pith. sign in

arxiv: 2604.18609 · v1 · submitted 2026-04-13 · 📊 stat.AP

The Broken Shield of European Palliative Care: Evidence from Synthetic Counterfactuals on Financial Toxicity and Informal Care

Pietro Grassi , Edoardo Paperi , Chiara Seghieri , Daniele Vignoli This is my paper

Pith reviewed 2026-05-10 15:35 UTC · model grok-4.3

classification 📊 stat.AP
keywords palliative carefinancial toxicityinformal caregivingsynthetic counterfactualscausal inferencequantile treatment effectsend-of-life careeuropean households
0
0 comments X

The pith

Palliative care reduces family financial and time burdens on average but leaves vulnerable households exposed in severe cases.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper deploys synthetic data generation to build digital twins of European households and compare outcomes with and without palliative care. It tests the claim that shifting to palliative care at the end of life either relieves economic pressure on families or simply transfers costs to unpaid caregivers. Average effects show reductions in both out-of-pocket spending and the value of informal care time, yet models at different points in the cost distribution reveal that these protections weaken sharply for certain groups. Readers should care because the work maps how medical transitions interact with pre-existing inequalities in income, family structure, and national welfare systems.

Core claim

Using Tabular Denoising Diffusion Probabilistic Models inside a Two-Learner framework on SHARE data from 2016-2021, the analysis finds that palliative care functions as a double shield that truncates out-of-pocket expenditures and informal caregiving shadow values for the average household. Quantile treatment effect estimates nevertheless identify a broken shield for vulnerable subgroups, where non-cancer trajectories, physical dependency, absence of a spouse, rigid gender roles, and financial distress produce large and escalating penalties that vary across high-wage Nordic and underfunded Eastern institutional settings.

What carries the argument

Tabular Denoising Diffusion Probabilistic Models within a Two-Learner architecture that generate high-fidelity synthetic counterfactuals, allowing isolation of palliative care effects while using 2020-2021 lockdowns to separate structural inequalities from transient shocks.

If this is right

  • Palliative care truncates average financial toxicity and time poverty across European households.
  • Non-cancer end-of-life paths generate massive structural penalties that grow with physical dependency.
  • Socio-demographic factors such as lack of spousal support, gender dynamics, and financial distress amplify exposure to high costs.
  • Institutional regimes matter: high-wage Nordic systems impose opportunity costs while underfunded Eastern systems drive resource exhaustion.
  • Expansion of palliative care must be decoupled from the oncological paradigm and paired with state-funded long-term care.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Targeted financial and respite-care supplements for non-cancer patients and households without spousal networks could reduce the tail risks identified in the quantile models.
  • The contrast between Nordic opportunity costs and Eastern exhaustion suggests that comparative welfare-state studies could test whether higher baseline funding levels buffer against the broken-shield pattern.
  • Re-running the synthetic twin procedure on post-2021 data would reveal whether recent policy adjustments have narrowed the quantile gaps for vulnerable subgroups.

Load-bearing premise

The synthetic digital twins created by the diffusion models accurately isolate the causal effect of palliative care without model-induced bias or unmeasured confounding.

What would settle it

If a new wave of real data or a hold-out validation set shows that matched households receiving palliative care do not exhibit lower out-of-pocket costs or lower informal care hours than their synthetic twins, especially at upper quantiles for non-cancer and financially distressed groups, the double-shield claim would be refuted.

read the original abstract

The transition of end-of-life care to palliative care (PC) sparks intense debate: does it provide economic relief or shift unremunerated labor costs onto families? Evaluating this is hindered by causal inference challenges and skewed healthcare costs. To overcome these limitations, we introduce a Synthetic Data Generation framework. Using pan-European SHARE data (2016-2021), we deploy Tabular Denoising Diffusion Probabilistic Models within a Two-Learner architecture to synthesize high-fidelity digital twins. By including the 2020-2021 lockdowns, we leverage the COVID-19 pandemic to isolate structural inequalities from transient market shocks. Our findings challenge the strict cost-shifting hypothesis: on average, PC acts as a "double shield", truncating out-of-pocket expenditures (financial toxicity) and informal caregiving shadow values (time poverty). However, quantile treatment models expose a "broken shield" for vulnerable households and severe tail events. Non-cancer trajectories drive massive structural penalties that escalate at the distribution's tail, mechanically compounded by physical dependency. Socio-demographics heavily modulate this exposure: lacking a spousal net inflates the burden, rigid gender dynamics trigger labor market ejection, and financial distress acts as a profound multiplier. Institutionally, high-wage Nordic regimes paradoxically impose opportunity costs, while severe penalties in underfunded Eastern systems, mediated by financial distress, drive families toward resource exhaustion. We conclude that while PC is an ethical imperative, its expansion must be decoupled from the oncological paradigm and matched with state-funded long-term care to protect against clinical decline and financial shocks.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper applies Tabular Denoising Diffusion Probabilistic Models within a Two-Learner architecture to SHARE data (2016-2021) to generate synthetic counterfactuals for palliative care (PC) receipt. Leveraging the 2020-2021 lockdowns to separate structural inequalities from transient shocks, it estimates average and quantile treatment effects on out-of-pocket expenditures and informal care hours. The central claim is that PC provides an average 'double shield' by truncating both financial toxicity and time poverty, but a 'broken shield' appears for vulnerable households and non-cancer trajectories at the right tail, modulated by socio-demographics and institutional regimes.

