Analysis Of End-Of-Life Care, Out-Of-Pocket Spending, And Place Of Death In 16 European Countries And Israel

Health care is delivered at the end of life to allow “all those with advanced, progressive, incurable illness to live as well as possible until they die.” ¹ This implies providing the right care at the right place, and in accordance with patients’ preferences. ^2,3 End-of-life care is a subset of long-term care and therefore is influenced by long-term care financing, organization, and provision. ⁴ It is projected that more than a quarter of the European population will be older than age sixty-five by 2050. ⁵ End-of-life care is therefore an urgent public health priority ⁵ in this context of an aging population, and given higher expectations for high-quality care from patients and their families. ⁶

Care delivery at the end of life involves a mix of public, private, and informal health and social care providers. Every European country has a system established to support older people who need care and assistance, ⁷ but the countries differ substantially in the organization, financing, and provision of long-term care—differences that have obvious repercussions on end-of-life care. Health care use and out-of-pocket spending in the last year of life and place of death depend largely on two main supply-side factors: financing mechanisms of end-of-life care and especially the mix of private and public providers and social support in the provision of informal end-of-life care. ^7–10

In approaching this subject, we formulated six hypotheses. We first hypothesized that in countries with predominantly public funding for long-term care, patients would rely less on hospital-based care and use more formal long-term care, such as home-based care funded by the health or social care system and nursing homes. Our second hypothesis was that in countries where long-term care is predominantly privately financed, patients would rely more on informal care and be more likely to use hospital-based care at the end of life.

We anticipated that end-of-life care funding mechanisms would have differential impacts on care pathways depending on patient age, cohabitation status, and primary diagnosis. Younger end-of-life patients are likely to receive higher amounts of care, compared to very old patients. ^11,12 Hence, we expected that younger patients would be exposed to more interventionist care with curative intent, compared to their older counterparts—who would be likely to receive more supportive and palliative care (our third hypothesis). We hypothesized that the effect of older patients receiving more palliative care would be stronger in predominantly publicly funded end-of-life care systems, because formal long-term care would be more widely available (our fourth hypothesis). We also hypothesized that people living alone would be more likely to die in a hospital because of the lack of support at home, especially in countries that rely mostly on private financing for end-of-life care (our fifth hypothesis).

Finally, since patients with chronic diseases experience a gradual decline in health, planning for an out-of-hospital death is more realistic for them than it is for people who die from an acute episode (such as cardiac arrest or stroke)—for whom death may be sudden. ¹³ We expected that in countries where public end-of-life care funding is predominant, end-of-life care pathways for chronically ill patients would be less hospital-based and more (nonhospital) institution-centric, compared to pathways for people in countries with predominantly privately funded end-of-life care systems (our sixth hypothesis).

The financing of end-of-life care also has an impact on households’ direct contributions to payment for the care their members receive. The amount of out-of-pocket spending depends on whether end-of-life care financing is predominantly private or public, the type of cost-sharing arrangement, the financial protection mechanisms in place (such as caps on patients’ out-of-pocket contributions), and the affordability of end-of-life care for households. We expected to find higher out-of-pocket spending where financing of end-of-life care is predominantly private, where protection mechanisms for patients’ contributions are weaker, and where end-of-life care is more affordable because of a favorable macroeconomic situation (such as high living standards and per capita gross domestic product).

There is limited population-based evidence on people’s end-of-life care needs and the circumstances in which they are dying. ^8,14 In this study we compared the experiences and outcomes of patients in their last year of life in sixteen European countries and Israel, and we discuss how these outcomes can be explained by funding and organizational differences and patient characteristics

Study Data And Methods

This study used data pooled from waves 2 (2006–07), 3 (2008–09), 4 (2011–12), and 5 (2013) of the Survey of Health, Ageing and Retirement in Europe (SHARE), which included an end-of-life care module. ^15–18 SHARE is a multidisciplinary, cross-national, and longitudinal survey, which contains individual-level data on the health, socioeconomic status, and social and family networks for more than 123,000 people ages fifty and older in twenty European countries and Israel. ¹⁹ The SHARE surveys included interviews concerning patients’ experiences with end-of-life care that were conducted with a proxy respondent in the case of a participant’s death and contain questions regarding the circumstances of the death. The proxy respondent could be a close relative, a household member, or any other person identified as part of the close social network of the deceased participant.

