data operating system: what it really means and why … 5.66 ehr w/dr 17.nextgen - ambulatory...

44
Data Operating System: What It Really Means and Why You Will Need It Imran Qureshi Chief Software Development Officer Health Catalyst Sean Stohl SVP - Platform Operations Product Development Health Catalyst Pete Hess VP Platform Engineering – DOS Health Catalyst Session #22

Upload: hakhue

Post on 19-Mar-2018

215 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Data Operating System:What It Really Means and Why You Will Need It

Imran QureshiChief Software Development OfficerHealth Catalyst

Sean StohlSVP - Platform Operations Product DevelopmentHealth Catalyst

Pete HessVP Platform Engineering – DOSHealth Catalyst

Session #22

Page 2: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Learning Objectives• Describe the major problems health systems are facing today.

• Demonstrate why current analytics solutions aren’t good enough.

• Analyze why we need to evolve from data warehouses.

• Show how a data operating system addresses common use cases.

Page 3: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

The State of Healthcare Today

Page 4: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Hospital Margins Are at RiskHospital Profit Margins 6%

2011

0.5%Annual growth/cost reduction needed to maintain margins

“Revenue growth will also temper amid declining reimbursement from both private and governmental payers.”

“..uncertainty over federal health care policy also poses a strong headwind (for hospitals' financial performance).”

CBO Publication 2016 – Projecting Hospitals’ Profit Margins

Moody's: Preliminary FY 2016 US NFP

Hospitals with negative margins

3.3%2025 (Projected)

41%2025 (Projected)

Page 5: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Clinicians Are Over-Worked and Over-Measured

“…the emphasis is really on data

collection, but what physicians ought to be

doing is data synthesis.”

50% Physicians penalized 1% of

payments by Meaningful Use

Steven J. Stack, MD, President, AMA

Gary Botstein, MD, DeKalb Medical

“Today’s search engines are better at

helping clinicians diagnose disease than

our EMRs.”

Lloyd B. Minor, MD, Dean, Stanford School of Medicine

Page 6: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Risk Management and Population Health Require More Data

Source: Chilmark Research – 2017 Healthcare Analytics Market Trends Report

Page 7: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

EMRs1. Affinity - ADT/Registration2. Allscripts - Ambulatory EMR

Clinicals3. Allscripts

Enterprise/Touchworks -Ambulatory EMR

4. Allscripts Sunrise - Acute EMR Clinicals

5. Aprima ERM6. Cerner - Acute EMR Clinicals7. Cerner - PowerWorks

Ambulatory EMR8. Cerner HomeWorks - Other9. CPSI - Acute EMR Clinicals10. eClinicalWorks - Ambulatory

EMR Clinicals11. Epic - Acute EMR Clinicals12. Epic - Ambulatory EMR

Clinicals13. GE (IDX) Centricity -

Ambulatory EMR Clinicals14. McKesson Horizon - Acute

EMR Clinicals15. McKesson Horizon Enterprise

Visibility16. Meditech 5.66 EHR w/DR17. NextGen - Ambulatory Practice

Management18. Quality Systems (Next Gen) -

Ambulatory EMR Clinicals19. Siemens Sorian Clinicals -

Inpatient EMR

Finance/Costing1. Affinity - Costing2. Allscripts (EPSi) - Budget3. Allscripts (EPSi) - Costing4. Allscripts (TSI) - Costing5. BOXI - GL6. Cost Flex - Costing7. Digimax Materials

Management - Inventory Management

8. IOS ENVI - Costing9. Kaufman Hall Budget Advisor

- Other10. Lawson - Accounts Payable11. Lawson - Accounts

Receivable12. Lawson - GL13. Lawson - Supply Chain14. McKesson - Accounts

Payable15. McKesson Enterprise

Materials Management16. McKesson HPM - Costing17. McKesson HPM - GL18. McKesson PFM - Accounts

Payable19. McKesson PFM - GL20. McKesson Series - Accounts

Receivable21. Meditech - GL22. Microsoft Great Plains - GL23. Oracle (Hyperion) - Costing24. Oracle (PeopleSoft) - GL25. Oracle (PeopleSoft) - Supply

