Future of Radiation Oncology EHR by liwenting

VIEWS: 217 PAGES: 39

									The Radiation Oncology EHR
       of the Future

       Joel W. Goldwein, MD
      Senior Vice President and Director
                Medical Affairs
                 Elekta, Inc.

               August 6, 2010
Future Radiation Oncology EHR:
• What’s the current state of the art?
• What’s driving progress?
• Where are we going?
State of the Art:
Radiation Oncology EHR
• RT used in approx 50% of 1.4M US cancer patients
• ~ 2,100 RT facilities in US
   – > 70% utilize dedicated information systems
   – Represents ~ 500,000 new RT patients/yr managed using RO EHR
• RO highly EHR-dependent clinical specialty (V&R…)
   – Degree of EHR use highly variable
• Two major EHR vendors (Elekta MOSAIQ, Varian Aria)
   – Well established / managed
   – Regulated QA systems support software development
   – Numerous interfaces to devices/machines, TPS, HIS, PACS, lab…
   – Largely standards-based (DICOM-RT, HL7, AJCC, CTC(AE)…)
   – IHE-RO participants (committed to interoperability)
RO EHR Features (Elekta MOSAIQ®)
• Clinical
   –   V&R
   –   CPOE
   –   Comprehensive noting/charting
   –   TPS interface/integration
   –   Medical oncology support
   –   Rule-based workflow management
   –   Numerous interfaces (lab, HIS,
       treatment machines…)
• Image Management (2D, 3D, 4D)
   – Import/export/manipulate
   – Treatment related image (IGRT)
   – Oncology PACS Integration
RO EHR Features (Elekta MOSAIQ®)
• Practice Management/Administrative
   –   Patient and Resource Scheduling
   –   Charge Capture/Billing
   –   Kiosk patient check in
   –   Inter/intra-product communications
• Research
   –   Data collection and reporting
   –   Data trending
   –   Outcome analyses support
   –   Graphical data exploration
   –   Clinical Trial management            Graphical Data Exploration Module
   –   Interfaces out to Cancer Registry
       and RO data aggregation products
RO EHR Environment
• Architecture
  – PC Based
     • Desktop workstations, laptop
       carts, limited mobile devices
  – Client/Server (MS-SQL)
     • Workstations - five to hundreds
     • LAN or WAN (Citrix)
• Device and imaging environment highly variable
  – Cadre or product mix innumerable (linacs, protons,
    orthovoltage, TomoTherapy, TPS, HIS…)
  – Workflow associated with evolving clinical trends
• Some FDA regulated EHR components
  – MLC control systems, Image management components
EHR Clinical Trial Support Features
• TrialCheck® (Coalition of Cancer Cooperative Groups)
    – Determine patient eligibility for clinical trials active at your facility
    – Based on clinical criteria (age, disease, stage, performance status…)
      available in EHR
• Trial data collection support
    – import/export
      assessment templates
• DICOM-RT image
  export to cooperative
  groups (RTOG…)
• Integrated National
  Comprehensive Cancer
  Network (NCCN)
                                Clinical Trial Eligibility Determination Screen
Web-centered Collaboration:
Established Community of EHR Users

• Shareable Reports
• Sharable Document
• Sharable Care Plans
• Listserver
• Infrastructure for
  future expansion
EHR Snapshots