Data Sample

We analyzed data on 5,343 participants from seventeen countries for which the end-of-life SHARE module was available: Austria, Belgium, the Czech Republic, Denmark, Estonia, France, Germany, Greece, Ireland, Israel, Italy, the Netherlands, Poland, Slovenia, Spain, Sweden, and Switzerland. The data set contains information on participants’ place and cause of death, health care use, out-of-pocket spending (in euros) adjusted for purchasing power parity, and difficulties with activities of daily living (ADLs) during the last year of life. Proxy respondents provided information on the following types of care and related out-of-pocket spending: general practitioner care, specialist care, hospital stay, care in a nursing home, hospice stay, medications, aids and appliances, and formal home care or help due to disability. Reported difficulties with ADLs (which typically are eating, bathing, dressing, using the toilet, and transferring from bed or chair) were limited to difficulties that persisted for at least three months during the last year of life. Of the proxy respondents, 73.1 percent were either a spouse or a child of the deceased, and 85.2 percent had had regular (that is, more than once a week) contact with the deceased.

In cases where exact out-of-pocket spending was unknown or proxy respondents did not report exact values, the unfolding brackets technique was used to reduce nonresponse. This technique was used in all waves of this study except wave 2. Using this technique, respondents were asked a series of closed-ended value range questions, and the resulting amount was a categorical second-best option to the initial continuous financial variable of interest. ²⁰ The respondents were able to choose between three different country-specific bracket values—representing low, middle, and high value. This approach was used to estimate 6.05 percent of the spending data.

Analyses

Descriptive statistics of the main variable of interest were produced for each country. Out-of-pocket spending was inflated to 2013 euros, and purchasing power parity was adjusted using 2011 price-level indices for the overall health sector to account for differences in the relative price levels of health goods and services. ^21,22 In the case of countries that do not use euros, out-of-pocket spending was initially converted to local currency using SHARE exchange rates and was then adjusted for purchasing power parity using appropriate price-level indices. ^21,22

To analyze the data, we used Stata, version 13.

Response rates in the end-of-life module varied across interview questions and countries. The missing values were highest for the questions related to out-of-pocket spending (ranging from 18.06 percent for spending on general practitioners to 81.19 percent for spending on nursing homes) and those related to health care use (ranging from 2.71 percent for care received from general practitioners to 32.49 percent for care received in a hospice). Some missing out-of-pocket spending values have been filled in using the unfolding brackets technique. For other variables, since the proportion of missing data was low, we assumed that the data were missing at random.

A set of logistic regressions was used to analyze the association between place of death, individual circumstances in the last year of life, and end-of-life care system typology. Depending on the predominant source of funding for end-of-life care, countries were classified into two groups: one with an end-of-life care system that was predominantly publicly funded (Belgium, Denmark, Germany, the Netherlands, and Sweden) and one with a system that was predominantly privately funded (Austria, the Czech Republic, Estonia, France, Greece, Ireland, Israel, Italy, Poland, Slovenia, Spain, and Switzerland). ^7,10,23,24 We first estimated a series of models to describe the determinants of dying in a hospital, in a care home, and at home. Care home refers to an institution that cares for elderly people and includes nursing homes and hospices. Independent variables were age (older than eighty), sex, living alone at the time of death, underlying cause of death (cancer, cardiovascular disease, or other), a set of variables indicating the type of health care service used during the last year of life (general practitioner, specialist physician, hospital, nursing home, hospice, home care, medications, and aids and appliances), the number of hours of help necessary during a typical day, the number of limitations with ADLs, the duration of illness (less than six months, more than six months but less than one year, and more than one year), country group, and wave dummy variables to control for fixed cross-national group differences and secular trends.

For estimating of the probability of dying in hospital, we used the number of hospital stays in the last year of life (none at all, one or two, three to five, or more than five). We then ran six additional logistic regressions to assess the effect of the type of end-of-life care system on the likelihoods of dying in a hospital and dying in a care home depending on the terminal illness, age group, and cohabitation status of the deceased. In these models, the underlying cause of death, age group, and living alone were interacted with the country group.

Limitations

Our study had several limitations. First, respondents to SHARE were sampled from private households, so the institutionalized population was not included—which might have resulted in the selection of healthier people. ^25,26

Second, the majority of countries had more males in the sample than females, while we would have expected the opposite because women live longer than men. It is possible that more women are institutionalized because their husbands or partners die before them. It is also possible that men have worse health status and need more care, compared to women. Both hypotheses might be valid. However, since we did not have data on the institutionalized populations, we could not test the hypotheses.