Chain26. PARExpress27. PPM - Costing28. Smartstream - GL29. StrataJazz - Costing

Billing1. Affinity - Hospital Billing2. CHMB 360+ RCM - Hospital

Billing3. CPSI - Hospital Billing4. Epic - Hospital Billing5. GE (IDX) Centricity - Hospital

Billing6. GE (IDX) Centricity -

Professional Billing7. HealthQuest - Patient

Accounting8. Keane - Hospital Billing9. McKesson Series - Patient

Billing10. McKesson STAR - Hospital

Billing11. MD Associates - Professional

Billing12. Siemens Sorian Financials -

Inpatient Registration and Billing

HR/ERP1. API Healthcare - Time and

Attendance2. iCIMS3. Kronos - HR4. Kronos - Time and

Attendance5. Lawson - HR6. Lawson - Payroll7. Lawson - Time and

Attendance8. Maestro9. MD People10. Now Solutions Empath - HR11. Oracle (PeopleSoft) - HR12. PeopleStrategy/Genesys -

HR13. PeopleStrategy/Genesys -

Payroll14. Ultimate Software Ultipro -

HR15. WorkDay

1. 835 – Denials2. Adirondack ACO Medicare3. Aetna - Claims4. Anthem - Claims5. Aon Hewitt - Claims6. BCBS Illinois7. BCBS Vermont8. Children's Community Health Plan

(CCHP) - Payer9. Cigna - Claims10. CIT Custom - Claims11. Cone Health Employee Plan (United

Medicare) - Claims12. Discharge Abstract Data (DAD)13. Hawaii Medical Service Association

(HMSA) - Claims14. HealthNet - Claims15. Healthscope16. Humana (PPO) - Claims17. Humana MA - Claims18. Kentucky Hospital Association (KHA) -

Claims19. Medicaid - Claims20. Medicaid - Claims - CCO21. Merit Cigna - Claims22. Merit SelectHealth - Claims23. MSSP (CMS) - Claims24. NextGen (CMS) - Claims25. Ohio Hospital Association (OHA) -

Claims26. ProHealth - Claims27. PWHP Custom - Claims28. QXNT - Claims29. UMR Claims Source30. Wisconsin Health Information

Organization (WHIO) - Claims

Claims1. Allscripts - Case Management2. Apollo - Lumed X Surgical System3. Aspire - Cardiovascular Registry4. Carestream - Other5. Cerner - Laboratory6. eClinicalWorks - Mountain Kidney Data

Extracts7. GE (IDX) Centricity Muse - Cardiology8. HST Pathways - Other9. ImageTrend10. ImmTrac11. Lancet Trauma Registry12. MacLab (CathLab)13. MIDAS - Infection Surveillance14. MIDAS - Other15. MIDAS - Risk Management16. Navitus - Pharmacy17. NHSN18. NSQIPFlatFile19. OBIX - Perinatal20. OnCore CTMS21. Orchard Software Harvest - Pathology22. PACSHealth - Radiology23. Pharmacy Benefits Manager24. PICIS (OPTUM) Perioperative Suite25. Provation26. Quadramed Patient Acuity Classification

System - Other27. QXNT/Vital - Member28. RLSolutions29. SafeTrace30. Siemens RIS - Radiology31. SIS Surgical Services32. StatusScope - Clinical Decisions33. Sunquest - Laboratory34. Sunrise Clinical Manager35. Surgical Information System36. TheraDoc37. TransChart - Other38. Varian Aria - Oncology39. Vigilanz - Infection Control

Clinical Specialty

1. Fazzi - Patient Satisfaction2. HealthStream - Patient

Satisfaction3. NRC Picker - Patient Satisfaction4. PRC - Patient Satisfaction5. Press Ganey - Patient

Satisfaction6. Sullivan Luallin - Patient

Satisfaction

Pat. Sat1. AHRQ Clinical Classification

Software (CCS)2. Charlson Deyo and Elixhauser

Comorbidity3. Clinical Improvement Grouper

(Care Process Hierarchy)4. CMS Hierarchical Condition

Category5. CMS Place Of Service6. LOINC7. National Drug Codes (NDC)8. NPI Registry9. Provider Taxonomy10. Rx Norm11. CMS/NQF Value Set Authority

Center

Terminology

1. Adirondack ACO Clinical Data from HIXNY (HIE)

2. ADT HIE Patient Programs 3. Vermont HIE

HIE

200+ Current Data Sources

Page 8: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Machine Learning Has Potential, but It’s Not Scaling

• Every machine learning project is a one-off requiring a lot of work.

