PERC: Envisioning the Next Generation of EPath

1
PERC Envisioning the Next Generation of EPath

• NAACCR Annual Meeting
• June 10, 2008, Denver

Jennifer Seiffert, MLIS, CTR, Northrop Grumman
Contractor to NPCR, CDC jenesei_at_comcast.net Jose
ph Rogers, NPCR, CDC jdr0_at_cdc.gov
2
Outline

• PERC History
• PERC Members
• PERC Goals
• Project Vision
• Cancer Care Ontario Activities

3
Note

• Presentation will emphasize Collaborative Staging
  (CS) because that is my role on PERC

4
History (1)

• PERT (Pathology Electronic Reporting Task Force)
  established in 2007
• PERT grew out of several prior activities
• SNOMED Surgical Pathology Working Group
• Reviewed and updated the SNOMED Anatomic
  Pathology subset (took over a year)
• College of American Pathologists (CAP) Cancer
  Committee

5
History (2)

• In 2008, PERT was upgraded to PERC
• A full Committee of CAP
• Reports to SNOMED STS SNOMED Terminology
  Solutions Oversight Committee

6
History (3)

• PERC has established two subcommittees
• Data Modeling (back end) group for modeling how
  CAP checklist data can best be collected
  electronically, interoperably, and mapped to
  other standards, including CS
• Includes CS and NAACCR representatives
• User Interface (UI) (front end) group for
  modeling interface for pathologists entering
  checklist data electronically

7
History (4)

• Priority being given to UI work, getting
  standardized electronic checklists to
  pathologists and LIS vendors ASAP
• Decision has been made, for current work, to
• Include mapping to CS for items with one-to-one
  non-algorithmic maps
• Postpone mapping to CS items requiring reference
  to multiple data items or algorithms to 2nd phase
  of project after easy mappings done

8
PERC Members May 2008

• John Madden, MD, PhDCo-ChairDuke University
• Monica de Baca, MD, Co-ChairPhysicians
  Laboratory
• George Birdsong, MDEmory University
• Kenneth Gerlach, MPH, CTRCDC/NPCR
• Lori Havener, CTRNAACCR
• Mary Kennedy, MPHCAP
• Robert Knapp, MDPathology Laboratory, P.C.

• Gemma LeeCancer Care Ontario
• Andrea MacLeanCancer Care Ontario
• Richard Moldwin, MD,PhDCAP
• Douglas MurphyCAP
• Wendy Scharber, RHIT, CTRRegistry Widgets
• Jennifer Seiffert, MLIS, CTRCollaborative
  Staging
• Mike Smith, MDCAP
• James Sorace, MD, MSASPE, HHS
• Henry Travers, MDPresident, WASPaLM

9
PERC Goals (1)

• To advance the computerized representation of the
  CAP checklists
• Create FRAMEWORK for
• electronic forms (standardized input)
• data repositories (retrievable output)
• Ensure electronic versions accurately represent
  the CAP Cancer Committees intended meaning

10
PERC Goals (2)

• To create an electronic representation of the CAP
  Cancer Protocol checklists which
• Is aligned with (upcoming) AJCC and Collaborative
  Staging Systems.
• Allows for historical representation
• Is amply designed for addition of medical
  specialties, coding, and/or staging systems

11
Project Vision (1)

• CAP provides software tools for computerized
  implementations of cancer checklists
• Path labs and vendors implement standardized
  checklist templates for synoptic cancer reports
  as part of full path reports

12
Project Vision (2)

• Central cancer registries receive standardized
  synoptic ePath and can automatically populate
  standard NAACCR items, including CS
• Registries consolidate ePath with other reports
  (clinical, demographic)

13
Aspects of Framework (1)
Template Editor

• Software application and database for creating
  and distributing computerized checklists
• Written specifically for PERC work
• Enables automated data capture
• Provides mapping to other standards incl. SNOMED,
  NAACCR, CS
• Facilitates reuse of data for CS, research,
  caBig, tissue banking, etc.

14
Aspects of Framework (2)

• Consistency across checklists to improve
  accuracy. Examples
• vascular invasion and lymphatic invasion
  standardized to lymph/vascular invasion
• mitotic activity vs. mitotic count
• tumor configuration vs. tumor features
• Standardizing flavors of null (indeterminate,
  no data, not reported, etc.)

