Organized alongside the French Archivists Association (AAF https://www.archivistes.org/) national workshop, on the theme “The Hospital Archivist and the Challenges of Digitalization: Issues, Opportunities, and Prospects (https://www.archivistes.org/larchiviste-hospitalier-et-les-defis-du-numeriques-enjeux-opportunites-et-perspectives/) , this interdisciplinary conference critically explores the electronic medical record and its impact on the generation, preservation and reuse of hospital health data. It interrogates digitization’s implications for record management and knowledge construction across medical, historical, communication, political and legal domains, and invites participants from varied national contexts.
Data collection must serve multiple objectives. It should document the patient’s hospital journey and support coordinated care. In addition, it depends on a digital application that produces quality indicators and feeds information systems and interoperable data repositories. In France, the transfer of data to the National Health Data System (SNDS) and, by extension, to the Health Data Platform (PDS) (https://www.snds.gouv.fr/SNDS/Accueil) underscores the strategic value of these data for medical research, health-policy development, and the progress of artificial intelligence.
The digitization of health information by healthcare professionals takes many forms, one of which for hospitals is the Electronic Medical Record (EMR), defined by the French Digital Health Agency as a function of a Hospital Information System (HIS) designed to store all documents related to a patient’s care pathway within the facility. It therefore contains the various documents needed to populate the EMR and to transmit them via MSSanté (Secure Health Messaging). The EMR centralizes all health data for patients treated within a healthcare facility and, as noted by the CNIL (National Commission for Information Technology and Civil Liberties): “Given the sensitivity and volume of the data it contains (consultation and hospital stay records, biological or radiological tests, medical prescriptions, etc.), the EMR must be subject to enhanced security measures” regarding access to this data. What is called “health data” is very diverse in content. The definition is broad according to the European General Data Protection Regulation: “data relating to the past, present, or future physical or mental health of a natural person (including the provision of health care services) that reveals information about that person’s health status” (CNIL : https://www.cnil.fr/fr/quest-ce-ce-quune-donnee-de-sante ), noting that health data “by nature,” data through cross-referencing with other data, or health data “by virtue of their intended use,” meaning due to their medical application (CNIL, op. cit.). They are characterized by how they are generated: within the EMR, at the hospital, during patient care, through healthcare professionals' work. However, the EMR is not just a collection of all documents related to patient care, each with a legally mandated retention period. It serves as a platform for two main types of data collection: evidence of care activities and information transfer of information among healthcare providers and to the patient (in part). It must fulfill the dual role of a medical record: “the traceability of all actions performed” and a “tool for communication, coordination, and information sharing among healthcare providers and with patients” (https://www.has-sante.fr/jcms/c_438115/fr/dossier-du-patient ). This raises the question of what should be retained, as there may be tension between these two functions. This tension between legal retention and memory relates to the two functions of the archive: evidence and historical record.
This tension is growing as digitization of records and documents becomes a reality in healthcare, just as it the case for all fields of knowledge and social life (Millerand et al. 2025). The EMR is likely to feed into data warehouses at various levels. According to the definition on the French government website “e-sante”: “A health data warehouse (HDW) is a structure that enables the collection of a large volume of data (data related to patient medical care, sociodemographic data, data from previous research, disease registries, etc.) intended to be reused primarily for management purposes (management, control, and administration of activities) and for research, and evaluations in the health sector” (https://esante.gouv.fr/doctrine/eds ). The reuse of data places the EMR at the center of building medical and managerial knowledge that it will enable. That is why the archivist must be involved in the process of creating the EMR. The archivist’s role cannot be limited to merely receiving documents accumulated during care.
However, the digitization of patient records and their preservation as digital archives fall outside the scope of French hospital archivists, who are restricted to preserving paper documents (from the pre-digital or hybrid era of patient records). The AAF has warned about the consequences of this situation, such as highlighting issues related to data migration during software updates—problems that expose the limitations of the current design of digitized records: “Every day, millions of health records are entered into the information systems of healthcare facilities (hospitals, clinics, etc.). However, when software is updated, this data is often not transferred due to technical or budget constraints. ‘Locked’ within their legacy software, which is no longer supported, this data risks being lost forever,” explains an AAF press release from December 2025 (https://www.archivistes.org/dossier-patient-informatise-vos-donnees-de-sante-sont-en-danger/ published on December 4, 2025). The AAF also urges the integration of archivists’ expertise, as they are specialists in data management. Moreover, confusion between storing and preserving digital data prevents any archival considerations from the start in the design of digitized patient records.
