CSSDH: An Ontology for Social Determinants of Health to Operational Continuity of Care Data Interoperability

Debashis Naskar; Sara Rodriguez Gonzalez; Subhashis Das

REVIEW 4 major objections 5 minor 13 references

CSSDH: An Ontology for Social Determinants of Health to Operational Continuity of Care Data Interoperability

T0 review · 4 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read CSSDH combines ISO/DIS 13940 ContSys with WHO social determinants of health to give continuity-of-care records a formal way to capture and query patients' social context.

desk verdict A plausible ontology extension whose own evaluation fails to support its interoperability claim; fix the evidence and release the artifact. read the letter →

arxiv 2412.09223 v1 pith:7USG5KE7 submitted 2024-12-12 cs.LO cs.AI

Subhashis Das , Debashis Naskar , Sara Rodriguez Gonzalez This is my paper · ORCID

classification cs.LOcs.AI

keywords socialdeterminantsofhealthcontinuitycareontologyOWLEHRinteroperabilityISO13940ContSyscompetencyquestionsprivacy

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

CSSDH, the Common Semantic Data Model for Social Determinants of Health, is a formal OWL ontology that combines ISO/DIS 13940:2024 ContSys with the WHO's social determinants of health (SDH) concepts. The paper's central claim is that this integration gives electronic health records a structured way to represent and query social factors such as employment, housing, and living conditions across care settings. If true, care teams and policymakers could ask questions like 'which patients are laid off and live in crowded homes?' directly against EHR data. The evidence offered is an internal consistency check, two author-written competency questions, a SPARQL sketch, and a clean pitfall scan.

What carries the argument

The central object is the CSSDH ontology itself, defined in OWL and built in Protégé. Its load-bearing mechanism is attaching SDH-related data properties directly to the subject-of-care class while anchoring clinical structure in ISO/DIS 13940 ContSys and aligning upper-level classes with DOLCE. The argument also depends on the competency-question method (CQ1 and CQ2) as the evaluation instrument, and on the HermiT reasoner and DL queries to test consistency and retrieval. Together these form a demonstration that SDH can be modeled inside a continuity-of-care ontology in a privacy-conscious, queryable way.

What would settle it

Encode a realistic set of patient records with known social determinants in CSSDH, run CQ1 and CQ2 plus independent clinical queries written by someone other than the authors, and attempt to exchange the data with a standards-based EHR system; if any expected patient is missed, any unexpected patient appears, or the social-determinant values are lost in exchange, the interoperability claim fails.

Watch

Extended reading notes

Core claim

Working in the authors' own terms, the discovery is an integrated formal information model: CSSDH. It reuses terminology from the SOHO ontology, the Social Prescribing Ontology, ICD-11 chapter 24, and the Gravity project value set, organizes it with OntoClean, and attaches social-determinant details to the subject-of-care class as boolean data properties (e.g., LayOffFromJob, CrowdingAtHome). This design is meant to preserve patient privacy while still enabling semantic reasoning and retrieval. The evaluation shows the ontology is consistent under the HermiT reasoner, returns answers for the two competency questions via a DL query and a SPARQL sketch, and raises no pitfalls in the oops! scanner. The intended consequence is that CSSDH operationalizes SDH capture in the continuity-of-care network, closing a gap left by existing SDH ontologies that are term lists rather than formal information models.

Load-bearing premise

The load-bearing premise is that passing a HermiT consistency check, one DL query, one SPARQL sketch, and an oops! scan against two author-written competency questions proves the model is logically sound and will support real EHR interoperability.

Editorial extensions

If this is right

If CSSDH is correct, EHR systems can add social determinants to their data model as boolean patient attributes, making queries about living conditions and employment routine.
The ontology's reuse of ISO 13940 and WHO SDH concepts means it can be aligned with existing continuity-of-care models and extended with other standard vocabularies.
Because social factors are stored as true/false flags, patient privacy is protected while still allowing semantic reasoning over those flags.
The model is positioned to be validated on large clinical datasets (NHS Scotland, HSE Ireland, MIMIC-III, Recoletas Red Hospital), which would test whether it performs outside the two competency questions.
A clean oops! scan and HermiT consistency suggest the ontology is at least internally well-formed, a necessary condition for any semantic interoperability claim.