Significance. If the synthetic counterfactuals validly isolate causal effects, the findings would challenge cost-shifting narratives in end-of-life care and supply policy-relevant evidence on heterogeneous burdens across Europe. The approach innovates by adapting diffusion models to tabular causal inference with skewed outcomes; successful validation could advance methods for estimating tail effects in healthcare economics where randomized trials are infeasible.

major comments (3)
  1. [Methods (Synthetic Data Generation)] Methods section on Tabular DDPM and Two-Learner architecture: no fidelity metrics (e.g., distributional distances, tail quantile preservation, or synthetic-vs-real overlap statistics) are reported for the generated counterfactuals. This is load-bearing for the quantile treatment effect claims, as diffusion models fitted to the same joint distribution used for evaluation can distort right-skewed cost tails and treatment heterogeneity without explicit checks.
  2. [Identification and Data] Identification strategy (lockdown leverage): the assumption that 2020-2021 lockdowns cleanly isolate structural inequalities requires conditional ignorability of PC assignment during the shock period. No robustness tests address potential violations from pandemic-induced changes in care access or reporting, which directly threatens the 'broken shield' interpretation at the tails for vulnerable groups.
  3. [Results (Quantile Treatment Effects)] Results (quantile models): the reported tail estimates lack sensitivity analyses to DDPM hyperparameters, training schedule, or post-hoc sample exclusions. Without these, it is impossible to determine whether the 'broken shield' for non-cancer and financially distressed households reflects data or modeling artifacts.
minor comments (2)
  1. [Introduction] The abstract and introduction would benefit from explicit comparison to prior synthetic data or counterfactual methods in health economics to clarify the incremental contribution of the Tabular DDPM approach.
  2. [Methods] Notation for the two-learner components and the mapping from diffusion outputs to treatment effects could be formalized with equations to improve reproducibility.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their detailed and insightful comments on our manuscript. Their feedback has prompted us to enhance the methodological transparency and robustness checks in the paper. Below, we provide point-by-point responses to the major comments and indicate the revisions we will make.

read point-by-point responses

  1. Referee: Methods section on Tabular DDPM and Two-Learner architecture: no fidelity metrics (e.g., distributional distances, tail quantile preservation, or synthetic-vs-real overlap statistics) are reported for the generated counterfactuals. This is load-bearing for the quantile treatment effect claims, as diffusion models fitted to the same joint distribution used for evaluation can distort right-skewed cost tails and treatment heterogeneity without explicit checks.

    Authors: We agree that reporting fidelity metrics is essential to substantiate the validity of the synthetic counterfactuals, particularly given the importance of tail behavior for our quantile treatment effect estimates. In the revised manuscript, we will add a dedicated subsection in the Methods section detailing several fidelity checks. These include: (i) distributional distances such as the Wasserstein-1 distance and maximum mean discrepancy for continuous variables like out-of-pocket expenditures and informal care hours; (ii) Kolmogorov-Smirnov tests for equality of distributions between real and synthetic data; (iii) overlap statistics and visual comparisons (e.g., histograms and Q-Q plots) focusing on the right tails to ensure preservation of extreme values. Our post-hoc analyses indicate that the Tabular DDPM within the Two-Learner framework maintains high fidelity, with tail quantiles closely matching the observed data, thereby supporting the reliability of the 'broken shield' findings at the upper quantiles. revision: yes

  2. Referee: Identification strategy (lockdown leverage): the assumption that 2020-2021 lockdowns cleanly isolate structural inequalities requires conditional ignorability of PC assignment during the shock period. No robustness tests address potential violations from pandemic-induced changes in care access or reporting, which directly threatens the 'broken shield' interpretation at the tails for vulnerable groups.