Third, observed country differences must be interpreted cautiously, as the sample design did not adjust for the fact that northern countries in the survey have more nursing homes per capita than southern countries do. ^25,26 For example, in Sweden the majority of people do not die in a hospital, ²⁷ but most Swedish respondents did die in a hospital. This suggests that the in-hospital deaths in SHARE are overreported for Sweden—probably because institutionalized people were not included, as noted above.

Fourth, in Europe most nonsurgical cancer treatments are administered in both inpatient and outpatient hospital settings, ²⁸ unlike in the United States—where most such treatments occur in the outpatient setting. ²⁹

Fifth, SHARE did not distinguish between emergency or inpatient admissions and outpatient hospital admissions. This could be a limitation when making international comparisons of use of end-of-life care in the hospital.

Sixth, the findings from interviews in SHARE are based on the responses provided by proxy respondents, which could result in a loss of information and biased estimates.

Seventh, this study does not provide insights into other health conditions and comorbidities that could affect out-of-pocket spending and insurers’ incurred costs during the last year of life.

Eighth, our analysis did not include a measure of socioeconomic status. Patients with low socioeconomic status are more likely to be diagnosed at a late disease stage and, on average, their survival period after diagnosis is shorter, compared to patients with high socioeconomic status. ³⁰ Therefore, patients with low socioeconomic status might have lower out-of-pocket spending because they had fewer months of end-of-life care, compared to patients with high socioeconomic status.

Ninth, in most countries, costs of care in nursing homes and hospices are the main drivers of total out-of-pocket health care spending (online Appendix Exhibit 1 presents these findings in graphical form). ³¹ However, these categories also had the highest share of missing data, which might have biased our analysis of out-of-pocket spending.

Finally, SHARE does not provide measures of the quality of end-of-life care.

Study Results

Study Population

We found substantial variation across countries regarding place of death. Even though the hospital is the most common place of death for the majority of countries included in the study, the share of hospital deaths varied from 31.07 percent for the Netherlands to 63.46 percent for Slovenia ( Exhibit 1 ). The countries with the highest proportion of decedents who died at home were Greece (49.49 percent), Italy (49.36 percent), and Poland (42.53 percent). The care home was not the prevalent place of death in any country: Sweden (36.43 percent), Denmark (30.58 percent), and Switzerland (30.56 percent) had the highest shares of deaths in that setting.

Cancer and cardiovascular diseases were the leading causes of death in all countries. Greece (53.06 percent) and Poland (48.28 percent) had a particularly high proportion of patients who died from cardiovascular diseases ( Exhibit 2 ). The countries with the highest shares of deaths from cancer were the Netherlands (38.93 percent), Italy (35.73 percent), and Switzerland (34.72 percent). (For additional patient characteristics and circumstances of death, see Appendix Exhibits 2 and 3.) ³¹

In most countries, a high proportion of survey participants received care from a general practitioner and a specialist physician and had a hospitalization during the last year of life (Appendix Exhibit 2). ³¹ A significant percentage of proxy respondents (23.08–46.76 percent) did not know whether the deceased had received care in a nursing home or hospice (Appendix Exhibit 3). ³¹ Among those whose proxy respondents provided information to interviewers, decedents in France (14.63 percent) and Sweden (14.13 percent) had the highest use of hospice services, while nursing home care was most prevalent in Switzerland (24.31 percent) and Denmark (23.31 percent). Denmark (55.89 percent) and Ireland (51.35 percent) had the highest shares of home care use, while Slovenia (11.54 percent) and Greece (18.88 percent) had the lowest shares of home care use.

The shares of decedents who incurred any out-of-pocket health care spending ranged from 90.44 percent in Sweden to 19.23 percent in Slovenia (Appendix Exhibit 3). ³¹ Among people who incurred out-of-pocket spending, the highest average spending was for decedents in Switzerland (4,687.0 euros, adjusted for purchasing power parity), while the lowest was in Estonia (159.1 euros, adjusted for purchasing power parity).

Determinants Of Place Of Death

Holding other factors constant, dying in a hospital was associated with settings where end-of-life care was predominantly privately funded and with patients who were younger than age eighty and male ( Exhibit 3 ). People who received any care in a nursing home or hospice or at home in the last year of life were less likely to die in a hospital, compared to people who did not receive such care. Dying in a care home was associated with settings where public funds are a predominant source of funding for end-of-life care, and with patients who were ages eighty and older and those struggling with ADL limitations. Type of funding mechanism was not associated with dying at home (data not shown).