Behind this situation, in which the EMR is seen as a technical and medical issue involving healthcare professionals and IT departments, the very concept of the “medical record” is undergoing a transformation and needs to be examined. What kind of health record can the EMR create under these conditions? Data is meant to be shared and integrated into information systems for further processing outside the original context. What happens to the patient’s history? How can we balance the individual level with the “datafication” needed to support these systems and the promises of artificial intelligence?
The conference aims to examine the effects of the datafication of health on knowledge production in this field, particularly (though not exclusively):
- What kind of knowledge does health data yield ?
- While “the need to establish robust data capture and exchange infrastructures capable of aggregating information from highly heterogeneous systems, such as medical data, patient data, and information related to hospital services and facilities” (ANRT, One Health, 2025) is increasingly recognized, how does the production of this data influence the knowledge we can develop? Across all sciences, information and knowledge infrastructures are studied both as technical systems that enable large-scale data sharing and as an evolution of scientific practice, a subject of research in the humanities and social sciences. What sets health apart in this shared perspective across all sciences?
- How is the patient portrayed in the EMR? What are the implications for choosing which documents to include and which data to extract?
- From paper records to the EHR, what are the effects of the “rematerialization” of data on screens via applications, forms, software, and interfaces?
- How should we approach data quality and its purposes of for reuse?
- What kind of health history is possible in the age of data? How can we write a history based on incomplete data i shaped by clinical practices and institutional constraints?
- Should we develop new healthcare data professions or adapt existing ones? What information and data roles are present along the “data journeys” (Bates et al. 2007) in medical practice?
- What lessons can be learned from international comparisons about digitizing patient records in the European Health Data Space era? How should health data be shared across nations?
References
- Anderson S., Blanke T. , (2015),"Infrastructure as Intermeditation - From Archives to Research Infrastructures", Journal of Documentation, Vol. 71 Iss 6.
- Association Nationale de la Recherche et de la Technologie, Veroye JP dir. Du groupe de travail (2025), « One Health. Enjeux du monitoring au service de la santé et du vivant. Comprendre, anticiper, décider ». Accessible à partir de : https://www.anrt.asso.fr/fr/rapport-one-health-enjeux-du-monitoring-au-service-de-la-sante-et-du-vivant-37294
- Assailly. Laurène (2024) « La santé des données : enquête sur un travail hospitalier de production, de contrôle et de mise en accès de données de santé face aux stratégies d’action publique pour les valoriser ». Thèse Science politique. Université de Strasbourg, 2024 https://theses.fr/2024STRAG012 HAL Id: tel-04984368 https://hal.science/tel-04984368v1
- Borgman C.L. (2000) From Gutenberg to The Global Information Infrastructures. MIT Press.
- Bowker G. C., Baker K. S., Millerand F., Ribes D., 2010. « Toward Information Infrastructure Studies: Ways of Knowing in a Networked Environmen »t, in J. Hunsinger, L. Klastrup, J. M. Allen (eds.), International Handbook of Internet Research, Londres, Springer, 97-118.