    Authors: The referee correctly identifies a key assumption in our identification strategy. While the lockdowns provide a natural experiment to separate structural from transient effects, we recognize the need for robustness against potential violations of conditional ignorability due to pandemic-related disruptions. In the revision, we will include additional robustness tests: (1) re-estimating the models using only pre-2020 data and comparing to the full sample; (2) incorporating country-specific pandemic severity measures as controls; (3) conducting placebo tests on non-PC related outcomes. These will be reported in a new appendix. Initial explorations suggest that the heterogeneous effects for vulnerable groups persist, but we will present the full set of checks to allow readers to assess the sensitivity of the 'broken shield' interpretation. revision: yes

  3. Referee: Results (quantile models): the reported tail estimates lack sensitivity analyses to DDPM hyperparameters, training schedule, or post-hoc sample exclusions. Without these, it is impossible to determine whether the 'broken shield' for non-cancer and financially distressed households reflects data or modeling artifacts.

    Authors: We appreciate the call for sensitivity analyses on the quantile treatment effect estimates. To address concerns about potential modeling artifacts, the revised version will feature an expanded sensitivity analysis section. This will include variations in DDPM hyperparameters (e.g., number of diffusion timesteps from 100 to 1000, different noise schedules), training schedules (learning rates and epochs), and post-hoc exclusions (e.g., trimming top 1% of costs or restricting to complete cases). Results will be presented in tables and figures showing that the tail estimates for non-cancer trajectories and financially distressed households remain qualitatively unchanged across these specifications. This demonstrates that the broken shield pattern is robust and not driven by specific modeling choices. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation relies on external SHARE data and standard synthetic counterfactual methods

full rationale

The paper trains Tabular DDPM models within a Two-Learner setup on the 2016-2021 SHARE dataset to produce synthetic digital twins and then applies quantile treatment models to estimate average and tail effects of palliative care. This is a conventional workflow for generating counterfactuals in observational data; the generated quantities are not shown by any quoted equation to equal the fitted parameters or input distributions by construction. No self-citations are invoked as load-bearing uniqueness theorems, no ansatz is smuggled, and no known empirical pattern is merely renamed. The central claims about a 'double shield' versus 'broken shield' therefore remain falsifiable against the external SHARE benchmark and are not forced by internal redefinition.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The central claim rests on the untested premise that the synthetic data generation accurately reproduces the joint distribution of costs, caregiving, and palliative care uptake under counterfactual no-PC scenarios, plus the assumption that COVID timing cleanly identifies structural effects.

free parameters (2)
  • Tabular DDPM hyperparameters and training schedule
    Parameters fitted to SHARE data to produce the digital twins that drive all counterfactual estimates.
  • Two-learner architecture tuning parameters
    Choices that determine how the synthetic controls are constructed and how treatment effects are isolated.
axioms (2)
  • domain assumption Synthetic digital twins generated by the diffusion model faithfully represent the counterfactual distribution of financial toxicity and informal care in the absence of palliative care.
    Invoked to justify treating the generated data as valid controls for causal inference.
  • domain assumption The 2020-2021 lockdown period isolates persistent structural inequalities from transient market or pandemic shocks.
    Used to leverage the pandemic timing for identification.

pith-pipeline@v0.9.0 · 5593 in / 1714 out tokens · 27862 ms · 2026-05-10T15:35:39.770072+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

80 extracted references · 80 canonical work pages

  1. [1]

    Alleviating the access abyss in palliative care and pain relief—an im- perative of universal health coverage: the Lancet Commission report.The Lancet, 391(10128): 1391–1454, 2018

    Felicia M Knaul et al. Alleviating the access abyss in palliative care and pain relief—an im- perative of universal health coverage: the Lancet Commission report.The Lancet, 391(10128): 1391–1454, 2018

    work page 2018

  2. [2]

    Pricivel M Carrera, Hagop M Kantarjian, and Victoria S Blinder. The financial burden and distress of patients with cancer: Understanding and stepping-up action on the financial toxicity of cancer treatment.CA: Cancer J Clin, 68(2): 153–165, 2018

    work page 2018

  3. [3]

    Exploring the financial im- pact of caring for family members receiving pal- liative and end-of-life care: a systematic review of the literature.Palliat Med, 28(3):375–390, 2014

    Clare Gardiner et al. Exploring the financial im- pact of caring for family members receiving pal- liative and end-of-life care: a systematic review of the literature.Palliat Med, 28(3):375–390, 2014

    work page 2014

  4. [4]

    The monetary valuation of informal care to cancer decedents at end-of-life: Evidence from a national census survey.Palliat Med, 35(4):750–758, 2021

    Sean Urwin et al. The monetary valuation of informal care to cancer decedents at end-of-life: Evidence from a national census survey.Palliat Med, 35(4):750–758, 2021

    work page 2021

  5. [5]

    Palliative care

    World Health Organization. Palliative care. https://www.who.int/news-room/ fact-sheets/detail/palliative-care, 2020. Accessed: 2026–03–19

    work page 2020

  6. [6]