Our analysis of the interaction between country group and age showed that younger patients were more likely to die in a hospital in countries with both predominantly private and public funding for end-of-life care, although the effect was stronger in the group of countries where end-of-life care was predominantly privately funded (see Appendix Exhibit 4 for results from models that included interaction terms). ³¹ In countries with predominantly privately funded end-of-life care, patients were more likely to use acute care and die in a hospital, regardless of cohabitation status. Finally, patients with cardiovascular diseases were more likely to die in a hospital and less likely to die in a care home, irrespective of the financing structure of end-of-life care. In contrast, the funding source played an important role in the place of death of cancer decedents, since they were more likely to use institutional care in areas with predominantly public support of end-of-life care.

Out-Of-Pocket Spending At The End Of Life

Our study findings show that the structure and financing of long-term care have an impact on health care use: Generally, countries with higher public expenditure on long-term care tend to have lower shares of in-hospital deaths ( Exhibit 4 ). More developed publicly funded long-term care institutions provide additional support for patients at the end of life and could be used as a substitute for hospital care at the end of life.

SOURCE Authors’ analysis of data from the Survey of Health, Ageing and Retirement in Europe; and OECD.Stat. Health expenditure and financing [Internet]. Paris: Organization for Economic Cooperation and Development; [cited 2017 May 26]. Available from: http://stats.oecd.org/Index.aspx?DataSetCode=SHA . NOTES The orange line is a fitted regression line ( $y=-4.919x+0.549$ ). The public expenditure for Israel is from 2012.

Neither less use of acute care nor more use of publicly funded long-term care services necessarily means lower patient out-of-pocket spending, for several reasons: Even if institutional care is publicly provided, its provision might be associated with high user charges (as is the case in Switzerland and Austria), ³² and our analysis did not capture out-of-pocket spending associated with informal care provision—which might involve higher total out-of-pocket spending in some of the countries where end-of-life care is predominantly privately funded. ^32,33 The variation in affordability of institutional care in such countries might explain the wide variation in out-of-pocket spending in this group.

Discussion

Our findings demonstrate the impact of the long-term care funding structure and patient characteristics on care pathways in the last year of life and on place of death. In countries with predominantly public financing of end-of-life care, institutional care is a substitute for hospital care, which confirms our first hypothesis. In countries with predominantly private financing, end-of-life care is hospital-based, which confirms our second hypothesis. Our findings reveal a north-south gradient among European countries in end-of-life care patterns—a gradient that has previously been observed for the organization and financing of health care and long-term care systems. ^34–37 Even though the Czech Republic, Estonia, Israel, and Poland are not in southern Europe, they are more similar to countries there than to those in northern Europe in terms of long-term care funding arrangements and provision and use of care. This is likely due to their historical heritage (except for Israel) of being former socialist economies, where welfare provision of long-term care was neglected, resulting in poor investments in residential facilities. ³⁸ Countries that spend more on long-term care also have more developed infrastructure for elderly care, so they offer an alternative to informal and acute care at the end of life. Northern European countries have significantly higher use of formal long-term care than do countries in the south, ³⁸ which is consistent with our findings.

Our results are in line with previous work that showed a higher intensity of care for younger patients, ^11,12 which supports our third hypothesis. In countries with predominantly public financing of end-of-life care, older patients are more likely to receive palliative and supportive care in institutions, compared to countries with mostly privately financed care—which is in line with our fourth hypothesis. In countries with predominantly privately financed end-of-life care, given the limited availability of formal end-of-life care, acute care is a substitute for institutional care. For people in such countries who are living alone, there is no alternative to acute end-of-life care, a finding that is in line with our fifth hypothesis. Cancer patients have a more predictable end-of-life care pathway, compared to people who die from an acute episode. In countries with predominantly public funding of end-of-life care, cancer patients use more institution-based and less hospital-based end-of-life care. In countries with predominantly privately funded end-of-life support, cancer patients are more likely to use costly hospital care, which supports our sixth hypothesis.

