- Electronic health records (EHRs) serve as repositories of documented data collected in a health care encounter. An EHR records information about who receives, who provides the health care and about the place where the encounter happens. We also observe additional elements relating to social relations in which the healthcare consumer is involved. To provide a consensus representation of common data and to enhance interoperability between different EHR repositories we have created a solution grounded in formal ontology. Here, we present how an ontology for the obstetric and neonatal domain deals with these general elements documented in health care encounters. Our goal is to promote the interoperability of information among EHRs created in different specialties. To develop our ontology, we used two main approaches: one based on ontological realism, the other based on the principles of the OBO Foundry, including reuse of reference ontologies.
INTRODUCTION
The motivation here grows out of the workings of the Stork
Network Program, a Brazilian Healthcare Program that aims to
provide comprehensive care of both woman and child during the
course of pregnancy and in the postpartum stage. The
organization of healthcare in Brazil involves health facilities at
different government levels each of which has certain autonomy.
These government levels are even free to adopt their own
electronic health record (EHR) information systems, which
creates challenges to data interoperability. To address some of
these challenges we are developing the Obstetric and Neonatal
Ontology (OntONeo) [
The scope of OntONeo was defined from a set of Brazilian and international EHR standards, for instance: the Woman’s Health Record and Antepartum Record and Postpartum Form provided by the American College of Obstetricians and Gynecologists (ACOG); the Children’s Electronic Health Record Format provided by the Agency for Healthcare Research and Quality (AHRQ). We also conducted interviews with Brazilian and American obstetricians to identify the workflow of the women's health clinic and the information needs of this medical practice. In the future, we will conduct interviews also with pediatricians to identify their needs.
The course of pregnancy, childbirth and child development involves a series of stages referred to as the prenatal, intrapartum and postnatal periods of care [2; 7]. Clinical care in each of these stages calls upon different medical specialties. Information pertaining to all stages of care is recorded in specific EHRs according to their specialty of origin.
EHRs across different medical specialties need information about elements observed in medical encounters. Each such encounter necessarily involves three elements: first, the location where clinical activity is performed, named health care facility; second, who performs or serves as the provider of health care, named the human health care provider; and third, who participates in an event as recipient of health care (consumers), named the human health care consumer.
In addition, during the health care encounter a provider
collects demographic data from a consumer. The demographic
data section of an EHR is more than just data to allow the
identification and categorization of the consumer [
Our representation of the demographic data section of the
EHR and the elements of the medical encounter are quite
generic. We follow the practice of the Ontology for General
Medical Science (OGMS) which deal with general terms of
medicine used across medical disciplines [
This paper describes the OntONeo approach to deal with the elements present in medical appointments: the consumer and the provider in a medical encounter, the place of the health care facility, and the social relations of the consumer that are important to document. We are extending OGMS and OMRSE to the specific case of pregnancy. (OGMS) and Ontology of Medically Related Social Entities (OMRSE).
To build OntONeo we adopt the principals of ontological
realism to foster semantic coherence, both for humans and
computers [
The top-level ontology used as starting point for OntONeo
is the Basic Formal Ontology 2.0 [
Fig. 1 and 2 show a sketch of the elements documented in a health care encounter. Terms in black represent entities reused from BFO, terms in gray are taken over from other ontologies; and terms in white are new entities of OntONeo. The term health care encounter is defined by OGMS ontology as a process in which at least one human being participates. Then we have in OMRSE ontology the term Homo sapiens that was reused from NCBI organismal. IV. CONCLUSION AND REMARKS
This paper described how OntONeo deals with the elements involved in a health care encounter. These elements defined by OntONeo has the potential to contribute to interoperability of information among EHR from different specialties. In addition, these definitions facilitate the understanding of how such information can be organized in EHRs for purposes of healthcare.
OntONeo is an on-going project in the early stages of development, and the current version is available on the web at http://ontoneo.wordpress.com. Its worth to mention that this work is justified above all by the lack of formal representations in the obstetric and neonatal domains.
ACKNOWLEDGMENT
We thank CAPES and CNPq for financial support to the primary author. This work is also supported in part by the NIH NCATS under CTSA award Number UL1TR001412.