Using Actuarial Analysis

to Impact Social
Determinants of Health
HFMA in partnership with the Society of Actuaries
Definition of Social Determinants of Health

Social determinants of health are conditions in the

environments in which people are born, live, learn, work,
play, worship and age that affect a wide range of health,
functioning and quality-of-life outcomes and risks.1

The ‘social determinants of health’ (SDOH) are the

conditions in which people are born, grow, work, live
and age, and the wider set of forces and systems
shaping the conditions of daily life.2
 Healthcare Contributions vs. Health Expenditures
• Healthcare accounts for 20% of factors contributing to health.
Social, environmental and behavioral factors account for 60%.
• Yet, healthcare represents 90% of national health expenditures.

Leveraging Healthcare Dollars to Address Health-Related Social Determinants

Factors Healthcare Genetics Social, Environment, and Behavioral Factors

Contributing
to Health 20% 20% 60%

U.S. Health Healthcare Behavioral Other

Expenditures* 90% 9% 1%

Adapted from James Rubin MD, TAV Health; * $3.2 Trillion in 2015 (CMS.gov Accessed 5/22/17
3
 Poll: Which of the following are true:

1. Patients with food insecurity are more likely to report multiple ER visits and
inpatient visits.
2. Patients with transportation needs are more likely to report multiple ER
visits and inpatient stay.
3. More social barriers are associated with increased healthcare needs
4. All of the above

HFMA National Update May 21, 2018

Poll: Which one of the following SDOH do you see as the largest
influencer on health in the patient population you serve?

1. Housing
2. Nutrition
3. Transportation
4. Physical/built environment
5. Social services support
6. Employment
7. Education
8. Income
9. Proximity/Access to healthcare
Fresh Food
Farmacy
Michelle Passaretti MSN BSN RN CCM
Senior Director of Innovations
Steele Institute of Health
Sarah MacDerment FSA, MAAA
Senior Director Actuarial Services
Geisinger Health Plan
About us
Geisinger is an integrated health care system
• We care for patients
11 hospital campuses
253 clinic sites
3000 providers
• We provide quality, affordable healthcare coverage
578,000 members
55,000 contracted facilities/providers
• We teach, research and innovate
523 MBS/MD students at GCSOM
48 GLH School of Nursing, 2,000+ nursing students
505 residents/fellows
1000 + active research projects
Polling Question

Americans with 5 or more chronic conditions

make up ______ of the population but account for
_______ of the total health care spending.1

1. 20% and 20%

2. 2% and 10%
3. 12% and 41%

1”Multiple Chronic Conditions in the United States”, Rand Corporation, 2017

Polling Question

Americans with 5 or more chronic conditions

make up 12% of the population but account for
41% of the total health care spending.1
US Population Health Care Costs

12% 10%

16% 40% 41%

23%

31% 26%

# of Chronic Conditions # of Chronic Conditions

0 1-2 3-4 5+ 0 1-2 3-4 5+

1”Multiple Chronic Conditions in the United States”, Rand Corporation, 2017

 Diabetes – A national call for help

Population Health Impacts Financial Impacts

Diabetes has highest healthcare

2018 2050 spend4
1 in 10 1 in 3
Adults Adults With
With Type Type II
II Diabetes Diabetes How do we solve the
problem?
$327 billion/year
HBA1c of 6.5-9 Diagnosed diabetes cost to America3
1 in 5 are food insecure2
2.3 times greater
HBA1c >9 Cost of healthcare with diabetes
1 in 4 are food insecure2 compared to without diabetes3

1DiabetesStatistics Report, 2014

2BerkowitzSA, Baggett TP, Wexler DJ, Huskey KW, Wee CC. Food insecurity and metabolic control among U.S. adults with diabetes. Diabetes Care. 2013;36:3093-3099
3ADA Website
4US Spending on Personal Health Care and Public Health, 1996–2013,” JAMA, 2016
A vicious bidirectional relationship: Which came first?
• Food insecurity can worsen a
person’s chronic conditions i.e.) o Reduced capacity to work
diabetes and vice versa o Higher stress and obesity
o Lower dietary quality o Poorer mental health
 Lack of funds or access to nutritious o Lack of opportunities to be
foods
physically active
 High availability of nutritiously poor
foods o Limited access to healthcare
o Increased spending on medical o Lack of transportation
care (“treat or eat”) o More emergency room visits
o Decreased capability for self-care

http://www.medscape.com/viewarticle/865606 &
https://www.diabetesselfmanagement.com/blog/are-you-food-insecure/
Innovative collaboration between clinical care and community
based organizations – New processes driving impactful change
How did we rethink our processes