On average, out-of-pocket spending for end-of-life care was particularly high in Austria, Belgium, Germany, and Switzerland (Appendix Exhibit 3). High out-of-pocket spending is linked to insufficient insurance coverage of long-term care, with a high proportion of care costs paid by patients ³⁹ and lack of sufficient family support to meet patients’ needs. ⁴⁰ People in richer countries have higher out-of-pocket spending, but these countries also use more formal long-term care because their citizens are better able to afford the higher costs of care. In Europe, hospital care has lower user charges than institutional long-term care, and in many European countries, there is no cost sharing for hospital treatment. ⁴¹ Given this difference in financial burden on patients and their families, there is a higher propensity in Europe to be admitted to an acute care setting—especially in countries with predominantly private financing of end-of-life care. Austria and Switzerland are exceptions here: They have significantly higher use of care homes but also patients who are better able to afford formal long-term care—which explains the high out-of-pocket spending in these two countries. Compared to countries with predominantly publicly funded end-of-life care, the Netherlands and Denmark have lower out-of-pocket spending for two reasons: They have better financial protection mechanisms; ^41,42 and the role of long-term care insurance is not only protective in these countries, but it also incentivizes people to use different types of nonhospital care. ⁴²

In the majority of European countries, most people prefer to die at home, ⁴³ but our analysis showed that most die in hospitals. Overall, policies to encourage out-of-hospital deaths are lacking. In the majority of countries analyzed in this study, the proportion of decedents who received formal home care in the last year of life was small (see Appendix Exhibit 3), ³¹ but there was substantially higher use of institutional end-of-life care in countries with significant public financing of that care. The development of policies and support to facilitate publicly funded end-of-life care would reduce hospitalizations and enable more people to die outside of the hospital. This could also decrease out-of-pocket spending for patients during their last year of life and enable them to choose their place of death.

Policy Implications

Understanding end-of-life care patterns is crucial for developing policies to address the challenges of providing that care. Countries with predominantly public financing of end-of-life care and developed long-term care infrastructures encourage more use of out-of-hospital care. Relieving the cost burden on patients and their families would stimulate more people to use formal care at the end of life and reduce the amount of unplanned hospitalizations. Implementing effective policies to facilitate the use of nonacute end-of-life care would reduce public health care costs.

Demand for end-of-life care is increasing as Europe’s population ages. Since births per woman are decreasing in all European countries, ⁴⁴ the aging of the population will affect the future availability of informal care. ^33,34 This is particularly relevant for Israel and countries in southern, central, and eastern Europe that—like Israel—rely primarily on private funding for end-of-life care, have underdeveloped long-term care infrastructures and low use of institutional long-term care, and therefore are less prepared to address the challenges of increasing elderly populations. Increased public financing through increased social health insurance contributions or taxation is necessary to improve the availability and quality of formal end-of-life care. Such an increase may be difficult to implement in less wealthy European countries, especially in the current context of economic crises and austerity.

The findings in this study are also relevant outside of the European context. For example, long-term care in the United States is mostly privately financed, but many people remain uninsured for such care because of the high cost of premiums. ⁴⁵ There is a clear message from Europe that investments in long-term care pay off, leading to more cost-effective care at the end of life and a lower financial burden for society.

Conclusion

Our findings provide valuable insights into the effects of long-term care funding and structure and of patients’ characteristics on resource use, out-of-pocket spending, and place of death at the end of life in seventeen countries. The findings reveal substantial differences in the use and provision of end-of-life care depending on the type of financing mechanism. Patients in countries with predominantly public financing of this care rely more on the welfare state, and end-of-life care is more institution-centric. Higher usage of outpatient long-term care settings in countries with higher public spending on long-term care leads to reduced hospitalizations and a higher share of out-of-hospital deaths, compared to countries with predominantly privately funded end-of-life care—which tend to have underdeveloped and underused formal long-term care facilities. As a result, end-of-life care in countries that rely on private funding is more hospital-centric.

ACKNOWLEDGMENTS

This work was funded by the Seragnoli Foundation. This article uses data from the Survey of Health, Ageing and Retirement in Europe (SHARE), Waves 2, 3 (SHARELIFE), 4, and 5 (DOIs: 10.6103/SHARE.w2.500, 10.6103/SHARE.w3.500, 10.6103/SHARE.w4.500, and 10.6103/SHARE.w5.500). The SHARE data collection has been primarily funded by the European Commission through the 5th Framework Programme (Project No. QLK6-CT-2001-00360 in the thematic program Quality of Life), through the 6th Framework Programme (projects SHARE-I3, RII-CT-2006-062193; COMPARE, CIT5-CT-2005-028857; and SHARELIFE, CIT4-CT-2006-028812), and through the 7th Framework Programme (SHARE-PREP, No. 211909; SHARE-LEAP, No. 227822; and SHARE M4, No. 261982). Additional funding from the German Ministry of Education and Research, the Max Planck Society for the Advancement of Science, the US National Institute on Aging (Grant Nos. U01 AG09740-13S2, P01 AG005842, P01 AG08291, P30 AG12815, R21 AG025169, Y1-AG-4553-01, IAG BSR06-11, OGHA 04-064, and HHSN271201300071C), and from various other national sources is gratefully acknowledged. For a full list of funding institutions, see the SHARE website, http://www.share-project.org/contact-organisation/funding.html .