Reading between the lines

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

The paper's own evaluation does not yet establish interoperability with real heterogeneous systems; a stronger test would map CSSDH instances to HL7 FHIR or openEHR and verify that SDH data round-trips between systems.
The boolean-only representation of SDH, chosen for privacy, discards detail such as income brackets or crowding severity; a testable extension is to compare CSSDH's query expressiveness against value-set-based approaches on realistic screening instruments.
Competency questions written by the authors may not reflect real clinical information needs; a robust evaluation would derive questions from standardized SDH screening tools (e.g., the Gravity project's own questionnaires) and measure precision and recall over annotated records.
If the planned large-scale validation succeeds, CSSDH could serve as a reference model for EHR vendors aiming to meet Healthy People 2030 SDH-capture goals while respecting patient privacy.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

Colleague,

The short version: this is a modest, plausible ontology-engineering contribution that overclaims what its evaluation shows. The authors extend their prior ContSys ontology with SDH concepts from SOHO, ICD-11 and the Gravity Project, and report an OWL model with 171 classes, 141 object properties, and 210 data properties. That is a legitimate extension of an established line of work, and the motivation — capturing social determinants in continuity of care — is real.

What the paper does well: it names the right gap, reuses existing standards rather than inventing terms, and runs two standard sanity checks, HermiT consistency and the oops! pitfall scanner, both of which pass. The DL query in Figure 3 may indeed be answerable in the model as built.

Now the soft spots, in order of importance. First, the evaluation does not actually answer its own competency questions. CQ1 asks for patients in a low-income area, but the text says the authors addressed it by associating LayOffFromJob and CrowdingAtHome with subjectOfCare. Those properties have nothing to do with income. Second, the SPARQL query shown for CQ2 uses OPTIONAL for CrowdingAtHome with no FILTER, so it returns laid-off patients even if crowding is unknown. The conjunctive intent of CQ2 is not enforced. Third, there is no OWL file, so the DL query and consistency check cannot be reproduced by a reader. Fourth, the competency questions were authored by the same team to match properties they added, so the evaluation is self-referential. There is no external dataset, no comparison with other SDH models, and no demonstration of interoperability with any real EHR system. The paper also has copy-paste inconsistencies, most visibly CSSDH versus CSSDM.

None of these flaws is fatal to the underlying idea. A corrected implementation, a released artifact, and at least one external dataset would make the interoperability claim testable. As published, the evidence supports only 'we built a coherent ontology extension,' not 'this enables continuity-of-care data interoperability.'

Who should read it: people building SDH-related ontologies or information models for EHRs will find the class and property choices useful as a starting point. I would not cite it in its current form because the artifact is unavailable and the query results are not reproducible.

Recommendation for peer review: yes, send it, but with a clear expectation of major revision — fix the evaluation, release the OWL file, and either correct the competency-question mapping or state the questions honestly. The manuscript is not ready as is, but the topic is important and the authors appear to be doing real work.

Referee Report

4 major / 5 minor

Summary. The paper proposes CSSDH, an OWL ontology that combines ISO/DIS 13940:2024 ContSys with WHO Social Determinants of Health concepts, with the stated aim of capturing social determinants of health in continuity-of-care data and enabling interoperability across EHR systems. The methodology section describes reuse of SOHO, the Social Prescribing Ontology, ICD-11 chapter 24, and Gravity project value sets, reporting 171 classes, 141 object properties, and 210 data properties, of which 171 are SDH-related. Evaluation consists of a HermiT consistency check, one DL query shown in Figure 3, a SPARQL sketch for two competency questions (CQ1 and CQ2), and an oops! pitfall scan. The authors state that future work will test the model with NHS Scotland, HSE Ireland, MIMIC-III, and Recoletas Red Hospital datasets.

Significance. If the central claim were fully supported, the contribution would be a practically relevant integrated ontology: it explicitly reuses established standards and vocabularies (ContSys, SOHO, ICD-11, Gravity), states concrete competency questions, and reports machine-checked consistency via HermiT and an oops! scan. Those are genuine strengths. However, the published evidence currently establishes only that a model of this design is internally consistent; it does not establish query-answering fidelity or interoperability, because the competency-question evaluation is directly contradicted by the implementation details given in Section 4, and no OWL artifact or external dataset is provided. The underlying taxonomy may well be salvageable, but the manuscript as written does not demonstrate its central assertion.

major comments (4)

