PROV-Health

method for managing provenance data in health information systems

Abstract

Information Science and Computer Science share interdisciplinary aspects when it comes to Data Provenance between different Health Information Systems (HIS). HIS store large volumes of data in their decentralized repositories, making the management and interoperability of this data a challenge. Thus, Data Provenance associated with computer technologies contributes to the traceability of source data in HIS, in order to identify errors and transformations in the data and attribute them to the sources, being useful for decision-making in the most varied health scenarios. This thesis presents a method for managing provenance data in HIS based on instancing the World Wide Web Consortium Provenance (W3C PROV) model in accordance with interoperability aspects between different decentralized health databases. In order to achieve the objectives defined for this thesis, an applied research methodology was used in a case study, characterized as experimental and exploratory, with a quantitative and qualitative approach. As for the technical procedures for developing and validating the method, experiments were carried out in a real health scenario with real data, which made it possible to identify not only benefits, but also challenges and threats during experimentation. The method was evaluated via an electronic form sent to experts in the field, who highlighted important points for reflection on the structure of the method. The conclusion is that at the end of the experiments and the stages of analysis of the results, the proposed method proved to be significant in terms of managing provenance data in different HIS and can be adapted to any healthcare scenario.