplinary recent (<3 care coordination months) Control group, Patient and management exacerbation of QoL: -Assessment of symptoms SF36, physical function (based on 100pt scale)- psychological, Slope over 9 months spiritual, and CHF: -1.39 emotional COPD: -0.95 needs of the SF36, general health (based on 100pt scale)- patient and family Slope over 9 months CHF: -0.17 Control group COPD: -1.67 (90) -MCO medical Overall control slope is declining while overall treatment for PhoenixCare slope is rising (for physical function, p disease <0.05) or remains stable (for general health, p<0.05) -Occasional home visits provided by There were no statistically significant between-group some MCOs differences in symptom severity at 6 months or health -Individual case resource utilization (ED visits) management to some patients Patients with COPD showed stronger responsiveness to the intervention. PhoenixCare Demo ©Institute for Clinical and Economic Review, 2016 Page 95 Bakitas Good RCT PC group (161) 58.4 weeks Newly diagnosed Age: 65.0 QoL: 2009 -Case (8-12 weeks) % female: 42.0 FACIT-Pal 184pt Scale difference in change scores management Monthly patients with 4.6, p=0.02(higher QoL in the intervention group) -Patient follow-up until advanced Indications: Psycholgical distress/depression: education death or study gastrointestinal, Advanced CES-D 60pt Scale difference in change scores - Symptom self- completion lung, cancer -1.8, p=0.02 (lower depression mood in the management and genitourinary, or intervention group) education breast cancer -Advance care PC better than control management Patients with 1-year Survival -Phone-based impaired HR=0.67 (95% CI, 0.50-0.91, p=0.009) format in rural cognition, a populations psychiatric There were no statistically significant between-group -Problem-solving disorder, or differences in health resource utilization, symptom -Monthly medical active substance severity, or survival rate (post-hoc) appointments use were with caregivers excluded -PC-certified nurse practitioner Control group (161) -All oncology and supportive services without restrictions -Referral to the institutions’ interdisciplinary palliative care service Project ENABLE Bakitas Fair RCT PC group (104) NR Adults (≥18 Age: 64.3 PC group, Patient 2015 -In-person PC years) with % female: 52.7 Survival rate: consultation Follow-up advanced-stage 63% -Structured PC every 6 weeks solid tumor or Indications: ©Institute for Clinical and Economic Review, 2016 Page 96 telehealth nurse until 24 weeks Hematologic Advanced Control group, Patient coaching sessions then every 12 malignancy, cancer Survival rate: - Problem solving weeks until oncologist- 48% -Symptom death determined management prognosis of 6 to PC > control -Symptom 24 months Survival rate: management p=0.038 at 1 year -Self-care Exclusions -Advance care included There were no statistically significant between-group planning impaired differences in symptom severity, QOL, psychological cognition, distress, resource utilization (ICU days, hospital days, Control group psychiatric or ED visits), overall median survival, or location of death (103) substance use (home) -Same as disorder, intervention uncorrectable group but hearing disorder, administered 3 or unreliable months later telephone service. Early vs. delayed (3 months) palliative oncology care Project ENABLE- telehealth approach Brumley Good prospective PC group (161) NR PC group- Kaiser Age: 60.0 PC group, Patient 2003 comparative -Interdisciplinary Permanente % female: 53.0 Health resource utilization: cohort team-based 60-day follow- members Physician visits: 5.3 approach, up interval presenting with Indications: Hospital visits: 2.4 Matched including the a life- CHF 27.4% ED visits: 0.9 groups? patient and family threatening COPD 18.3% Skilled nursing care visits: 0.9 ☐ members plus a disease Cancer 50.8% Total home health visits: 35.0 physician, (primarily COPD, Location of Death: Deceased nurse, and social CHF, or cancer) Palliative care % died at home: nearly 90% subgroup worker with and a prognosis group was selected expertise in members were Control group, Patient in order to Health resource utilization: ©Institute for Clinical and Economic Review, 2016 Page 97 attempt to symptom of approximately more likely to Physician visits: 11.1 control for management and 1 year or less to be married, Hospital visits: 9.4 baseline biopsychosocial live not have CHF, ED visits: 2.3 differences intervention have fewer Skilled nursing care visits: 4.6 -Patient and Control group- days on Total home health visits: 13.2 family education Kaiser service, and Location of Death: medication use, Permanente to be sicker as % died at home: 57% and self- members determined by management presenting with a lower score PC better than control (based on reduced resource -24-hour a life- on utilization) telephone threatening the PPS (0- Physician visits: p=0.001 support disease death, 100- Hospital visits: p<0.001 (primarily COPD, normal) ED visits: p<0.001 Control group CHF, or cancer) Skilled nursing care visits: p=0.005 (139) and a prognosis -Various amounts of approximately Resource utilization analysis included physician, and levels of 24 months or hospital, ED, and skilled nursing visits controlled for home less to live days on service, CHF diagnosis, and severity of illness health services, as covariates using MANCOVA acute care services, primary Patient satisfaction improved significantly in the PC care services and group and stayed the same in the control group based hospice care on the Reid-Gundlach Satisfaction with Services instrument but these differences were not compared statistically Brumley Fair RCT PC group (145) PC group: 196 Patients with a Age: 73.8 PC group, Patient 2007 -Pain control and days primary % female: 73.2 Satisfaction: other symptom Control group: diagnosis of CHF, 93.4% at 90 days management 242 days COPD, or cancer Indications: Health resource utilization: -Pain control and and a life Cancer 46.6% ED visits: 20% other symptom Outcomes expectancy of 12 CHF 32.5% Hospitalizations: 36% management measured months or less, COPD 30.8% Location of Death: -Palliative care every 30 days have visited the Home death: 71% physician who up to 90 days emergency Only those coordinates care (outcomes at department or analyzed, not Control group, Patient from a variety of 120 days were hospital at least ITT Satisfaction: health care excluded due once within the 33% at 90 days providers to sample previous year; Health resource utilization: -Educations focus attrition) and scored 70% ED visits: 33% at 90 days on identifying or less on Hospitalizations: 59% Location of Death: ©Institute for Clinical and Economic Review, 2016 Page 98 goals of care and the PPS (0-death, Home death: 51% the expected 100-normal) course of the PC better than control disease Satisfaction: OR=3.37 (95% CI: 0.65-4.96, p=0.003) Control group Health resource utilization: ED visits- p=0.01; Cramer’s (152) V=0.15 -Various amounts Hospitalizations- p<0.001; Cramer’s V =0.23 and levels of (no significant changes in services use after adjusting home health for survival, services, acute age, and severity of illness) care services, Location of Death:home death (after controlling primary for age, survival time, and medical conditions): care services, and OR=2.20, (95% CI: 1.3–3.7, p<0.001) hospice care Cheville Fair RCT PC group (49) NR Adult patients Age: 59.5 PC group, Patient 2010 Secondary - Physical therapy undergoing % female: 35.9 QoL: Analysis of with educational Outcomes radiation LASA physical well-being change score at week 4: Rummans components assessed at therapy for Indications: +0.4 2006 -A psychiatrist or week 4, 8, and advanced cancer (tumor type) psychologist with 27 with prognoses GI 37.9% Control group, Patient co-facilitation ≥6 months and Head and neck QoL: provided 5-year survival 17.5% LASA physical well-being change score at week 4: by an advanced estimates ≤50% Lung 14.6% -10.0 practice nurse, and no evidence Brain 11.6% licensed social of alcohol or Other 18.4% PC better than control at week 4, p=0.02 worker, or nontobacco certified hospital substance Fatigue and vigor (on POM) were not significantly chaplain dependence different between the groups and all intergroup -Goal settings and differences were no longer significant at 8 and 27 coping strategies Patients with weeks recurrent Control group disease (54) following a 6- -Regular month disease- assessments for free interval treatment were excluded toxicities and weight loss by an advanced ©Institute for Clinical and Economic Review, 2016 Page 99 practice nurse or radiation oncologist Ciemins Fair retrospective PC group (140) NR Adult patients Age: (based on PC group, Patient 2006 comparative -Patients and with life groups) cohort families not ready expectancy of 6 <65, 16.6% Health resource utilization: for hospice but months or less 65-75, 17.5% Discharge to hospice: 47% Matched eligible for home and two of the 76-86, 32.5% groups? care following: >85, 33.5% Control group, Patient ☒ -Disease process 1) Diagnosis of % female: 62.6 Health resource utilization: education, advanced Discharge to hospice: Control group 1: 33% Patients in symptom cancer, Significant Control group 2: 16% the Usual management, advanced CHF, differences for Care cohorts goal clarification, end-stage some PC better than control were advance care pulmonary characteristics, group 1: p=0.003 matched to planning and disease, end- e.g., life group 2: p<0.0001 the AIM contingency stage hepatic expectancy of cohort on planning disease, 6 months, AIM was particularly effective in African Americans for selected -Pain and end-stage gender, race, referral to hospice symptoms symptom neurologic primary and management disease, other caregiver prognosis end-stage Control group 1 medical No differences (68) diagnosis, or between - Standard advanced groups at home health debility and baseline for services, without decline pain, the programmatic 2) Failing or breathlessness, focus on palliative reduced depression or care effectiveness of anxiety non-palliative Control group 2 treatment of (227) primary disease - Same as Control process 1 3) Decline in functional status PC group vs. and/or Control 1: nutritional status patients from the in past 30 days. same home health site ©Institute for Clinical and Economic Review, 2016 Page 100 Control 2 patients were drawn from another site (demographic-ally similar but without an AIM program) Dionne- Good RCT PC group (61) NR Patients >18 Age: 60 PC group, Caregiver Odom Same -Patients and CGs years; new % female: 78.7 Psycholgical distress/depression: 2015 intervention each were Minimum diagnosis, Indications: Change in CESD Score for depression from enrollment as Bakitas assigned a follow-up time recurrence, or Advanced to 3 months: -3.2 2015 but different nurse was 24 weeks progression cancer measuring coach or until patient of an advanced- Control group, Caregiver caregiver -Sessions were death if it stage cancer Psycholgical distress/depression: outcomes guided by the occurred within The early Change in CESD Score for depression from enrollment Charting Your during that approximately group had a to 3 months: -1.8 Course: Caregiver period. Not all 30 to 60 days of higher (CYC-C) caregivers the proportion PC better than control mean between group guidebook that were followed date the patient employed, difference in change from baseline (change represents focused on topics beyond the was informed of fewer retired, average follow-up minus baseline), -3.4; SE, 1.5; d=- such as cultivating initial 24-week the diagnosis by and fewer 0.32; p=0.02 communication period. his or her unemployed skills with patient oncology (p=0.05). There were no statistically significant differences in and health care clinician QOL or burden subscale clinicians and oncologist- determined Control group prognosis of 6 to (61) 24 months -Same as PC group but Exclusions received the included intervention 3 impaired months later cognition, psychiatric or Early vs. delayed substance use (3 months) disorder, palliative uncorrectable oncology care hearing disorder, ©Institute for Clinical and Economic Review, 2016 Page 101 or unreliable Project ENABLE- telephone telehealth service. approach Greer Fair RCT PC group (n=77) 18 months Patients Age: 64.9 PC group, Patient 2012 Secondary -At least monthly receiving % female: 52 Chemotherapy Analysis of consults with a PC treatment at % Caucasian: Within 60 days of death: Temel 2010 team member MGH for 97 PC: 52.5% implementing metastatic % Hispanic: 1.3 Days between last IV chemo and death: National NSCLC diagnosed % Married: 62 PC: median 64, mean 100.6, SD 89.6 Consensus Project in previous 8 for Quality weeks, ECOG PS Indications: Hospice received >7 days before death: Palliative Care 0-2 Non-small cell PC: 60% guidelines lung cancer -Routine oncology Control group, Patient care Chemotherapy Within 60 days of death: Control group Control: 70.1% (n=74) -Routine oncology Days between last IV chemo and death: care Control: median 40.5, mean 75.5, SD 42.6 -Palliative care Hospice received >7 days before death: upon request 33.3% PC better than control (based on reduced resource utilization) Chemotherapy Within 60 days of death: p=0.05 Days between last IV chemo and death: p=0.02 Hospice received >7 days before death: p=0.004 There was no statistically significant between-group difference in death at home Kerr 2014 Good retrospective PC group (n=149) NR Home Age: PC group, Patient comparative -Team included a Connection PC group Health resource utilization: cohort palliative care- palliative care 0–50: 4 (3%) Enrolled in hospice (n, %)- trained registered program 105, 70% ©Institute for Clinical and Economic Review, 2016 Page 102 Matched nurse (RN), social participants 51–65: 20 Median days spent in hospice- groups? worker, enrolled for (13%) 34 (0-276) ☒ volunteers, and more than 30 66–75: 36 palliative care days; Controls (24%) Control group, Patient physician identified from 76–90: 80 Health resource utilization: -Services included enrollment and (53%) Enrolled in hospice (n, %)- symptom claim data from ≥91: 9 (6%) 133 (25%) management, Independent Median days spent in hospice- education, Health, a Control group 9 (0-606) supportive commercial 0–50: 9 (2%) discussions about insurance 51–65 71 (13%) PC better than control for enrollment in hospice heath care provider. 