General Principle Subteam Minutes
February 2011
Subgroup Meeting Minutes
General Principles (Feb 17, 2011)
Attendees: Susan Duke (Chair), Mat Soukup, Mary Banach, Richard Forshee, Larry Gould, Qi Jiang, Fabrice Bancken, Brenda Crowe, Andreas Krause.
Meeting Purpose: (1) Demo proof of concept of the public wiki (Mary), (2) get feedback from team on the first conference presentation (Fabrice), (3) discuss actions needed and date for launching the public wiki (Mat).
Key Points:
- The team had helpful feedback for Mary's proof of concept, which is now ready to be demo'd to the full group. There has been a lot of work to get this far, especially by Mary, Mat, and a number of folks from Vanderbilt. Link to (eventual) public site: http://www.ctspedia.org/do/view/CTSpedia/StatGraphHome
- Fabrice received some useful feedback on his slides for the first conference. It was also recognized by many on the call that this first one will be helpful for subsequent conferences, as well as for what gets published on the wiki.
- The focus of Mat's topic was the classes of graphs. These will be needed prior to launch. It was agreed these should be completed by end of March. A launch date for the public wiki itself was not discussed, as we ran out of time. So far, five people have volunteered to create some of the other classes needed. Other volunteers are welcomed! Link to table of class graphs needed: http://www.ctspedia.org/do/view/CTSpedia/StatGraphHome#SeeBelow
October 2010
Meeting Date/Time: October 19, 2010, 11-12 EDT
Present: Rich, Mary, Mat, Fabrice, Brenda, Andreas, Frank, Susan, Qi
Actions:
- Set up next month's General Principles meeting - Susan
- Set up meeting times for 2011 - Susan
- Set up offline meeting with Mary, Mat, Rich and Susan to be held prior to the next monthly meeting - Susan
- Create initial draft of the glossary - Susan
- Review Tags, Hyperlinks, etc information before next meeting (see agenda item 4 below) before next meeting - Team
- Review Giving Credit information from another CTSpedia group (see agenda item 5 below) before next meeting - Team
Agenda:
1. Welcome new team member, Andreas Krause
Andreas introduced himself, as did everyone on the team briefly.
2. The categorical flow chart page - Fabrice
Fabrice presented his concepts on graphing qualitative data, along with reference to the work he'd already completed. [Choice_of_graph_rules_20101019.ppt]]
He also described how these pages and the context pages would be laid out. [VisioProcess_ChooseGraph_20101019.pdf]]
There was general consensus that Fabrice's work achieved the goals very well. Mary commented that she has already used it in another setting (secondary school teachers) with good results (they wanted to know when the whole thing will be ready). The team commended Fabrice for his fine work.
If there is any feedback please send it to Fabrice and/or Susan.
3. Template for context (intermediate) page - Rich
Rich walked through the context page template he created using histograms as an example. There was general consensus that Rich's work achieved the goals for this page very well.
Rich made mention that terms such as 'mouseover' will need to be included in the glossary. The team discussed two different ways one will be able to navigate to the context page - either from the flowchart (as laid out by Fabrice) or from the glossary (which is an action still to be completed by Susan).
Mary recommended that a form be used for the context page. We agreed that an offline meeting would be needed to better understand how this will work. Susan will set up a TC with Mary, Mat, Rich and Susan prior to the next General Principles meeting.
Concept_page_example_Histograms.doc
4. Tags, forms, hyperlinking - Mary
We ran short of time, so Mary suggested that we review the links below prior to the next meeting. She encouraged the team to create tags wherever they wish.
5. Giving credit example from research ethics on CTSpedia - Mary
We ran short of time, so Mary suggested that we review the links below prior to the next meeting.
6. Putting this info on the public part of the wiki - Mat
Mat suggested that we plan to put the work described today on the public wiki and continue to flesh it out from there. Will discuss in the offline meeting how to do that.
The team agreed on these next steps:
- Confirm the structure/content of the pieces described by Fabrice and Rich in the offline meeting
- At the next meeting, ask for volunteers to create context pages for the other graph types needed.
August 2010
Meeting Date/Time: August 16, 2010, 2-3pm EDT Present: Rich Forshee, Larry Gould, Markus Yap, Fabrice Bancken, Qi Jiang, Mary Banach, Frank Harrell, Susan Duke
Apologies: Mat Soukup, Brenda Crowe
Agenda:
1. Review of Choosing the Right Graph document - 40 mins
a. Does it include all graph types of interest?
2. Do we want to have something ready for full team by Aug 31? - 10 mins
2.1 added topic - volunteers for presentations at upcoming meetings
3. Glossary - 10 mins
Actions:
Choosing the Right Graph flow chart, based on Fabrice's & Rich's email comments and comments from today's meeting. The result of this further work is to be recommended for presentation at the next full team meeting on 31 Aug.