NOTES

• 1 Department of Health . End of life care strategy: promoting high quality care for all adults at the end of life [Internet]. London : The Department ; 2008 Jul [cited 2017 May 26 ]. Available from: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/136431/End_of_life_strategy.pdf Google Scholar
• 2 National Institute for Health Research Dissemination Centre . Better endings: right care, right place, right time [Internet]. Southampton (UK) : The Centre 2015 Dec [cited 2017 May 26 ]. Available from: http://www.dc.nihr.ac.uk/themed-reviews/Better%20endings%20FINAL%20DH%20single%20page.pdf Google Scholar
• 3 Nicholas LH , Langa KM , Iwashyna TJ , Weir DR . Regional variation in the association between advance directives and end-of-life Medicare expenditures . JAMA . 2011 ; 306 ( 13 ): 1447 – 53 . Crossref, Medline, Google Scholar
• 4 Huskamp HA , Kaufmann C , Stevenson DG . The intersection of long-term care and end-of-life care . Med Care Res Rev . 2012 ; 69 ( 1 ): 3 – 44 . Crossref, Medline, Google Scholar
• 5 World Health Organization, Regional Office for Europe . Palliative care for older people: better practices [Internet]. Copenhagen ; WHO ; c 2011 [cited 2017 May 26 ]. Available from: http://www.euro.who.int/__data/assets/pdf_file/0017/143153/e95052.pdf Google Scholar
• 6 Bähler C , Huber CA , Brüngger B , Reich O . Multimorbidity, health care utilization, and costs in an elderly community-dwelling population: a claims data based observational study . BMC Health Serv Res . 2015 ; 15 : 23 . Crossref, Medline, Google Scholar
• 7 Mot EFaber RGeerts JWillemé P editors. Performance of long-term care systems in Europe [Internet]. Brussels : Centre for European Policy Studies ; 2012 Dec [cited 2017 May 26 ]. (ENEPRI Research Report No. 117). Available from: https://www.ceps.eu/system/files/ENEPRI%20RR117%20_ANCIEN_%20Evaluation%20Final%20Report.pdf Google Scholar
• 8 Bekelman JE , Halpern SD , Blankart CR , Bynum JP , Cohen J , Fowler R , et al. Comparison of site of death, health care utilization, and hospital expenditures for patients dying with cancer in 7 developed countries . JAMA . 2016 ; 315 ( 3 ): 272 – 83 . Crossref, Medline, Google Scholar
• 9 Penders YW , Rietjens J , Albers G , Croezen S , Van den Block L . Differences in out-of-pocket costs of healthcare in the last year of life of older people in 13 European countries . Palliat Med . 2017 ; 31 ( 1 ): 42 – 52 . Crossref, Medline, Google Scholar
• 10 Kraus M , Riedel M , Mot E , Willemé P , Röhrling G , Czypionka T . A typology of long-term care systems in Europe [Internet]. Brussels : European Network of Economic Policy Research Institutes ; 2010 Aug [cited 2017 May 26 ]. (ENEPRI Research Report No. 91). Available from: http://aei.pitt.edu/32248/1/ENEPRI_RR_No_91_Typology_of_LTC__Systems_in_Europe.pdf Google Scholar
• 11 Lindskog M , Tavelin B , Lundström S . Old age as risk indicator for poor end-of-life care quality—a population-based study of cancer deaths from the Swedish Register of Palliative Care . Eur J Cancer . 2015 ; 51 ( 10 ): 1331 – 9 . Crossref, Medline, Google Scholar
• 12 Miesfeldt S , Murray K , Lucas L , Chang CH , Goodman D , Morden NE . Association of age, gender, and race with intensity of end-of-life care for Medicare beneficiaries with cancer . J Palliat Med . 2012 ; 15 ( 5 ): 548 – 54 . Crossref, Medline, Google Scholar
• 13 Murray SA , Kendall M , Boyd K , Sheikh A . Illness trajectories and palliative care . BMJ . 2005 ; 330 ( 7498 ): 1007 – 11 . Crossref, Medline, Google Scholar
• 14 Thompson G , McClement S . Defining and determining quality in end-of-life care . Int J Palliat Nurs . 2002 ; 8 ( 6 ): 288 – 93 . Crossref, Medline, Google Scholar