• Hospitals can’t find enough data scientists.

• Clinicians don’t trust “black box” machine learning.

2xReadmissions predicted by UNC machine learning model vs LACE

Advisory Board 2017 - The Rise of Machine Learning.

So, machine learning feels out of reach…

20% Decrease in CLABSI rate

Indiana University System – Scottsdale Institute, Inside Edge (2016)

Page 9: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

50+ Health Catalyst Machine Learning Models

Page 10: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

What Is the Solution?

Page 11: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

First, Let Me Tell You a StoryA decade ago, best-of-breed applications ruled healthcare.• Each fit the individual workflows well.• But data remained in silos and apps did not

interoperate.

Today, we have one monolithic application (EHR).• Clinical data is in one place.

- Although non-clinical data remains separate.• But cited as:

- Biggest cause of physician burnout.- Forces one-size-fits-all workflows.- Stifles innovation.

Page 12: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Siloed apps don’t work.

App (EHR)

Data

One app for everything doesn’t work.

Data Data Data

App (EHR) App

Best of breed apps working together.

Data Operating System

AppApp

Data

App

Data

App

Data

Single Monolithic App Model Isn’t Working

Page 13: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

We Already Have a Model for This

Your iPhone has best-of-breed apps that all work together and share your data (e.g., your contacts).

Imagine if you were forced to use your email application to order Uber.

Page 14: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Precision Software

“EHRs would become commodity components in a larger platform that would include other transactional systems and data warehouses running myriad apps, and apps could have access to diverse sources of shared data beyond a single health system’s records.”

“A 21st-Century Health IT System — Creating a Real-World Information Economy”, Kenneth D. Mandl, MD, MPH; Isaac S. Kohane, MD, MPH; NEJM, 18 May 2017.

Best-of-breed apps

Many software vendors

Data operating

system

Centralized security

Machine learning

In clinical workflow

Page 15: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Health Catalyst Workflow Apps

Health Catalyst Data Operating System (DOS™)

Ope

ratio

ns &

Wor

kflo

w

Impr

ovem

ent

Patie

nt R

elat

ions

hip

& C

M

Popu

latio

n H

ealth

& A

C

Fina

ncia

l Im

prov

emen

t

Com

para

tive

Anal

ytic

s

Res

earc

h In

form

atic

s

Precise Patient Registries

Patient Safety

Clin

ical

Impr

ovem

ent

Leading Wisely

MACRA Measures & Insights

Measures Business Library

CAFÉ

Care Management

CORUS

HCC Insights

Uncompensated Care Suite

Machine Learning Text Analytics

Page 16: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

70+ Health Catalyst Analytics Accelerators

Acute Coronary Syndrome (ACS)Blood Utilization DashboardBreast Milk FeedingCatheter Associated Urinary TractInfection (CAUTI) PreventionCentral Line Associated Blood StreamInfections (CLABSI) PreventionColorectal SurgeryEarly Mobility in the ICUGlycemic Control in the HospitalHeart FailureJoint Replacement - Hip & KneeKey Process Analysis (KPA)Labor and DeliveryPatient flight Path - DiabetesPatient Safety ExplorerPediatric AppendectomyPediatric AsthmaPediatric ExplorerPediatric SepsisPneumoniaPopulation ExplorerReadmission ExplorerSepsis PreventionSpine SurgeryStroke (Acute Ischemic & TIA)Surgical Site Infection Prevention

