Integrating clinical information from a personal health record into the virtual healthcare record
This article explores one of the most sensitive issues of a patient-centered electronic health record: the relationship between the information entered and managed directly by the patient in the Personal Health Record and the validated, longitudinal clinical information integrated into the Virtual HealthCare Record. It marks an important step in my research direction, because it explicitly links the concept of patient empowerment to the VHR architecture and attempts to resolve technically and semantically the transition between the patient's perspective and the healthcare professional's perspective.
Bibliographic Reference
Authors:
Luca Dan Șerbănați, Andrei Vasilățeanu, Mihai Isăroiu
Title:
Integrating clinical information from a personal health record into the virtual healthcare record
Published in:
Proceedings of HEALTHINF 2011
Year: 2011
Pages: 521–524
Type: conference paper
Article Figures
Thumbnails (click to enlarge)
Fig. 1. The CCR standard schema used in the Google Health-based demonstration.
Fig. 2. The validation flow for information entered in the PHR before its integration into the VHCR.
Fig. 3. PHR notification through a proxy that selects and translates events from the VHCR.
Conceptual Contribution
The article shifts the discussion about the Virtual HealthCare Record into a very concrete area: how can the patient be involved in the care process without compromising the clinical validity of the information.
Its main contributions are:
- explicitly links patient empowerment to the VHCR architecture;
- addresses interoperability in both directions: VHCR → PHR and PHR → VHCR;
- introduces a filtering and validation of data entered by the patient prior to clinical integration;
- proposes a semantic proxy that translates between different representations, for example HL7 and CCR;
- prepares the transition towards smarter mechanisms, based on adaptive agents and PHR-dedicated ontologies.
Description
The article starts from two ideas that, in the early 2010s, were beginning to gain more and more importance in e-health. The first was the consolidation of the Electronic Health Record as the central infrastructure for continuity of care. The second was the emergence of the patient empowerment concept, meaning the involvement of the patient as an active part in monitoring and managing their own health. From the meeting of these two lines arises the problem treated here: what do we do with the information produced and administered by the patient in a Personal Health Record and how do we integrate it into a longitudinal clinical record without losing medical rigor.
In the article, the Virtual HealthCare Record is presented again as a central, unifying, internet-based entity, capable of integrating distributed and heterogeneous sources of medical information into a longitudinal record. The VHCR is not an inert repository of documents, but a service provider that maintains a coherent view of the patient's clinical condition, using a model inspired by HL7 RIM v3 and European standards for continuity of care. Information is organized into episodes of care and is accessed through dedicated services.
The novelty brought by this article is the focus on the Personal Health Record. The PHR is described as the tool through which the patient can manage their clinical information, track treatments, consultations, medical conditions, self-monitoring data, and even diet and lifestyle-related elements. The article insists that this direction is not just a technical one, but a mindset shift: the patient must become a pro-active part of the care process, especially in the case of chronic diseases, where continuous observation and daily decisions matter a lot.
From this perspective, the relationship between VHCR and PHR is bidirectional. On the one hand, relevant information from the VHCR should reach the PHR, so the patient has a useful and intelligible picture of their own health condition. On the other hand, the PHR contains data that can be clinically important and should not be ignored. The problem is that these two worlds do not have the same logic. VHCR works with relatively formal medical representations, while the PHR rather expresses the patient's perspective, with different levels of health literacy and a varying degree of semantic precision.
The article discusses the two directions separately. In the VHCR → PHR direction, the main difficulty is translating from the service provider's perspective to the patient's perspective. The organization into health issues, episodes of care, and contacts can be too technical or less intuitive for the end-user, and the level of detail must be adapted to the health literacy degree of each patient. In other words, simply copying the data is not enough: it must be reorganized and reinterpreted into an intelligible form.
In the reverse direction, PHR → VHCR, the problem is even more delicate. Information entered by the patient cannot be automatically integrated into the clinical record, because, once signed and validated, it becomes the professional responsibility of a doctor. That is why the article very well describes the practical obstacles: the doctor may consider that filtering such data consumes unpaid time, may be reserved towards "e-visits", and may refuse to assume responsibility for unverified data. However, the authors also underline the reverse: ignoring relevant information entered by the patient can itself become dangerous for the patient.
The proposed solution is an elegant one and very representative of our research direction: a combination of service-oriented architecture, notifications, and semantic filtering. In the proof-of-concept demonstration, Google Health is used, which partially implemented the Continuity of Care Record (CCR) standard. The system adds a rule-based filtering subsystem on top of POS, capable of analyzing data consistency and managing notifications coming from the PHR. The flow is different for symptoms and for medication: symptoms can generate alerts for the doctor and possibly a new encounter, while medications are checked against known interactions with other drugs or with the patient's existing conditions.
This section is particularly important, because it shows how the problem of patient participation is translated into a realistic technical logic. The patient is not left completely outside the system, but neither is given the power to directly modify the validated clinical record. Between the two levels, a mediation and filtering mechanism is introduced, meaning an intermediate space where data meaning is verified, and the clinical decision remains with the professional.
In the VHCR → PHR direction, the article proposes a proxy service between the two worlds. The proxy registers to the notifications generated by the VHCR, selects which events should be sent to the PHR, and translates them from HL7 into CCR, the format interpretable by Google Health. It is a typically "intermediary" solution, very much in the spirit of my VHR research: it does not force the complete unification of all models, but builds controlled bridges between them.
The conclusion of the article is very well formulated. The PHR is seen as an important source of information for a truly comprehensive view of the patient, and the filtering system and semantic proxy are described as mediators between the patient's perspective and the professional's perspective. The authors already foresee the next steps: increasing the autonomy of the filtering mechanism through the use of intelligent, adaptive agents that learn from patient preferences and past choices, as well as extending the ontologies and vocabularies necessary for more types of PHR applications.
Viewed retrospectively, the article is important not only because it addresses a concrete problem, but also because it marks a significant opening of the VHR towards the patient's world. If our other works take the VHR towards agents, security, or digital ecosystems, this text takes it towards participation, co-production of information, and the reconciliation between two types of knowledge: medical knowledge and that lived by the patient.
Place in the Research Journey
Within my overall research in the Virtual Healthcare Record area, this article occupies an important place by explicitly opening the VHR to the direct contribution of the patient. It complements the articles about multi-agent architecture and organizational security with a different but essential problem: integrating the patient's perspective into a coherent clinical system.
At the same time, the article shows very well how two major directions of my research meet here: on the one hand, the interest in continuity of care and the longitudinal record; on the other hand, the idea that the patient should not remain a mere object of the record, but must become an active participant in the construction of their digital image.
For this reason, the work deserves to be clearly placed in the Virtual Healthcare Record branch, but also read as a step towards later directions related to adaptive agents, context awareness, and, in a broader sense, to increasingly rich forms of digital representation of the person in the health ecosystem.
Links