     Patient Face Sheet (Chart Cover)
EHR Snapshots

   Patient Assessments (Labs, Vital Signs…)
EHR Snapshots

    Treatment setup image verification (2D)
EHR Snapshots

        X-ray Volumetric Image Guided RT
 RO EHR: Current Status Summary
• Widely deployed systems across entire specialty
• Largely standards based
• Device, HIS, TPS interconnectivity and
  interoperability comprehensive
• Modern UI, RO image-enabled
• Support clinical, research and administrative
  aspects of RO practice robust
• Mature, established vendors with regulated &
  structured development control systems
Seems like it’s all there!
What’s next?
How can we leverage
today’s products to build
the system of the
Some Rad Onc EHR Drivers
• Health Information Technology for Economic and Clinical Health Act
       • Meaningful use and ARRA certification ► $$$
• Rapidly emerging technologies (device, imaging, techniques,
   – All expensive; proof of utility often limited
   – Associated workflows will require EHR support
   – Associated high costs implicate Comparative Effectiveness Research
• Recent safety-related incidents
   – Incent development of “safer” EHR systems
• Rising costs with potential diminishing reimbursement
   – Incents increased efficiency by improving workflow and
     the Human-Computer Interface
 Meaningful Use (HITECH)
 Ambulatory Provisions
 • Qualified Eligible Provider Requirements
    – Meet ~ 20 MU criteria AND use certified EHR system
          • e.g. - CPOE medications (e-prescribing), clinical rules engine (decision
            support), capture vital signs, provide patient EHR access, supply
            summary of care record…
 • Payout to qualified/eligible providers begin 2011
                          MEDICARE PAYMENT SCHEDULE
                                         Initial Qualifying Year

Calendar Yr        2011        2012        2013                    2014   > 2015
   2011          $18,000
   2012          $12,000      $18,000
   2013          $8,000       $12,000    $15,000
   2014          $4,000        $8,000    $12,000              $12,000
   2015          $2,000        $4,000     $8,000               $8,000       $0
   2016                        $2,000     $4,000               $4,000       $0
  TOTAL          $44,000      $44,000    $39,000              $24,000       $0
  Meaningful Use (HITECH)
  Effect on Future of RO EHR
  • Certification is current major Vendor focus
  • Expected Results
         Core                                           Import/Export                Administrative
        – Acceleration of EHR system enhancements
Demographics           Vital Signs        LabAppt Reminders
                                                                   Public registry
    CPOE              Problem lists     Discharge   Pt. summary                      Follow-up & scheduling
        – Improvements in scopeexport (meaningful)migration and transfer
            Drug interactions &
Allergies/Alerts                Smoking
                                        Import of
                                                       use         export
                   Med reconciliation               standards      import/export
        – Overall:Security (but important) effect on RO EHR
        Privacy and Limited         Vocabulary     Decision Support
          features, development, and end-user adoption
         Authentication  Encryption  ICD-9/10; CPT      Guidelines
       Disclosure Compliance              Audit                 SNOMED                Interlocks and Alerts

                                         Already available or in progress
Emerging technologies as RO EHR
• Are we moving too quickly?
• Will there ever be sufficient supportive
• Can EHR systems help
  us obtain it?
• Will EHR’s be able to
  accommodate ever
  changing workflow?
RO: Fodder for Comparative
Effectiveness Research
 Economic Scene
 In Health Reform, a Cancer Offers an Acid Test
 By DAVID LEONHARDT   Published: July 7, 2009

