Context: Researchers, hospices, and government agencies administer standardized questionnaires to caregivers for assessing end-of-life care quality. Caregiving experiences may influence end-of-life care quality reports, which have implications for caregiver outcomes, and are a clinical and policy priority.
Objectives: This study aims to determine whether and how caregivers' end-of-life care assessments depend on their burden and benefit perceptions.
Methods: This study analyzes data from 391 caregivers in the 2011 National Study of Caregiving and their Medicare beneficiary care recipients from the 2011–2016 National Health and Aging Trends Study. Caregivers assessed five end-of-life care aspects for decedents. Logistic regression was used and predicted probabilities of caregivers positively or negatively assessing end-of-life care based on their burden and benefit experiences calculated. Analyses adjusted for caregiver and care recipient demographic and health characteristics.
Results: No or minimal caregiving burden is associated with =0.70 probability of caregivers reporting they were always informed about the recipient's condition and that the dying person's care needs were always met, regardless of perceived benefits. High perceived caregiving benefit is associated with =0.80 probability of giving such reports, even when perceiving high burden.
Conclusion: Caregiver burden and benefit operate alongside one another regarding two end-of-life care evaluations, even when years elapse between caregiver experience reports and care recipient death. This suggests that caregiver interventions reducing burden and bolstering benefits may have a positive and lasting impact on end-of-life care assessments.
Objectives: To map current practice regarding discussions around resuscitation across England and Scotland in patients with cancer admitted acutely to hospital and to demonstrate the value of medical students in rapidly collecting national audit data.
Methods: Collaborators from the Macmillan medical student network collected data from 251 patient encounters across eight hospitals in England and Scotland. Data were collected to identify whether discussion regarding resuscitation was documented as having taken place during inpatient admission to acute oncology. As an audit standard, it was expected that all patients should be invited to discuss resuscitation within 24 hr of admission.
Results: Resuscitation discussions were had in 43.1% of admissions and of these 64.0% were within 24 hr; 27.6% of all admissions. 6.5% of patients had a “do not attempt resuscitation” order prior to admission with a difference noted between patients receiving palliative and curative treatment (8.5% and 0.39%, respectively, p < .05). Discussions regarding escalation of care took place in only 29.3% of admissions.
Conclusions: These data highlight deficiencies in the number of discussions regarding resuscitation that are being conducted with cancer patients that become acutely unwell. It also demonstrates the value of medical student collaboration in rapidly collecting national audit data.
Background: Data quality is fundamental to the integrity of quantitative research. The role of external researchers in data quality assessment (DQA) remains ill-defined in the context of secondary use for research of large, centrally curated health datasets. In order to investigate equity of palliative care provided to Indigenous Australian patients, researchers accessed a now-historical version of a national palliative care dataset developed primarily for the purpose of continuous quality improvement.
Objectives: (i) To apply a generic DQA framework to the dataset and (ii) to report the process and results of this assessment and examine the consequences for conducting the research.
Method: The data were systematically examined for completeness, consistency and credibility. Data quality issues relevant to the Indigenous identifier and framing of research questions were of particular interest.
Results: The dataset comprised 477,518 records of 144,951 patients (Indigenous N = 1515; missing Indigenous identifier N = 4998) collected from participating specialist palliative care services during a period (1 January 2010–30 June 2015) in which data-checking systems underwent substantial upgrades. Progressive improvement in completeness of data over the study period was evident. The data were error-free with respect to many credibility and consistency checks, with anomalies detected reported to data managers. As the proportion of missing values remained substantial for some clinical care variables, multiple imputation procedures were used in subsequent analyses.
Conclusion and implications: In secondary use of large curated datasets, DQA by external researchers may both influence proposed analytical methods and contribute to improvement of data curation processes through feedback to data managers.
OBJECTIVES: We investigated trends in end-of-life hospitalizations among nursing home residents (NHR) over 10 years and looked at differences between age groups and sexes as well as the length of terminal hospital stays.
STUDY DESIGN: Retrospective cohort study based on health insurance claims data of the AOK Bremen/Bremerhaven. All NHR aged 65 years or more who died between 2006 and 2015 were included.
MAIN OUTCOME MEASURES: We assessed the proportions of decedents who were in hospital on the day of death and during the last 3, 7, 14 and 30 days of life, stratified by two-year periods. Multiple logistic regressions were conducted to study changes over time, adjusting for covariates.