    Economics of palliative care for hospitalized adults with serious illness: A meta- analysis.JAMA Intern Med, 178(6):820–829, 2018

    Peter May et al. Economics of palliative care for hospitalized adults with serious illness: A meta- analysis.JAMA Intern Med, 178(6):820–829, 2018

    work page 2018

  7. [7]

    Cost trajectories at the end of life: the canadian experience.J Pain Symptom Manage, 38(1):75–80, 2009

    Konrad Fassbender, Robin L Fainsinger, Mary Carson, and Barry A Finegan. Cost trajectories at the end of life: the canadian experience.J Pain Symptom Manage, 38(1):75–80, 2009

    work page 2009

  8. [8]

    Building integrated palliative care programs and services.Catalonia Collaborating Centre for Pal- liative Care, 2017

    Xavier Gómez-Batiste and Stephen Connor. Building integrated palliative care programs and services.Catalonia Collaborating Centre for Pal- liative Care, 2017

    work page 2017

  9. [9]

    Equity and the financial costs of informal caregiving in palliative care: a critical debate.BMC Palliat Care, 19(1):71, 2020

    Clare Gardiner et al. Equity and the financial costs of informal caregiving in palliative care: a critical debate.BMC Palliat Care, 19(1):71, 2020

    work page 2020

  10. [10]

    Inequalities in access to palliative care in europe: A multilevel analysis

    Pietro Grassi, Arianna Bellini, Chiara Seghieri, and Daniele Vignoli. Inequalities in access to palliative care in europe: A multilevel analysis. Mimeo, 2026

    work page 2026

  11. [11]

    Illness trajectories and pallia- tive care.BMJ, 330(7498):1007–1011, 2005

    Scott A Murray, Marilyn Kendall, Kirsty Boyd, and Aziz Sheikh. Illness trajectories and pallia- tive care.BMJ, 330(7498):1007–1011, 2005

    work page 2005

  12. [12]

    Redefining palliative care—a new consensus-based definition.J Pain Symptom Manage, 60(4):754–764, 2020

    Lukas Radbruch et al. Redefining palliative care—a new consensus-based definition.J Pain Symptom Manage, 60(4):754–764, 2020

    work page 2020

  13. [13]

    Katherine E Sleeman et al. The escalating global burden of serious health-related suffering: pro- jections to 2060 by world regions, age groups, and health conditions.Lancet Glob Health, 7(7): e883–e892, 2019

    work page 2060

  14. [14]

    Statistics and causal inference

    Paul W Holland. Statistics and causal inference. JASA, 81(396):945–960, 1986

    work page 1986

  15. [15]

    Estimat- ing log models: to transform or not to transform? J Health Econ, 20(4):461–494, 2001

    Willard G Manning and John Mullahy. Estimat- ing log models: to transform or not to transform? J Health Econ, 20(4):461–494, 2001

    work page 2001

  16. [16]

    Review of statistical methods for analysing healthcare re- sources and costs.Health Econ, 20(8):897–916, 2011

    Borislava Mihaylova, Andrew Briggs, Anthony O’Hagan, and Simon G Thompson. Review of statistical methods for analysing healthcare re- sources and costs.Health Econ, 20(8):897–916, 2011

    work page 2011

  17. [17]

    Methodological challenges and potential solutions for economic evaluations of palliative 15 and end-of-life care: A systematic review.Palliat Med, 38(1):85–99, 2024

    Claudia Fischer, Damian Bednarz, and Judit Simon. Methodological challenges and potential solutions for economic evaluations of palliative 15 and end-of-life care: A systematic review.Palliat Med, 38(1):85–99, 2024

    work page 2024

  18. [18]

    The state of applied econometrics: Causality and policy evaluation.Journal of Economic Perspectives, 31(2):3–32, 2017

    Susan Athey and Guido W Imbens. The state of applied econometrics: Causality and policy evaluation.Journal of Economic Perspectives, 31(2):3–32, 2017

    work page 2017

  19. [19]

    Tabddpm: Mod- elling tabular data with diffusion models

    Akim Kotelnikov, Dmitry Baranchuk, Ivan Rubachev, andArtemBabenko. Tabddpm: Mod- elling tabular data with diffusion models. In International Conference on Machine Learning, volume 40, pages 17564–17579. PMLR, 2023

    work page 2023

  20. [20]

    Metalearners for estimating hetero- geneous treatment effects using machine learn- ing.Proc Natl Acad Sci USA, 116(10):4156–4165, 2019

    Sören R Künzel, Jasjeet S Sekhon, Peter J Bickel, and Bin Yu. Metalearners for estimating hetero- geneous treatment effects using machine learn- ing.Proc Natl Acad Sci USA, 116(10):4156–4165, 2019

    work page 2019

  21. [21]