- Bates J, Lin Y & Goodale P (2016) « Data journeys: Capturing the socio-material constitution of data objects and flows ». Big Data and Society, 3 (2).https://doi.org/10.1177/2053951716654502
- Bucki, Elaine (2022), « La dématérialisation en établissement de santé : analyses juridiques ». Thèse de doctorat, Université de Lorraine. https://theses.fr/2022LORR0303
- Edwards, P. N., Jackson, S. J., Chalmers, M. K., Bowker, G. C., Borgman, C. L., Ribes, D., Burton, M., & Calvert, S. (2013) Knowledge Infrastructures: Intellectual Frameworks and Research Challenges. Ann Arbor: Deep Blue. http://hdl.handle.net/2027.42/97552
- Edwards Paul N, Mayernik Matthew S. Batcheller Archer L., Bowker Geoffrey C., Borgman Christine L. (2011) “Science friction: Data, metadata, and collaboration”. Social Studies of Science 41(5) p.667–690
- Erard F. (2022) « Le dossier électronique du patient : révolution ou désillusion ? » In : Métille S. (édit.), L’informatique en nuage, Berne 2022, p. 219 ss
- Favier L. (2015) « Les Humanités numériques et l’évolution des infrastructures de recherche : quels enjeux pour l’organisation des connaissances ? » 10ème Colloque ISKO 2015, LISEC (EA 2310), Nov 2015, Strasbourg, France. pp.3-16. ⟨hal-01643159⟩
- Gekiere, C., & Soudan, S. (2015). « Dossier patient informatisé et confidentialité : évolution des modèles et des pratiques. « Le diable gît dans les détails » ». L’Information psychiatrique, 91(4), 323-330. https://doi.org/10.1684/ipe.2015.1336
- Gekiere Claire, « Écrire ou saisir, peut-on choisir? », Empan, 2018/2 (n° 110), p. 64-67. DOI : 10.3917/empa.110.0064. URL : https://www.cairn.info/revue-empan-2018-2-page 64.htm
- Guillemain H. et Hanafi N., « Pour une histoire des données médicales (xviie-xxie siècle) », Histoire, médecine et santé, 2022, no 22, p. 31‑46.
- Hochlaf, A. (2021). « L’usage du dossier patient informatisé dans un hôpital public : une approche qualitative sur les affordances et les contraintes ». Journal de gestion et d'économie de la santé, 6(6), 347-363. https://doi.org/10.54695/jges.216.0347 5
- Juven, P-A. (2013) « Produire l’information hospitalière: Du codage des dossiers au gouvernement de l’activité ». Revue d’anthropologie des connaissances, vol. 7, no 4, décembre 2013. DOI.org (Crossref), https://doi.org/10.3917/rac.021.0815.
- Le Bonhomme, F. (2022). Le patient dit « qu’il ne peut pas penser de manière marxiste » Faire des dossiers psychiatriques des sources d’histoire sociale (RDA, Berlin-Est, années 1960) Genèses, 126(1), 34-55. https://doi.org/10.3917/gen.126.0034
- Leigh Star Susan,(1999) « The Ethnography of Infrastructure », American Behavioral Scientist, 1999, vol. 43, no 3, p. 377‑391.
- Lenay, Olivier, et Jean-Claude Moisdon (2003) « Du système d’information médicalisée à la tarification à l’activité. Trajectoire d’un instrument de gestion du système hospitalier ». Revue française de gestion, vol. 29, no146, octobre 2003, p. 131-41. https://doi.org/10.3166/rfg.146.131-141
- Mayère, A., & Grosjean, S. (2016). L’« échafaudage de soins » dans un hôpital: Un entrelacement de technologies, de conversations et d’écrits multiples. Communication, vol. 34/1. https://doi.org/10.4000/communication.6719
- Mathieu-Fritz, A., et Laurence E. (2013)« Les médecins et le dossier santé informatisé communiquant: Analyse d’une expérimentation du dossier médical personnel (DMP) ». Réseaux, n° 178-179, no 2, juillet 2013, p. 223-55. DOI.org (Crossref), https://doi.org/10.3917/res.178.0223.
- Mayère, A., Bazet, I. et Roux, A. (2012). « Zéro papier » et « pense-bêtes » à l'aune de l'informatisation du dossier de soins. Revue d'anthropologie des connaissances, . 6, n° 1(1), 115-139. https://doi.org/10.3917/rac.015.0154
- Millerand F, Delias L et Millette M. ( « La datafication de la société », Communication [En ligne], Vol. 42/2 | 2025, mis en ligne le 01 décembre 2025, consulté le 02 mars 2026. URL : http://journals.openedition.org/co)mmunication/21331 ; DOI : https://doi.org/10.4000/15bo2
- Odier, B. (2007). Du PMSI à la T2a, de la fiche par patient au RIMpsy : la grande peur de la VAP. L’Information psychiatrique, 83(7), 539-549. https://doi.org/10.1684/ipe.2007.0226
- Tolonen HM, Kaukovuori J, Airaksinen M, Holmström AR (2025). « Are electronic health record big data ready for secondary use in research? Exploring potential limitations with opioids as a case study ». Health Informatics J. 2025 Jul-Sep;31(3). Doi:10.1177/14604582251363501.
-