2. Susan to communicate to Bruce what the team is comfortable presenting at the Aug 31 meeting (see item 2 in minutes, below).
Minutes:
1. There was quite good discussion on content for Choosing the Right Graph flowchart. The team walked through Fabrice's & Rich's email comments, which generated further productive discussion and agreement: Audience: people who are interested in making an effective graph (more likely to be clinicians than statisticians but can be used by both.
Linkage with glossary: Graph types referred to in both glossary and this flowchart will reference the same hyperlinked location (eg, the line plot graph entry will hyperlink to both the flowchart and the glossary).
Hyperlinks will be two-levels deep for families of graphs. For example, there are many types of boxplots, so the initial boxplot hyperlink from the flowchart would describe the various types of boxplots. The individual boxplots described may have their own specific graphics entries.
Non-recommended graphs on the flowchart can remain on the flowchart. The hyperlink to the pie chart entry can link to a page that describes why the graph is not recommended and suggestions for what to use instead.
Several graph types will be added to the flowchart based on suggestions from Rich (Frank Harrell's extended boxplots, violin plots, sunflower plots) and Fabrice (hexbin, hazard function). These will be added to the flowchart (extended & violin under boxplots, sunflower and hexbin plots under scatterplots, hazard function under survival).
Some flowchart elements refer to types of group, not types of graphs. Fabrice & Susan will consider how to handle this as part of their action and bring back to the team.
1a. Larry suggested that a similar approach may be taken by each of the other subgroups. They could create a flowchart of their clinical questions and have them hyperlink to the appropriate graphic entries.
2. Team agreed that the flowchart should be presented at the next general meeting, along with the idea that the other subteams may want to consider it for their subteams to more easily format the clinical questions and the appropriate graphs. We will also mention the glossary, how terms in the glossary can be linked to each team's deliverables, and also the relationship between flowchart and glossary.
2.1 Qi asked for volunteers for several upcoming conferences. [Qi, pls fill in outcome here]
3. We ran out of time to discuss glossary, to be discussed further at the next General Principles meeting.
July 2010
Minutes for General Principles subteam of
FDA/Industry/Academia Safety Graphics
Meeting Date/Time: July 6, 2010, 2-3pm EDT
Present: Mat Soukup, Qi Jiang, Susan Duke, Brenda Crowe, Frank Harrell
Actions:
- All: If you haven't posted your biosketches to CTSpedia, you will find below the text each of you already provided (it's a good way to get started with editing on the wiki). It should be posted here: http://www.ctspedia.org/do/view/GraphicsWorkingGroup/BackgroundSafetyGraphics (just click on the ‘edit' button in upper right corner, then scroll down to the bio section, then save when you're done. Amazingly simple.)
- Fabrice and Susan: Team agrees that adding a glossary is a good idea. Fabrice has some ideas and Susan will ask to use the glossary GSK has already created.
- Mat: Will work with Mary to complete the graphics entry template for all of us to use.
- All: I would also like to challenge each member of the team to add an item to the General Principles Reference Links section prior to our next subteam meeting (as more links are added, we can rearrange things as needed). Everyone who does so will get a gold star! (Not much, I know but if you need a reward to get you there, there you go... The wiki is only as valuable as our contributions.)
- Susan: re-post ‘Choosing the Right Graph' visual as a pdf, as not everyone could see it
- Susan: send a reminder 2-3 days in advance for subsequent meeting for any materials to be reviewed in advance of the meeting.
June 2010
Minutes for General Principles subteam of
FDA/Industry/Academia Safety Graphics
Meeting Date/Time: June 7, 2010, 2-3pm EDT
Present: Mat Soukup, Richard Forshee, Qi Jiang, Susan Duke, Brenda Crowe
Actions:
- Review revisions of the graphics template for finalization before next general meeting
- Susan to add minutes to the wiki
- Susan to offer ‘Choosing the Right Graph’ visual to the team for consideration of further development/refinement
- There was not an agenda for this meeting. The time was used for Mat to demonstrate how to interact with the wiki for adding in content, revising content, and including attachments.
- Discuss template for graphics entries (GSK’s template may be used as a starting point if the team so chooses)
Susan provided an example entry, which generated good discussion.
Mat subsequently sent out a version for discussion, which is progressing with the subteam in an email stream.