    Evidence on the cost and cost-effectiveness of palliative care: a literature review.Palliative Medicine, 28(2):130–150, 2014

    Samantha Smith, Aoife Brick, Sinéad O’Hara, and Charles Normand. Evidence on the cost and cost-effectiveness of palliative care: a literature review.Palliative Medicine, 28(2):130–150, 2014

    work page 2014

  22. [22]

    How to include informal care in eco- nomic evaluations.Pharmacoeconomics, 31(12): 1105–1119, 2013

    Renske J Hoefman, Job van Exel, and Werner Brouwer. How to include informal care in eco- nomic evaluations.Pharmacoeconomics, 31(12): 1105–1119, 2013

    work page 2013

  23. [23]

    Economic valuation of informal care

    Bernard Van den Berg, Werner Brouwer, and Marc Koopmanschap. Economic valuation of informal care. an overview of methods and ap- plications.Eur J Health Econ, 5(1):36–45, 2004

    work page 2004

  24. [24]

    Effect of a home-based palliative care program on healthcare use and costs.J Am Geriatr Soc, 64(11):2288–2295, 2016

    J Brian Cassel et al. Effect of a home-based palliative care program on healthcare use and costs.J Am Geriatr Soc, 64(11):2288–2295, 2016

    work page 2016

  25. [25]

    Comparison of site of death, health care utilization, and hospital expenditures for patients dying with cancer in 7 developed countries.JAMA, 315(3):272–283, 2016

    Justin E Bekelman et al. Comparison of site of death, health care utilization, and hospital expenditures for patients dying with cancer in 7 developed countries.JAMA, 315(3):272–283, 2016

    work page 2016

  26. [26]

    End-of-life medical spending in last twelve months of life is lower than previ- ously reported.Health Affairs, 36(7):1211–1217, 2017

    Eric B French et al. End-of-life medical spending in last twelve months of life is lower than previ- ously reported.Health Affairs, 36(7):1211–1217, 2017

    work page 2017

  27. [27]

    Family health behaviors.American Economic Review, 109(9):3162–3191, 2019

    Itzik Fadlon and Torben Heien Nielsen. Family health behaviors.American Economic Review, 109(9):3162–3191, 2019

    work page 2019

  28. [28]

    Medicaid insurance in old age

    Mariacristina De Nardi, Eric French, and John Bailey Jones. Medicaid insurance in old age. American Economic Review, 106(11):3480–3520, 2016

    work page 2016

  29. [29]

    Partners’ health and silver splits in europe: A gendered pattern?Journal of Mar- riage and Family, 87(4):1639–1663, 2025

    Daniele Vignoli, Giammarco Alderotti, and Ce- cilia Tomassini. Partners’ health and silver splits in europe: A gendered pattern?Journal of Mar- riage and Family, 87(4):1639–1663, 2025

    work page 2025

  30. [30]

    Partner care ar- rangements and well-being in mid-and later life: The role of gender across care contexts.J Geron- tol B Psychol Sci Soc Sci, 77(2):435–445, 2022

    Ginevra Floridi, Nekehia T Quashie, Karen Glaser, and Martina Brandt. Partner care ar- rangements and well-being in mid-and later life: The role of gender across care contexts.J Geron- tol B Psychol Sci Soc Sci, 77(2):435–445, 2022

    work page 2022

  31. [31]

    Short- and medium-term effects of informal eldercare on labor market outcomes.Feminist Economics, 26 (4):205–227, 2020

    Dörte Heger and Thorben Korfhage. Short- and medium-term effects of informal eldercare on labor market outcomes.Feminist Economics, 26 (4):205–227, 2020

    work page 2020

  32. [32]

    Gender dis- crepancies in the outcomes of schedule control on overtime hours and income in germany.European Sociological Review, 32(6):752–765, 2016

    Yvonne Lott and Heejung Chung. Gender dis- crepancies in the outcomes of schedule control on overtime hours and income in germany.European Sociological Review, 32(6):752–765, 2016

    work page 2016

  33. [33]

    Flexi- ble working, work-life balance, and gender equal- ity: Introduction.Soc Indic Res, 151(2):365–381, 2020

    Heejung Chung and Tanja van der Lippe. Flexi- ble working, work-life balance, and gender equal- ity: Introduction.Soc Indic Res, 151(2):365–381, 2020

    work page 2020

  34. [34]

    Long term care insurance and family norms.The B.E

    Chiara Canta and Pierre Pestieau. Long term care insurance and family norms.The B.E. Jour- nal of Economic Analysis & Policy, 14(2):401– 428, 2013

    work page 2013

  35. [35]

    Financing long-term care: ex-ante, ex-post or both?J Health Econ, 24(S1): 45–57, 2015