66–75: 116 (p<0.0001 and median days spent in hospice decision making (22%) (p=0.0003).). and goals, social 76–90: 280 work visits to (52%) Utilization costs within the last year of life were also facilitate access ≥91: 61 (11%) reported to community support services, Gender respite care, and PC group 24/7 on-call Male: 70 (47%) palliative care Female: 79 nurse support (53%) -Nurses and social workers visit PC group participants a Male: 70 (47%) minimum of Female: 79 every 30 days, (53%) and a biweekly interdisciplinary Control group group meeting is Male: 222 held to discuss (41%) each patient case Female: 315 with the palliative (59%) care physician Baseline Control group characteristics (n=537) were not -Propensity- statistically matched patients significant with similar between clinical and groups with ©Institute for Clinical and Economic Review, 2016 Page 103 demographic the exception characteristics of CHF (40% of controls versus 28% of PC patients [p=0.01]) and CRD (28% of controls versus 14% of PC patients [p<0.0001]). This was controlled for in a multivariate analysis. Keefe 2005 Poor RCT PC group (41) 1 week Inclusion criteria: Patient PC group - Three pain Advanced cancer Age: 60.5 Caregiver Matched management Mean 7.56 diagnosis (i.e., % female: 44 Other: groups? training sessions days, range 0 metastatic or % white: 78 a caregiver ☐ with patient and to 31 days disseminated Partner version of the Chronic Pain Self-Efficacy Scale, mean: partner at home disease) with Age: 58.5 pain: - Patient and disease-related % female: 62 baseline: 53.2 partner education pain, a worst % white: 79 post-tx: 62.6 about pain Other symptoms (Self-Efficacy for Other Symptoms cancer pain and rating>3 on the Indications: Scales): its management Brief Pain advanced baseline: 63.4 - Patient and Inventory, a life cancer post-tx: 68.8 partner education expectancy of Control group about a variety of less than six Caregiver pain coping months, no Other: strategies change in a caregiver - Partner disease version of the Chronic Pain Self-Efficacy Scale, mean: education about treatment pain: how to planned, and baseline: 44.1 over 18 years of post-tx: 39.6 age. All patients other symptoms: ©Institute for Clinical and Economic Review, 2016 Page 104 help the patient met the baseline: 61.0 acquire and Medicare post-tx: 52.5 maintain coping hospice benefit skills definition for PC better than control hospice Other: Control group eligibility, i.e., a a caregiver (37) prognosis of six version of the Chronic Pain Self-Efficacy Scale, mean: - usual care months or less. pain: p=0.006 through their other symptoms: p=0.012 medical outpatient or There were no statistically significant between-group hospice program differences in patient QOL or pain or caregiver strain or caregiver mood Lindell Poor RCT PC group (10) 6 weeks Patients were Patient PC group 2010 required 1) to be Age: 66 Patient Matched -Group sessions aged %Female: 34 QoL: groups? discussing more than 21 % Caucasian: Health-related quality of life (HRQoL) measured by SF- ☐ cognitive years; 95 36 v2-physical: behavior 2) to be able to 31.06 techniques read and Caregiver Caregiver -Care planning understand Age: 65 Psycholgical distress/depression: -Symptom English; %Female: 76 Stress measured by Perceived Stress Scale: management 3) to be % Caucasian: 17.61 diagnosed with 95 Control group IPF; and PC group (11) 4) to have an Indications: Patient -Usual care: seen FVC reflecting Idiopathic QoL: by clinical team of moderate (FVC pulmonary Health-related quality of life (HRQoL) measured by SF- nurse specialist 55%-70% fibrosis 36 v2-physical: and physicians predicted) or 36.04 -Psychological severe (FVC < ANCOVA used Caregiver counseling if 55% predicted) to control for Psycholgical distress/depression: indicated disease. significant Stress measured by Perceived Stress Scale: Care partners differences at 20.99 *counted as were required baseline patient/care giver 1) to be aged dyads more PC worse than control than 21 years; Patient QoL: p=0.038 2) to be able to Caregiver stress: p=0.018 read and ©Institute for Clinical and Economic Review, 2016 Page 105 understand There were no statistically significant between-group English; and differences in patient symptom severity or patient 3) to live with or anxiety or patient depression or caregiver QoL or care for the caregiver anxiety or caregiver depression patient with IPF. McMillan Poor RCT PC group (111) 4 weeks Inclusion Caregiver Caregiver 2006 Caregiver training Criteria: Age: 61.5 QoL: Matched on 30 days follow- caregivers had to %Female: 85.5 Caregiver Quality of Life Index-Cancer (CQOL-C) groups? -Creativity: to up be providing 0.096, p=0.042 ☐ view problems care Patient Psycholgical distress/depression: from different for adult Age: 70.7 Burden of cancer symptoms measured by Memorial perspectives and patients with %Female: 40.0 Symptom Assessment Scale (MSAS): develop cancer, and both -0.14, p<0.001 strategies to solve had to consent Other: problems to participate, Indications: Caregiver Demands Scale (CDS): 0.01, p=0.033 -Optimism: to have at least a Advanced have a positive sixth grade cancer *Only comparison between intervention and standard but realistic education, be care control were presented. Statistics are Estimate of attitude able to read and Group*Time Interaction and p-values from Random -Planning: to set understand effects model reasonable English, and caregiving goals achieve a There was no statistically significant between-group and steps to minimum score difference in general caregiver mastery reach goals of seven on the -Expert Short Portable information: Mental Status knowledge about Questionnaire. the nature of Exclusion problems and criteria: symptom caregivers were assessment in active -Standard hospice treatment for care cancer themselves. Control group Primary I(109) caregivers were -Standard hospice identified by the care hospice, and dyads were Control group II excluded if it was (109) unclear who the ©Institute for Clinical and Economic Review, 2016 Page 106 -Standard hospice primary care caregiver would - three supportive be or if patient visits providing performance individual support status suggested to caregivers, that patients discussing would not feelings, fear and survive more relationship than a few days. Meyers Poor prospective PC group (44) NR All patients