• 15 Börsch-Supan A . Survey of Health, Ageing and Retirement in Europe (SHARE) Wave 2. Release version: 5.0.0 . Munich : Max Planck Institute for Social Law and Social Policy, Munich Center for the Economics of Aging ; 2016 . Google Scholar
• 16 Börsch-Supan A . Survey of Health, Ageing and Retirement in Europe (SHARE) Wave 3—SHARELIFE. Release version: 5.0.0 . Munich : Max Planck Institute for Social Law and Social Policy, Munich Center for the Economics of Aging ; 2016 . Google Scholar
• 17 Börsch-Supan A . Survey of Health, Ageing and Retirement in Europe (SHARE) Wave 4. Release version: 5.0.0 . Munich : Max Planck Institute for Social Law and Social Policy, Munich Center for the Economics of Aging ; 2016 . Google Scholar
• 18 Börsch-Supan A . Survey of Health, Ageing and Retirement in Europe (SHARE) Wave 5. Release version: 5.0.0 . Munich : Max Planck Institute for Social Law and Social Policy, Munich Center for the Economics of Aging ; 2016 . Google Scholar
• 19 Survey of Health, Ageing and Retirement in Europe [home page on the Internet]. Munich : Max Planck Institute for Social Law and Social Policy, Munich Center for the Economics of Aging ; c 2017 [cited 2017 May 26 ]. Available from: http://www.share-project.org/data-access-documentation.html Google Scholar
• 20 Börsch-Supan A , Hunkler C , Gruber S , Orban A , Stuck S , Brandt M . SHARE Release Guide 5.0.0 . Munich : Max Planck Institute for Social Law and Social Policy, Munich Center for the Economics of Aging ; 2016 . Google Scholar
• 21 Organization for Economic Cooperation and Development . Inflation (CPI) [Internet]. Paris : OECD ; c 2016 [cited 2017 May 26 ]. Available from: https://data.oecd.org/price/inflation-cpi.htm Google Scholar
• 22 Koechlin F , Konijn P , Lorenzoni L , Schreyer P . Comparing hospital and health prices and volumes internationally: results of a Eurostat/OECD project [Internet]. Paris : Organization for Economic Cooperation and Development ; 2014 Aug 26 [cited 2017 Jun 5 ]. (OECD Health Working Papers No. 75). Available for download from: http://www.oecd-ilibrary.org/docserver/download/5jxznwrj32mp-en.pdf Google Scholar
• 23 Kraus M , Czypionka T , Riedel M , Mot E , Willemé P. How European nations care for their elderly: a new typology of long-term care systems [Internet]. Brussels : European Network of Economic Policy Research Institutes ; 2011 Jul [cited 2017 Jun 5 ]. (ENEPRI Policy Brief No. 7). Available from: http://www.ancien-longtermcare.eu/sites/default/files/ENEPRI%20PB%20No%207%20_ANCIEN%20WP1_%20final.pdf Google Scholar
• 24 Costa-Font J , Courbage C , Swartz K . Financing long-term care: ex ante, ex post, or both? Health Econ . 2015 ; 24 ( Suppl 1 ): 45 – 57 . Crossref, Medline, Google Scholar
• 25 Schulz A , Doblhammer-Reiter G . Validity of mortality follow-up in SHARE . Poster presented at: European Population Conference ; 2010 Sep 1–4 ; Vienna . Google Scholar
• 26 Börsch-Supan A , Brugiavini A , Jürges H , Mackenbach J , Siegrist J , Weber G , editors. Health, ageing, and retirement in Europe: first results from the Survey of Health, Ageing and Retirement in Europe [Internet]. Mannheim : Mannheim Research Institute for the Economics of Aging ; 2005 Apr [cited 2017 May 26 ]. Available from: http://www.share-project.org/uploads/tx_sharepublications/SHARE_FirstResultsBookWave1.pdf Google Scholar
• 27 Martinsson L , Fürst CJ , Lundström S , Nathanaelsson L , Axelsson B . Registration in a quality register: a method to improve end-of-life care—a cross-sectional study . BMJ Open . 2012 ; 2 ( 4 ): e001328 . Crossref, Medline, Google Scholar
• 28 Numico G , Cristofano A , Mozzicafreddo A , Cursio OE , Franco P , Courthod G , et al. Hospital admission of cancer patients: avoidable practice or necessary care? PLoS One . 2015 ; 10 ( 3 ): e0120827 . Crossref, Medline, Google Scholar