[§4, CQ1] The text states: 'To address CQ1, we associated the data properties LayOffFromJob and CrowdingAtHome with the class subjectOfCare.' Neither property represents residence in a low-income area, so CQ1 ('patients lives in a low-income area') is not answered by the described implementation. This is load-bearing because CQ1 and CQ2 are the only query-answering evidence for the paper's central claim. The authors must either add and query a property that actually represents low-income area, or restate CQ1 to match the implemented properties.
[§4, CQ2 SPARQL] The displayed SPARQL query for CQ2 places CrowdingAtHome inside an OPTIONAL clause without any FILTER. Consequently, a laid-off patient with no crowding data is returned, which contradicts the conjunctive reading of CQ2 ('who are laid off from their job and whose houses are also crowded'). The query should require the crowding property and filter its value to true if the query is to reflect the stated competency question.
[§3.2, §4] No OWL artifact, ontology file, or query artifact is provided, so the DL query in Figure 3 and the claimed HermiT consistency result cannot be independently reproduced or checked. Because the entire evaluation rests on internal query-answering behavior, the missing artifact is a central reproducibility gap. A persistent URL to the OWL file, the exact DL query, and the version of Protégé/HermiT should be supplied.
[§4] The paper's interoperability claim is stated as an aim ('CSSDH aims to achieve interoperability') and the evaluation section explicitly defers real data testing to future work ('Our future plan is to test our model with a large dataset from NHS Scotland and HSE Ireland, MIMIC-III ...'). No external EHR data, integration scenario, or cross-system query is used. The title and abstract claim 'Operational Continuity of Care Data Interoperability,' but the evidence supports only internal consistency; the interoperability claim needs either a data-level demonstration or an explicit reframing as a design goal.

minor comments (5)

[§3.2, §4] The acronym is inconsistent: the paper uses CSSDH in the title, abstract, and Section 3.1, but CSSDM in Sections 3.2 and 4. Please use one consistent name throughout.
[§4] The phrase 'obtained no pitfall badge' is unclear; oops! reports pitfalls rather than badges. Please state the actual number and types of pitfalls detected, or report explicitly that zero pitfalls were found.
[References] Several cited resources are missing bibliographic entries: ISO/DIS 13940:2024 ContSys, ICD-11 chapter 24, the Gravity project SDH value set, and the Healthy People 2030 initiative. Please add references or persistent URLs for these.
[§4, SPARQL listing] The SPARQL listing is difficult to read because tokens are broken across lines (e.g., 'La y o f f f ro m jo b'). Provide a clean, copyable query listing, preferably in an appendix or auxiliary file.
[§3.2] The Boolean-only encoding for SDH data is justified by privacy considerations, but the paper should also discuss the resulting loss of information granularity, for example how 'crowded' or 'low-income' is defined and whether a binary flag is clinically meaningful for the stated competency questions.

Circularity Check

1 steps flagged · score 4.0 of 10

Self-referential CQ evaluation: the ontology is built to answer the authors' own competency questions, so the query-based validation reduces by construction; no other circularity found.

self definitional [Section 3.2 and Section 4 (Implementation; Results and Discussion), CQ1/CQ2 and SPARQL evaluation]
"We evaluated our ontology-driven CSSDM by modeling the Competency Queries elicited in Section 3.2 as SPARQL queries and assessing the returned answers. To address CQ1, we associated the data properties LayOffFromJob and CrowdingAtHome with the class subjectOfCare and set their data types as Boolean."

The competency questions used as the evaluation target were authored by the same team in Section 3.2 ('CSSDM should be capable of answering Competency Questions (CQ) like CQ1... CQ2...'), and the ontology was explicitly constructed to include the very data properties those CQs reference (LayOffFromJob, CrowdingAtHome on subjectOfCare). The SPARQL 'evaluation' then selects those properties and presents the returned tuples as evidence that the model 'is logically sound in retrieving information.' The retrieval is therefore entailed by the construction: the query is over the properties added to satisfy the CQs, so the 'answer' is not an independent prediction or test.

full rationale

The only defensible circularity is the self-referential competency-question evaluation. The base model from the authors' prior work [4,3] is cited as the foundation, but this is not circular: the model content is independently sourced from external standards (ISO/DIS 13940:2024 ContSys, WHO SDH, SOHO, ICD-11), and no claim is derived solely from the self-citation. The broader interoperability claim is not mathematically derived from the CQs; it is an untested design goal. However, the paper's central evidence that the model 'is logically sound in retrieving information' reduces to the authors' own CQs and the properties they attached to subjectOfCare. There is also an internal mismatch (CQ1 is about low-income area but is 'addressed' by LayOffFromJob and CrowdingAtHome properties, and the CQ2 SPARQL uses OPTIONAL without enforcing crowding), which weakens even the self-referential check. Because the evaluation is not independent, score is 4.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

