Datasheets for Healthcare AI: A Framework for Transparency and Bias Mitigation

Harshvardhan J. Pandit; Marjia Siddik

REVIEW 4 major objections 4 minor 30 references

Datasheets for Healthcare AI: A Framework for Transparency and Bias Mitigation

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that existing dataset documentation is inadequate for healthcare AI and proposes a 55-field 'Healthcare AI Datasheet' that records bias categories, risk assessments, and GDPR/AI Act compliance in a machine-readable format.

desk verdict A modest but honest extension of Datasheets for Datasets; the main weakness is that 'GDPR-aligned' is asserted rather than demonstrated field-by-field. read the letter →

arxiv 2501.05617 v1 pith:ZWNO53EW submitted 2025-01-09 cs.CY cs.DL

Marjia Siddik , Harshvardhan J. Pandit This is my paper · ORCID

classification cs.CYcs.DL

keywords AIinHealthcareDatasetDocumentationDataTransparencyBiasMitigationEthicalGDPREUActRiskAssessment

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper argues that current dataset documentation practices, such as Datasheets for Datasets and Dataset Nutrition Labels, are too shallow to support bias mitigation and legal compliance in healthcare AI. To close that gap, it proposes the Healthcare AI Datasheet, a 55-field structured template that records demographics, data provenance, temporal information, bias mitigation steps, personal data characteristics, and risk and compliance details, aligned with the GDPR and the EU AI Act. The authors also provide a machine-readable JSON schema so that datasheets can be checked automatically and passed along the AI value chain. A sympathetic reader would see this as a concrete mechanism for making bias and legal risk visible at the point where data is created and reused, rather than after a model has already caused harm.

What carries the argument

The central object is the Healthcare AI Datasheet, a structured documentation template with 55 information fields in 10 sections (metadata, purpose, source, temporal, demographic, data characteristics, bias mitigation, personal data, risk and compliance, and usage restriction). It functions as a checklist that forces dataset creators to record demographic distributions and their potential bias implications, data completeness and provenance, anonymisation techniques and re-identification risk, applicable laws and impact assessments, and usage restrictions. The machine-readable JSON encoding is what lets this checklist be validated, shared, and used as input to automated risk assessment.

What would settle it

Take a real clinical dataset with a documented bias, such as the data behind the well-known racial-bias finding in population health management, and have independent data stewards complete the Healthcare AI Datasheet for it. If the completed datasheets do not flag the known bias, or if two annotators reach materially different risk levels, then the framework's claim to enable systematic bias recognition fails.

Watch

Extended reading notes

Core claim

The central claim is that existing dataset documentation practices are inadequate for healthcare AI because they omit specific bias categories, systematic risk assessment, and regulatory alignment, and the Healthcare AI Datasheet fills those gaps. The paper supports this by comparing its 55-field, 10-section schema against three established approaches (Datasheets for Datasets, Dataset Nutrition Label, Data Statements for NLP), showing that the new schema adds fields for demographic distributions, temporal characteristics, bias mitigations, personal data and re-identification risk, risk levels, applicable laws, impact assessments, and usage restrictions. The authors further show the datasheet can be expressed in JSON, making it machine-readable and amenable to automated validation and risk assessment. The claim is not that the datasheet eliminates bias, but that it makes the information needed to recognise, assess, and legally enforce bias mitigation part of the dataset record itself.

Load-bearing premise

The framework's bias-mitigation value depends on dataset creators filling in the 55 fields accurately and completely; if they provide vague or self-serving answers, the datasheet becomes a box-ticking exercise and does not prevent biased AI.

