The Power and Promise of Unstructured Patient Data

Unstructured search capabilities, superior natural language processing, and healthcare ontology capabilities will help distinguish the leading products information and data-driven decision making.

Transcript

1. © 2014 Healthline Networks Inc. Confidential and Proprietary. The Power and Promise of Unstructured Patient Data
2. 2 BIG DATA U.S. Spending on Healthcare Little Insight
3. 3 Data-Driven Solutions Can Improve Outcomes and Bend Cost Curves Source: JEGI, Gartner, McKinsey, ADA, AHA, HealthPartners Research Foundation, Healthline analysis McKinsey estimates the U.S. can save $300B-$450B per year from investments in Big Data analytics 2.5 2.6 2.7 2.8 2.9 3.0 3.1 3.2 3.3 3.4 3.5 1   2   3   4   $ Trillions What curve would look like with savings from successful use of Big Data U.S. Spending on Healthcare 2012 2013 2014 2015
4. 4 Driving Data from Descriptive to Prescriptive/Predictive Analytics Source: Liquid Analytics Tech investments shifting from collecting data to understanding it to making it actionable at the point of care Data Latency Reporting Analytics What happened? What will happen? Why did it happen? What is happening? What should we do? What can we offer? Data Information Knowledge Data Freshness
5. 5 Clinical Analysis, Data Mining, and Predictive Modeling Top of Mind Source: SearchHealthIT.com's business intelligence survey 0 10 20 30 40 50 60 70 80 other none administrative business intelligence predictive analysis data mining clinical data analysis Which advanced analytics tools does your organization plan to you use in the next 2 years? Results based on 243 responses from CIOs and senior IT executives at medical centers, health systems and physician practices across U.S.
6. 6 Goal: Making Unusable Data Actionable 90% of healthcare data over the next decade will be unstructured (IDC, Kaiser Family Foundation) •  Healthcare is moving to a value based model •  Providers need to make investments in data-driven technologies to manage the health of their patient populations more effectively •  A major factor mitigating the power of these analytics solutions is access to information-rich unstructured data (e.g., physician notes, family histories, etc.) •  Leveraging data—structured and unstructured—from disparate sources is key Leveraging Unstructured Data and Data from Disparate Sources Is Critical
7. 7 Unstructured search capabilities, superior natural language processing, and healthcare ontology capabilities will help distinguish the leading products in the category (information and data-driven decision making). Robust Health Informatics is the Key to Unlocking the Unusable Data “ “ Source:  JEGI  HCIT  Issues,  Trends  and  M&A  Outlook  2014  
8. 8 IMPROVE PATIENT CARE BETTER PRIORITIZE AND FOCUS HEALTHCARE RESOURCES UNDERSTAND AND REDUCE RISK Understanding Unstructured Patient Data Can Provide New Insights
9. 9 For Instance: Risk Assessment for Readmission Source:  CMS,  Healthcare  Cost  U7liza7on  Project,  AHA,  Healthline  analysis   Seven conditions / procedures account for 30 percent of potentially preventable readmissions: 1.  Heart failure (HF) 1 2.  Chronic obstructive pulmonary disease (COPD) 2 3.  Pneumonia 1 4.  Acute myocardial infarction 1 5.  Coronary artery bypass graft surgery 6.  Percutaneous transluminal coronary angioplasty 7.  Other vascular procedures Heart Failure Readmissions Average 300-bed hospital at 90% occupancy •  27,000 stays •  1,755 HF stays (~6.5%) •  439 HF readmissions (25%) •  $15,000 average cost of HF readmission •  $6.6M total HF readmission costs BY THE NUMBERS Note: Hospitals with high avoidable readmission for highlighted conditions/procedures currently penalized by CMS 1 Currently part of CMS Readmission Measures 2 COPD added to CMS Readmission measures for October 2014
10. 10 UNLOCKING UNSTRUCTURED DATA CAN ENABLE SYSTEMS TO IDENTIFY WHO IS IN THE HIGHEST RISK CATEGORY BASED ON A VARIETY OF FACTORS: 1.  Medical / Health Factors 2.  Psycho-Social Factors 3.  Socio-Economic Factors Understanding who is a highest risk for readmission makes the targeting of scare resources in terms of interventions and support possible at scale.
11. 11 Risk Assessment for Heart Failure (HF) Readmission Assump7ons:  6.5%  HF  stays  /  total  hospital  stays;  25%  HF  readmission  rate;  $15,000  avg  cost  of  HF  readmission;  75%  of  HF  readmits  theore7cally  avoidable  (CMS)   Source:  CMS,  Healthcare  Cost  U7liza7on  Project,  AHA,  Healthline  analysis         HF READMISSION – CUSTOMER ECONOMICS Average 300 Bed Hospital (90% Occupancy) 27,000 stays 1,755 HF stays 439 HF readmits $15,000 per readmit $6.6M total 15% reduction in readmits ~$1M cost savings $564 savings per admit Patients Costs Potential Cost Savings
12. 12 Important to a Growing Array of Risk-Bearing Entities (RBEs), Especially Providers Life Science (21%) Insurance (25%) Provider (54%) Physicians (9%) Hospital (45%) Source:  JEGI,  Gartner,  McKinsey,  Nuance,  Healthline  Analysis     U.S. HCIT Market ~$72B (2014) ~5% CAGR “Main driver of HCIT spending in U.S. can be attributed to hospitals, clinics and private practices implementing health IT solutions.” – VP Healthcare Solutions, Nuance 0.0   5.0   10.0   15.0   20.0   25.0   1   2   3   4   5   6   7   8   Spending on Healthcare Analytics $  Billions   2013                2014              2015              2016              2017            2018                2019              2020   ~25% CAGR ~65% from providers
13. © 2014 Healthline Networks Inc. Confidential and Proprietary. press@healthline.com