“The prostate cancer test will determine
 whether President Obama and
 Congress put together a bill that begins
 to fix the fundamental problem with our
 medical system: the combination of
 soaring costs and mediocre results. If
 they don’t, the medical system will
 remain deeply troubled, no matter what
 other improvements they make.”
[Radiation] Oncology:
A CER Priority?
• Cancer focus of 6 primary
  IOM CER topics
• Direct US cancer costs
  $80 Billion
• RO consumes approx.
  10%; growing fast!
   – Burden likely to fall on us to
     provide evidence basis
• A real-time registry
  capturing patterns-of-care
  data may represent part of
  the solution                        Distribution of the IOM's Recommended CER Priorities
                                             Iglehart J. N Engl J Med 2009;361:325-328
Radiation Oncology Data Registry
• Pilot Program Background
  – Est. 10/2008 as derivative of more
    mature Med Onc data program
• Aim
  – Demonstrate proof of concept for
    establishment of central RO data
    warehouse derived from live EHRs
  – Establish basis for CER….
  – Reduce costs of data collection
• Method
  – Leverage widely deployed RO EHRs
  – Aggregate de-identified data collected
    in EHRs in routine course of care
  – Extensible/scalable design
Registry Architecture
•   Small program installed on
    local EHR PC
•   Scheduled service runs                             Registry
    program at some regular
    interval                                                    EHR
                                                              Facility 5
                                                   EHR                       EHR
                                                 Facility 3                Facility 4
•   Program runs => De-identified     EHR                                                 EHR
                                    Facility 1
    dataset created                                                                     Facility n
•   Dataset uploaded to central                         Facility 2
    data warehouse
•   Dataset
RO Registry Pilot Results
ASTRO 2009 Analysis
    18 RO participants
    121,000 patient records, 108,000 patient treatments
Data completeness (quality)
    Date of Diagnosis: 29% complete
    Overall Stage: 21% complete
    Successful, but quality issues need to be addressed
       HITECH/MU may incent quality improvements
    If scaled, could represent real-time model dataset for CER,
    advocacy, administrative and clinical research…
    Could serve as feed for higher-level multi-purpose registries and
    complement US Cancer Registry program
  RO Registry Multi-Purpose Model
• Opportunities
  – Specialized
    registries (SBRT,
  – CER registry
  – Safety (near-hit)
  – caBIG Grid
    RO EHRs and Safety

 • RO info systems have incorporated safety features such
   as V&R for decades
 • Evidence suggests these features do indeed enhance
 • Contemporary systems designed top-down to minimize
 • This has not always been the case

*Frass et al – IJROBP - 1998
RO Control Systems:
Vehicles of “Safety” ?

  Nov ‘05

• 2 of 10 (and the only “deaths by software”) were RO-related
   – 1985 – Therac 25 Linear Accelerator
       • 5 deaths
       • Significant changes in way RO software developed
   – 2000 – Multidata TPS (Panama)
       • 8 deaths
    RO EHR Systems and Safety

The Radiation Boom
Radiation Therapy Offers New Cures, and Ways to do Harm
By WALT BOGDANICH         Published: January 24, 2010

 NY State Records 2001-2008
     – 621 events, 1,264 causes, 2 notable deaths
         • 46% - missed target       #      Cause
         • 41% - wrong dose          352    Flawed Q/A plan
                                     252    Human data entry/calculation error
         • 8% - wrong patient        174    Wrong patient, wrong site
                                     133    Wedge or collimator misused
                                      60    Hardware malfunction
                                      24    Software bug
                                      19    Erroneous software override
 RO Safety Record

                     Event Type               Events per million RT courses*
                                          Any ~ 10,000 - 20,000
  Errors w/ significant clinical consequences ~ 1,000 – 10,000

     Errors w/ serious clinical consequences ~ 5 - 10

                Lots of caveats (e.g. – under-dosing, under-reporting)

    • Most events ≠ serious injury
    • RT safe, but could/should always be safer

                       [~ 1M RT courses yearly in US (IMV Report – 2007)]

* Munro – BJR 2007
  Why, in era of sophisticated EHRs,
  does this still happen?
• RO Highly Complex!
  – Diseases varied
  – Processes fluid
  – Humans involved
    (both sides of table)
  – EHRs can’t anticipate it all

• Failure Mode and Effect
  Analysis (FMEA)
                                     Simplified RO Workflow Illustrating the
                                   USING FAILURE MODE AND EFFECTS ANALYSIS
  – 127 high risk failure modes      multitude of opportunities to manage
                                   ERIC C. FORD, PH.D.,* RAY GAUDETTE, M.S.,* LEE MYERS, PH.D.,* BRUCE VANDERVER, M.D., LILLY
                                   ENGINEER, DR.P.H., M.D., M.H.A., RICHARD ZELLARS, M.D.,* DANNY Y. SONG, M.D.,* JOHN WONG, PH.D.,*
                                   AND THEODORE L. DEWEESE, M.D.*
                                     the RO process
                                   IJROBP, 2008
Hazard Mitigation
                                                 Training &
• Hierarchy of effectiveness          Rules
                                    & Policies

