Biomedical ontologies are consensus-based controlled vocabularies of terms and relations with associated definitions, which are logically formulated to promote automated reasoning. They thus go further than the systems just mentioned in providing support for computational analysis of data. The Ontology of Adverse Events (OAE) is a community-based biomedical ontology in the field of adverse events.
A medical intervention is a medical procedure or application intended to relieve or prevent illness or injury. The medical intervention can be an administration of a drug, a vaccine, a special nutritional product (for example, medical food), or it can be the use of a medical device. As a result of a medical intervention, events may occur which lie outside the intended consequences of the intervention. Some of these events are adverse events, in the sense that they are pathological bodily processes. For many such adverse events it is unclear whether they are causal consequences of the medical intervention which preceded them, since adverse events may also occur due to other reasons (for example a natural viral infection). Current adverse event representation systems, such as MedDRA (http://www.meddramsso.com/), and CTCAE http://ctep.cancer.gov/reporting/ctc.html), focus on the representation of symptoms or diseases which are the outcomes of adverse events of various sorts. These systems thus do not represent the whole process which goes from initial medical intervention to subsequent outcomes. OAE seeks to rectify this shortfall.
Dr. Yongqun "Oliver" He, an associate professor from the University of Michigan Medical School, developed the first version of OAE (previously named Adverse Event Ontology or AEO). Oliver was funded by a NIH-NIAID R01 grant (#1R01AI081062) in 2009 to develop the Vaccine Ontology (VO). As a part of the VO development, Oliver initially generated a few hundred VO terms to represent the vaccine adverse events commonly reported in the VAERS system. Later it was recommended that these VO adverse event terms might be general and applicable for other adverse events. Then Oliver created an Adverse Event Ontology AEO using OWL and moved those VO adverse event terms to AEO. Later, Drs. Werner Ceusters and Luca Toldo joined the discussion and development o f the AEO.
The first paper of introducing the OAE system (with the previous nam AEO) was presented by Oliver He in the 2011 Adverse Event (AE) Workshop in the International Conference in Biomedical Ontology (ICBO) in 2011. Oliver's presentation received a lot of attentions and constructive feedback. In this paper, we defined an adverse event as a pathological bodily process that is induced by a medical intervention. OAE also first defines the term 'medical intervention' (OAE_0000002). One novelity of our system is to treat adverse event as a process instead of an outcome as described above. Our paper provides detailed introduction on this whole process and introduces an example as well. This AEO adverse event assumes causal association between a medical intervention and the pathological bodily process outcome. One feedback is that due to many reasons and in most cases, it is difficult to identify the causal assocition. In addition, the causal assumption is not consistent with the definition of the term in VAERS and FAERS. Therefore, after the 2011 AE workshop, we extended the definition of 'adverse event' (OAE_0000001) to include those adverse bodily events that occur after a medical intervention and may not be caused by the medical intervention. One goal of our study is to identify the causal association or the probabiity of the causal association. In the VDOSME 2012 workshop, Oliver He presented an updated version of the OAE (see Slides).
The first use case of OAE was its application on the analysis of adverse events associated with killed or live attenuated influenza vaccines. Ms. Sirarat (Sira) Sarntivijai worked on this topic as one of her PhD thesis research projects in 2010-2012. Sira's PhD research was co-mentored by Drs. Brian Athey and Oliver He. Sira's research resulted in the development of the novel strategy called combinatorial, ontology-based detection of AEs (CODAE). Using this study case as an example, CODAE uses a series of combinatorial bioinformatics methods to identify signficantly associated MedDRA-vaccine term pairs from the VAERS data. The MedDRA is the terminology system used in VAERS. After mapping the MedDRA terms to OAE terms, we will be able to use the OAE ontology term hierarchy and axioms to classify the significantly enriched adverse events associated with killed or live attenuated influenza vaccines. Many interesting findings have been observed in this study. The results have been published in a PLoS ONe paper (URL: http://dx.plos.org/10.1371/journal.pone.0049941). After graduation, Sira becomes a postdoctoral research fellow in Dr. Darrell R. Abernethy's laboratory in USA FDA. The CODAE approach is being used in Sira's current study associated with drug adverse events (or called adverse drug reactions).
OAE is also being used in many other areas. For example, OAE is being used for literature mining by different groups internationally now. OAE can also be used to support termporal analysis of the processes in between a medical intervention and a display of a symptom or other pathological bodily outcomes. Different variables may affect the adverse event outcome. It is possible to use OAE to model different intermediate processes and analyze them by combining clinical data or literature reports.
The OAE will be continuously improved. International collaboration is required to ensure its continuous success. The OAE development and application team is becoming bigger. Your participation is more than welcome. Thank you!