RESULTS: A total of 10,781 decedents were included (mean age 86.1 years, 72.1 % females). Overall, 29.2 % died in hospital, with a slight decrease from 30.3 % in 2006-2007 to 28.3 % in 2014-2015 (OR 0.86; 95 % CI 0.75-0.98). Of the 3150 terminal hospitalizations, 35.5 % lasted up to 3 days and the mean length of stay decreased from 9.0 (2006-2007) to 7.5 days (2014-2015). When looking at the last 7, 14 and 30 days of life, no changes over time were found. Male sex and younger age were associated with a higher chance of end-of-life hospitalization in almost all analyses.
CONCLUSIONS: End-of-life hospitalizations of NHR are common in Germany. There has been a small decrease during recent years in the proportion of in-hospital deaths, but not of hospitalizations during the last 7, 14 and 30 days of life. This might be explained by shorter durations of hospital stays.
BACKGROUND: There are no processes that routinely assess end-of-life care in Australian general practice. This study aimed to develop a data collection process which could collect observational data on end-of-life care from Australian general practitioners (GPs) via a questionnaire and clinical data from general practice software.
METHODS: The data collection process was developed based on a modified Delphi study, then pilot tested with GPs through online surveys across three Australian states and data extraction from general practice software, and finally evaluated through participant interviews.
RESULTS: The developed data collection process consisted of three questionnaires: Basic Practice Descriptors (32 items), Clinical Data Query (32 items) and GP-completed Questionnaire (21 items). Data extraction from general practice software was performed for 97 decedents of 10 GPs and gathered data on prescriptions, investigations and referral patterns. Reports on care of 272 decedents were provided by 63 GPs. The GP-completed Questionnaire achieved a satisfactory level of validity and reliability. Our interviews with 23 participating GPs demonstrated the feasibility and acceptability of this data collection process in Australian general practice.
CONCLUSIONS: The data collection process developed and tested in this study is feasible and acceptable for Australian GPs, and comprehensively covers the major components of end-of-life care. Future studies could develop an automated data extraction tool to reduce the time and recall burden for GPs. These findings will help build a nationwide integrated information network for primary end-of-life care in Australia.
Background: The location of death is an important component of end-of-life care. However, contemporary trends in the location of death for cardiovascular deaths related to heart failure (CV-HF) and comparison to cancer deaths have not been fully examined.
Methods: We analyzed data from the Centers for Disease Control and Prevention’s Control Wide-Ranging Online Data for Epidemiologic Research database between 2003 and 2017 to identify location of death for CV-HF and cancer deaths. The proportions of deaths that occurred in a hospice facility, home, and medical facility were tested for trends using linear regression. Odds ratios were calculated to determine the odds of death occurring in a hospice facility or home (versus a medical facility) stratified by sex and race.
Results: We identified 2 940 920 CV-HF and 8 852 066 cancer deaths. Increases were noted in the proportion of CV-HF deaths in hospice facilities (0.2% to 8.2%; Ptrend<0.001) and at home (20.6% to 30.7%; Ptrend<0.001), whereas decreases were noted in the proportion of deaths in medical facilities (44.5% to 31.0%; Ptrend<0.001) and nursing homes (30.8% to 25.7%; Ptrend<0.001). The odds of dying in a hospice facility (odds ratio, 1.79 [1.75–1.82]) or at home (odds ratio, 1.55 [1.53–1.56]) versus a medical facility was higher for whites versus blacks. The rate of increase in proportion of deaths in hospice facilities was higher for cancer deaths (ß=1.05 [95% CI, 0.97–1.12]) than for CV-HF deaths (ß=0.61 [95% CI, 0.58–0.64]).
Conclusions: The proportion of CV-HF deaths occurring in hospice facilities is increasing but remains low. Disparities are noted whereby whites are more likely to die in hospice facilities or at home versus medical facilities compared with blacks. More research is needed to determine end-of-life preferences for patients with HF and identify the basis for these differences in location of death.
BACKGROUND: Global annual deaths are rising. It is essential to examine where future deaths may occur to facilitate decisions regarding future service provision and resource allocation.
AIMS: To project where people will die from 2017 to 2040 in an ageing country with advanced integrated palliative care, and to prioritise recommendations based on these trends.
METHODS: Population-based trend analysis of place of death for people that died in Scotland (2004-2016) and projections using simple linear modelling (2017-2040); Transparent Expert Consultation to prioritise recommendations in response to projections.