Age: median 62 PC group 2004 comparative entered onto a in PC, 57 in Patient cohort -Nurse focused on Patients were University of usual Health resource utilization: chemo toxicity traced until California %Female: 45 % entering hospice in those who died: 92 Matched and symptom death Davis Phase I or Control group groups? management and Phase II cancer Indications: Patient ☐ care coordination investigational Advanced Health resource utilization: -Social worker therapy protocol cancer % entering hospice in those who died: 53 focused on were considered emotional eligible for PC > control support, family Simultaneous Health resource utilization: and interpersonal Care protocol p=0.034 issues and end of entry. life planning Randomized There were no statistically significant between-group -Services Phase III studies differences in patient QoL or cycles of chemo or days provided in were allowed if in hospice patients’ homes they compared -Investigational different therapy trials chemotherapy regimens for Control group advanced (20) disease. -Standard supportive care -Investigational therapy trials UC-Davis Cancer Center ©Institute for Clinical and Economic Review, 2016 Page 107 Rabow Fair RCT PC group (50) 52 weeks Inclusion criteria: Age: 68.6 PC group 2004 Cluster RCT Physicians in %Female: 64 Patient -Consultation Questionnaires both practice % White: 53 Psycholgical distress/depression: Matched with PCPs based at enrollment, modules were % live alone: 43 Anxiety (Profile of Mood States) groups? on assessments 6 months and invited to refer At 6 months: 6.8 ☐ conducted by 12 months, adult Indications: At 12 months: 5.3 social works patients who patients with Cancer, Symptom severity: -Symptom completed all diagnoses of advanced Dyspnea (UCSD Shortness of Breath Q-degree dyspnea management three were cancer, COPD, interferes): -Case analyzed advanced COPD, advanced CHF At 6 months: 32.6 management or advanced At 12 months: 32.6 -Family caregiver CHF, whom they Other: training and believed had a Sleep quality (6 items from Medical Outcomes Study): support life expectancy At 6 months: 11.9 -Chart review of of 1 to 5 years At 12 months: 12.5 medication and who were Spiritual well-being Scale (overall): -Spiritual and not yet ready for At 6 months: 98.0 psychological hospice care. At 12 months: 105.5 support Exclusion Health resource utilization: -Support groups criteria: Primary care visits: 7.5 discussion Patients with Urgent care visits: 0.3 symptom non-melanoma management and skin cancers, Control group care planning dementia, or Patient -Telephone psychosis; those Psycholgical distress/depression: support and enrolled in Anxiety (Profile of Mood States) advocacy hospice care; At 6 months: 5.5 and those At 12 months: 5.9 unable to Control group complete a Symptom severity: (40) written survey in Dyspnea (UCSD Shortness of Breath Q-degree dyspnea -Usual primary English or interferes): care Spanish. At 6 months: 40.3 At 12 months: 40.6 Other: Sleep quality (6 items from Medical Outcomes Study): At 6 months: 10.0 At 12 months: 11.0 Spiritual well-being Scale (overall): At 6 months: 91.2 ©Institute for Clinical and Economic Review, 2016 Page 108 At 12 months: 92.4 Health resource utilization: Primary care visits: 10.6 Urgent care visits: 0.6 *means were adjusted for baseline, ANOVA F and p value for Between Groups were presented as following: PC better than control Psycholgical distress/depression: Anxiety (Profile of Mood States) 4.09, p=0.05 Symptom severity: Dyspnea (UCSD Shortness of Breath Q-degree dyspnea interferes): 7.06, p=0.01 Health resource utilization: Primary care visits: p=0.03 Urgent care visits: p=0.04 Other: Sleep quality (6 items from Medical Outcomes Study): 4.05, p=0.05 Spiritual well-being Scale (overall): 8.21, p=0.007 There were no statistically significant between-group differences in patient depression or patient pain or patient satisfaction or patient QoL or patient survival Radwany Fair RCT PC group (40) 1 year Inclusion criteria: Age: 69.2 2014 All new %Female: 75 Patient Matched -Home visit by PASSPORT % White: 85 Other: groups? interdisciplinary enrollees > 60 Palliative Care Outcome Scale: ☐ team for years old who Mean difference between groups at 12 month: assessment and passed a mental -4.546 (95% CI, -7.853 to -1.238) care planning status screening -Symptom and had one of There were no statistically significant between-group management the following: differences in patient symptom severity or patient congestive anxiety or patient depression or patient QoL or patient heart failure and survival or health care utilization ©Institute for Clinical and Economic Review, 2016 Page 109 -Psychological being actively support and treated); chronic spiritual need obstructive communication pulmonary and legal support disease and on home oxygen; Control group diabetes with (40) renal disease, -Usual PASSPORT neuropathy, care visual problems, or coronary artery disease; Ohio’s PASSPORT end-stage liver program disease or (community- cirrhosis; cancer based long term (active, not care Medicaid history of) waiver program) except skin cancer; renal disease and actively receiving dialysis; amyotrophic lateral sclerosis with history of aspiration; Parkinson’s disease stages 3 and 4; or pulmonary hypertension. Exclusion criteria: Active alcoholics (i.e., those who drink > 2 drinks per day on average) and illegal substance users; clients who have ©Institute for Clinical and Economic Review, 2016 Page 110 schizophrenia or are psychotic; consumers who could not pass the Mental Status Questionnaire; consumers who were already enrolled in hospice Ranganathan Good retrospective PC group (392) 4 weeks Inclusion criteria: Age: 67 PC group 2013 comparative Patients % Female: 51.4 Patient cohort -Palliative home Follow up for discharged from *demographics Health resource utilization: care staffed by 30-day one of the three in PC group, 30-day readmission (%): Matched hospice nurses readmission health system controls were 9.1 groups? - Biweekly hospitals and matched ☒ discussion in were admitted Control group interdisciplinary to either a home Indications: Patient team meetings care program or 67.3% cancer Health resource utilization: propensity that include the a palliative home diagnosis 30-day readmission, n (%): score and patient’s nurse as care program. 