    Joan Costa-Font, Christophe Courbage, and Katherine Swartz. Financing long-term care: ex-ante, ex-post or both?J Health Econ, 24(S1): 45–57, 2015

    work page 2015

  36. [36]

    Marco Albertini et al. Caring in the xxi cen- tury: the sustainability of long-term care in ag- ing societies-mapping challenges and developing solutions within the age-it research program.J Gerontol B Psychol Sci Soc Sci, 2025

    work page 2025

  37. [37]

    Uncertainty and the welfare economics of medical care.The American Eco- nomic Review, 53(5):941–973, 1963

    Kenneth J Arrow. Uncertainty and the welfare economics of medical care.The American Eco- nomic Review, 53(5):941–973, 1963

    work page 1963

  38. [38]

    The asset cost of poor health.The Journal of the Economics of Ageing, 9:172–184, 2017

    James M Poterba, Steven F Venti, and David A Wise. The asset cost of poor health.The Journal of the Economics of Ageing, 9:172–184, 2017

    work page 2017

  39. [39]

    Princeton University Press, Princeton, NJ, 1990

    Gøsta Esping-Andersen.The Three Worlds of Welfare Capitalism. Princeton University Press, Princeton, NJ, 1990

    work page 1990

  40. [40]

    Health inequalities and welfare state regimes: theoretical insights on a public health ‘puzzle’.J Epidemiol Community Health, 65(9):740–745, 2011

    Clare Bambra. Health inequalities and welfare state regimes: theoretical insights on a public health ‘puzzle’.J Epidemiol Community Health, 65(9):740–745, 2011. 16

    work page 2011

  41. [41]

    The ‘southern model’ of wel- fare in social europe.Journal of European Social Policy, 6(1):17–37, 1996

    Maurizio Ferrera. The ‘southern model’ of wel- fare in social europe.Journal of European Social Policy, 6(1):17–37, 1996

    work page 1996

  42. [42]

    The soviet legacy in diagno- sis and treatment: implications for population health.Journal of Public Health Policy, 32(3): 293–304, 2011

    Boika Rechel, Colin Kennedy, Martin McKee, and Bernd Rechel. The soviet legacy in diagno- sis and treatment: implications for population health.Journal of Public Health Policy, 32(3): 293–304, 2011

    work page 2011

  43. [43]

    Nordic eldercare—weak universalism becoming weaker? Journal of European Social Policy, 28(3):294–308, 2018

    Marta Szebehely and Gabrielle Meagher. Nordic eldercare—weak universalism becoming weaker? Journal of European Social Policy, 28(3):294–308, 2018

    work page 2018

  44. [44]

    Family policies in OECD countries: A comparative analysis.Population and Development Review, 37(1):57–87, 2011

    Olivier Thévenon. Family policies in OECD countries: A comparative analysis.Population and Development Review, 37(1):57–87, 2011

    work page 2011

  45. [45]

    Walter Korpi, Tommy Ferrarini, and Stefan Englund. Women’s opportunities under differ- ent family policy constellations: Gender, class, and inequality tradeoffs in western countries re- examined.Social Politics: International Studies in Gender, State & Society, 20(1):1–40, 2013

    work page 2013

  46. [46]

    How many people will need palliative care in 2040? past trends, future projections and implications for services.BMC Medicine, 15(102), 2017

    Simon Noah Etkind et al. How many people will need palliative care in 2040? past trends, future projections and implications for services.BMC Medicine, 15(102), 2017

    work page 2040

  47. [47]

    Long-term care policy after covid- 19 — solving the nursing home crisis.New Eng- land Journal of Medicine, 383(10):903–905, 2020

    Rachel M Werner, Allison K Hoffman, and Norma B Coe. Long-term care policy after covid- 19 — solving the nursing home crisis.New Eng- land Journal of Medicine, 383(10):903–905, 2020

    work page 2020

  48. [48]

    The covid-19 pandemic and health inequalities.J Epidemiol Community Health, 74(11):964–968, 2020

    Clare Bambra, Ryan Riordan, John Ford, and Fiona Matthews. The covid-19 pandemic and health inequalities.J Epidemiol Community Health, 74(11):964–968, 2020

    work page 2020

  49. [49]

    Nurs- ing home care in crisis in the wake of covid-19

    David C Grabowski and Vincent Mor. Nurs- ing home care in crisis in the wake of covid-19. JAMA, 324(1):23–24, 2020

    work page 2020

  50. [50]

    Evi- dence in palliative care research: how should it be gathered?Med J Aust, 183(5):264–266, 2005

    Samar M Aoun and Linda J Kristjanson. Evi- dence in palliative care research: how should it be gathered?Med J Aust, 183(5):264–266, 2005

    work page 2005

  51. [51]