                                  & Checklists
                              & standardization
                              & Computerization
                              Forcing functions
                           & Constraints (Interlocks)
                                Mitigation Strategies
                    The EHR Solution?
Leverage Hazard
Mitigation Strategies to
Reinforce EHR Safety                                                 Training &
Where human                                                          Education
intervention is                                          Rules
required…Reduce                                        & Policies

                           Less Automation
                                                      & Checklists
                                                        & standardization
         Increase                               Human

   dependence on
                                             Intervention  Automation
     most reliable                                     & Computerization
           hazard                                       Forcing functions
        strategies                              & Constraints (Interlocks)
         Iterate!                                   Safer System =>
    Forcing Functions in Medicine:
    Interlocks, Timeouts and Checklists
   They work!
   Supplied throughout
    EHR, devices and
   Next generation
       Roll your own!
                                                 Design Toolkit
                          Graphical Workflow Console Checklist
                           Gamma Knife Treatment
                          MOSAIQ Universal Timeout
    MOSAIQ Workflow Manager
   Incorporate customized
    checkpoints w/ optional
    interlocks and alerts at
    any point
   Special attention to
    any/all high risk hazard
    nodes in work flow
   Requires iteration via
    ongoing FMEA

                               USING FAILURE MODE AND EFFECTS ANALYSIS
                               ERIC C. FORD, PH.D.,* RAY GAUDETTE, M.S.,* LEE MYERS, PH.D.,* BRUCE VANDERVER, M.D., LILLY
                               ENGINEER, DR.P.H., M.D., M.H.A., RICHARD ZELLARS, M.D.,* DANNY Y. SONG, M.D.,* JOHN WONG, PH.D.,*
                               AND THEODORE L. DEWEESE, M.D.*
                               IJROBP, 2008
MOSAIQ Workflow Manager Process


                 Check Modality
 Automatic Script
                    Customer Defined
Ask for More Data with   Task
 User-Defined Form
              Create QCL Item
              To Verify Info, etc.
The Human-EHR Interface: A few words
• Pen/paper – hard to compete, but also hard to
• PC / keyboard / structured data entry
   – slow, awkward, interferes with MD-patient relationship
   – BUT, structured data necessary for true benefits of
• Dictation / voice recognition
   – Faster, easier, more natural
   – BUT, natural language processing that would provide
     for creation of structured data far off
• A number of solutions are on the horizon
   The Human-EHR Interface: Teamwork!
     Software usability Improvements

Hardware usability improvements                     Kiosks and
                                                    PRO Interfaces
           (Lab, HIS,

       Interfaces galore
                                                  Digital pen & paper
                            Dictation + Natural
                            Language Processing
What’s the Long Term Vision?
EHRs as the Cornerstone
• Remote Care Monitoring and Delivery
      – Driven by efficiency, cost containment, and Rx
        consistency requirements
      – Facilitated via EHR
• Radiation Oncology “Medical Home”*
      – Our “processes” are far to diverse,
        uncontrolled and unstructured
      – Demands on our time will only increase
      – Patients will demand more control of their care
*An approach to providing comprehensive evidence-based, guideline-directed care... that facilitates
 partnerships between patients and providers
RO EHR Systems of the Future

            RO EHR systems are
            already (arguably) most
            sophisticated in medicine

            Managing complex tasks
            in an increasingly
            complicated environment

            The unique mix of
            technical and clinical
            specialists immersed in
            the explosion of
            advanced technologies
            will continue driving this   Which will be great for
            trend….                          our patients!
The Radiation Oncology EHR
       of the Future

       Joel W. Goldwein, MD
      Senior Vice President and Director
                Medical Affairs
                 Elekta, Inc.

               August 6, 2010

To top