RESULTS: Deaths are projected to increase by 15.9% from 56,728 in 2016 (32.8% aged 85+ years) to 65,757 deaths in 2040 (45% aged 85+ years). Between 2004 and 2016, proportions of home and care home deaths increased (19.8-23.4% and 14.5-18.8%), while the proportion of hospital deaths declined (58.0-50.1%). If current trends continue, the numbers of deaths at home and in care homes will increase, and two-thirds will die outside hospital by 2040. To sustain current trends, priorities include: 1) to increase and upskill a community health and social care workforce through education, training and valuing of care work; 2) to build community care capacity through informal carer support and community engagement; 3) to stimulate a realistic public debate on death, dying and sustainable funding.
CONCLUSION: To sustain current trends, health and social care provision in the community needs to grow to support nearly 60% more people at the end-of-life by 2040; otherwise hospital deaths will increase.
La recherche en santé autochtone au Canada a été négligée dans le passé et qualifiée de problématique, notamment en raison du manque de collaboration avec les peuples autochtones. L'Énoncé de politique des trois Conseils sur l'éthique de la recherche avec des êtres humains décrit au chapitre 9 la conduite éthique de la recherche axée sur les Premières nations, les Inuits et les Métis. Les principes PCAP® des Premières nations (propriété, contrôle, accès et possession) soulignent l'importance majeure de l'engagement et de la gouvernance autochtones. En vue d'assurer que les buts et les activités de la recherche développée soient réalisés en partenariat complet et significatif avec les peuples et les communautés autochtones, il est possible de faire appel à des méthodes de recherche participative communautaire (RPC) intégrant leur plein engagement. Les recherches utilisant des ensembles de données secondaires, telles que les données administratives sur la santé recueillies en routine, ne devraient plus être exclues de cette approche. Notre objectif était de décrire comment notre équipe de chercheurs universitaires, alliée à un organisme national de santé autochtone, a adapté les méthodes de RPC dans le cadre d'un projet de recherche utilisant des données recueillies antérieurement pour examiner les lacunes dans la prestation de soins de fin de vie aux peuples autochtones en Ontario. Nous décrivons le processus d'élaboration de ce partenariat de recherche et expliquons comment l'intégration des principes de base et des processus de formation du savoir autochtones ont guidé cette collaboration. Notre partenariat de recherche, qui implique l'adaptation de méthodes de RPC, illustre un processus d'engagement qui pourrait guider d'autres chercheurs désirant mener des recherches en santé autochtone à l'aide de données déjà recueillies. Nous faisons aussi état d'une entente de recherche transparente, négociée équitablement entre un organisme national de santé autochtone et des chercheurs, qui pourrait servir de cadre pour des collaborations de recherche similaires. Il est essentiel de s'assurer que les perspectives autochtones soient au cœur des processus de recherche et qu'elles soient reflétées dans ceux-ci lorsque des données administratives sur la santé sont utilisées.
PURPOSE: Clinical data warehouses (cDWHs) and cancer registry databases have enabled researchers to conduct clinical analytics with structured electronic health record data. However, these secondary electronic health record sources are often limited in scope because they do not capture the clinical information needed to understand complex clinical questions. Thus, we evaluated the effect of additional curation of data.
MATERIALS AND METHODS: Clinical data sets of 149 patients with prostate cancer with biochemical recurrence after radical prostatectomy treated with salvage or palliative radiotherapy between 2008 and 2017 from our institutional cDWH and Gießener Tumor Documentation System (GTDS) were linked (data warehouse [DWH] population) for analyzing treatment outcomes. The linked data sets were manually curated (manual postprocessing [MPP], eg, incorporate data from established urologists). The primary outcomes were the impact on data quality of treatment outcomes and the time spent on data curation.
RESULTS: We obtained significantly more information on disease progression and patient survival (nonsignificant) when using curated data; the biochemical progression-free survival rate at 5 years for the DWH and DWH plus MPP populations was 63% v 30% (P = .001) and the overall survival rate was 84% v 81% (P = .479), respectively. The median deviation of completeness and the median concordance of clinical data values were 21.47% (range, 55.38%-100%) and 95.00% (range, 63.40%-100%), respectively. We spent 121 hours, 42 minutes on data curation, with most time required for laboratory values, accounting, for a total of 45 hours, 20 minutes (37.26%).
