RESUMO
Clinical Practice Guidelines (CPGs) contain a set of schematic plans for the treatment and management of patients who have a particular clinical condition. CPGs are increasingly being used to support physician decision making. Many groups develop tools for the representation of CPGs. These differ in their approaches to addressing particular modeling challenges. Despite this strong effort, physicians still primarily rely on free-text narrative descriptions. Thus, a core challenge is to develop a formal representation of CPGs that physicians can easily read and verify, yet a machine can process, analyze and apply directly to a patient's EHR data. Our paper proposes a solution to this fundamental problem by describing an approach to CPG formalization using the Natural Rule Language (NRL), coupled with transformation to Object Constraint Language (OCL) constraints that are applied on a patient's clinical data record, in our case an HL7 Continuity of Care Document (CCD). We illustrate our approach on a simple guideline directive for Essential Hypertension.
Assuntos
Algoritmos , Sistemas de Apoio a Decisões Clínicas/normas , Modelos Teóricos , Guias de Prática Clínica como Assunto , Garantia da Qualidade dos Cuidados de Saúde/normas , Software , Terminologia como Assunto , Processamento de Linguagem Natural , Padrões de ReferênciaRESUMO
A core challenge in biomedical data integration is to enable semantic interoperability between its various stakeholders as well as other interested parties. Promoting the adoption of worldwide accepted information standards along with common controlled terminologies is the right path to achieve this. Our paper describes a solution to this fundamental problem by proposing an approach to semantic data integration based on information models serving as a common language to represent health data coupled with technology that is able to represent the data semantics. We used the HL7 v3 Reference Information Model (RIM) [1] to derive a specific data model for the integrated data, the Web Ontology Language (OWL) [2] to build an ontology that harmonizes the metadata from the disparate data sources, the Unified Modeling Language (UML) [3] to model the data representation, and the Object Constraint Language (OCL) [4] to specify UML model constraints. To illustrate the approach, we use the Essential Hypertension Summary CDA document and related models from Hypergenes, a European Commission funded project [5] exploring the Essential Hypertension disease model.