    Cambridge University Press, Cam- bridge, UK, 2nd edition, 2009

    Judea Pearl.Causality: Models, Reasoning, and Inference. Cambridge University Press, Cam- bridge, UK, 2nd edition, 2009

    work page 2009

  52. [52]

    Double/debiased machine learning for treatment and structural parameters.The Econometrics Journal, 21(1): C1–C68, 2018

    Victor Chernozhukov et al. Double/debiased machine learning for treatment and structural parameters.The Econometrics Journal, 21(1): C1–C68, 2018

    work page 2018

  53. [53]

    Cambridge University Press, 2015

    Guido W Imbens and Donald B Rubin.Causal inference in statistics, social, and biomedical sci- ences. Cambridge University Press, 2015

    work page 2015

  54. [54]

    Why propensity scores should not be used for matching.Political Analysis, 27(4):435–454, 2019

    Gary King and Richard Nielsen. Why propensity scores should not be used for matching.Political Analysis, 27(4):435–454, 2019

    work page 2019

  55. [55]

    Data resource profile: the survey of health, ageing and retirement in europe (share).International Journal of Epi- demiology, 42(4):992–1001, 2013

    Axel Börsch-Supan, Martina Brandt, Christian Hunkler, Thorsten Kneip, Julie Korbmacher, Frederic Malter, Barbara Schaan, Stephanie Stuck, and Sabrina Zuber. Data resource profile: the survey of health, ageing and retirement in europe (share).International Journal of Epi- demiology, 42(4):992–1001, 2013

    work page 2013

  56. [56]

    mice: Multivariate imputation by chained equations in r.Journal of Statistical Software, 45(3), 2011

    Stef van Buuren and Catharina GM Groothuis- Oudshoorn. mice: Multivariate imputation by chained equations in r.Journal of Statistical Software, 45(3), 2011

    work page 2011

  57. [57]

    The proportion of missing data should not be used to guide decisions on multiple imputation.J Clin Epidemiol, 110:63– 73, 2019

    Paul Madley-Dowd, Rachael Hughes, Kate Till- ing, and Jon Heron. The proportion of missing data should not be used to guide decisions on multiple imputation.J Clin Epidemiol, 110:63– 73, 2019

    work page 2019

  58. [58]

    Purchasing power parities, price level indices, nominal and real expenditures by analytical categories - based on coicop

    Eurostat. Purchasing power parities, price level indices, nominal and real expenditures by analytical categories - based on coicop

    work page

  59. [59]

    Accessed: 2026-03-17

    https://ec.europa.eu/eurostat/ databrowser/view/prc_ppp_ind_1, 2026. Accessed: 2026-03-17

    work page 2026

  60. [60]

    Fest: A unified framework for evaluating synthetic tabular data

    Weijie Niu, Alberto Huertas Celdran, Karoline Siarsky, and Burkhard Stiller. Fest: A unified framework for evaluating synthetic tabular data. InProceedings of the International Conference on Information Systems Security and Privacy, pages 434–444, 2025

    work page 2025

  61. [61]

    Better bootstrap confidence in- tervals.Journal of the American Statistical As- sociation, 82(397):171–185, 1987

    Bradley Efron. Better bootstrap confidence in- tervals.Journal of the American Statistical As- sociation, 82(397):171–185, 1987

    work page 1987

  62. [62]

    A prac- titioner’s guide to cluster-robust inference.Jour- nal of Human Resources, 50(2):317–372, 2015

    A Colin Cameron and Douglas L Miller. A prac- titioner’s guide to cluster-robust inference.Jour- nal of Human Resources, 50(2):317–372, 2015

    work page 2015

  63. [63]

    Bias reduction in standard errors for linear regression with multi-stage samples.Survey Methodology, 28(2):169–181, 2002

    Robert M Bell and Daniel F McCaffrey. Bias reduction in standard errors for linear regression with multi-stage samples.Survey Methodology, 28(2):169–181, 2002

    work page 2002

  64. [64]

    Regres- sion quantiles.Econometrica, 46(1):33–50, 1978

    Roger Koenker and Gilbert Bassett Jr. Regres- sion quantiles.Econometrica, 46(1):33–50, 1978

    work page 1978

  65. [65]

    Quan- tiles for counts.Journal of the American Statis- tical Association, 100(472):1226–1237, 2005

    José AF Machado and JMC Santos Silva. Quan- tiles for counts.Journal of the American Statis- tical Association, 100(472):1226–1237, 2005. 17

    work page 2005

  66. [66]

    Making sense of sensitivity: Extending omitted variable bias

    Carlos Cinelli and Chad Hazlett. Making sense of sensitivity: Extending omitted variable bias. JRSS: Series B (Statistical Methodology), 82(1): 39–67, 2020

    work page 2020

  67. [67]

