Re: [Oae-devel] Name space change to OAE.

[He, Y. <yon...@me...>]

Hi Melanie,

Good discussion. I have added an editor's note about the time associated as you described. Sorry that I forget to add this part before. Basically, I think it is associated with 'adverse event incubation time', an OAE term I coined based on the 'incubation time' from an infectious disease. Practically, when we assign some AEs to a medical intervention for which AE information is not available, we (a clinician or a patient or a researcher) tend to think that potentially (even with low chance) this AE is caused by the medical intervention. Therefore, there is a possibility involved in the assignment process. This is what I tried to convert with my previous version of AE definition. We  can rethink about this if possible.

For the 'side effect' definition, I'm still not convinced that side effect is not an alternative term for adverse event. I think the wiki definition is right in a sense. However, we can discuss more about this later. For now, I have removed the term as an alternative term in OAE. Instead, I added an editor's note about this issue:

" There has been discussion regarding whether the term 'side effect' is an alternative term for 'adverse event'. In AERO, the term 'AERO:adverse event' represents a subset of those adverse events for which causality has been established. In OAE, an adverse event for which causality has been established is called 'causal adverse event'."

I hope that these make sense.

Also, as suggested by Luca Toldo from Merck KGaA, Germany, I have added to OAE four causal adverse event terms due to the following factors:

A) drug-drug interactions and

B) food-drug interactions

C) genetic predisposition (eg SNP)
Note: SNP is a subtype of the 'genetic predisposition'.
Note: Since Luca is from a company, sometimes he emails me directly instead of to the email list due to some company policy.

Thanks!
Oliver


From: Melanie Courtot [mailto:mco...@gm...]
Sent: Friday, November 18, 2011 3:20 PM
To: He, Yongqun
Cc: Xiang, Allen; oae...@li...
Subject: Re: [Oae-devel] Name space change to OAE. RE: [obo-admin] Request the addition of Adverse Event Ontology (AEO) into OBO foundry library

Hi Oliver,

Please see my comments inline.

On 2011-11-18, at 11:47 AM, He, Yongqun wrote:


Hi Melanie,

Time is for sure important for adverse events. The importance of time is also indicated by the current definition of 'adverse event'. The vaccine injury table provides some good information. I have previously generated many AE terms specific to different vaccines (e.g., many killed or live flu vaccines). I envision that a large knowledgebase of vaccine- and drug-specific AEs would help a lot to the AE research community.

In my opinion, the current definition of adverse event doesn't reflect the importance of time, as it reads "a pathological bodily process that occurs after a medical intervention." When we discussed how to reconcile AERO and AEO, we agreed that we would had an editor note to that effect, and  that it was important to work on adding time appropriately in that definition. The editor note reads "Some work is needed on how to restrict the scope of a term to be an 'adverse event', notably regarding temporal association. When is an appropirate time interval between a medical intervention and a adverse event observed? One week, one month, one year, or a lifetime? For some well studied medical interventions (e.g., many vaccines or drugs), we probably have a general idea. For some new intervention, we probably don't know much." I am wondering how we could work towards integrating this more formally in the above mentioned definition.


I am not sure if I agree with you on the difference between "side effect" for "adverse event". On website: http://en.wikipedia.org/wiki/Adverse_effect
"An adverse effect may be termed a "side effect", when judged to be secondary to a main or therapeutic effect."
On VAERS website: http://vaers.hhs.gov/index
"VAERS is a post-marketing safety surveillance program, collecting information about adverse events (possible side effects) that occur after the administration of vaccines licensed for use in the United States."
All these references indicate that "side effect" and "adverse event" are synonyms.

I don't have the same reading of those 2 sources.

On the page you cite, http://en.wikipedia.org/wiki/Adverse_effect, the full citation reads: "An adverse effect may be termed a "side effect", when judged to be secondary to a main or therapeutic effect. If it results from an unsuitable or incorrect dosage or procedure, this is called a medical error and not a complication. Adverse effects are sometimes referred to as "iatrogenic" because they are generated by a physician/treatment. Some adverse effects only occur only when starting, increasing or discontinuing a treatment." Note that they also deem adverse effects to be iatrogenic, which is certainly not desired for our adverse event case.
I would like to add that while wikipedia is a useful source of information, I don't think it it the best one for very specific (and potentially controversial) topics - for example there are pages for adverse effect and adverse event, with suggestion of merging (or not). http://en.wikipedia.org/wiki/Adverse_event also refers specifically to clinical trials, whereas our scope is clearly larger.

VAERS states indeed that adverse events are possible side effects, which matches the AERO definition: adverse events are those such as reported in the VAERS system, and side effect are a subset of those adverse events for which causality has been established. You may remember that in our past discussions I had mentioned that a bridge between AERO-AEO could be done by asserting equivalence between AERO:side effect and AEO:adverse event (note that I refer to AEO:adverse event, i.e. the old version in which causality was asserted)



It's good that you will develop the causality assessment part within AERO as suitable. I believe that the causality assessment is important for both AERO and OAE. As I also mentioned in a couple of my previous emails, we will provide knowledge-oriented representation of AE causality assessment in OAE. Indeed, we have generated many such terms in OAE out of our collaborations with Dr. Richard Boyce and others. Our current focus is on causality assessment of adverse drug events. For example, we have represented the 'Naranjo ADR probability score' and 'DIPS score' for causal assessment of ADEs out of the reactions to drug and drug-drug interaction, respectively. Sira's PhD study focuses on making statistic association between AEs and different type of vaccines based on MedDRA annotation of VAERS data. We have developed a novel combinatorial biostatistical/bioinformatics pipeline to assess the significance of vaccine-AE association. Sira's paper is currently at the final stage of preparation. I guess that your research focus is on ontology representation of causality assessment of vaccine adverse events (VAE) and VAE data reporting based on Brighton's definitions. Whenever possible, let's communicate more on this topic. I have no problem of transferring some existing OAE terms (e.g., a current OAE term called 'adverse event reporting') to AERO if appropriate and needed. More people working on this important AE representation area will make this study progress further and faster.

Sure, happy to discuss any of those further. I have been busy setting up our Brighton working group and am now able to dedicate more time to the AERO editing - feel free to submit reporting terms directly under the tracker at http://code.google.com/p/adverse-event-reporting-ontology/issues/list and I'll add them asap.



I will soon add the annotation you suggested to the AE definition in OAE, "adverse reaction is more specifically used by the drug ( in the sense "drug and not vaccine") community".

I would suggest to add it to the "adverse reaction" annotation rather than to the adverse event class itself to make it clearer.

Thanks,
Melanie



Thanks!
Oliver


**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues