integrating data sources to support care coordination and
TRANSCRIPT
1
Integrating Data Sources to Support Care Coordination and Delivery
INT4, March 5, 2018
Edwin Miller, CTO and Cofounder, Aledade,
Dan Chavez, CEO, San Diego Health Connect
2
Daniel J. Chavez, MBA
Edwin Miller
Has no real or apparent conflicts of interest to report.
Conflict of Interest
3
Learning Objectives• Classify the sources of data required to provide coordinated care in
serving complex patients.
• Evaluate how new approaches to improving interoperability between stakeholders to support improved approaches in caring for complex patients.
• Describe the challenges in workflow integration for providers and payers when supported by an interoperable system which supports new data sources from disparate sources.
4
4
Introduction to Aledade
• Founded: June 2014
• Headquarters: Bethesda, MD
• Funded by Venture Capital
• Venrock
• ARCH Venture Partners
• Biomatics Capital
• Google Ventures
• 175 employees
• In-House Analytics, Technology, Regulatory and Practice Transformation Services
Good for Doctors
Good for Patients
Good for Society
5 5
18 States
20 ACOs
20 Value-Based MSSP Contracts
300+ Practices
1,400+ Physicians with 3.0M+ patients
330,000+ Attributed Patients230,000+ Medicare
100,000+ Commercial
60+ different Electronic Health Records
~$2.5 billion annual medical spend under
management
6
It’s All About Practice Transformation
7
End of Life Palliative Care
Transitions in Care Transitional Care Mgmt
High Risk Patients Care Management
Chronic Conditions
Clinical Quality
Medication Management
Referral Management
Healthy & Home
Access
Wellness
Immunization
Screening
Aledade Population Health Model
8
8Core Competencies –Get the Data
Practice(s)
Payers
Health Information
Exchange(s)
3rd Party
ADT Clinical Attribution ClaimsSchedulin
g
✔
✔ ✔ ✔✔
✔✔
✔✔
9
9Workflow and Interop
• Workflow for TCM
• Emphasis on TCM of high
priority patients
• Tracking of workflow
• Ongoing surveillance of
data feeds vs. claims
10
10 Transition of Care (ADT/ENS) Use Case
• Currently Connected to 18 HIEs
• Supplement with “FAX 2 ADT” approach
• Panel approach yields improved patient matching
• Drives TCM workflow tool
11
11Confidential & ProprietaryWe improved transitions of care
Source: Quarterly performance data reported to
Aledade from CMS
1. “National” includes all Medicare Fee-for-Service
beneficiaries. “MSSP” includes all beneficiaries in the
Medicare Shared Savings Program. Both are displayed
as “% Change, Relative to 2016 Q4.”
2. Each ACO has an individualized benchmark from
CMS assessment based on historical data.
In 2017, Aledade ACOs
have demonstrated:
• Fewer ED visits leading
to hospitalization
• Fewer readmissions
• Fewer hospital visits
12
12Confidential & Proprietary
We reduced utilization of Skilled Nursing Facilities
Source: Quarterly performance data reported to
Aledade from CMS
1. “National” includes all Medicare Fee-for-Service
beneficiaries. “MSSP” includes all beneficiaries in the
Medicare Shared Savings Program. Both are displayed
as “% Change, Relative to 2016 Q4.”
2. Each ACO has an individualized benchmark from
CMS assessment based on historical data.
In 2017, Aledade ACOs
have demonstrated:
• Decreased SNF length-
of-stay
• Decreased SNF visits
• Decreased SNF spend
13
Geographic Reach13
Utah
Idaho
Nevada
Arizona
Colorado
Nebraska
Current Landscape
5.1 million Lives
58 organizations
318+ Facilities
Connected
3.5 million Lives
14
Connected Data
ADT (Admits, Discharges and Transfers) – This HL7 type message is
designed to support the identification of patients moving in and out of different
healthcare facilities upon receipt of treatment.
ORU (Laboratory Results) –to share results from laboratory procedures
on patients. Reduce duplicate lab procedures and quickly access critical
lab data from IEHIE. Discharge Summaries, transcribed reports
VXU (Vaccinations) –report administered and historical vaccinations. VXU
messages may also be used to identify patients who have not received
vaccinations.
RDE (Medications) –communicate medications that have been prescribed by
the rendering provider. View history of medication on a patient, as well as to
avoid negative drug interactions and duplicate medication orders.