Editorial extensions

If this is right

If adopted, healthcare AI datasets will come with documented bias categories and risk levels, making it harder to deploy biased models unknowingly.
The machine-readable format enables automated checks for GDPR and AI Act obligations, such as the presence of a data protection impact assessment or the applicable law.
The explicit demographic fields turn missing information, such as an unknown gender distribution, into a visible risk rather than an overlooked gap.
The datasheet can support secondary reuse of health data under initiatives like the Irish Health Information Bill and the European Health Data Space.
The comparison indicates that existing frameworks would need substantial extension to match this level of bias and compliance detail.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Editorial inference: The same schema could be extended outside healthcare, but the paper itself notes its healthcare focus; a testable extension is adapting the fields to other high-risk domains such as finance or criminal justice.
Editorial inference: Because the authors could not find any real dataset with sufficient existing documentation to populate the fields, the schema's real-world usability remains unvalidated; applying it retrospectively to a well-documented benchmark dataset like MIMIC would show which fields are feasible to fill.
Editorial inference: The explicit demographic and re-identification fields could serve as a checklist for data protection impact assessments, but only if regulators accept them as evidence; that adoption is not yet tested.
Editorial inference: The framework's bias-mitigation value depends on how faithfully creators populate the fields; a natural next study, which the paper does not report, is measuring inter-rater reliability of datasheet completion.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The actual new thing here is the Healthcare AI Datasheet itself: 55 fields in 10 sections, plus a machine-readable JSON schema and an open GitHub repository. That is a legitimate artifact, and the extension is sensible. The authors pulled bias categories and legal considerations from the literature, and the structure is coherent. They are also unusually transparent about the limits: no real dataset with adequate documentation could be found, so the validation is a manual exercise on hypothetical scenarios, and they concede in Section 5 that effectiveness depends on creators providing accurate and complete information. Credit is due for shipping the schema and the comparison tables rather than leaving everything as prose.

The soft spots are real but not fatal. The biggest one: the claim that the datasheet 'is aligned with the GDPR' is not backed by a traceable mapping from individual fields to specific GDPR articles or AI Act provisions. The literature review mentions Articles 9, 32, and 35 at a high level, and the Risk and Compliance section references DPIAs, but you cannot check whether the 55 fields actually cover all relevant obligations. So 'aligned' is doing a lot of work. Table 1 is a self-scored comparison with no rubric; the proposed approach naturally has more checkmarks because it was designed to add those fields. That doesn't make the contribution worthless, but it makes the superiority argument weaker than the text suggests. Also, the title says 'Bias Mitigation,' but the datasheet can only support identification and documentation of bias, not remove it. The paper's body mostly says 'support,' which is accurate, but the framing overpromises slightly.

One smaller point: the paper recommends DPV, an ontology co-authored by Pandit, for representing regulatory information. Self-citation is not automatically a flaw, but here the affiliation is not disclosed in that context. It is a future-work suggestion rather than a load-bearing step, so I treat it as minor.

The stress-test note holds up: the sufficiency of the 55 fields is unestablished, and the GDPR alignment claim is not traceable. But this is a framework proposal in a short conference paper, not a quantitative claim. The paper is honest about its validation limits, and the missing mapping is an addressable revision, not a fundamental error.

Who should read this: people working on dataset documentation, especially in healthcare AI, and anyone looking for a concrete template to adapt. It deserves a serious referee. I would want the revision to include a field-to-obligation mapping, a rubric for the comparison, and at least one test on a real dataset. The artifact is useful and the authors are clear-eyed about what they have and have not shown.

Referee Report

4 major / 4 minor

Summary. The paper investigates whether current dataset documentation practices are adequate for healthcare AI, argues that they fall short on bias categorisation, risk assessment, and regulatory alignment, and proposes the 'Healthcare AI Datasheet'—a framework with 55 information fields in 10 sections, a JSON-based machine-readable representation, and a comparison with three existing documentation approaches. It also discusses application in the Irish healthcare context, including the Health Information Bill and the EU AI Act. The central claim is that the proposed datasheet advances the state of the art by supporting bias categorisation, risk assessment, and GDPR/AI Act alignment, and that machine-readability enables automated risk assessments.