A new model with 5 basic elements

1. Identification
2. Food as medicine
3. Education/Clinical support
4. Care beyond health
5. Community partnerships
How we do it.
1. Identification
• Medical record clinical criteria and screening questions
• 18 years old
• Dx Type II
• HbA1c 8.0 or higher
• Food insecure-collaborative definition
• With the past 12 month, we worried whether our food would run
out before got money to buy more (Y/N)
• Within the past 12 months, the food bought just didn’t last and
we didn’t have money to get more (Y/N)
• Physician or care team member referral
• Collect baseline outcomes and closure of care gaps
How we do it.
2. Food as medicine
• Stock only fruit and vegetables, lean meats,
whole grains, proteins and healthy options
• Provide enough food to create healthy meals for
five days a week, two meals a day. (10 meals a
week, every week)
• Provide food for the entire household
How we do it.
3. Education
• Care team providing in person and/or telephonic
support
• Screen for complications and close care gaps
• Lifestyle changes: Cornerstone of treatment
• In person evidence based programming
• Meal planning and recipes
• Additional supplemental lifestyle management programs
including tobacco cessation, physical activity programs,
grocery tours, food demonstrations, etc.
How we do it.
4. Care beyond health
• Access to transportation, housing and food stamp
programs

How we do it.
5. Community Partnerships
The Current Patient Experience – Meet Tom

Began regularly
Worked with checking sugars,
Enrolled in watching diet, &
Case Manager on
Fresh Food walking for exercise.
diet, exercise, &
Farmacy Began teaching classes
proper diabetes
management for the program.
Age 57
Condition(s) Diabetes

About Tom:
• Single
• Lives alone
9.1 A1C
• Multiple ED visits / admissions
450 lbs.
related to poor diabetes care
355 Triglycerides
• Doesn’t know how to cook
• Reported feeling alone & depressed 6.6 A1C
400 lbs.
152 Triglycerides
FFF Outcomes
All-In
Currently Enrolled 747

Average Family Size 3

Food Provisions 1,263,316 pounds of food

1,067,718 meals

Outcomes: 64% Compliant on eye exams

66% Compliant on foot exams
83% Blood Pressure in goal

2.4 A1c reduction in A1c

 ~27 PCP &
14% difference
in Endo visits

~13%
difference in
FOOD AS ED visits

MEDICINE IS ~49%
HAVING A REAL difference in
admission rates
FINANCIAL
IMPACT

Non-enrollees Enrollees
Thank you
 Poll: _____ % of US households have no
emergency savings:
1. 20
2. 45
3. 53
4. 75

HFMA National Update May 21, 2018

SDOH ANALYTICS
Use Case: Improving coding
and documentation for MA
members

Jodi Haddad
Jim Dolstad
Coding and documentation: Key success factor for MA
plans
• Situation • Solution
• Enrollment in MA plans growing • Excel at coding and documentation
year over year • Payer/provider collaboration
• Providers under value based • Risk sharing arrangements
contracting growing year over year • Interoperability
• Invest in analytics
• Members prefer
• Plans with Star ratings of 4 or higher
• Leverage traditional and new data
sources
• No/low premium cost
• Understand the members’ needs
• Low copays and supplemental benefits (SDOH)
• CMS sets county level rates • Transportation insecurity
adjusted at payer level for • Social isolation
• Star rating • Primary language
• Risk adjustment factor • Health literacy
• Financial insecurity

© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 24

Do not distribute or reproduce without express permission from Optum.
Holistic View of the Member - SDOH
Enables everyone to work smarter by knowing the customer not just the patient