Demographics
Encounters
Problem Lists
Diagnosis Codes
Allergies
Medications
Immunizations
Labs
Progress Notes
Discharge Summaries
Imaging Narratives Reports
Providers
Hospitals
Health Plans
Laboratories
EMSImaging
SNF/Long Term Care
Medical Groups
State and County
Reporting
History and Physical
Vitals
15
Integrating Data Sources15
16
200
100
0
.
Prepared by County of San Diego, Health & Human Services Agency, Public Health Services, Epidemiology & Immunization Services, 01/22/18
(including California)
500
400
300
600
700
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
20
12
20
13
20
14
20
15
20
16
20
17
20
18
Hepatitis A Cases, San Diego County1994 ‐ 2018
Vaccineintroduced
Routine vaccination for children in high‐incidence states
Routine vaccination for all U.S. children
Non-Current Outbreak
Current Outbreak
68%Hospitalized
20Deaths
HEPATITIS A OUTBREAK IN SAN DIEGO COUNTY
Outbreak-
related
cases
Data updated 1/22/2018
49% - Homeless
46% - Drug Users
19
19
Vaccination
Sanitation
Education
20
20
21
Hep A Alert21
Hepatitis A Case Example
• Homeless, age 30
• 10 ED Visits (Mar–Aug)
• 1 Hospitalization
• 1 Hep A Vaccination
• Hx ETOH
March
April
May
June
August
ED Visit 3/31
July
ED Visit 4/3
ED Visit 5/6
ED Visit 5/19
ED Visit & Hospitalization 7/16
ED Visit 7/12
ED Visit 8/20
ED Visit 8/31
ED Visit 6/19
ED Visit 6/22
Hep A Diagnosis 7/16
Hep A Vaccination 6/7
Cellulitis & abscess of lung
Blisters on hand
Abdominal pain
Cough & brown phlegm
Vomiting
Convulsions
Rib pain
ETOH withdrawal
Weakness, cough, fever, chills
Onset of illness 6/16
ETOH withdrawal & Seizure
SYSTEMS OVERVIEW
Communicable
Disease Registry
Syndromic Surveillance
System
• Local system
• Web based
• Manual entry &
ELR interfaces
• 31,000 new cases
each year
• HIE Interfaces
• Local system
• SAS based
• 12 of 16 hospitals
sending HL7 data
• 2.5 million HL7
messages a month
• ED, IP, Outpatient
• HIE Interfaces
MRN MRN
Linkage
Potential
METHODS
• Both data sources prepped prior to linkage
• Only hospital organizations sending syndromic HL7
data through San Diego Health Connect Health
Information Exchange (HIE) used.
• Linkage performed using SAS
• Linkage: Where MRN&DOB(HepA) = MRN&DOB(Syndromic HL7)
• Syndromic HL7 messages combined by HL7 Visit ID to
construct a health care “visit” record
• Discrepant linkages assessed for project criteria
inclusion
0
20
100
80
60
40
120
Hepatitis A Outbreak CasesData Time Period (N=546 cases)
Syndromic Data Time Period(11,953,866 HL7 A01,A03,A04 Messages)
90Days
30Days
Distribution of Hepatitis A Outbreak Cases
METHODS (cont.)
Data as of 12/15/2017
26
27
27
28
28
29
14,289231 ICU CASES
206106
DEATHS
TOTAL REPORTED
INFLUENZA CASES
CURRENT UPDATEReported Influenza Cases Since July 1, 2017
OUTBREAKS77.6%
2.7%
0.3%
18.2%
0.2%0.1% 0.9%
Influenza A
Influenza A (H1N1)pdm09
Influenza A (H3) Seasonal
Influenza B
Influenza B/Yamagata
Influenza B/Victoria
Influenza A/B
Preliminary Results
Data Source: Reported Influenza Case Reports
Prepared by County of San Diego, Health & Human Services Agency,
Public Health Services, Epidemiology and Immunization Services Branch
1
/
3
0
/
2
0
1
8
30
INFLUENZA SURVEILLANCEUPDATE, 2017-18 YTD
Week Week
4 3 FYTD#Week
4 FYTD#
Week
4 FYTD#
All influenza detections reported (rapid or PCR) 601 1,168 14,289 288 2,398 445 2,450
Percent of emergency department visits for ILI 6% 9% 4% 5%
Percent of deaths registered with pneumonia and/or influenza 15% 14% 8% 9%
Number of influenza-related deaths reported^ 32 32 206 12 33 8# FYTD=Fiscal Year To Date (FY is July 1- June 30, Weeks 27-26). Total deaths reported in prior years: 87 in 2016-17, 68 in 2015-16, and 97 in 2014-15.