Significance. The paper addresses a timely and practical problem: translating bias and regulatory considerations into structured dataset documentation for healthcare AI. It does useful work in Section 2 by bringing together bias categories and legal frameworks, and the proposed 55-field datasheet, the open repository, and the JSON prototype are concrete artifacts that could be built upon. The discussion of the Irish healthcare context gives a plausible deployment path. However, as a contribution the paper is a framework proposal: the comparative evaluation is self-assessed, the validation is on hypothetical data, and no independent evidence shows that the fields are sufficient, correctly populated, or actually used for risk assessment. The significance will remain conditional until at least one real-world application or a traceable field-to-obligation mapping is supplied.

major comments (4)

[Section 3.3 / Table 1] The central comparison in Table 1 is self-scored by the authors without a rubric, and it is not a valid basis for the claim that the proposed framework 'advances the state of the art.' The paper does not define the decision rule for ∙, ∘, or ✕, does not report who assigned the symbols, and does not provide inter-rater agreement or external validation. The row 'Demographics' also appears to mischaracterize the baseline: Datasheets for Datasets includes demographic composition questions in its Composition section, so if the row is meant to cover only the bias-likelihood fields introduced here, that needs to be stated explicitly. As it stands, the table gives the reader no way to verify the comparison, and it is the main evidence for the paper's central claim.
[Sections 2.2, 3.1, 3.2] The assertion that the datasheet is 'aligned with the GDPR' and supports AI Act obligations is not traceable. The paper identifies legal requirements (GDPR Articles 9, 32, 35; the AI Act's risk-based approach) but never maps its 55 information fields to the specific provisions they are supposed to support. For example, the 'Risk and Compliance' fields record the existence of an impact assessment but the paper does not explain which fields would enable the GDPR Article 35 DPIA content or the AI Act Article 11 technical documentation. I would need a field-by-field mapping to legal obligations, even in an appendix, to accept the regulatory-alignment claim; absent this, the sufficiency of the 55 fields is unverified.
[Section 3.1 and Section 5] The framework is validated only through manual exercises on hypothetical datasets, and the paper itself concedes in Section 5 that 'its effectiveness relies on the accuracy and completeness of the information provided by dataset creators.' These two limitations are load-bearing for the bias-mitigation claim: if creators omit or misreport information, the datasheet becomes a box-ticking exercise, and the paper presents no evidence from a real dataset or real documentation process to show that the fields would be populated accurately or would change downstream risk assessment. I would like to see at least one real case study or user study, or failing that, a clearly labeled design-evaluation claim rather than an effectiveness claim.
[Section 3.1] The machine-readable and automated-risk-assessment benefits are asserted rather than demonstrated. While a JSON structure is described and a repository is linked, the paper does not include the JSON schema, a validation mechanism, or an example of a datasheet instance being consumed by an automated risk-assessment tool. Because 'machine-readable' and 'enabling automated risk assessments' are part of the stated contribution (RO4 and the abstract), the reader cannot evaluate whether the format actually supports these functions. Please include the schema or a concrete snippet and a small trace of how the data would be used in an automated check.

minor comments (4)

[Section 2.3 and Section 3.3] The comparison table does not include Open Datasheets, DataDoc Analyzer, or MetaReader, although Section 2.3 discusses them as part of the current practices; the caption and text should state that Table 1 covers only the three selected approaches.
[Sections 1, 2.3, 3.2, 4] There are several typos that should be corrected: 'healthare' in Section 1, 'spacial' in Section 2.3, 'sensitity' in Section 3.2, and 'organsiation' and 'landscaope' in Section 4.
[Section 3.2] The sentence 'any missing value (e.g. gender distribution unknown) becomes a risk in itself' conflates a missing metadata field with a measured risk; the claim should be clarified to say that absence of documentation prevents assessment, rather than being a risk itself.
[Section 3.1] The recommendation to use DPV (reference [26]) cites the second author's own work; please add a conflict-of-interest or competing-interests statement to the footnote or acknowledgments.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular reduction: the datasheet is a constructive design, but a minor non-load-bearing self-citation (DPV) and self-assessed comparisons appear.