17.2 the well as a Mahalanobis physician, PC better than control matching chaplain, and Mean ATT (the average procedure social worker treatment effect on treated patients) 8.3% (95% CI, -Access to a 8.0-8.6) telephone triage line that provides There was no statistically significant between-group 24-hour difference in mortality access to a hospice nurse Control group (890) -Home care program staffed by home care nurses ©Institute for Clinical and Economic Review, 2016 Page 111 Penn Homecare and Hospice Services Rummans Fair RCT PC group (49) NR Inclusion criteria: Age: 59.5 PC group 2006 A diagnosis % Female: 36 Patient Matched -Cognitive Primary within the last QoL: groups? intervention outcomes 12 months, an Indications: Change in Linear Analog Scale of Assessment (LASA) at ☐ -Emotional measured at 4 expected Advanced 4 weeks: intervention weeks survival time of cancer 3 -Physical at least 6 intervention months, a 5-year Symptom severity: -Social survival Change in LASA-physical symptom subconstruct at 4 intervention probability of no weeks: -Spiritual more than 0.4 intervention 50% (as routinely determined by Other: Control group the primary Change in LASA-emotional wellbeing subconstruct at 4 (54) radiation weeks: 2.8 -Standard medical oncologist), and Change in LASA-social wellbeing subconstruct at 4 care as a weeks: 7.0 recommended by Treatment Change in LASA-legal concerns subconstruct, at 4 their radiation recommendation weeks: 6.7 oncologist. of radiation -Interactions with therapy of at Control group their oncologist least 2 weeks. Patient -Referrals to Exclusion QoL: specialists when criteria: Change in Linear Analog Scale of Assessment (LASA) at indicated Previous 4 weeks: -Opportunities for radiation -9 receiving support therapy, Symptom severity: through a range recurrent Change in LASA-physical symptom subconstruct at 4 of outside disease after a weeks: agencies disease-free -10.0 period of greater Change in LASA-emotional wellbeing subconstruct at 4 Mayo Clinic than 6 months, weeks: -5.4 Rochester and a previous Change in LASA-social well-being subconstruct at 4 cancer diagnosis weeks: -5.4 within 5 years; Change in LASA-legal concerns subconstruct, at 4 a Mini Mental weeks: -4.7 Status Examination All scores on 0-100 scale ©Institute for Clinical and Economic Review, 2016 Page 112 score of less than 20,an PC>control Eastern QoL: Cooperative Change in Linear Analog Scale of Assessment (LASA) at Oncology Group 4 weeks: performance p=0.009 score of 3 or more, active Symptom severity: alcohol or Change in LASA-physical symptom subconstruct at 4 substance weeks: dependence p=0.022 (except nicotine), active Other: thought Change in LASA-emotional wellbeing subconstruct at 4 disorder, or weeks: p=0.046 suicidal plans or Change in LASA-social well-being subconstruct at 4 were weeks: p=0.025 participating in Change in LASA-legal concerns subconstruct, at 4 another weeks: p=0.048 psychosocial research trial; receiving psychotropic medications or counseling Scheffey Fair retrospective PC group (342) NR Inclusion criteria: Age: PC group 2014 comparative All patients who 40-54, 17% Patient cohort -Symptom died under 55-64, 21% Health resource utilization: assessment using CC hospice 65-74, 24% LOS, median (day): Matched a new Edmonton services during 75 and older, 24 groups? Symptom the five-year 38% ☒ Assessment period 2008 to Control group System tool 2012. % Caucasian Patient -Advance care Exclusion not Hispanic: Health resource utilization: matched by planning criteria: 62 LOS, median (day): age, gender, -Discussion about Patients with a 15 median goals of care hospice *numbers from income of diagnosis PC group, PC>control their zip Control group of HIV or who controls were Difference between group medians: 6 (95% CI, 5-13, (1368) were younger matched p<0.001) than 18 years ©Institute for Clinical and Economic Review, 2016 Page 113 code, and - no PC prior to Indications: diagnostic hospice 17% group nonmalignant conditions Capital Caring (CC), a large nonprofit hospice and palliative care organization in the Washington, District of Columbia, area. Seow Good retrospective PC group (3109) 0.33 years Exposed Age: median 74 PC group 2014 comparative -Palliative care Inclusion criteria: in PC, 75 in Patient cohort team to manage The study Formal palliative control Health resource utilization: symptoms duration is 4 care specialist % female: 51.7 Being in hospital in the last 2 weeks of life, n (%): Matched -Provision of months teams that 970 (31.2), p<0.001 groups? education and (a) provided Indications: RR=0.68 (95% CI, 0.61-0.76) ☒ care interdisciplinary, about 80% had Having an emergency department visit in the last 2 -Coordination of home based, cancer weeks of life, n (%): services palliative and 78% 896 (28.9), p<0.001 Propensity care; (b) were received end of RR 0.77 (95% CI, 0.69-0.86) score based Control group the only such life homecare Location of Death: on age at (3109) team in their services (after Dying in hospital, n (%): death, sex, - no specialist respective matching) 503 (16.2), p<0.01 comorbidity, palliative care region; (c) had RR=0.46 (95% CI, 0.40-0.52) cancer dx, team available little or no Control group hospital/ED change in Patient use before staffing between Health resource utilization: intervention, 2009 until 2012; Being in hospital in the last 2 weeks of life, n (%): then (d) had broad 1219 (39.3) matched by admission Having an emergency department visit in the last 2 propensity criteria not weeks of life, n (%): score, region limited to one 1070 (34.5) and time, disease (such as Location of Death: homecare cancer); (e) Dying in hospital, n (%): service type admitted more 887 (28.6) than 50 patients ©Institute for Clinical and Economic Review, 2016 Page 114 and time in a year; (f) were PC better than control homecare available to Health resource utilization: patients 24/7; Being in hospital in the last 2 weeks of life: and RR 0.68 (95% CI, 0.61-0.76) (g) had the same Having an emergency department visit in the last 2 core members of weeks of life: their team as the RR 0.77 (95% CI, 0.69-0.86) past randomized Location of Death: trials. Dying in hospital: Exclusion RR 0.46 (95% CI, 0.40-0.52) criteria: Patients who were alive after fiscal year 2011, were <18 years old, or had an invalid or missing provincial health insurance number. Unexposed were sampled using historical and geographical methods. Strasser Poor retrospective PC group (80) NR The MD clinic at Age: 54 PC group 2004 comparative - Assessment by The University of % female: 54.7 Patient cohort an In the MD Texas M. D. Survival rate: interdisciplinary clinic, follow- Anderson Cancer Indications: Median: 10 weeks Matched team up data Center was Advanced groups? -On-site obtained a established five cancer Control group ☐ counseling, median of nine years after the Patient specific patient days (mean pain and Survival rate: and family 26; range 3– symptom Median: 51 weeks education, and 63) management simple In the PSM (PSM) clinic. The PC<control interventions clinic 32 (42%) investigators p<0.0001 -Appropriate patients were identified the follow-up seen for first 138 follow-up consecutive within a ©Institute for Clinical and Economic Review, 2016 Page 115 that is tailored to median of 28 patients seen at the needs of the days the MD patients and after the initial clinic and their caregivers visit (mean 26; selected a range 3–56). consecutive Control group sample of 77 (32) patients seen at -The PSM clinic the PSM clinic