Upgrade to Pro — share decks privately, control downloads, hide ads and more …

DHIN Summer Summit, Finding the Humans in Health IT

Shahid N. Shah
June 14, 2018
270

DHIN Summer Summit, Finding the Humans in Health IT

A landscape review of what HHS, ONC, and other gov't agencies are focused on vs. what the health IT industry is working on and how ML, AI fit into the overall picture.

Shahid N. Shah

June 14, 2018
Tweet

Transcript

  1. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    1
    Finding the Human(s) in
    Health and IT
    By Shahid N. Shah
    @ShahidNShah

    View Slide

  2. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    2
    Why is it so hard to find
    the Human in Health IT?
    EVERYONE WANTS TO DO THE RIGHT THING, RIGHT?
    @ShahidNShah

    View Slide

  3. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    3
    There are no business
    models to do the “right
    thing” in healthcare.
    BUT HIEs ARE AS CLOSE AS WE GET THESE DAYS
    @ShahidNShah

    View Slide

  4. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    4
    HHS & ONC Focus
    TEFCA
    DATA / APIs
    PX / ACCESS
    REDUCE BURDEN
    @ShahidNShah
    SAFETY
    USABILITY
    TRANSPARENCY
    OUTCOMES COST
    VALUE
    CONSUMERISM
    EXPERIENCE
    FHIR

    View Slide

  5. OUTCOMES MATTER

    View Slide

  6. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    6
    CMS QPP is good but are APMs the better driver?

    View Slide

  7. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    7
    Can Meaningful
    Measures allow
    us to make real
    progress?
    http://www.modernhealthcare.com/article/20180120/NEWS/180129995

    View Slide

  8. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    8
    Meaningful Measures shows some helpful directions

    View Slide

  9. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    9
    RESPONSIBLE
    AND
    ACCOUNTABLE

    View Slide

  10. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    10
    Industry Focus
    HIEs
    Cybersecurity
    Population Health
    Telemedicine
    @ShahidNShah
    APIs / FHIR
    Conversational UX
    Productivity
    ML / AI
    BLOCKCHAIN

    View Slide

  11. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    11
    Who’s working on answering these questions?
    • Should we go into at-risk (downside, too not just
    upside) contracts? Is the uncertainty tolerable?
    • Will our current pricing allow us to maintain the
    margins necessary for us to stay in business?
    • Can we bear the scrutiny of pricing transparency?
    THIS IS THE HARD STUFF!
    @ShahidNShah

    View Slide

  12. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    12
    Does new Health IT
    create “interruptions at
    scale”?
    AI / ML / Telemedicine / Messaging
    creates significant opportunity for increased
    unnecessary interruptions if not properly designed for
    multiple stakeholder workflows
    @ShahidNShah

    View Slide

  13. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    13
    The ultimate goals of health IT should be to
    increase patient education, answer questions in
    a timely manner, improve patient outcomes,
    reduce patients’ administrative burdens, and
    increase staff productivity by allowing staff to
    focus on the neediest patients.
    @ShahidNShah

    View Slide

  14. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    14
    The unintended
    consequences of health
    IT systems have been to
    increase staff workloads
    and reduce productivity.
    @ShahidNShah

    View Slide

  15. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    15
    What would quality
    measurement look like if
    MU silliness didn’t make
    us take our eye off the
    innovation ball?

    View Slide

  16. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    16
    We’d focus on quality
    improvement (QI) and
    continuous quality assurance
    (CQA) not data collection and
    quality measurement.

    View Slide

  17. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    17
    Insurer | Payer
    Insurance
    Product 1
    Insurance
    Product 2
    Provider 2
    Provider 1
    Health Systems
    But we can’t… because of “Institution First” IT

    View Slide

  18. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    18
    How do we move from “institution
    first” to “patient first” or “patient
    centered” health IT?
    Hint: it’s not about patient engagement. It’s
    about transparency and outcomes.
    @ShahidNShah

    View Slide

  19. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    19
    Let’s reimagine quality improvement
    for a .real-time patient-first, digital-
    first quality experience.
    How? AI & ML

    View Slide

  20. Machine Learning and AI in healthcare will be slowed by intermediated business
    models, misunderstood regulations such as HIPAA / FDA QSR and protective
    regulations such as licensure and credentialing.

    View Slide

  21. How DHIN Can Help Constituents Traverse the
    Digital Transformation Spectrum
    Manual Data
    Collection
    Systems
    Integration
    Reporting
    and Analytics
    Data Mining
    Predictions
    Machine
    Learning
    Augmented
    Intelligence
    Artificial
    Intelligence
    Docs and nurses
    as clerical staff
    TODAY

    View Slide

  22. How DHIN Can Help Constituents Traverse the
    Digital Transformation Spectrum
    Manual Data
    Collection
    Systems
    Integration
    Reporting
    and Analytics
    Data Mining
    Predictions
    Machine
    Learning
    Augmented
    Intelligence
    Artificial
    Intelligence
    PGHD, Med Device
    Connectivity
    TODAY, ACCELERATING

    View Slide

  23. How DHIN Can Help Constituents Traverse the
    Digital Transformation Spectrum
    Manual Data
    Collection
    Systems
    Integration
    Reporting
    and Analytics
    Data Mining
    Predictions
    Machine
    Learning
    Augmented
    Intelligence
    Artificial
    Intelligence
    Automating
    retrospective
    visibility
    TODAY

    View Slide

  24. How DHIN Can Help Constituents Traverse the
    Digital Transformation Spectrum
    Manual Data
    Collection
    Systems
    Integration
    Reporting
    and Analytics
    Data Mining
    Predictions
    Machine
    Learning
    Augmented
    Intelligence
    Artificial
    Intelligence
    Pattern matching
    mastery (unsupervised
    and supervised)

    View Slide

  25. How DHIN Can Help Constituents Traverse the
    Digital Transformation Spectrum
    Manual Data
    Collection
    Systems
    Integration
    Reporting
    and Analytics
    Data Mining
    Predictions
    Machine
    Learning
    Augmented
    Intelligence
    Artificial
    Intelligence
    Use past knowledge to
    make rudimentary
    predictions about the
    future

    View Slide

  26. How DHIN Can Help Constituents Traverse the
    Digital Transformation Spectrum
    Manual Data
    Collection
    Systems
    Integration
    Reporting
    and Analytics
    Data Mining
    Predictions
    Machine
    Learning
    Augmented
    Intelligence
    Artificial
    Intelligence
    Finding known needles
    in haystacks and pop
    health
    TODAY, MAY SKIP FOR ML

    View Slide

  27. How DHIN Can Help Constituents Traverse the
    Digital Transformation Spectrum
    Manual Data
    Collection
    Systems
    Integration
    Reporting
    and Analytics
    Data Mining
    Predictions
    Machine
    Learning
    Augmented
    Intelligence
    Artificial
    Intelligence
    Semi autonomous
    intelligence which
    needs humans
    ARRIVING SOON

    View Slide

  28. How DHIN Can Help Constituents Traverse the
    Digital Transformation Spectrum
    Manual Data
    Collection
    Systems
    Integration
    Reporting
    and Analytics
    Data Mining
    Predictions
    Machine
    Learning
    Augmented
    Intelligence
    Artificial
    Intelligence
    Real intelligence
    indistinguishable from
    humans and fully
    autonomous
    YEARS AWAY

    View Slide

  29. Data ushering in Scientific Method 3.0 .
    1.0
    Identify
    phenomenon
    Think about
    nature
    Fit to known
    patterns
    Guess at
    answers
    3.0
    Identify
    data
    Generate
    questions
    Mine data
    Answer
    questions
    2.0
    Identify
    problem
    Ask
    questions
    Collect
    data
    Answer
    questions