Clinical Analytics & Decision SupportDepartment Explorer: Emergency ServicesDepartment Explorer: Surgical ServicesLabor Management ExplorerLeading WiselyPatient Experience ExplorerPatient Flow ExplorerPractice Management: Patient AccessProvider ProductivitySupply Chain ExplorerMACRA Measures & InsightsCommunity Care

Operations & Performance Management

CORUS (Clinical & Operational Resource Utilization System)Advanced Billing MonitoringFinancial Management ExplorerGeneral Ledger ExplorerHIM Documentation Workflow AnalyzerRevenue Cycle Advisor: HospitalRevenue Cycle Advisor: ProfessionalRevenue Gap Finder

Financial Decision Support

ACO ExplorerACO MSSP MeasuresAttribution ModelerBundled PaymentsHCC InsightsPMPM AnalyzerRisk Model Analyzer

Accountable Care

Patient StratificationPatient IntakeCare CoordinationCare CompanionCare Team InsightsCatalyst 4 Health

Care Management & Patient Relationships

Cohort Definition Framework

Research Informatics

Collective Analytics for Excellence (CAFÉ)

Comparative Analytics

Page 17: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

The Evolution from Data Warehouse

to a Data Operating System

Page 18: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Evolution from a Data Warehouse to a Data Operating System

Data Warehouse Data Operating System

1 Collects data from EHR and claims. Collects data from many sources.

2 Enables creating reports. Enables creating reports and web/mobile apps.

3 Enables SQL queries. Enables SQL, machine learning (R/Python) queries.

4 Data is updated nightly. Data is updated in real-time.

5 Not available in the EHR workflow. Insights are easily available in the EHR workflow.

6 Requires replacing your existing EDW. Works with your existing EDW (or use our EDW).

7 Proprietary schemas. Industry standard schemas (e.g., FHIR).

8 Text analytics is a separate process. Text analytics is built-in.

9 Works with rows and columns. Works with rows, columns and reusable healthcare logic like registries, measures, risk, insights.

10 Provides centralized security at app and data levels.

11 Makes machine learning as easy to use as SQL.

12 Content Marketplace to share executable content with other health systems.

Page 19: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

The Health Catalyst Data Operating System (DOS)

Data Ingest Real-time Streaming

Source Connectors

Catalyst Analytics Platform Fabric Data Services

Real time Processing

Health Catalyst Applications

Data Quality

Data Governance

Pattern Recognition

Hadoop/ Spark

Data Export

RegistryBuilder

Leading Wisely

Care Management

Atlas

Client BuiltApplications

Fabric Real Time Services

NLP

CAFÉBenchmarks

Choosing Wisely

Patient Safety

MeasureLibrary

ACO Financials

Patient Engagement

and more …

HL7

Data Pipelines Metadata

Data Lake

Reusable Content ML Models

3rd Party Apps

Machine Learning Pipelines

Marketplace

SAMD & SMD

Fabric Application Services

Registries Terminology & Groupers

FHIREHR Integration

Security, Identity& Compliance

Patient & Provider Matching

Measures

Standard Data Models

Page 20: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Monolithicsystem

Fabric - Building Blocks for Healthcare• Microservices.• Open APIs.• FHIR-based.• Open-source.• Reusable clinical and financial logic.• Fast access to ALL aggregated data.• Interoperable and portable apps.• Install only the components you want.• Replace any component with your own.

Page 21: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Who Needs a Data Operating System?

Clinicians Hospital IT Hospital Leaders

Data Administrators

Software Vendors Patients Data

Scientists

Page 22: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Clinicians Need a Data Operating System

Mergers & Acquisitions have left health systems with fragmented sources of data.

“I have to login to multiple EHRs just to see bits of patient data. How can I make a good decision about this patient in 10 minutes?”

What if you could access the data in a single place from all the fragmented sources of data?

Page 23: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Data Sources

Data Aggregation

Clients

Reusable logic

EHRs Claims Care Mgmt

Social Determinants of HealthDevice

G/L Cost (ERP)

Retail

Notes

Patient Generated

Patient Satisfaction

Reports Web Apps Mobile Apps

Data Warehouse

Registries Measures Cost Accounting Insights Risk Trends Predictions

Machine Learning Models

Data Aggregation + Reusable Logic + Open APIs

EHR Apps

Page 24: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Clinicians Need a Data Operating System (cont.)