team (physician during the same and nurse) time period can request an assessment by any of the members of the MD clinic team -No handwritten or audiotaped recommendations are routinely given to patients The MD clinic at The University of Texas M. D. Anderson Cancer Center Temel Good RCT PC group (77) 9.8 months Eligible patients Age: 64.9 PC group 2010 -Meeting with a Median were enrolled % female: 52.1 Patient Matched member survival 9.8 within 8 weeks Psycholgical distress/depression: groups? of the palliative months in the after diagnosis Indications: % patient with mood symptoms @ 12 weeks ☐ care team within entire sample and were non–small-cell HADS-D: 16 3 weeks after randomly lung cancer PHQ-9: 4 enrollment and at assigned to one least monthly of the two Survival rate: thereafter in the groups in a 1:1 Median survival, month: 11.6 (95% CI, 6.4-16.9) outpatient setting ratio without until death stratification. Control group -Assessing Patient physical Psycholgical distress/depression: % patient with mood symptoms @ 12 weeks ©Institute for Clinical and Economic Review, 2016 Page 116 and psychosocial HADS-D:38 symptoms PHQ-9: 17 -Establishing goals of care Survival rate: -Assisting with Median Survival, month: 8.9 (95% CI, 6.3-11.4) decision making regarding PC better than control treatment HADS-D: p=0.01 Coordinating care PHQ-9:p=0.04 on the FACT-L scale basis of the difference between intervention and control at 12 individual needs weeks of the patient 6.5 (95% CI, 0.5-12.4, p=0.03) Symptom severity: Control group the lung cancer (74) subscale (LCS) of the FACT-L scale -Not scheduled to difference between intervention and control at 12 meet with the weeks palliative care 1.7 (95% CI, 0.1-3.2, p=0.04) service unless a Survival rate: p=0.02 meeting was Adjusted HR for death in controls: 1.70 (95% CI, 1.14- requested by the 2.54, p=0.01) patient -Continuation of There was no statistically significant between-group routine oncologic difference in anxiety care throughout the study period MGH ©Institute for Clinical and Economic Review, 2016 Page 117 Zimmermann Good RCT PC group (228) 16 weeks Inclusion criteria: Age: 60.7 QoL: 2014 Cluster -Assessment of Ambulatory % female: 56.6 FACIT-Sp symptoms, The study patients with adjusted difference between change scores Matched psychological duration is 4 newly diagnosed Indications: at 4 months groups? distress, social months metastatic non– advanced 6.44 (95% CI, 2.13-10.76, p=0.006; d= 0.44, ICC=0.024) ☐ support, and small-cell lung cancer, home services cancer; European QUAL-E scale -Routine aged 18 years or Cooperative adjusted difference between change scores telephone older, had stage Oncology at 4 months contact from a IV cancer (for Group 3.51 (95% CI, 1.33-5.68, p=0.003;d=0.45, ICC=0.015) palliative breast or performance care nurse prostate cancer, status of 0–2, Symptom severity: -Outpatient refractory to and a clinical ESAS palliative care hormonal prognosis 4 month follow-up therapy was an of 6–24 -4.41 (95% CI, -8.76 to -0.06, p=0.05; d=-0.31, -A 24-h on-call additional months ICC=0.034) service for criterion; telephone patients with Satisfaction: management of stage III cancer FAMCARE-P16 urgent issues and 4 month poor clinical 6.00 (95% CI, 3.94-8.05, p<0.0001;d=0.73, ICC=-0.018) Control group prognosis was (233) included at the Sensitivity analysis - Last value forward -No formal discretion of Adjusted difference between change scores intervention the oncologist); -Palliative care if an estimated FACIT-Sp requested survival of 6–24 4 months 4.34 (95% CI, 0.70-7.89, p=0.02) months (assessed by QUAL-E at 4 months Princess Margaret their main 2.75 (95% CI, 0.56-4.95, p=0.02) Cancer Centre oncologist); and (Toronto, ON, Eastern FAMCARE-P16 at 4 mo Canada) Cooperative 5.59 (95% CI, 3.65-7.52, p<0.0001) Oncology Group (ECOG) There was no statistically significant between-group performance difference in ESAS at 4 months status of 0, 1, or 2 (assessed by Sensitivity analysis using complete cases and multiple their main imputation were not presented here oncologist). ©Institute for Clinical and Economic Review, 2016 Page 118 Exclusion criteria: Insufficient English literacy to complete questionnaires and inability to pass the cognitive screening test (Short Orientation- Memory- Concentration Test score <20 or >10 errors). ©Institute for Clinical and Economic Review, 2016 Page 119 A9. Key Informant Interviews Affiliation Research University of Alabama at Birmingham Center to Advance Palliative Care Harvard Pilgrim Health Care Institute University Health Network Mt. Sinai Hospital University of California San Francisco Palliative Care Commonwealth Care Alliance Programs Beth Israel Deaconess Medical Center Coalition for Compassionate Care of California Massachusetts General Hospital The Conversation Project Payers Blue Cross Blue Shield of Massachusetts Aetna Professional Hospice and Palliative Nurses Association Societies Other MedEthics Consulting To develop a list of potential interviewees, we reviewed the policy literature and identified the key groups of stakeholders relevant to palliative care. Within each group, we relied on input from the New England CEPAC Advisory Board to identify key individuals and organizations to interview as part of our research process. When conducting interviews with initial contacts, we sought recommendations for additional regional and national experts to include as part of our assessment. We conducted 17 30-minute telephone interviews using a semi- structured guide. We attempted contact with a range of stakeholders within each New England state, though due to time limitations and scheduling challenges, were unable to interview all relevant stakeholders. To help ensure that key barriers and solutions were not left out of our assessment, we performed a scan of the existing policy literature. ©Institute for Clinical and Economic Review, 2016 Page 120 A10. Policy Roundtable Participants Policy Roundtable Participants Marie A. Bakitas, DNSc, NP-C, FAAN Professor, Marie O’Koren Endowed Chair School of Nursing/Department of Medicine, University of Alabama at Birmingham Associate Director, Center for Palliative and Supportive Care Lori-Jane Higgins Caregiver, Data Collector for Palliative Care Research Joseph Kozachek, MD Senior Medical Director, Custom Care Unit, Aetna Diane E. Meier, MD Vice-Chair for Public Policy, Professor of Geriatrics and Palliative Medicine, Catherine Gaisman Professor of Medical Ethics, Icahn School of Medicine at Mount Sinai Hospital Director, Center to Advance Palliative Care Rob Zavoski, MD, MPH Medical Director, Connecticut Department of Social Services 1. National Consensus Project for Quality Palliative Care. Clinical Practice Guidelines for Quality Palliative Care. Third ed. Pittsburgh, PA: NCP. 2013. 2. Kelley AS, Morrison RS. Palliative Care for the Seriously Ill. The New England journal of medicine. 2015;373(8):747-755. 