full rationale

The paper's derivation chain is a needs analysis and a design proposal rather than a mathematical or data-derived prediction. RO1 and RO2 identify bias categories and legal considerations from external sources such as GDPR, the AI Act, and the Health Information Bill; RO3 surveys existing practices; RO4 explicitly builds on 'Datasheets for Datasets' [11] as an acknowledged baseline and extends it to 55 fields; RO5 applies the result to Ireland. No parameter is fitted, no equation relates input to output by construction, and no known result is merely renamed. The central claim that the Healthcare AI Datasheet 'supports information requirements regarding bias categorisation and risk assessment, and is aligned with the GDPR' is a design claim supported by qualitative argument and a self-assessed comparison table, not a prediction reduced to its inputs. The only self-citation is the recommendation of DPV [26], co-authored by Pandit, as a possible future standard for representing regulatory information: 'we recommend using existing standards such as DCAT with ODRL for expressing usage policies and DPV [26] to represent regulatory information.' This is explicitly a future-work suggestion and is not used to justify the sufficiency of the 55 fields, so it does not make the central claim circular. The reliance on hypothetical datasets for validation and on accurate creator-supplied information, conceded in Section 5 as 'its effectiveness relies on the accuracy and completeness of the information provided by dataset creators,' are limitations in evidence strength, not circularity.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No fitted parameters or new physical entities are introduced. The framework rests on three domain assumptions: documentation improves bias mitigation, the identified field requirements are sufficient, and dataset creators supply accurate information. The comparison table supporting the gap claim is self-assessed.

assumptions (4)

domain assumption Structured documentation of bias, risk, and legal fields leads to mitigation of bias and support for regulatory compliance.
The entire paper is built on this causal claim, but no empirical evidence is provided. Section 3 frames the fields as requirements without testing their effect.
domain assumption The bias categories in Section 2.1 and legal requirements in Section 2.2 are sufficiently complete to capture healthcare AI risks.
Derived from a literature review; no validation or domain-expert audit is provided.
domain assumption Dataset creators will populate the datasheet accurately and completely.
Explicitly identified as a limitation in Section 5. If false, the framework cannot deliver its claimed benefits.
ad hoc to paper The comparison in Table 1, scored by the authors using their own categories, is a valid measure of how existing frameworks cover the requirements.
The table is the main evidence for the gap the paper claims to fill; it is self-assessed and not independently audited.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Datasheets for Healthcare AI: A Framework for Transparency and Bias Mitigation." pith.science (2026). https://pith.science/paper/ZWNO53EW

@misc{pith2026250105617,
  author       = {Pith},
  title        = {Pith review of: Datasheets for Healthcare AI: A Framework for Transparency and Bias Mitigation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZWNO53EW}},
  note         = {Machine review of arXiv:2501.05617}
}

read the original abstract

The use of AI in healthcare has the potential to improve patient care, optimize clinical workflows, and enhance decision-making. However, bias, data incompleteness, and inaccuracies in training datasets can lead to unfair outcomes and amplify existing disparities. This research investigates the current state of dataset documentation practices, focusing on their ability to address these challenges and support ethical AI development. We identify shortcomings in existing documentation methods, which limit the recognition and mitigation of bias, incompleteness, and other issues in datasets. We propose the 'Healthcare AI Datasheet' to address these gaps, a dataset documentation framework that promotes transparency and ensures alignment with regulatory requirements. Additionally, we demonstrate how it can be expressed in a machine-readable format, facilitating its integration with datasets and enabling automated risk assessments. The findings emphasise the importance of dataset documentation in fostering responsible AI development.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 27 canonical work pages

[1]

E. J. Topol, High-performance medicine: the convergence of human and artificial intelli- gence, Nature medicine 25 (2019) 44–56

work page 2019
[2]

P. Shah, F. Kendall, S. Khozin, R. Goosen, J. Hu, J. Laramie, M. Ringel, N. Schork, Artificial intelligence and machine learning in clinical development: a translational perspective, NPJ digital medicine 2 (2019) 69

work page 2019
[3]

Obermeyer, E

Z. Obermeyer, E. J. Emanuel, Predicting the future—big data, machine learning, and clinical medicine, New England Journal of Medicine 375 (2016) 1216–1219

work page 2016
[4]

Obermeyer, B

Z. Obermeyer, B. Powers, C. Vogeli, S. Mullainathan, Dissecting racial bias in an algorithm used to manage the health of populations, Science 366 (2019) 447–453

work page 2019
[5]

