Unlock Insights Enabling Data-driven Decisions with Databricks, Precisely & CMA
Download as PPTX, PDF0 likes192 views
The document outlines how data-driven approaches are transforming healthcare and governmental operations, focusing on two primary use cases: Medicaid program redesign and critical COVID-19 relief distribution. It highlights significant improvements achieved through the utilization of data analytics, such as reducing avoidable healthcare admissions and efficiently distributing federal funding. The integration of analytics with secure data portals has facilitated actionable insights, optimizing care delivery and enhancing transparency among stakeholders.
Unlock Insights Enabling Data-driven Decisions with Databricks, Precisely & CMA
- 1. Unlock Insights Enabling Data- Driven Government Showcasing Lessons Learned from Public Sector and Health & Human Services Agencies
- 2. Gary Davis - CMA
- 3. Unlocking Data to Transform Healthcare
- 4. Unlocking data to transform healthcare Two Use Cases • Transform a $70 Billion annual program serving more than 6.5M Members • Distribute billions in CARES COVID relief funding for public healthcare Unlock Insights Enabling Data-Driven Government 4 • https://youtu.be/P6IyJRXTunM
- 5. Medicaid transformation journey 5 Use Case #1: Transform a $70 Billion annual healthcare program serving more than 6.5M Members • 5 Year DSRIP Waiver Outcome Reduce Potentially Preventable Admissions by 26% and Potentially Preventable Readmissions by 18% Distribute 86% of the $6.8 Billion in Incentive Payments available across 25 Performing Provider Systems Facilitated the fundamentable and sustainable redesign of how Medicaid services are delivered and financed • DSRIP transitioned a greater percentage of NYS Medicaid Managed Care portfolio to Value Based Payment Pilots Shifting accountability to Value Based Arrangements Move 80% of Medicaid Managed Care Contracts to VBP, 35% which featured upside/downside risk Medicaid spending per beneficiary in 2019 was less than 2011 Unlock Insights Enabling Data-Driven Government
- 6. 6 Focus on Operational Performance • Develop and deploy new MDW • Orchestrate of data workflows between MDW and downstream Data Marts • Enhance analytic capabilities for Fraud & Abuse, Rate Setting Focus on Medicaid Program Redesign • Develop Data Enrichment & Analytics for 1115 (DSRIP) Waiver • Develop and Deploy - Medicaid Analytics & Performance Portal (MAPP) • Integrate the State’s - Health Homes Case Management System • Develop Data Enrichment & Analytics for Value Based Payment - Pilots Medicaid Data Warehouse has provided the flexibility needed to address an evolving set of requirements Unlock Insights Enabling Data-Driven Government
- 7. 7 • Attributing Medicaid Beneficiaries to one of 25 Performing Provider Systems • Valuation of DSRIP Specific Projects for Potential Incentive Payments • Enrichment data via Location, Grouper and MDM tools • Baselining Cost/Quality for Attributed Populations on a PPS-specific basis • Monitoring & Reporting Project-Specific Performance Metrics (HEDIS, AHRQ) • Calculating the distribution of up to $6.8 Billion in Incentive Payments • Enabling internal and external stakeholders with highly-directive, interpretive dashboards with drill down capabilities via MAPP Data, Analytics and Visualizations have been integral to success to unlock data and impact change Trust and Transparency, Safe and Secure, User- Driven, Visual Discovery Unlock Insights Enabling Data-Driven Government
- 8. 8 Solution Overview: • Highly secure and scalable infrastructure integrate with Medicaid Data Warehouse (MDW) • Facilitates workflows and reporting associated with Provider Network Creation, Member Attribution, Reform Project Plans, Project Valuation Tool, Performance Goals and Quarterly Metrics • Leverages CMA technology accelerators with Tableau embedded as the standard analytic tool available for use by all authorized knowledge workers Outcomes: • Centralized asset enhanced level of trust and transparency between providers and NYS Medicaid • Facilitates the actionable, population-level insights needed to address gaps in care and improve outcomes Create a secure, web-based portal to facilitate access to trusted and transparent data and insights Unlock Insights Enabling Data-Driven Government
- 9. Unlocking Data to Transform Healthcare Sample Performance Dashboards
- 10. MAPP has served as the common portal for a variety of standard, DSRIP - Reports and Visualizations including Statewide Measures 10 Unlock Insights Enabling Data-Driven Government
- 11. MAPP also publishes the trending of Performance Measures for each Performing Provider System’s DSRIP Projects 11 Unlock Insights Enabling Data-Driven Government
- 12. Example: The Performance Overview dashboard also provides a deeper look into the episodic arrangements and the distribution of respective costs 12 1 MCO 006 might want to focus on performance improvements in Arrhythmia/Heart Block/CondnDis,sinceth e PAC Score Ration is so high. Same with Bipolar and Trauma & Stressors Disorders. 3 The Newborn episode is the main driver for high (bald) efficiency ratio in overall Maternity Care episodes. 2 Depression & Anxiety is one of the main drivers for high (bad) efficiency ratio in Chronic Bundle episodes. Unlock Insights Enabling Data-Driven Government
- 13. Example: The Efficiency Dashboard allows a user to view a peer to peer comparison of cost efficiency at the MCO level 13 1 2 3 Functions Structure Unlock Insights Enabling Data-Driven Government