15
Aspects of Framework (3)
Use appropriate dialog controls to improve
accuracy
Group of Radio Buttons
Check Boxes

• Choose one
• Squamous cell carcinoma
• Adenocarcinoma
• Large cell carcinoma
• Small cell carcinoma

• Choose as many as apply
• Embryonal carcinoma
• Choriocarcinoma
• Yolk sac tumor
• Teratoma

16
Aspects of Framework (4)
Increased Atomicity
Current Checklist Format
Possible modification

• Tumor size _______ cm
• Extraparenchymal extension
• Positive
• Negative
• Indeterminate

___ Tumor 2cm or less without extra-parenchymal
extension (T1). ___ Tumor 2cm but not more than
4cm without extracapsular extension (T2). ___
Tumor more than 4 cm and/or tumor having
extra-parenchymal extension (T3).
CS algorithm could derive pT.
17
Aspects of Framework (5)
Context Sensitivity

• ___ Cannot be assessed
• ___ Benign glands at surgical margin
• _X_ Margins uninvolved by invasive CA
• ___ Margin(s) involved by invasive CA
• ___ Unifocal
• ___ Multifocal
• ___ Apical
• ___ Bladder neck
• ___ Anterior
• ___ Lateral
• ___ Postero-lateral
• ___ Posterior
• ___ Other(s) (specify)

• ___ Cannot be assessed
• ___ Benign glands at surgical margin
• ___ Margins uninvolved by invasive CA
• _X_ Margin(s) involved by invasive CA
• ___ Unifocal
• _X__ Multifocal
• ___ Apical
• _X_ Bladder neck
• ___ Anterior
• _X_ Lateral
• ___ Postero-lateral
• ___ Posterior
• ___ Other(s) (specify)

18
NPCRs Goal

• Increase quality of cancer surveillance data and
  efficiency of collection by increasing use of CAP
  checklists and synoptic reporting
• How? Make them easier to use by pathologists,
  lab system vendors, registries

19
Project Vision Review(1)

• CAP provides software tools for computerized
  implementations of cancer checklists
• Path labs and vendors implement standardized
  checklist templates for synoptic cancer reports
  as part of full path reports

20
Project Vision Review (2)

• Central cancer registries receive standardized
  synoptic ePath and can automatically populate
  standard NAACCR items, including CS
• Registries consolidate ePath with other reports
  (clinical, demographic)

21
PERC Activities for 2008

• Weekly teleconferences of each subgroup
• Goal to have model electronic data entry forms
  for selected checklists published in 2008, using
  TNM 6th ed., for pathologists and LIS vendors to
  begin implementing
• Will probably recommend simple interim database
  approach for LIS vendorslimited mapping

22
Sample Early Draft Prototype of Lung Electronic
Checklist
Check boxes Optionality Radio buttons
23
Lung Draft Prototype Detail

• Pathologist checks all that apply
• Choices are nested, grouped

45
20
70
55
40

• Mapping to CS Extension available behind the
  scenes
• Database can be built behind the scenes
• Algorithm needed for combinations checked.

60
56
60
70
70
71
70
70
75
70
65
45
24
Cancer Care Ontario Activities

• Real-world test of some of the PERC concepts
• 90 of path reports in province of Ontario are
  received electronically
• Modified CAP/CS aligned checklists implemented in
  participating hospitals
• Central registry will capture CS items from
  synoptic reports mapped to CS and derive stage
• Feasibility and accuracy will be assessed
• Also testing a computerized path requisition form
  for surgeons to complete

25
Cancer Care Ontario Activities

• Collaborative Staging Colorectal Pilot Project.
• Modified front end of checklist to align w/CS.
• Pathologists complete checklist, data are
  extracted from hospital information system, CCO
  coders supplement staging data w/radiology,
  surgery and clinical info
• Other Sites to come breast, lung, prostate,
  endometrium
• eSurgery and eRadiology feasibility assessment
  underway

26
Issues from Cancer Care Ontario (1)

• Multiple path reports on same patient/tumor
• Need guidelines for pathologistsregistry would
  prefer separate synoptic report for each
  reportable tumor
• Need guidelines for registries, consolidating
  multiple reports from multiple specimens
• Need to determine if reports pertain to stage AT
  DIAGNOSIS

27
Issues from Cancer Care Ontario (2)

• Determining which lab results to capture from
  multiple tests for SSFs/tumor markers
• Can algorithm be written to select the
  appropriate test results from lab information
  system?
• Usually need highest value prior to treatment
• Rules are site- and test-specific

28

• Information about CDCs Cancer Prevention
  Control Programs
• www.cdc.gov/cancer
• The findings and conclusions in this presentation
  are those of the presenter, and do not
  necessarily represent the views of the Centers
  for Disease Control and Prevention.