D. S. Char, N. H. Shah, D. Magnus, Implementing machine learning in health care—addressing ethical challenges, New England Journal of Medicine 378 (2018) 981–983

work page 2018
[6]

Mehrabi, F

N. Mehrabi, F. Morstatter, N. Saxena, K. Lerman, A. Galstyan, A survey on bias and fairness in machine learning, ACM computing surveys (CSUR) 54 (2021) 1–35

work page 2021
[7]

Benjamin, Assessing risk, automating racism, Science 366 (2019) 421–422

R. Benjamin, Assessing risk, automating racism, Science 366 (2019) 421–422

work page 2019
[8]

Vayena, A

E. Vayena, A. Blasimme, I. G. Cohen, Machine learning in medicine: addressing ethical challenges, PLoS medicine 15 (2018) e1002689

work page 2018

Show all 30 references

[9]

Rajkomar, M

A. Rajkomar, M. Hardt, M. D. Howell, G. Corrado, M. H. Chin, Ensuring fairness in machine learning to advance health equity, Annals of internal medicine 169 (2018) 866–872

2018
[10]

Jiang, Y

F. Jiang, Y. Jiang, H. Zhi, Y. Dong, H. Li, S. Ma, Y. Wang, Q. Dong, H. Shen, Y. Wang, Artificial intelligence in healthcare: past, present and future, Stroke and vascular neurology 2 (2017)

2017
[11]

Gebru, J

T. Gebru, J. Morgenstern, B. Vecchione, J. W. Vaughan, H. Wallach, H. D. Iii, K. Crawford, Datasheets for datasets, Communications of the ACM 64 (2021) 86–92

2021
[12]

Paullada, I

A. Paullada, I. D. Raji, E. M. Bender, E. Denton, A. Hanna, Data and its (dis) contents: A survey of dataset development and use in machine learning research, Patterns 2 (2021)

2021
[13]

URL: http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=OJ:L:2016:119:TOC

Regulation2016, Regulation (eu) 2016/679 of the european parliament and of the council of 27 april 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing directive 95/46/ec (general data pro...

2016
[14]

URL: http://data.europa.eu/eli/reg/2024/1689/oj

Regulation2024, Regulation (eu) 2024/1689 of the european parliament and of the coun- cil of 13 june 2024 laying down harmonised rules on artificial intelligence and amend- ing regulations (ec) no 300/2008, (eu) no 167/2013, (eu) no 168/2013, (eu) 2018/858, (eu) 2018/1139 and ...

2024
[15]

of Ireland, Health information bill, Oireachtas (No

G. of Ireland, Health information bill, Oireachtas (No. 61 of 2024) (2024). URL: https: //www.oireachtas.ie/en/bills/bill/2024/61/

2024
[16]

L. A. Celi, J. Cellini, M.-L. Charpignon, E. C. Dee, F. Dernoncourt, R. Eber, W. G. Mitchell, L. Moukheiber, J. Schirmer, J. Situ, et al., Sources of bias in artificial intelligence that perpetuate healthcare disparities—a global review, PLOS Digital Health 1 (2022) e0000022

2022
[17]

Russo, M.-E

M. Russo, M.-E. Vidal, Leveraging ontologies to document bias in data, arXiv preprint arXiv:2407.00509 (2024)

2024
[18]

Gaonkar, K

B. Gaonkar, K. Cook, L. Macyszyn, Ethical issues arising due to bias in training ai algorithms in healthcare and data sharing as a potential solution, The AI Ethics Journal 1 (2020)

2020
[19]

M. Ganz, S. H. Holm, A. Feragen, Assessing bias in medical ai, in: Workshop on Inter- pretable ML in Healthcare at International Connference on Machine Learning (ICML), 2021

2021
[20]

K. S. Chmielinski, S. Newman, M. Taylor, J. Joseph, K. Thomas, J. Yurkofsky, Y. C. Qiu, The dataset nutrition label (2nd gen): Leveraging context to mitigate harms in artificial intelligence, arXiv preprint arXiv:2201.03954 (2022)