22
* Previous weeks case counts or percentages may change due to delayed processing or reporting.
** Includes FYs 2014-15, 2015-16, and 2016-17.
^ Current FY deaths are shown by week of report; by week of death for prior FYs.
Indicator
FY 2017-18* FY 2016-17
Prior 3-Year
Average**
Table 1. Influenza Surveillance Indicators
Preliminary Results
Data Source: Reported Influenza Case Reports
Prepared by County of San Diego, Health & Human Services Agency,
Public Health Services, Epidemiology and Immunization Services Branch
31
Influenza Linked to Myocardial InfarctionEvidence of an association between influenza and acute myocardial infarction (AMI) was
reinforced in a new study published last week in the New England Journal of Medicine
(NEJM). Canadian researchers compared databases of respiratory virus tests with those of
hospital admissions in Ontario and found that the risk of AMI increases six-fold during the
first seven days after a lab-confirmed flu diagnosis in adults 35 and older.
The researchers noted that the AMI risk may be higher for older adults, those sick with
influenza B, and people experiencing their first AMI. The team indicated that there is
stronger evidence that flu can trigger cardiac events. These findings may drive more efforts
to elevate vaccine coverage in people who are at risk for AMI. This supports a study
published last year that estimated the efficacy of influenza vaccine, in preventing AMI,
ranges from 15% to 45%. This is a similar range of efficacy compared with the accepted
routine coronary prevention measures, such as smoking cessation (32–43%), statins (19–
30%) and antihypertensive therapy (17–25%).
Influenza vaccine should be considered an integral part of cardiovascular disease
management and prevention. The NEJM study can be found here: Acute Myocardial
Infarction after Laboratory-Confirmed Influenza Infection.
1/30/2018
INFLUENZA SURVEILLANCEUPDATE, 2017-18 YTD
32
INTEGRATION IS THE ESSENCE OF WPW
33
Social Services
Behavioral Health
Health Care
34
Whole Person Wellness Target Outcomes
34
Health
Outcomes
Access to
housing and
supportive
services
Data sharing
and collection
across entities
Inappropriate
ED use and
hospitalizations
System
coordination to
improve access
35
Social Determinants of Health Data Will Soon Overtake Healthcare Data
35
Housing Stability Primary Care and
Prevention
Health Management
Nutrition & Food Security
Financial Wellness and Benefits
Activities of Daily Living
Social &
Community Connection
Legal & Criminal Justice
Safety & Disaster Utility & Technology
Transportation Education &
Human Development
Personal Care & Household Goods
Employment Development
36
• There is a new doc flowsheet available in the Flowsheet activity to assess a patient’s risk based on ten areas of focus rather than individual risk factors. This is also available as a MyChart Questionnaire.
• This is to assist clinicians in assessing their patient’s potential risk based on their access to social and economic opportunities; the resources and support available in the home, neighborhood, and community; the availability of food and water; and the nature of our social interactions and relationships.
36 New Flowsheet to Capture SDoH
and Compute Risk Score
37
Population Health is the Future of HIE
37
Patient Practice PublicPopulation
Informatics, standards, workforce, business drivers, governance
Source: AMIA, Fridsma
38
Integrated Data Sources for Care Coordination
Providers
Hospitals
Health Plans
Laboratories
EMSImaging
SNF/Long Term Care
Medical Groups
State and County
Reporting
PCDH
Arizona
Utah
Colorado
Nevada
Nebraska
Idaho
eHealthExchange
DOD
VA
Kaiser
DaVita
38
Housing Stability
Primary Care
and
Prevention
Health Management
Nutrition & Food Security
Financial
Wellness and Benefits
Activities of Daily Living
Social &
Community Connection
Legal &
Criminal Justice
Safety & Disaster
Utility & Technology
Transportation
Education &
Human Development
Personal Care
& Household Goods
Employment Development
38
Social Services
Behavioral Health
Health Care
39
QuestionsDan Chavez
CEO
San Diego Health Connect
619.573.4445
• Please complete the online session evaluation