No fitted numerical parameters appear. The central claim rests on domain assumptions about the adequacy of reused standards and vocabularies, on a Boolean modeling choice, and on trust in the HermiT and oops evaluations. No independently verifiable external benchmark is used.

assumptions (5)

standard math OWL DL description-logic semantics and the HermiT reasoner are sound for consistency checking.
Section 4 relies on HermiT for consistency and subsumption reasoning; this is standard unproved background for ontology tooling.
domain assumption ISO/DIS 13940:2024 ContSys concepts are a valid foundation for continuity of care semantics.
Sections 1 and 3 state the model combines ContSys and WHO SDH, but no independent validation of ContSys coverage is given.
domain assumption SOHO, ICD-11 chapter 24, the Social Prescribing Ontology, and the Gravity Project value set provide adequate SDH terminology.
Section 3.1 lists these as the four resources for concept collection, but no systematic coverage evaluation is presented.
ad hoc to paper Boolean encoding of SDH information preserves enough information for interoperability and clinical use.
Section 3.2 says SDH data is collected as true/false for privacy, but this loses granularity and is not justified against alternative encodings.
domain assumption DOLCE top-level ontology alignment remains appropriate for the extended model.
Section 2 refers to prior work aligned with DOLCE; the paper relies on that alignment without re-justifying it for SDH concepts.

how reviews work

0 comments

Cite this review

Pith. "Pith review of CSSDH: An Ontology for Social Determinants of Health to Operational Continuity of Care Data Interoperability." pith.science (2026). https://pith.science/paper/7USG5KE7

@misc{pith2026241209223,
  author       = {Pith},
  title        = {Pith review of: CSSDH: An Ontology for Social Determinants of Health to Operational Continuity of Care Data Interoperability},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7USG5KE7}},
  note         = {Machine review of arXiv:2412.09223}
}

read the original abstract

The rise of digital platforms has led to an increasing reliance on technology-driven, home-based healthcare solutions, enabling individuals to monitor their health and share information with healthcare professionals as needed. However, creating an efficient care plan management system requires more than just analyzing hospital summaries and Electronic Health Records (EHRs). Factors such as individual user needs and social determinants of health, including living conditions and the flow of healthcare information between different settings, must also be considered. Challenges in this complex healthcare network involve schema diversity (in EHRs, personal health records, etc.) and terminology diversity (e.g., ICD, SNOMED-CT) across ancillary healthcare operations. Establishing interoperability among various systems and applications is crucial, with the European Interoperability Framework (EIF) emphasizing the need for patient-centric access and control of healthcare data. In this paper, we propose an integrated ontological model, the Common Semantic Data Model for Social Determinants of Health (CSSDH), by combining ISO/DIS 13940:2024 ContSys with WHO Social Determinants of Health. CSSDH aims to achieve interoperability within the Continuity of Care Network.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

13 extracted references · 12 canonical work pages

[1]

Health Affairs 37(4), 585–590 (2018)

Cantor, M.N., Thorpe, L.: Integrating data on social determinants of health into electronic health records. Health Affairs 37(4), 585–590 (2018)

work page 2018
[2]

Journal of the American Medical Informatics Association 30(9), 1465–1473 (2023)

Dang, Y., Li, F., Hu, X., Keloth, V.K., Zhang, M., Fu, S., Amith, M.F., Fan, J.W., Du, J., Yu, E., et al.: Systematic design and data -driven evaluation of social deter - minants of health ontology (sdoho). Journal of the American Medical Informatics Association 30(9), 1465–1473 (2023)

work page 2023
[3]

In: pHealth 2021, pp

Das, S., Hussey, P.: Contsonto: A formal ontology for continuity of care. In: pHealth 2021, pp. 82–87. IOS Press (2021)

work page 2021
[4]

Journal of Personalized Medicine 13(7), 1024 (2023)

Das, S., Hussey, P.: Hl7 -fhir-based contsys formal ontology for enabling continuity of care data interoperability. Journal of Personalized Medicine 13(7), 1024 (2023)

work page 2023
[5]

IEEE Access 7, 91265 –91290 (2019)