2022 arXiv
[21]

A. C. Roman, J. W. Vaughan, V. See, S. Ballard, N. Schifano, J. Torres, C. Robinson, J. M. L. Ferres, Open datasheets: Machine-readable documentation for open datasets and respon- sible ai assessments, arXiv preprint arXiv:2312.06153 (2023)

2023 arXiv
[22]

A. K. Heger, L. B. Marquis, M. Vorvoreanu, H. Wallach, J. Wortman Vaughan, Understanding machine learning practitioners’ data documentation perceptions, needs, challenges, and desiderata, Proceedings of the ACM on Human-Computer Interaction 6 (2022) 1–29

2022
[23]

E. M. Bender, B. Friedman, Data statements for natural language processing: Toward mitigating system bias and enabling better science, Transactions of the Association for Computational Linguistics 6 (2018) 587–604

2018
[24]

Giner-Miguelez, A

J. Giner-Miguelez, A. Gómez, J. Cabot, Datadoc analyzer: A tool for analyzing the docu- mentation of scientific datasets, in: Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, 2023, pp. 5046–5050

2023
[25]

H. M. Jannah, Metareader: A dataset meta-exploration and documentation tool, 2014

2014
[26]

H. J. Pandit, B. Esteves, G. P. Krog, P. Ryan, D. Golpayegani, J. Flake, Data privacy vocabulary (dpv) – version 2, The 23rd International Semantic Web Conference (ISWC) (in-press) (2024). URL: https://arxiv.org/abs/2404.13426

2024 arXiv
[27]

M.-A. Wren, S. Connolly, A european late starter: lessons from the history of reform in irish health care, Health Economics, Policy and Law 14 (2019) 355–373

2019
[28]

Craig, N

S. Craig, N. Kodate, Understanding the state of health information in ireland: A qualitative study using a socio-technical approach, International journal of medical informatics 114 (2018) 1–5

2018
[29]

Walsh, C

B. Walsh, C. Mac Domhnaill, G. Mohan, Developments in healthcare information systems in ireland and internationally, Dublin: The Economic and Social Research Institute (2021)

2021
[30]

Hickey, R

D. Hickey, R. O’Connor, P. McCormack, P. Kearney, R. Rosti, R. Brennan, The data quality index: improving data quality in irish healthcare records, ICEIS, 2021

2021

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

E. J. Topol, High-performance medicine: the convergence of human and artificial intelli- gence, Nature medicine 25 (2019) 44–56

work page 2019

[2] [2]

P. Shah, F. Kendall, S. Khozin, R. Goosen, J. Hu, J. Laramie, M. Ringel, N. Schork, Artificial intelligence and machine learning in clinical development: a translational perspective, NPJ digital medicine 2 (2019) 69

work page 2019

[3] [3]

Obermeyer, E

Z. Obermeyer, E. J. Emanuel, Predicting the future—big data, machine learning, and clinical medicine, New England Journal of Medicine 375 (2016) 1216–1219

work page 2016

[4] [4]

Obermeyer, B

Z. Obermeyer, B. Powers, C. Vogeli, S. Mullainathan, Dissecting racial bias in an algorithm used to manage the health of populations, Science 366 (2019) 447–453

work page 2019

[5] [5]

D. S. Char, N. H. Shah, D. Magnus, Implementing machine learning in health care—addressing ethical challenges, New England Journal of Medicine 378 (2018) 981–983

work page 2018

[6] [6]

Mehrabi, F

N. Mehrabi, F. Morstatter, N. Saxena, K. Lerman, A. Galstyan, A survey on bias and fairness in machine learning, ACM computing surveys (CSUR) 54 (2021) 1–35

work page 2021

[7] [7]

Benjamin, Assessing risk, automating racism, Science 366 (2019) 421–422

R. Benjamin, Assessing risk, automating racism, Science 366 (2019) 421–422

work page 2019

[8] [8]

Vayena, A

E. Vayena, A. Blasimme, I. G. Cohen, Machine learning in medicine: addressing ethical challenges, PLoS medicine 15 (2018) e1002689

work page 2018