CREDO

1
CREDO

• Review 22nd November 2003

2
Credo objectives and plan

• Goal is a RCT of clinical decision support,
  workflow and other services in breast cancer care
• Looking at clinical benefit, patient satisfaction
  and technical/sustainability issues
• Planned in 3 stages
• Period I, start-up and piloting (2003-4)
• Period II, small scale testing (2004-5)
• Period III, integrated trials (2005-7)

3
Purpose of this meeting

• CREDO team to present various aspects of project,
  to identify strengths, weaknesses, issues
• Role of review group is to identify critical
  problems and advise CREDO team on
• Whether or not we are ready to go
• What areas need focus
• What problems need to be addressed
• We need to empower the review group with the
  information they need to give this advice

4
Credo core service model
Demo
5
System infrastructure (illustrative)
Clinical UI
Special services
6
Collaborators _at_ September 2003

• Guys Hospital Breast Unit
• Ian Fentiman, David Miles, Corrado dArrigo,
  Shirley Hodgson
• Web-based system hosted by CRUK ACL?
• Addenbrookes Hospital Breast Unit
• Arnie Purufshotham et al
• Built on top of Joint Clinical Information System
  (in development)
• Carol Franc Buck Breast Care Center, UCSF and UC
  Davis
• US Army COE project
• Laura Esserman, Michael Kamerick, Mike Hogarth,
  Jon Showstack et al
• Seen as national project, commercial aspirations

7
LRI team _at_ August 2003

• John Fox
• Clinical
• Alyssa Alabassi, breast surgeon (1/2003 to
  10/2003)
• Vivek Patkar, breast surgeon (11/2003-5/2004)
• Matt Williams, clinical fellow (From march 2004)
• RD
• Tony Rose, development manager
• Ali Rahmanzadeh,
• Rory Steele,
• Chris Hurt
• Other contributors Liz Black (communication),
  David Glasspool (reactive planning), Martin
  Beveridge (speech dialogue)

8
Start-up (2003/4)

• Build development team and approach
• Refine decision support, workflow and other
  Tallis capabilities
• Establish clinical collaborators (buy in)
• Consensus on clinical requirements
• Technical demonstrator
• Core service definition
• Go, no-Go decision, end 2003

9
2004-5 testing

• Individual service trials
• Learning curve
• Reflecting local priorities and interests
• Paper patients to simple clinical testing
• Requires pre-clinical infrastructure
• Patient data capture
• Workflow management (over time)
• Communication and messaging
• Issues of security and safety, governance
  addressed

10
Clinical trial 2004-7

• Individual service trials (poss. 2004)
• Integrated service trials (starting 2005)
• Patient journey
• Data collection and analysis
• Publish interim results 2005, final report 2007
• Plan hand over to ongoing service provider (or
  decommissioning)

11
Critical technical issues for 2004

• Technology, methods sufficient to achieve 2004
  objectives
• Platform and integration (Tallis composer,
  engine, web)
• Content development (data set, services, media
  etc.)
• Scalability (repository, reusability, versioning)
• Lifecycle (platform and content)
• Staffing
• Platform development
• Content development
• Support for clinical collaborators
• Support
• team may be increased through academic and
  commercial collaborations (e.g. UCSF, InferMed)
  and further CRUK funding
• Note ACL Quinquennial Review, November 2004.

12
Critical clinical issues for 2004

• Credible demonstrator (content and usability)
• Core Service Model carefully reviewed and agreed
• Defined data set for complete patient journey
• Low hanging fruit for early trials
• Triple assessment
• Genetics
• Two-week referral
• Systemic therapy
• Clinical Evidence Interactive style breast cancer
  application
• Multi-disciplinary meeting
• Sound design of trials, address governance issues

13
Presentations

• 1. Do we have a clear, practical plan for phase
  2, including milestones and deliverables? Review
  John Fox, Tony Rose and Vivek Patkar
• 2. Is the Tallis/PROforma authoring system
  adequate to build and test pilot applications
  (e.g. triple assessment, genetic risk
  assessment)? What problems need to be given
  priority for resolution during 2004? Review
  Chris Hurt, Vivek Patkar and Ali Rahmanzadeh
•  
• 3. Will the Tallis delivery software (engine, web
  server, clinical middleware etc.) have sufficient
  functionality, performance and resilience to
  permit test trials to be successfully carried out
  at Guys, Addenbrookes and UCSF during 2004? What
  issues need to be addressed on what timescales
   Review Rory Steele, Tony Rose
• 4. Can we be confident of meeting any
  infrastructure needs that will arise during 2004
  (e.g. secure patient data capture, clinical order
  entry, messaging) and anticipate being able to
  solve infrastructure needs for a clinical
  strength system during phase 3 of the project
  Review Ali Rahmanzadeh, Tony Rose, Liz Hurt
• 5. Will we be in a position during 2004 to plan a
  full CREDO trial starting 2005 (ethical
  permission, trial design, governance) and if not
  what are the major problems to be addressed?
  Review John Fox, Chris Hurt, Vivek Patkar
•  
• 6. Do we have sufficient staff to provide
  adequate manpower to support our clinical
  partners during pilot trials during 2004 (eg. for
  customising triple assessment and integrating
  Tallis into the local infrastructure)? If not,
  what are our options? Review Tony Rose and John
  Fox
•  
• 7. Other questions and issue. Review John Fox
  and Tony rose