    Generation and evaluation of privacy preserving synthetic health data.Neu- rocomputing, 416:244–255, 2020

    Andrew Yale et al. Generation and evaluation of privacy preserving synthetic health data.Neu- rocomputing, 416:244–255, 2020

    work page 2020

  68. [68]

    Nice technology appraisal and highly spe- cialised technologies guidance: the manual

    National Institute for Health and Care Excel- lence. Nice technology appraisal and highly spe- cialised technologies guidance: the manual. Tech- nical Report PMG36, NICE, 2025. Accessed: 2026–03–24

    work page 2025

  69. [69]

    Does informal care from children to their elderly parents substitute for formal care in europe?J Health Econ, 28(1):143–154, 2009

    Eric Bonsang. Does informal care from children to their elderly parents substitute for formal care in europe?J Health Econ, 28(1):143–154, 2009

    work page 2009

  70. [70]

    Female labor supply: Why is the united states falling behind?The American Economic Review, 103 (3):251–256, 2013

    Francine D Blau and Lawrence M Kahn. Female labor supply: Why is the united states falling behind?The American Economic Review, 103 (3):251–256, 2013

    work page 2013

  71. [71]

    Progress on catastrophic health spending in 133 countries: a retrospective observational study.Lancet Glob Health, 6(2): e169–e179, 2018

    Adam Wagstaff et al. Progress on catastrophic health spending in 133 countries: a retrospective observational study.Lancet Glob Health, 6(2): e169–e179, 2018

    work page 2018

  72. [72]

    The socioeconomic status of older adults: How should we measure it in studies of health inequalities?Journal of Epidemiology and Community Health, 55(12): 895–904, 2001

    Emily Grundy and G Holt. The socioeconomic status of older adults: How should we measure it in studies of health inequalities?Journal of Epidemiology and Community Health, 55(12): 895–904, 2001

    work page 2001

  73. [73]

    Stefania Ilinca, Ricardo Rodrigues, and An- drea E Schmidt. Fairness and eligibility to long- term care: An analysis of the factors driving inequality and inequity in the use of home care for older europeans.Int J Environ Res Public Health, 14(10):1224, 2017

    work page 2017

  74. [74]

    Does long-term care sub- sidization reduce hospital admissions and utiliza- tion?J Health Econ, 58:43–66, 2018

    Joan Costa-Font, Sergi Jimenez-Martin, and Cristina Vilaplana. Does long-term care sub- sidization reduce hospital admissions and utiliza- tion?J Health Econ, 58:43–66, 2018

    work page 2018

  75. [75]

    The economic value of time of informal care and its determinants (the cuidarse study).PLoS One, 14(5):e0217016, 2019

    Juan Oliva-Moreno et al. The economic value of time of informal care and its determinants (the cuidarse study).PLoS One, 14(5):e0217016, 2019

    work page 2019

  76. [76]

    Economic burden of informal caregiving associated with history of stroke and falls among older adults in the u.s.AJPM, 53 (6):S197–S204, 2017

    Heesoo Joo et al. Economic burden of informal caregiving associated with history of stroke and falls among older adults in the u.s.AJPM, 53 (6):S197–S204, 2017

    work page 2017

  77. [77]

    An overview of methods and applications to value informal care in economic evaluations of healthcare.Pharmacoeconomics, 26(4):269–280, 2008

    Marc A Koopmanschap, Job N van Exel, Bernard van den Berg, and Werner BF Brouwer. An overview of methods and applications to value informal care in economic evaluations of healthcare.Pharmacoeconomics, 26(4):269–280, 2008

    work page 2008

  78. [78]

    The valuation of informal care in cost- of-illness studies: A systematic review.Pharma- coEconomics, 35(3):331–345, 2017

    Juan Oliva-Moreno, Marta Trapero-Bertran, Luz Maria Peña-Longobardo, and Raúl Del Pozo- Rubio. The valuation of informal care in cost- of-illness studies: A systematic review.Pharma- coEconomics, 35(3):331–345, 2017

    work page 2017

  79. [79]

    double shield

    Michael Bergmann and Melanie Wagner. The im- pact of covid-19 on informal caregiving and care receiving across europe during the first phase of the pandemic.Front Public Health, 9:673874, 2021. Code A vailability Statement The R code used for the data cleaning, multiple im- putation, and statistical modeling is openly available at https://github.com/pietr...

    work page doi:10.6103/share.w1 2021

  80. [80]

    institutional shield

    periods. Crucially, the underlying clinical and micro- socioeconomic determinants of the EoL bur- den—specifically age, the ADL dependency score, baseline comorbidities, and subjective financial dis- tress—exhibit near-perfect alignment and overlap- ping confidence intervals between the pre-pandemic and pandemic cohorts. Furthermore, the massive eco- nomi...

    work page 2020