• 29 American Society of Clinical Oncology . The state of cancer care in America, 2015: a report by the American Society of Clinical Oncology . J Oncol Pract . 2015 ; 11 ( 2 ): 79 – 113 . Crossref, Medline, Google Scholar
• 30 Cavalli-Björkman N . Implications of patients’ socioeconomic status—what oncologists should be aware of . Acta Oncol . 2014 ; 53 ( 2 ): 161 – 3 . Crossref, Medline, Google Scholar
• 31 To access the Appendix, click on the Appendix link in the box to the right of the article online.
• 32 Allen K , Bednárik R , Campbell L , Dieterich A , Durrett EH , Emilsson T , et al. Work package 6: governance and finance of long-term care across Europe: overview report [Internet]. Birmingham (UK) : European Centre for Social Welfare Policy and Research ; 2011 Sep [cited 2017 May 30 ]. Available from: http://interlinks.euro.centre.org/sites/default/files/1%20WP6_GOV_FIN_final.pdf Google Scholar
• 33 Courtin E , Jemiai N , Mossialos E . Mapping support policies for informal carers across the European Union . Health Policy . 2014 ; 118 ( 1 ): 84 – 94 . Crossref, Medline, Google Scholar
• 34 Bolin K , Lindgren B , Lundborg P . Informal and formal care among single-living elderly in Europe . Health Econ . 2008 ; 17 ( 3 ): 393 – 409 . Crossref, Medline, Google Scholar
• 35 Organization for Economic Cooperation and Development . Long-term care for older people . Paris : OECD ; 2005 . Google Scholar
• 36 Blank RH , Burau VD . Comparative health policy . Basingstoke (UK) : Palgrave Macmillan ; 2004 . Google Scholar
• 37 Reher DS . Family ties in Western Europe: persistent contrasts . Popul Dev Rev . 1998 ; 24 ( 2 ): 203 – 34 . Crossref, Google Scholar
• 38 Bettio F , Verashchagina A . Long-term care for the elderly: provisions and providers in 33 European countries [Internet]. Luxembourg : Publications Office of the European Union ; 2010 Nov [cited 2017 May 30 ]. Available from: http://ec.europa.eu/justice/gender-equality/files/elderly_care_en.pdf Google Scholar
• 39 Mossialos E , Wenzl M , Osborn R , Sarnak D , editors. 2015 international profiles of health care systems [Internet]. New York (NY) : Commonwealth Fund ; 2016 Jan [cited 2017 May 30 ]. Available from: ww.commonwealthfund.org/~/media/files/publications/fund-report/2016/jan/1857_mossialos_intl_profiles_2015_v7.pdf Google Scholar
• 40 Hank K , Jürges H . The last year of life in Europe: regional variations in functional status and sources of support . Ageing Soc . 2010 ; 30 ( 6 ): 1041 – 54 . Crossref, Google Scholar
• 41 Strban G . MISSOC analysis 2014: cost sharing for health and long-term care benefits in kind: final version [Internet]. Brussels : Mutual Information System on Social Protection in the EU Member States, the EAA, and Switzerland ; 2014 Oct [cited 2017 May 30 ]. Available from: http://www.missoc.org/INFORMATIONBASE/OTHEROUTPUTS/ANALYSIS/2014/MISSOC%20Analysis%202014_EN.pdf Google Scholar
• 42 Mossialos E , Dixon A , Figueras J , Kutzin J , editors. Funding health care: options for Europe [Internet]. Buckingham (UK) : Open University Press ; 2002 [cited 2017 May 30 ]. Available from: http://www.euro.who.int/__data/assets/pdf_file/0003/98310/E74485.pdf Google Scholar
• 43 Brazil K , Howell D , Bedard M , Krueger P , Heidebrecht C . Preferences for place of care and place of death among informal caregivers of the terminally ill . Palliat Med . 2005 ; 19 ( 6 ): 492 – 9 . Crossref, Medline, Google Scholar
• 44 European Commission . Demography report [Internet]. Luxembourg : Publications Office of the European Union ; 2015 [cited 2017 May 30 ]. Available from: http://ec.europa.eu/eurostat/documents/3217494/6917833/KE-BM-15-003-EN-N.pdf/76dac490-9176-47bc-80d9-029e1d967af6 Google Scholar
• 45 Feder J , Komisar HL , Niefeld M . Long-term care in the United States: an overview . Health Aff (Millwood) . 2000 ; 19 ( 3 ): 40 – 56 . Go to the article, Google Scholar