Extend life and value of current EHR investments.

“Our clinicians are over-worked and over-measured, but under-informed”

What if the EHR was not just a place to record data but a place where clinicians see synthesized data and are informed?

Page 25: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

How Can We Be Non-intrusive?

No more “black box” yes or no recommendationsFocus on data synthesis and let clinicians choose the right path

Page 26: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Refer to Pain Clinic

Send Summary to PCP15x

How likely are clinicians to follow care recommendations if they are in an EHR vs reports.

Page 27: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Evolution from a Data Warehouse to Data Operating System for Clinicians

EHR Data Operating System

1 Contains EHR data. Contains aggregated data from all sources.

2 Designed for data entry. Designed for data synthesis.

3 Each EHR is an island. Works with all the EHRs in your health system.

4 Apps are not portable. Apps are portable: standard APIs and data models.

5 No access to machine learning. Has built-in machine learning.

6 No reusable logic. Has reusable logic like registries, measures, risk.

7 Each integration requires making changes in the EHR.

Integrate once into your EHR. Insights from many apps can show up without reconfiguring the EHR.

8 Information overload for clinicians. Hospital IT can centrally define rules to control what is shown in EHR workspace.

Page 28: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Fabric.EHR, the GPS for Clinicians

Analyst(SQL)

Apps(FHIR)

Data Scientists(REST)

Fabric.EHR Hospital IT(Rules)

Target by Patient, User, and Encounter

Epic Cerner Allscripts Other

Page 29: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Poll Question #1

How many EHRs does your health system use?

1) Zero2) One3) Two4) Three5) Four or more6) Unsure or not applicable

Page 30: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Hospital IT Needs a Data Operating System

Scaling existing data warehouses

“We have the data, but it is impossible for us to scale to more than a few reports or apps with the team we have.”

What if you could enable lots of reports and apps without having to hire more people?

Page 31: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

SQL or Big Data?

SQL is great for most cases Simpler. Existing skills. Existing code.

You need big data Volume (typically > 5TB). Velocity (typically > 1/day). Variety (typically 10+ sources).

A data operating system allows you to use the same tools with SQL or big data.

False choice: You don’t need to choose!

Page 32: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Evolution from Data Warehouse to Data Operating System for IT

Data Warehouse Data Operating System

1 Source Mart Designer (SMD) creates source marts.

Source Mart Designer (SMD) creates source marts and data lakes.

2 Subject Area Mart Designer (SAMD) creates SQL bindings.

Subject Area Mart Designer (SAMD) creates SQL bindings and Hadoop/Spark bindings.

3 SAM Engine runs pipelines via SQL Server.

SAM Engine runs pipelines via SQL Server or via Hadoop.

4 Atlas tracks meta-data for SQL data.

Atlas tracks metadata for SQL data and Hadoop data with smart search across both.

5 Machine learning is a separate process.

Create machine learning (R/Python) bindings and SAM engine runs them at scale.

6 Only deals with data. myCatalyst portal to access all your analytical assets: data, reports, apps, re-usable logic, permissions with smart search across all of them.

Page 33: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

But I Already Have a Data Warehouse

Page 34: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

The Health Catalyst DOSRuns on Any Data Warehouse

Real-time Streaming

Data Warehouses Fabric Data Services

Real time Processing

Health Catalyst Applications

Data Quality

Data Governance

Pattern Recognition

RegistryBuilder

Leading Wisely

Care Management

Atlas

Client BuiltApplications

Fabric Real Time Services

NLP

CAFÉBenchmarks

Choosing Wisely

Patient Safety

MeasureLibrary

ACO Financials

Patient Engagement

and more …

HL7

Metadata

Reusable Content ML Models

3rd Party Apps

Machine Learning Pipelines

Marketplace

SAMD & SMD

Fabric Application Services

Registries Terminology & Groupers

FHIREHR Integration

Security, Identity& Compliance

Patient & Provider Matching

Measures

Standard Data Models

HC EDW

Epic Cerner

Teradata

Home grown

IBMOracle

Hadoop/ Spark

Allscripts

Page 35: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Hospital Leaders Need a Data Operating System

Machine learning is too hard.