- 14. Medicaid Transformation Journey 14 Use Case #2: Providing critical relief to Healthcare during COVID pandemic • The Challenge As part of the CARES Act, Congress allocated billions in critical relief to hospitals, physicians, and other health care providers including Medicaid providers. In April 2020, the CARES Act allocated $30 billion, and CMS required to providers within each state based on the percentage of Medicaid payments each provider received during the past two years. • The Results Leveraging the MDW, CMA was able to meet this complex challenge in a matter of days to apply complex Medicaid payment criteria to calculate managed care and lump-sum payments. Billions of dollars in relief from the federal government to State healthcare who have also gone above and beyond performing critical work to save lives during this health crisis. Unlock Insights Enabling Data-Driven Government
- 15. Mike Maxwell - Databricks
- 16. A unified data analytics platform for accelerating innovation across data engineering, data science, and analytics • Global company with over 5,000 customers and 450+ partners • Original creators of popular data and machine learning open source projects
- 17. Healthcare & Life Sciences Public Sector Financial Services Manufacturing Media & Entertainment Technology Retail, CPG, and eCommerce Data-driven innovation across industries Education 17 Unlock Insights Enabling Data-Driven Government
- 18. 18 Use Case • Opioid Epidemic Research • Study maternal-fetal outcomes across 3 petabytes of data • Fraud analytics • Analytical and statistical studies of chronic conditions • Evaluation & scoring of alternative payment models Why Databricks? • Medicaid claims are received in different formats from states and territories: Databricks accelerates ETL across many pipelines and processes • Databricks allows for experiments to run at 5x scale and 3x frequency, enabling rapid data exploration across large volumes of data Impact • Data-driven research used to support public health programs Application(s): Public health Primary Data Source(s): Claims, Geo, Social data Unlock Insights Enabling Data-Driven Government
- 19. Machine Learning Analytics DATA Unlocking mission value These institutions combined their massive data with machine learning and analytics capabilities Unlock Insights Enabling Data-Driven Government 19
- 21. Open Data Lake 21 Structured Semi-structured Unstructured Streaming Lakehouse Platform Data Engineering BI & SQL Analytics Real-time Data Applications Data Science & Machine Learning Data Management & Governance Open Data Lake SIMPLE OPEN COLLABORATIVE
- 22. Data Warehouses were purpose-built for BI and reporting, however… • No support for video, audio, text • No support for data science, ML • Limited support for streaming • Closed & proprietary formats Therefore, most data is stored in data lakes & blob stores ETL External Data Operational Data Data Warehouses BI Reports 22 Unlock Insights Enabling Data-Driven Government
- 23. Data Lakes could handle all your data for data science and ML, however… • Poor BI support • Complex to set up • Poor performance • Unreliable data swamps BI Data Science Machine Learning Structured, Semi-Structured and Unstructured Data Data Lake Real-Time Database Reports Data Warehouses Data Prep and Validation ETL 23 Unlock Insights Enabling Data-Driven Government
- 24. Coexistence is not a desirable strategy BI Data Science Machine Learning Structured, Semi-Structured and Unstructured Data Data Lake Real-Time Database Reports Data Warehouses Data Prep and Validation ETL ETL External Data Operational Data Data Warehouses BI Reports 24 + Unlock Insights Enabling Data-Driven Government
- 25. Lakehouse architecture offers a new way Open format and architecture on the data lake First class support for data science and SQL-based BI Excellent recency of data with one source of truth Analytics DS/ML Operational and External Structured, Semi-Structured, and Unstructured Data 25 Unlock Insights Enabling Data-Driven Government
- 26. provides a solution An approach to building curated data lakes BI/ Reporting DS/ML Operational and External Structured, Semi- Structured, and Unstructured Data Delta Lake • Adds reliability, performance, governance, and quality to existing data lakes • Based on open data format (Parquet) Real-time Applications 26 Unlock Insights Enabling Data-Driven Government
- 27. Databricks SQL Delivering analytics on the freshest data with data warehouse performance and data lake economics • Query your lakehouse with better price / performance • Simplify discovery and sharing of new insights • Connect to your preferred BI tools, like Tableau or Power BI • Simplify administration and governance Analytics & BI Machine Learning Operational and External Structured, Semi- Structured, and Unstructured Data Data Science Workspace Databricks SQL Delta Lake 27 Unlock Insights Enabling Data-Driven Government
- 28. BREAKING NEWS! Databricks SQL - Faster, Lower Cost 28 Unlock Insights Enabling Data-Driven Government
- 29. Open Data Lake 29 Lakehouse for Healthcare SIMPLE OPEN COLLABORATIVE Structured Semi-structured Unstructured Streaming Data Engineering BI & SQL Analytics Real-time Data Applications Data Science & Machine Learning Data Management & Governance Open Data Lake Claims Processing & Reporting (Quality, Cost Transparency) Predictive Care Mgmt (Readmission, Chronic Disease Programs) Admin Automation (Prior Auth, FWA with Human in Loop) Telemetry Health Risk Surveys Member Calls Webclicks Chatbots Social Behavior Claims Clinical (eg., ADT) Lab / Image Genomics
- 30. 30 Unlock Insights Enabling Data-Driven Government
- 31. Tim McKenzie - Precisely
- 32. Our Mission: Make Location Intelligence instant, powerful, and simple for everyone. 32 Unlock Insights Enabling Data-Driven Government
- 33. Superior understanding of your members and providers helps you to promote member health, identify, and engage ideal customers and prospects.