3. Gillick MR. Rethinking the central dogma of palliative care. Journal of palliative medicine. 2005;8(5):909-913. 4. Meier DE, McCormick E. Benefits, services, and models of subspecialty palliative care. UpToDate; Dec. 2015. 5. De Lima L, Bennett MI, Murray SA, et al. International Association for Hospice and Palliative Care (IAHPC) List of Essential Practices in Palliative Care. Journal of pain & palliative care pharmacotherapy. 2012;26(2):118-122. 6. Gomes B, Calanzani N, Curiale V, McCrone P, Higginson IJ. Effectiveness and cost-effectiveness of home palliative care services for adults with advanced illness and their caregivers. The Cochrane database of systematic reviews. 2013;6:Cd007760. 7. Rabow MW, Dahlin C, Calton B, Bischoff K, Ritchie C. New Frontiers in Outpatient Palliative Care for Patients With Cancer. Cancer control : journal of the Moffitt Cancer Center. 2015;22(4):465-474. 8. Quill TE, Abernethy AP. Generalist plus specialist palliative care--creating a more sustainable model. The New England journal of medicine. 2013;368(13):1173-1175. 9. Traue DC, Ross JR. Palliative care in non-malignant diseases. Journal of the Royal Society of Medicine. 2005;98(11):503-506. 10. Bakitas M, Lyons KD, Hegel MT, et al. Effects of a palliative care intervention on clinical outcomes in patients with advanced cancer: the Project ENABLE II randomized controlled trial. Jama. 2009;302(7):741-749. 11. Bakitas MA, Tosteson TD, Lyons KD, et al. Early versus delayed initiation of concurrent palliative oncology care: Patient outcomes in the ENABLE III randomized controlled trial. Journal of clinical oncology. 2015;33(13 // (ACS) *National Institute of Nursing Research* // (NIH) *National Institute of Nursing Research* // (NINR) *National Institute of Nursing Research*):1438-1445. ©Institute for Clinical and Economic Review, 2016 Page 121 12. Brumley R, Enguidanos S, Jamison P, et al. Increased satisfaction with care and lower costs: results of a randomized trial of in-home palliative care. Journal of the American Geriatrics Society. 2007;55(7):993-1000. 13. Brumley RD, Enguidanos S, Cherin DA. Effectiveness of a home-based palliative care program for end-of-life. Journal of palliative medicine. 2003;6(5):715-724. 14. Cheville AL, Girardi J, Clark MM, et al. Therapeutic exercise during outpatient radiation therapy for advanced cancer: Feasibility and impact on physical well-being. American journal of physical medicine & rehabilitation. 2010;89(8):611-619. 15. Dionne-Odom JN, Azuero A, Lyons KD, et al. Benefits of early versus delayed palliative care to informal family caregivers of patients with advanced cancer: Outcomes from the ENABLE III randomized controlled trial. Journal of clinical oncology. 2015;33(13):1446-1452. 16. Greer JA, McMahon PM, Tramontano A, et al. Effect of early palliative care on health care costs in patients with metastatic NSCLC. Journal of clinical oncology. 2012;30(15 SUPPL. 1):CONFERENCE START: 2012 Jun 2011 CONFERENCE END: 2012 Jun 2015. 17. Kerr CW, Donohue KA, Tangeman JC, et al. Cost savings and enhanced hospice enrollment with a home-based palliative care program implemented as a hospice-private payer partnership. Journal of palliative medicine. 2014;17(12):1328-1335. 18. Radwany SM, Hazelett SE, Allen KR, et al. Results of the promoting effective advance care planning for elders (PEACE) randomized pilot study. Popul Health Manag. 2014;17(2):106-111. 19. Scheffey C, Kestenbaum MG, Wachterman MW, et al. Clinic-based outpatient palliative care before hospice is associated with longer hospice length of service. J Pain Symptom Manage. 2014;48(4):532-539. 20. Seow H, Brazil K, Sussman J, et al. Impact of community based, specialist palliative care teams on hospitalisations and emergency department visits late in life and hospital deaths: a pooled analysis. Bmj. 2014;348:g3496. 21. Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients with metastatic non-small-cell lung cancer. New England Journal of Medicine. 2010;363(8):733-742. 22. Zimmermann C, Swami N, Krzyzanowska M, et al. Early palliative care for patients with advanced cancer: a cluster-randomised controlled trial. Lancet (London, England). 2014;383(9930):1721-1730. 23. Ciemins EL, Stuart B, Gerber R, Newman J, Bauman M. An Evaluation of the Advanced Illness Management (AIM) Program: Increasing hospice utilization in the San Francisco Bay area. Journal of palliative medicine. 2006;9(6):1401-1411. 24. Rabow MW, Dibble SL, Pantilat SZ, McPhee SJ. The comprehensive care team: a controlled trial of outpatient palliative medicine consultation. Arch Intern Med. 2004;164(1):83-91. 25. Rummans TA, Clark MM, Sloan JA, et al. Impacting quality of life for patients with advanced cancer with a structured multidisciplinary intervention: a randomized controlled trial. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2006;24(4):635-642. 26. Ranganathan A, Dougherty M, Waite D, Casarett D. Can palliative home care reduce 30-day readmissions? Results of a propensity score matched cohort study. Journal of palliative medicine. 2013;16(10):1290-1293. 27. Cohen SR, Boston P, Mount BM, Porterfield P. Changes in quality of life following admission to palliative care units. Palliative medicine. 2001;15(5):363-371. 28. Hughes MT, Smith TJ. The growth of palliative care in the United States. Annual review of public health. 2014;35:459-475. 29. Kerr K, Beresford L. Next Generation of Palliative Care: Community Models Offer Services Outside the Hospital. California Health Care Foundation. November 2012. http://www.chcf.org/publications/2012/11/next- generation-palliative-care. 30. Rangachari D, Smith TJ. Integrating palliative care in oncology: the oncologist as a primary palliative care provider. Cancer journal (Sudbury, Mass.). 2013;19(5):373-378. 31. Medicaid.gov. Fee-for-Service. https://www.medicaid.gov/medicaid-chip-program-information/by- topics/delivery-systems/fee-for-service.html. 32. Meier DE. Increased access to palliative care and hospice services: opportunities to improve value in health care. The Milbank quarterly. 2011;89(3):343-380. ©Institute for Clinical and Economic Review, 2016 Page 122 33. Arias E. United States Life Tables, 2011. Center for Disease Control. National Vital Statistics Reports. September 22, 2015;64(11). http://www.cdc.gov/nchs/data/nvsr/nvsr64/nvsr64_11.pdf. Accessed January 28, 2016. 34. The Dartmouth Atlas of Health Care. Care of Chronic Illness in the Last Two Years of Life. The Dartmouth Institute for Health Policy and Clinical Practice2005. 35. Morrison RS. Models of palliative care delivery in the United States. Current opinion in supportive and palliative care. 2013;7(2):201-206. 36. Gelfman LP, Kalman J, Goldstein NE. Engaging heart failure clinicians to increase palliative care referrals: overcoming barriers, improving techniques. Journal of palliative medicine. 2014;17(7):753-760. 