“I would love to use machine learning, but I would have to build a huge infrastructure and

hire a bunch of expensive data scientists.”

What would a data platform look like if it was designed to make machine learning as easy

as SQL?

Providers need to manage risk.

“How can I manage risk if I don’t have all the data and can’t easily calculate risk?”

What if it was easy to calculate risk and trend it over time both at population level

and patient level?

Page 36: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Machine Learning as Easy as SQL

Model

Real DataThe data operating system takes care of running your machine learning algorithms at scale in production.

Predictions

Predictions can be routed to EHR workflow via the data operating system.

Feature Extraction

Tools for analysts to create “features” using SQL (or R/ Python).

Feature library for re-use.

Train and Model

Selection

Data operating system tools make it easy to train models.

Training Data

Access data as longitudinal patient records.

Access registries, measures, terminology hierarchies, cost accounting, etc.

Model Creation

Point-and-click wizards for common machine learning algorithms.

For advanced scenarios, include R, Python and deep learning scripts.

Page 37: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Poll Question #2

How does your health system calculate per-member, per-month costs for an insurance contract?

a) Machine learning modelsb) Extrapolating from last yearc) We negotiate up from what the insurance company offersd) Rough calculationse) We need to but can’t todayf) Unsure or not applicable

Page 38: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Data Administrators Need a Data Operating System

Security that is secure, but allows access to the data.

“I need to be HIPAA compliant so I can’t allow anyone to access the data.”

What if you could easily implement common healthcare security models in a central place?

Are you protecting the right thing?

Page 39: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

We Don’t Need More Locks

We need intelligent security…

Imagine your hospital having these on every door and passing out keys

Page 40: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Security Should Not Be an Afterthought

Secure by design• At app level: Fabric.Identity and Authorization

- Integrates with existing identity providers like AD.- Centralized permissions.

• At data level: Fabric.FHIR data service- Implements permissions at the data layer.- At conceptual layer (patient, medications, diagnosis,

claims) instead of table and column level.

Centralized Security Console• Set permissions for all your applications in one

place.

Implements HL7 Security Reference Model

Built-in support for common healthcare use cases:• By affiliates.• By role.• By care program.• By sensitivity of data.• By insurance company.• Define your own security

template.

Page 41: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Lessons Learned• Healthcare industry faces significant challenges

• Hospital margins are at risk.• Clinicians are over-worked and over-measured.• One monolithic app (EHR) model is not working.• Machine learning has potential, but is not scaling.

• The future of healthcare is an app ecosystem• Best-of-breed applications from many companies all working together.

• The data operating system is an evolution of the data warehouse• Enables best-of-breed applications.• Integrates into the EHR workflow.• Leverages big data, machine learning, and text analytics.

Page 42: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

ScalableAll services are designed to run in multiple nodes and cluster themselves automatically.

Easy Install and UpdatesAll services install via Docker containers.

Microservices ArchitectureREST-based services that can be called from web, mobile, or BI tools.

Secure by DesignSecurity services make it easy to build security into any application.

Open and Modular• All APIs will be publicly published.• Customers can pick and choose from the

Health Catalyst components or replace any component with their own or from a third party.

Can Health Catalyst Be a Role Model?Open Source and Collaborative Development• All code is available on

github.com/HealthCatalyst.• External developers can submit enhancements.

Page 43: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Join Us

• Help us evolve the data operating system concept.

• Start using the available Health Catalyst DOS components.

• Use the Health Catalyst DOS to start putting insights into EHR workflow for your clinicians.

• Build new components for the Health Catalyst DOS or build better versions of existing components.

Together we can make software be a catalyst not an obstacle [email protected]

Page 44: Data Operating System: What It Really Means and Why … 5.66 EHR w/DR 17.NextGen - Ambulatory Practice Management 18.Quality Systems (Next Gen) - ... Supply Chain 14.McKesson - Accounts

Thank You