- 34. Sound familiar? Can we improve our speed of digital innovation? How do we ensure availability of care to our constituents? Can we identify “food deserts” within our population? How can we modernize our Member & Provider Portals? How can we analyze population health? How will housing migration patterns affect our network? How can I improve Big Data Analytics? How do we improve find the nearest provider? How do we increase our STAR ratings? Where do we need to place vaccination/ testing? 34 Unlock Insights Enabling Data-Driven Government
- 35. Location data prep slows data science Forbes, Mar 23, 2016 “Cleaning Big Data: Most Time-Consuming, Least Enjoyable Data Science Task, Survey Says” 3% 60% 19% 9% 4% 5% What data scientists spend the most time doing Building datasets Cleaning and organizing data Collecting datasets Mining data for patterns Refining algorithms Other Data preparation accounts for about 80% of the work of data scientists 35
- 36. Challenges in adding location to Data Science Massive Scale • Wearables & mobile trace • Social determinants • People Data: Thousands of demographic attributes Unique Data Types • Addresses, lat/long, shapes, lines, formats • Difficult to join different formats and data types with accuracy • Data types must be modified to be used in ML algorithms & analytical models Geospatial Calculations • Computationally intense to join and enrich spatial data at scale • Enriching with routing and catchment is critical, but highly time consuming 36 Unlock Insights Enabling Data-Driven Government
- 37. A location centric approach to data puts you in control 03. Analyze Apply data science at scale to gain a competitive advantage 02. Enrich Leverage trusted ID to join massive amounts of your own and 3rd party data sources 01. Organize Assign a trusted ID that is unique and persistent to each address 37 Unlock Insights Enabling Data-Driven Government
- 38. Business Locations Location Risk Data Consumer Demographics Property Attributes Admin Boundaries Schools & Neighborhood Add your data to analytics PreciselyID Hospital & Emergency Care Claims History Available Fresh Food Modeled Scores Available Transportation Nearest Family Practice & Specialist 38 Unlock Insights Enabling Data-Driven Government
- 39. 39 Precisely Big Data/Cloud SDKs Leverage cloud native solutions to create scalable location intelligence. Component Features Global Addressing & Geocoding • Address Validation • Forward and Reverse geocoding • Unique and Persistent Address Key (PreciselyID) Spatial Processing • Geohash ID Assignment • Point and polygon • Spatial join • Distance to point, shape, line Street Routing • Isochrone/Isodistance • Walk time/drivetime • Point-to-Point calculations 39 Unlock Insights Enabling Data-Driven Government
- 40. Location-enabled Lakehouse Care Facilities Connected Devices Provider Network Historical Claims Hospital Visits IoT Sensors Lakehouse Platform Geocoding and location intelligence capabilities to organize and enrich your data Clinical Information All of your sources Any structure or frequency Analytics capabilities for any use case or persona Ad Hoc Data Science Low-cost, rapid experimentation with new data and models. Explainable Machine Learning High volume, fine-grained analysis at scale served in the tightest of service windows. BI Reporting & Dashboarding Power real-time dashboarding directly, or feed data to a data warehouse for high-concurrency reporting. Real-time Applications Provide real-time data to downstream applications or power applications via APIs.. PreciselyID PUBLIC TRANSPORT VIDEO/ AERIAL IMAGERY EVERY ADDRESS WEATHER EVENTS HAZARD & RISK DATA AMENITIES & COMPETITION APARTMENT BUILDINGS MEDICAL FACILITIES PROPERTY ATTRIBUTES SCHOOLS & NEIGHBORHOODS POPULATION DEMOGRAPHICS PARCELS & BUILDINGS 40