(FYI for folks on the public-semweb-lifesci list who are not
participating in the task force [1])
Hello,
Below I have pasted the minutes from the last meeting of the content
sub-team for the DDI Minimum Information Model task force [1]. I
apologize that I was forgot to record the meeting but I have tried to
write detailed minutes. The full set of minutes for the task force are
available [2].
1. https://sites.google.com/site/ddikrandir/home/ddi_info_model_taskforce
2. https://goo.gl/TKWfNN
------------
Minutes for 2/24/2016 (Content subgroup)
In Attendance : Dan Malone, Evan Draper, John Horn, Oktie Hassanzadeh,
Richard Boyce, Sam Habiel
Meeting recording: None taken (Rich forgot to start the recording, Sorry!)
Meeting:
*
Brief discussion of progress on the proposed task force activities:
1.
Clarify definitions for the seven minimum information items
recommended by prior AHRQ drug interaction conference series
*
A sub-team has been working on creating definitions. Initial
ideas have been developed and will be discussed over the next
month: https://goo.gl/N0MxPd
*
Some questions were raised by the sub-team that were discussed
later during the meeting.
2.
Propose additional attributes
*
Depends on Activity 1
3.
Create deliverablesusing an interesting and non-trivial set of
potential drug-drug interactions:
*
Depends on Activities 1 and 2
4.
Create a foundation for further collaborative workby disseminating
results through an interest group note, a website, and an online
discussion forum
*
Ratnesh Sahay has created a project that will be used for
advancing development of the interest group note:
https://github.com/W3C-HCLS/w3c-ddi
1.
A very initial outline can be seen here: http://goo.gl/ZMU1Wm
2.
All persons who have participated in the calls have been
listed as contributors for now (in alphabetical order by
first name). Please email Rich Boyce with spelling
corrections and the affiliation you would like (or if you
would like your name removed)
*
Scott Nelson submitted a panel proposal on 3/10 to AMIA to
discuss the topic of including seriousness as a minimum
information item
*
Discussion of what DDI content to include in our deliverables.
o
Sub-team members had made progress on suggesting interacting
pairs based on known issues with DDI evidence and knowledge
+
See https://goo.gl/rYpmjt
o
Rich created a survey to increase the participation:
https://pitt.co1.qualtrics.com/jfe/form/SV_0GPfzm7UbGDYV5X
+
Survey closes EOB Wednesday 3/30
o
It was decided that only one or two PDDIs will be selected for
each category. The group agreed that, for each PDDI, a decision
tree will be created that provides 1) information that can help
contextualize the interaction depending on drug and patient
factors, 2) management recommendations depending on the
contextualizing factors in (1), and 3) evidence supporting or
refuting each scientific claim made for (1)
+
Such decision trees have already been created by John Horn
and Phil Hansten for several interactions as part of Dan
Malone’s AHRQ grant (R21HS023826).
#
These can be used for the project as long as we provide
citation to the authors and grant.
#
These interactions might fit as representative for
certain information categories. The group agreed that,
for those cases, additional interactions will be chosen
to provide heterogeneity
*
Discussion of stakeholders
o
The question of who the stakeholders are for the min info model
came up from the Standard sub-team. The content sub-team agreed
that the stakeholders included primarily the following:
+
Clinicians (doctors, nurses, pharmacists, and various
practitioners) who must use PDDI information to make
clinical decisions
+
Persons who create systems that provide PDDI information to
clinicians in various settings (e.g., database vendors, drug
information editorial boards)
+
Organizations that provide publicly available lists of
drug-drug interactions (e.g., developers of RxNorm)
o
Clarification of this in writing is planned
*
Discussion of use cases:
o
Members of the standard sub-team had suggested using Medication
Reconciliation as a target use case for showing the value of the
info model (e.g., competency questions). Rich asked the content
sub-team for comment
+
It was agreed that med rec is an important use case but one
that can probably be broken down into a number of smaller
use cases depending on the phase of the medication therapy
process in question (e.g., ordering, consulting, monitoring)
o
Clarification of this in writing is planned and also a focused
discussion at our “all team” meeting in May.
*
Action Items
+
Rich will merge the final results of the survey on suggested
PDDIs
(https://pitt.co1.qualtrics.com/jfe/form/SV_0GPfzm7UbGDYV5X)
into the google sheet (https://goo.gl/rYpmjt)
+
He will then go through and highlight the PDDIs for which
decision trees have been created as part of Dan Malone’s
AHRQ grant
+ The next meeting will finalize the list and future meetings
will finalize decision trees for the chosen interactions. In
parallel, the standard sub-team will work on representing
the chosen interactions, seeking feedback as they go.
--
Richard D Boyce, PhD
Assistant Professor of Biomedical Informatics
Faculty, Center for Pharmaceutical Policy and Prescribing
Faculty, Geriatric Pharmaceutical Outcomes and Gero-Informatics Research and Training Program
University of Pittsburgh
rdb20@pitt.edu
Office: 412-648-9219
Twitter: @bhaapgh