Dhayne, H., Haque, R., Kilany, R., Taher, Y.: In search of big medical data inte - gration solutions -a comprehensive survey. IEEE Access 7, 91265 –91290 (2019)

work page 2019
[6]

In: International conference on knowledge engineering and knowledge management

Gangemi, A., Guarino, N., Masolo, C., Oltramari, A., Schneider, L.: Sweetening ontologies with dolce. In: International conference on knowledge engineering and knowledge management. pp. 166–181. Springer (2002)

work page 2002
[7]

In: Benchmarking —Theory and practice, pp

Gru¨ninger, M., Fox, M.S.: The role of competency questions in enterprise engineer- ing. In: Benchmarking —Theory and practice, pp. 22–31. Springer (1995)

work page 1995
[8]

Handbook on ontologies pp

Guarino, N., Welty, C.A.: An overview of ontoclean. Handbook on ontologies pp. 201–220 (2009)

2009

Show all 13 references

[9]

Journal of Medical Internet Research 22(12), e23721 (2020)

Jani, A., Liyanage, H., Okusi, C., Sherlock, J., Hoang, U., Ferreira, F., Yonova, I., de Lusignan, S.: Using an ontology to facilitate more accurate coding of social prescriptions addressing social determinants of health: feasibility study. Journal of Medical Internet Research...

2020
[10]

In: 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

Kollapally, N.M., Chen, Y., Xu, J., Geller, J.: An ontology for the social determi - nants of health domain. In: 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). pp. 2403–2410. IEEE (2022)

2022
[11]

American journal of public health 110(S2), S219–S221 (2020)

Meyer, D., Lerner, E., Phillips, A., Zumwalt, K.: Universal screening of social determinants of health at a large us academic medical center, 2018. American journal of public health 110(S2), S219–S221 (2020)

2020
[12]

World Health Organization (2024)

Organization, W.H.: Operational framework for monitoring social determinants of health equity. World Health Organization (2024)

2024
[13]

Informatics for Health and Social Care 46(3), 333 –344 (2021)

Pecoraro, F., Luzi, D., Pourabbas, E., Ricci, F.L.: A system of concepts to sup- port the integration of health and social care and assistive domotics services: The health@ home project. Informatics for Health and Social Care 46(3), 333 –344 (2021)

2021

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [1]

Health Affairs 37(4), 585–590 (2018)

Cantor, M.N., Thorpe, L.: Integrating data on social determinants of health into electronic health records. Health Affairs 37(4), 585–590 (2018)

work page 2018

[2] [2]

Journal of the American Medical Informatics Association 30(9), 1465–1473 (2023)

Dang, Y., Li, F., Hu, X., Keloth, V.K., Zhang, M., Fu, S., Amith, M.F., Fan, J.W., Du, J., Yu, E., et al.: Systematic design and data -driven evaluation of social deter - minants of health ontology (sdoho). Journal of the American Medical Informatics Association 30(9), 1465–1473 (2023)

work page 2023

[3] [3]

In: pHealth 2021, pp

Das, S., Hussey, P.: Contsonto: A formal ontology for continuity of care. In: pHealth 2021, pp. 82–87. IOS Press (2021)

work page 2021

[4] [4]

Journal of Personalized Medicine 13(7), 1024 (2023)

Das, S., Hussey, P.: Hl7 -fhir-based contsys formal ontology for enabling continuity of care data interoperability. Journal of Personalized Medicine 13(7), 1024 (2023)

work page 2023

[5] [5]

IEEE Access 7, 91265 –91290 (2019)

Dhayne, H., Haque, R., Kilany, R., Taher, Y.: In search of big medical data inte - gration solutions -a comprehensive survey. IEEE Access 7, 91265 –91290 (2019)

work page 2019

[6] [6]

In: International conference on knowledge engineering and knowledge management

Gangemi, A., Guarino, N., Masolo, C., Oltramari, A., Schneider, L.: Sweetening ontologies with dolce. In: International conference on knowledge engineering and knowledge management. pp. 166–181. Springer (2002)

work page 2002

[7] [7]

In: Benchmarking —Theory and practice, pp

Gru¨ninger, M., Fox, M.S.: The role of competency questions in enterprise engineer- ing. In: Benchmarking —Theory and practice, pp. 22–31. Springer (1995)

work page 1995

[8] [8]

Handbook on ontologies pp

Guarino, N., Welty, C.A.: An overview of ontoclean. Handbook on ontologies pp. 201–220 (2009)

2009