    View Slide

  30. How will we know if we’ve reached 3.0 ?

    View Slide

  31. How will medical roles evolve or be
    eliminated?
    Find other opportunities within 2-3 years
    Clerical
    Focus on value-added tasks such as revenue
    growth or new business development within 3-
    5 years
    Administrative
    •Learn data science
    •Seek digital imaging and digital pathology
    integration opportunities
    •Become telehealth native
    Early career
    clinical
    • Learn to teach computers
    • Become a digital
    transformation change
    agent / leader
    • Drive telehealth across
    the multiple institutions
    Mid career
    clinical • Learn to teach computers your wisdom
    • Ignore the transformation and retire early
    Late career
    clinical

    View Slide

  32. Where ML and AI are applicable (care)
    Therapies
    Therapeutic
    Tools
    Diagnostics
    Diagnostic
    Tools
    Patient
    Administration
    Payer Admin
    Clinical
    Professional
    Education
    Public Health
    Education
    Patient
    Education
    Most Regulation
    Least Regulation
    Cohort specific
    Personalized
    Risk Data Sharing

    View Slide

  33. Where ML and AI are applicable (care)
    Therapies
    Therapeutic
    Tools
    Diagnostics
    Diagnostic
    Tools
    Patient
    Administration
    Payer Admin
    Clinical
    Professional
    Education
    Public Health
    Education
    Patient
    Education
    Most Regulation
    Least Regulation
    Auto Literature Review
    Specialty-specific Content

    View Slide

  34. Where ML and AI are applicable (care)
    Therapies
    Therapeutic
    Tools
    Diagnostics
    Diagnostic
    Tools
    Patient
    Administration
    Payer Admin
    Clinical
    Professional
    Education
    Public Health
    Education
    Patient
    Education
    Most Regulation
    Least Regulation
    Auto Adjudication
    Fraud Detection
    Quality Compliance
    Contract Adherence

    View Slide

  35. Where ML and AI are applicable (care)
    Therapies
    Therapeutic
    Tools
    Diagnostics
    Diagnostic
    Tools
    Patient
    Administration
    Payer Admin
    Clinical
    Professional
    Education
    Public Health
    Education
    Patient
    Education
    Most Regulation
    Least Regulation
    Patient Self Diagnostics
    Unlicensed Pro Diagnostics
    Digitally and Heuristically Guided Diagnostics
    Images (self, guided, consulted)
    Labs and Chemistry (self, guided, consulted)
    Multi-omics (self, guided, consulted)
    Molecular Biology

    View Slide

  36. Where ML and AI are applicable (care)
    Therapies
    Therapeutic
    Tools
    Diagnostics
    Diagnostic
    Tools
    Patient
    Administration
    Payer Admin
    Clinical
    Professional
    Education
    Public Health
    Education
    Patient
    Education
    Most Regulation
    Least Regulation
    Auto Triage for Low-risk
    Augmented Triage for Higher risk
    Infection control / Anti-microbial Stewardship
    Consulted Tele Diagnostics
    Med Device Continuous Diagnostics

    View Slide

  37. Where ML and AI are applicable (care)
    Therapies
    Therapeutic
    Tools
    Diagnostics
    Diagnostic
    Tools
    Patient
    Administration
    Payer Admin
    Clinical
    Professional
    Education
    Public Health
    Education
    Patient
    Education
    Most Regulation
    Least Regulation
    Physical
    Mental (chat, VR, etc.)
    Digital (nutritional, etc.)
    Clinical Research ( “systematic review automation”)
    Drug Development
    Clinical Discovery (unattended and digital)

    View Slide

  38. Where ML and AI are applicable (data)
    Proteomics
    Genomics
    Biochemical
    Imaging
    Behavioral
    Phenotypics
    Admin
    Economics
    Connectivity Integration Transformation Comprehension Enrichment Insights Cognition
    No ML or AI possible without these

    View Slide

  39. View Slide

  40. View Slide

  41. How should machines go through medical training?
    Which medical school will have the first machine
    learning algorithm training department?

    View Slide

  42. www.netspective.com
    © 2017 Netspective. All Rights Reserved.
    42
    THANK YOU
    Shahid N. Shah
    @ShahidNShah
    Finding the Human(s) in
    Health and IT

    View Slide