CONCLUSION: Our analysis indicates that time-to-event outcomes for patients with prostate cancer cannot be extracted using secondary data sources (cDWH plus GTDS) only. Outcomes data differed between the electronic data (DWH) and the second manual extraction (DWH plus MPP) because of a lack of follow-up data. When using such unique database resources, only baseline characteristics can reliably be extracted.
The number of children in the UK with life-limiting conditions and the demand for home-based palliative care is increasing. Children's hospices remain a dominant provider of palliative care. This study aimed to determine the approaches taken by children's hospices across the UK in meeting the planned and unplanned health needs of children and their families who receive palliative care at home. In addition, the survey aimed to identify the professional composition of community teams and the number of children and families supported by each service. An internet-based questionnaire survey was sent to all children's hospices in the UK, comprising ten questions exploring the size of the team, geographical areas covered, workforce composition, services offered and approaches to managing unplanned, out of hours care. Responses were received from 14 (26%) of the hospices. A total of 1,618 children and their families were being cared for by these hospices, of whom 825 received care at home. Registered nurses constituted the greatest proportion of staff and were employed by all teams. Care provided at home was broadly split into two categories: planned short breaks and responsive palliative nursing. The latter comprised advance care planning, anticipatory prescribing and active symptom control. Out of hours care was usually offered in the form of telephone support. Models of community-based care are evolving to include nurses practising at specialist and advanced levels, allowing more children with increasingly complex conditions to be cared for at home.
OBJECTIVES: In the Netherlands, the use of continuous deep sedation at the end of life has sharply increased from 8.2% of all deaths in 2005 to 12.3% in 2010 to 18.3 % in 2015. We describe its clinical characteristics in 2015 and compare it with 2010 and 2005.
DESIGN: Questionnaire study in random samples of death reported to a central death registry.
SETTING AND PARTICIPANTS: A nationwide study in the Netherlands among physicians attending reported deaths.
METHODS: Continuous deep sedation characteristics (patient characteristics, drugs, duration, estimated shortening of life, and palliative consultation) from the Netherlands in 2015 were compared with continuous deep sedation characteristics of 2010 and 2005.
RESULTS: The response rate was 78% (n = 7277) in 2015, 74% (n = 6263) in 2010, and 78% (n = 6860) in 2005. The increased frequency of continuous deep sedation was notable in all patient subgroups, but mainly occurred among deaths attended by general practitioners, particularly in patients older than 80 years and patients with cancer. In 2015, continuous deep sedation was performed in 93% of the patients through administration of benzodiazepines. In 3% of the patients, the sedation lasted more than 1 week. Furthermore, 60% of the physicians reported that they had no intention to hasten death, 38% reported that they have taken hastening of death into account, and 2% reported their intention was to hasten death. For 1 in 5 patients, a palliative care expert was consulted prior to the start of sedation. These characteristics were comparable between 2015 and 2010.
CONCLUSIONS AND IMPLICATIONS: The increase in continuous deep sedation mainly occurred in deaths attended by general practitioners, especially in older patients and patients with cancer. As there are no major shifts in demographic and epidemiologic patterns of dying, future studies should investigate possible explanations for the increase predominantly in societal developments, such as increased attention to sedation in education and society, a broader interpretation of the concept of refractoriness, and an increased need of patients and physicians to control the dying process.
The present doctoral thesis focuses on evaluating the use, trends in use, factors that influence use, and the impact of using policy measures to support patients and informal caregivers to provide palliative care in the home or community setting in Belgium. Using full-population administrative databases containing information on all decedents in Belgium between 2010 and 2015, we provided insights into these patterns that allow formulating points of improvement, identification of focal groups for policy, but also allow evaluating the societal impact of the measures in tems of quality and costs of care.
[Extrait résumé éditeur]
Background: Palliative care program service delivery is variable, and programs often lack data to support and guide program development and growth.
Objective: To review the development and key features of the National Palliative Care Registry™ ("the Registry") and describe recent findings from its surveys on hospital palliative care.
Description: Established in 2008, the Registry data elements align with National Consensus Project (NCP) guidelines related to palliative care program structures and operations. The Registry provides longitudinal and comparative data that palliative care programs can use to support programmatic growth.