37. Loscalzo MJ. Palliative care: an historical perspective. Hematology / the Education Program of the American Society of Hematology. American Society of Hematology. Education Program. 2008:465. 38. Parikh RB, Kirch RA, Smith TJ, Temel JS. Early specialty palliative care--translating data in oncology into practice. The New England journal of medicine. 2013;369(24):2347-2351. 39. American Academy for Hospice and Palliative Medicine. ABMS subspecialty certification in hospice and palliative medicine. 2014; http://aahpm.org/certification/subspecialty-certification. 40. Barbour LT, Cohen SE, Jackson V, et al. Models for Palliative Care Outside the Hospital Setting. The IPAL‐OP Project, Center to Advance Palliative Care. 2012. 41. Lupu D. Estimate of current hospice and palliative medicine physician workforce shortage. J Pain Symptom Manage. 2010;40(6):899-911. 42. Center to Advance Palliative Care. America's Care of Serious Illness. 2015 State-By-State Report Card on Access to Palliative Care in Our Nation's Hospitals. https://reportcard.capc.org/. Accessed February 1, 2016. 43. Johnson KS. Racial and ethnic disparities in palliative care. Journal of palliative medicine. 2013;16(11):1329-1334. 44. Aldridge MD, Kelley AS. The Myth Regarding the High Cost of End-of-Life Care. American journal of public health. 2015;105(12):2411-2415. 45. Center to Advance Palliative Care. Palliative Care Billing. 2016; https://www.capc.org/topics/palliative-care- billing/. Accessed March 8, 2016. 46. Meier D, Kelley A. Expanding Access To Palliative Care Services: The Tide Is Turning. Health Affairs Blog. August 6, 2015. 47. Institute of Medicine. Dying in America: Improving Quality and Honoring Individual Preferences Near the End of Life. September 17, 2015: http://iom.nationalacademies.org/Reports/2014/Dying-In-America-Improving- Quality-and-Honoring-Individual-Preferences-Near-the-End-of-Life.aspx. 48. Meier DE. The development, status, and future of palliative care. Palliative care: Transforming the care of serious illness,. 2010:18-19. 49. American Cancer Society Cancer Action Network. How do you measure up? A Progress Report on State Legislative Activity to Reduce Cancer Incidence and Mortality. 13th ed;2015. 50. Kirch RA. Building A Quality Of Life National Movement: Igniting Advocacy To Integrate Palliative Care In Our US Health System. 2015(March 30, 2015). http://healthaffairs.org/blog/2015/03/30/building-a-quality-of-life- national-movement-igniting-advocacy-to-integrate-palliative-care-in-our-us-health-system/. 51. Brumley RD, Hillary K. The TriCentral palliative care program toolkit. Downey CA: Kaiser Permanente. 2002;11:03-07. 52. Commonwealth Care Alliance. A Look Back At 2013. http://www.commonwealthcarealliance.org/wp- content/uploads/2013/04/Annual-Report-2013.pdf. 53. Complex Care Innovation Lab. Profiles in Innovation: Commonwealth Care Alliance, Massachusetts. Center for Health Strategies, Inc. May 2015. http://www.chcs.org/media/Commonwealth_Care_Alliance_Profiles-in- Innovation_November-2015.pdf. Accessed February 1, 2016. 54. Cook DJ, Mulrow CD, Haynes RB. Systematic reviews: synthesis of best evidence for clinical decisions. Ann Intern Med. 1997;126(5):376-380. 55. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta- analyses: the PRISMA statement. Int J Surg. 2010;8(5):336-341. 56. U.S. Preventive Services Task Force. Procedure Manual. Agency for Healthcare Research and Quality;2008. ©Institute for Clinical and Economic Review, 2016 Page 123 57. Currow DC, Plummer JL, Kutner JS, Samsa GP, Abernethy AP. Analyzing phase III studies in hospice/palliative care. a solution that sits between intention-to-treat and per protocol analyses: the palliative-modified ITT analysis. J Pain Symptom Manage. 2012;44(4):595-603. 58. Ollendorf DA, Pearson SD. An integrated evidence rating to frame comparative effectiveness assessments for decision makers. Med Care. 2010;48(6 Suppl):S145-152. 59. Aiken LS, Butner J, Lockhart CA, Volk-Craft BE, Hamilton G, Williams FG. Outcome evaluation of a randomized trial of the PhoenixCare intervention: program of case management and coordinated care for the seriously chronically ill. Journal of palliative medicine. 2006;9(1):111-126. 60. Keefe FJ, Ahles TA, Sutton L, et al. Partner-guided cancer pain management at the end of life: a preliminary study. Journal of pain and symptom management. 2005;29(3):263-272. 61. Lindell KO, Olshansky E, Song MK, et al. Impact of a disease-management program on symptom burden and health-related quality of life in patients with idiopathic pulmonary fibrosis and their care partners. Heart & lung : the journal of critical care. 2010;39(4):304-313. 62. McMillan SC, Small BJ, Weitzner M, et al. Impact of coping skills intervention with family caregivers of hospice patients with cancer: a randomized clinical trial. Cancer. 2006;106(1):214-222. 63. Meyers FJ, Linder J, Beckett L, Christensen S, Blais J, Gandara DR. Simultaneous care: a model approach to the perceived conflict between investigational therapy and palliative care. J Pain Symptom Manage. 2004;28(6):548- 556. 64. Strasser F, Sweeney C, Willey J, Benisch-Tolley S, Palmer JL, Bruera E. Impact of a half-day multidisciplinary symptom control and palliative care outpatient clinic in a comprehensive cancer center on recommendations, symptom intensity, and patient satisfaction: a retrospective descriptive study. J Pain Symptom Manage. 2004;27(6):481-491. 65. Bakitas M, Lyons KD, Hegel MT, et al. The project ENABLE II randomized controlled trial to improve palliative care for rural patients with advanced cancer: baseline findings, methodological challenges, and solutions. Palliative & supportive care. 2009;7(1):75-86. 66. Lukas L, Foltz C, Paxton H. Hospital outcomes for a home-based palliative medicine consulting service. Journal of palliative medicine. 2013;16(2):179-184. 67. Rhee Y, Degenholtz HB, Lo Sasso AT, Emanuel LL. Estimating the quantity and economic value of family caregiving for community-dwelling older persons in the last year of life. J Am Geriatr Soc. 2009;57(9):1654-1659. 68. Yu M, Guerriere DN, Coyte PC. Societal costs of home and hospital end-of-life care for palliative care patients in Ontario, Canada. Health & social care in the community. 2015;23(6):605-618. 69. Chai H, Guerriere DN, Zagorski B, Coyte PC. The magnitude, share and determinants of unpaid care costs for home-based palliative care service provision in Toronto, Canada. Health & social care in the community. 2014;22(1):30-39. 70. Dumont S, Jacobs P, Fassbender K, Anderson D, Turcotte V, Harel F. Costs associated with resource utilization during the palliative phase of care: a Canadian perspective. Palliative medicine. 2009;23(8):708-717. 71. Colorado Department of Health Care Policy & Financing. ACC Incentive Payments Fact Sheet. 2015; https://www.colorado.gov/pacific/sites/default/files/KPI%20Incentive%20Fact%20Sheet.pdf. ©Institute for Clinical and Economic Review, 2016 Page 124