May 2010
Agenda for General Principles subteam of
FDA/Industry/Academia Safety Graphics
Meeting Date/Time: May 3, 2010, 2-3pm EDT
Present: Larry Gould, Markus Yap, Mat Soukup, Richard Forshee, Qi Jiang, Susan Duke, Brenda Crowe, Frank Harrell, Fabrice Bancken
Agenda:
- Welcome to Frank Harrell and Fabrice Bancken
- Last month’s minutes – any discussion or accept as written
- Actions
- Update from Mat on private area, progress on setting up of the site itself
- Update from Susan on future standing meeting times
- Discuss template for graphics entries (GSK’s template may be used as a starting point if the team so chooses)
Actions:
- Mat offered to send out a template for the team to discuss, based on today’s discussion (agenda item 4)
- Susan offered to draft brief biosketches on each of us, based on notes from each of our introductions
- Welcome to Frank Harrell and Fabrice Bancken
- Frank, Fabrice and Brenda (who was on holiday last month) gave a brief introduction. Susan offered to write up a biosketch based on notes from intros at both meetings
- Last month’s minutes – any discussion or accept as written
- Minutes accepted as written
- Actions
- Update from Mat on private area, progress on setting up of the site itself At the last general safety graphics meeting, it was decided that a private area was not needed. So no discussion of this topic was needed in this meeting.
- Update from Susan on future standing meeting times Standing meetings will take place on the first Monday of the month from 2-3 Eastern (unless there is a holiday, etc – will be rescheduled to another time)
- Discuss template for graphics entries (GSK’s template may be used as a starting point if the team so chooses)
Susan provided an example entry, which generated good discussion.
Mat subsequently sent out a version for discussion, which is progressing with the subteam in an email stream.
Draft biosketches (please review/revise as needed – I wasn’t planning to share these so they are obviously incomplete!)
Larry Gould , Merck – has taught short courses on graphics. Believes it is important to frame the question – what are you trying to do, what are the questions the graph is to answer?
April 2010
Minutes from General Principles subteam of
FDA/Industry/Academia Safety Graphics
Meeting Date/Time: April 5, 2010, 2-3pm EDT
Present: Larry Gould, Markus Yap, Mat Soukup, Richard Forshee, Qi Jiang, Susan Duke
(Brenda Crowe was on holiday)
Agenda:
- Introductions
- Examples of items to consider including on the wiki
- Brief review of Safety Graphics team objectives
- Brainstorm/discuss scope, breadth, depth of what Graphics Principles subteam wants to cover (ie, subteam objectives)
Actions:
- Mat to find out if a private area can be set up in the wiki for the subteam’s work (and if so, to set it up and communicate to the subteam)
- Susan to request availability for subsequent subteam meetings and set them up
Minutes:
- Introductions were made by all. Collectively we have a broad and deep breadth and depth of graphics knowledge and interest. It was nice for all to meet and consider motives and interests for being part of the subteam.
- Items considered for inclusion on the wiki:
- Reference links (eg, Tufte, Cleveland, Robbins, Heiberger, etc)
- Best practice (right way to ‘do graphs’)
- Guide audience in appropriate ways of graphing for common purposes, eg count data
- Common mistakes in graphing (eg too many bells/whistles, chart junk, etc)
- Graphical perception
- Risk communication
- A brief review of Safety Graphics objectives was undertaken to identify which objectives our subteam is aligned. We propose the subteam is aligned to the team’s objectives in the following ways:
- Identify areas particularly applicable or useful to regulatory review in which graphics can enhance understanding of safety information.
- Develop a palette of statistical graphics for reporting on clinical trials safety data. Post-meeting suggestion for this team from Mat: Define a standard template to be used for graphics entries on the wiki.
- Recommend graphics for clinical data based on good scientific principles and best practices. Identify general graphics principles that will aid readers of graphs to better design and interpret them, especially those principles that may aid in safety data interpretation.
- Create a publicly-available repository of sample graphics (ensuring appropriate credits are given for contributions), including data sets and code.
- Educate and engage stakeholders through outreach activities.
- Consider publishing with authorship/acknowledgments as is consistent with contributions and policy of the affiliated institution. A possible outcome of the General Principles subteam’s output is a white paper (to be determined after the wiki takes shape).
- Discussion of scope, etc for this subteam is included in italics above (after each Safety Graphics objective). In addition, Mat offered to follow up on whether the wiki can be made available for an initially private area for the team to begin their work.
- There was discussion of software and interactive graphics. Several members noted that the purpose of the Safety Graphics team is to identify good graphic design for particular safety questions. If vendors or users choose to create interactive versions of these designs, that will be a beneficial additional outcome of Safety Graphics team’s identification of the graph design. However creation of interactive graphs or identification of a specific software as the only way to render the graph is outside the scope of this team.
- The team agreed to meet on a monthly basis, preferably during the first week of the month. Susan has an action to request availability and set up future meetings.
- Post-meeting note: Frank Harrell has asked to join this subteam. Welcome Frank! Good to have you on board.
ppt 
docx 
pdf