Results: As of 2018, >1000 hospitals and 120 community sites have submitted data on their palliative care programs to the Registry. Over the past decade, the percentage of hospital admissions seen by palliative care teams (penetration) has increased from 2.5% to 5.3%. Higher penetration is correlated with teaching hospital status, having a palliative care trigger, and hospital size (p < 0.05). Although overall staffing has expanded, only 42% of Registry programs include the recommended four key disciplines: physician, advanced practice or other registered nurse, social worker, and chaplain. Compliance with NCP guidelines on key structures and processes vary across adult and pediatric programs.
Conclusions: The Registry allows palliative care programs to optimize core structures and processes and understand their performance relative to their peers.
Planners, actuaries, and others involved in forecasting capacity and costs must manipulate historical data. Data from calendar/financial year totals have been assumed to be adequate and reliable. This relies on the assumption that year-to-year differences do not arise from patterns concealed in the data. While the seasonal cycle is widely recognized, longer term patterns such as disease outbreaks will act to modify annual demand and costs. Monthly data relating to deaths in local government areas in England and Wales are used to demonstrate curious semipermanent bursts of high behavior. There is no seasonal pattern for the start of these events, and the sudden switch to high deaths can occur at any time, even in immediately adjacent areas. Higher deaths and related demand and costs endure for around 12 months before they suddenly revert to the former level where they stay until the next of these curious high events. In England and Wales (and many other countries), a period of unexplained higher deaths, reduced life expectancy, and health care and life insurance costs since 2011 appears to be coming to an end and looks to have arisen from a coincidence of these events at sub-national level.
Objective To examine the association of the use of hospice care on patient experience and outcomes of care. Promoting high-value, safe and effective care is an international healthcare imperative. However, the extent to which hospice care may improve the value of care is not well characterised.
Methods A secondary analysis of variations in care was conducted using the Dartmouth Atlas Report, matched to the American Hospital Association Annual Survey Database to abstract organisational characteristics for 236 US hospitals to examine the relationship between hospice usage and a number of variables that represent care value, including hospital care intensity index, hospital deaths, intensive care unit (ICU) deaths, patient satisfaction and a number of patient quality indicators. Structural equation modelling was used to demonstrate the effect of hospice use on patient experience, clinical and efficiency outcomes.
Results Hospice admissions in the last 6 months of life were correlated with a number of variables, including increases in patient satisfaction ratings (r=0.448, p=0.01) and better pain control (r=0.491, p=0.01), and reductions in hospital days (r=-0.517, p=0.01), fewer hospital deaths (r=-0.842, p=0.01) and fewer deaths occurring with an ICU admission during hospitalisation (r=-0.358, p=0.01). The structural equation model identified that use of hospice care was inversely related to hospital mortality (-0.885) and ICU mortality (-0.457).
Conclusions The results of this investigation demonstrate that greater use of hospice care during the last 6 months of life is associated with improved patient experience, including satisfaction and pain control, as well as clinical outcomes of care, including decreased ICU and hospital mortality..
BACKGROUND: To enhance the quality of hospice care and to facilitate consumers' choices, the Centers for Medicare and Medicaid Services (CMS) began the Hospice Quality Reporting Program, in which CMS posted the quality measures of participating hospices on its reporting website, Hospice Compare. Little is known about the participation rate and the types of nonparticipating hospices.
OBJECTIVE: To examine the factors associated with hospices' nonparticipation in Hospice Compare.
RESEARCH DESIGN: We analyzed data from the CMS 2016 Hospice Compare. "Nonparticipants" were those who did not submit any quality measure. With the data of the Provider of Service file, the Healthcare Cost Report Information System, and the Area Health Resources File, multivariate logistic regressions estimated the association between nonparticipants and hospice and market characteristics, including ownership, size, nurse staffing ratio, and market competition intensity.
RESULTS: Among the 4123 certified hospices subject to penalty from nonparticipation, 259 did not participate in Hospice Compare. California, New Mexico, Texas, and Wyoming had participation rates lower than 80%. Hospices that were for-profit, had no accreditation, had few nurses per patient day, provided no inpatient care, and were located in competitive markets were less likely to participate than other hospices.
CONCLUSIONS: Hospice Compare successfully motivated hospice in participating in the quality report program in most of states. For-profit hospices, hospices with less quality, and hospices located in competitive markets were less likely to participate. Further research is warranted to examine the quality of these nonparticipants, especially in the 4 states with a lower participation rate.
Specialized palliative care teams improve outcomes for the steadily growing population of people living with serious illness. However, few studies have examined whether the specialty palliative care workforce can meet the growing demand for its services. We used 2018 clinician survey data to model risk factors associated with palliative care clinicians leaving the field early, and we then projected physician numbers from 2019 to 2059 under four scenarios. Our modeling revealed an impending "workforce valley," with declining physician numbers that will not recover to the current level until 2045, absent policy change. However, sustained growth in the number of fellowship positions over ten years could reverse the worsening workforce shortage. There is an immediate need for policies that support high-value, team-based palliative care through expansion in all segments of the specialty palliative care workforce, combined with payment reform to encourage the deployment of sustainable teams.
BACKGROUND: Inadequate description of palliative care cancer patients in research studies often leads to results having limited generalizability. To standardize the description of the sample, the European Association for Palliative Care basic data set was developed, with 31 core demographic and disease-related variables.
AIM: To pilot test the data set to check acceptability, comprehensibility and feasibility.
DESIGN: International, multi-centre pilot study at nine study sites in five European countries, using mixed methods.
SETTING/PARTICIPANTS: Adult cancer patients and staff in palliative care units, hospices and home care.
RESULTS: In all, 191 patients (544 screened) and 190 health care personnel were included. Median time to fill in the patient form was 5 min and the health care personnel form was 7 min. Ethnicity was the most challenging item for patients and requires decisions at a national level about whether or how to include. Health care personnel found weight loss, principal diagnosis, additional diagnoses and stage of non-cancer diseases most difficult to respond to. Registration of diagnoses will be changed from International Statistical Classification of Diseases and Related Health Problems, 10th version code to a predefined list, while weight loss and stage of non-cancer diseases will be removed. The pilot study has led to rewording of items, improvement in response options and shortening of the data set to 29 items.
CONCLUSION: Pilot testing of the first version of the European Association for Palliative Care basic data set confirmed that patients and health care personnel understand the questions in a consistent manner and can answer within an acceptable timeframe. The pilot testing has led to improvement, and the new version is now subject to further testing.
OBJECTIVES: This study aims to identify heterogeneous depressive symptom trajectories among the widowed elderly in China, to explore predictive variables of latent class membership, and to detect between-classes differences in life satisfaction across time.
METHOD: Data of 198 individuals widowed between 2011 and 2013 were drawn from The China Health and Retirement Longitudinal Study (CHARLS), a nationally representative survey. Latent class growth analysis was employed to identify depressive symptom trajectories across 2011, 2013, and 2015, and a three-step auxiliary approach was applied to detect predictive variables of latent class membership. Mixed Analysis of Variances was followed to make between-class comparisons on life satisfaction across time.
RESULTS: Four grieving trajectories were identified: resilient (54.6%), chronic grief (23.7%), depressed-improved (11.6%), and chronic depression (10.1%). Older age, living exclusively with spouse before widowhood, and agricultural Hukou were significant predictors of depressed-improved, chronic grief, and chronic depression patterns, respectively. Life satisfaction in all groups except for the resilient one remained stable across time, and that of the chronic depression group was significantly lower than those of all the remaining groups.
CONCLUSION: Consistent depressive symptom trajectories during late-life widowhood exist across nations while the specific culture, values, and resources in the Chinese context may have contributed to a particularly high proportion of the chronic grief trajectory. More efforts should be made to identify patterns with predictors before support are provided, and interventions need to be tailored to target specific needs in each subgroup of the elderly during their transitions to widowhood.
As the Medicare program struggles to control expenditures, there is increased focus on opportunities to manage patient populations more efficiently and at a lower cost. A major source of expense for the Medicare program is beneficiaries at end of life. Estimates of the percentage of Medicare costs that arise from patients in the last year of life differ, ranging from 13% to 25%, depending on methods and assumptions. We analyze the most recently available Medicare Limited Data Set to update prior studies of end-of-life costs and examine different methods of performing this calculation. Based upon these findings, we conclude that higher estimates that take into account the spending over the 12 months leading up to death more accurately reflect the full cost of a patient's last year of life. Comparing current year costs of decedents with Medicare's current year costs understates the full budgetary impact of end-of-life patients. Because risk-taking entities such as Medicare Advantage plans and Accountable Care Organizations (ACOs) need to reduce costs while improving the quality of care, they should initiate programs to better manage the care of patients with serious or advanced illness. We also calculate costs for beneficiaries dying in different settings and conclude that more effective use of palliative care and hospice benefits offers a lower cost, higher quality alternative for patients at end of life.