99.9% 80%

Percentage of time members spend Percentage of member healthcare

outside of a clinician’s office costs driven by lifestyle and behaviors

© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 25

Do not distribute or reproduce without express permission from Optum.
What is possible by leveraging SDOH of a member
Incorporating SDOH into analytics can improve the following opportunity areas:

Understanding SDOH variances across

Growth & Retention provider groups

Improved clinical, financial and Benefit design that better

operational performance aligns with member needs

Coding and documentation to Better geographic alignment of

drive CMS quality and risk community resources and member
adjustment revenue needs

Member outreach that reflects Measurable understanding of health

who is willing to be impacted disparities
and how

© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 26

Do not distribute or reproduce without express permission from Optum.
Moving from Collection of SDOH to Prediction
Gaining knowledge on 85% of the membership rather than less than 5% of membership through Z codes

Health Ownership
Identifies the members’ level of interaction with the health care Housing Security
system and personal health choices based on their health status. Predicts the likelihood of experiencing housing security issues
It’s a measure of a member’s level of engagement with their own and being homeless.
health.

Financial Security Social Isolation

Predicts the likelihood of deferring treatment due to concerns Predicts the likelihood for someone to have lower social ties to
family and friends; a measure of social frailty or involuntary
over cost; this can have down-the-line impact on outcome and loneliness which we know can have a profound impact on
utilization. utilization.

Propensity to Engage Out of Network Utilization

Predicts the likelihood for someone to engage in care & disease
Predicts the likelihood of using providers outside of the
management programs, or to utilize a specific communication
members’ indicated network to get the care they need.
channel, such as phone.

Food Security Transportation Access

Predicts the likelihood of suffering from food security issues Predicts the likelihood of experiencing transportation access
either because of financial reasons or because of lack of access to issues, for financial reasons or due to the lack of access to reliable
healthy food. public transportation.

© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 27

Do not distribute or reproduce without express permission from Optum.
Provider group patient bases have varying degrees of SDOH
Understanding the differences between provider group patient bases enables focused strategies and support for each provider group
based on the needs of a given provider group’s patient base.

© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 28

Do not distribute or reproduce without express permission from Optum.
From data collection to measurement of success
Data Assimilation
Integrating COVID and SDOH data sets Analytics & Recommendations
and analytics with Quality and RAF
data and analytics Develop member & provider targets based on
missed opportunities -County level analysis

Market Data Sharing, Collaboration Execution

Adapt strategies to address specific market initiatives Member level lists integrated into existing provider
and provider group variances & plan targeting workflows

Measure Success
Pre/Post member level assessment. Actual
dollar impact leveraging MMR files.

© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 29

Do not distribute or reproduce without express permission from Optum.
Payer Provider Collaboration –
Leveraging SDoH Data to Optimize Results
Case Study – Midwest market
What did we do?
• We took our list of high priority members (members without a visit, who have multiple
chronic conditions) and pulled in data around the member’s propensity to engage with
their provider

How did the data drive action?

• Understanding the member’s propensity to engage helped us focus limited resources
on the areas where we had the opportunity to make the greatest impact
• Predictive SDoH data told us how a member would most likely engage in care: In
office Annual Care Visit, Telehealth visit, or In-home visit
• Created Member Outreach Campaigns based on members’ preferences to more
quickly schedule need care

What were the results?

• Member Outreach calls were more efficient because the caller started with the
member’s preferred visit type
• The Midwest market saw a 30% increase in scheduled visits

© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 30

Do not distribute or reproduce without express permission from Optum.
Payer Provider Collaboration –
Leveraging SDoH Data to Optimize Results
Case Study – Southern market
What did we do?
• Shared SDoH data with providers to help address medication adherence issues

How did the data drive action?

• Identified patients that might have financial insecurity preventing them from filling
necessary medication prescriptions. This insight allowed the provider to prescribe a
more affordable prescription, and increase Part D Med Adherence compliance

What were the results?

• Providers loved having this information, understanding the SDoH barriers helped
drive med adherence discussions and improved patient compliance on these critical
Part D measures

© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 31

Do not distribute or reproduce without express permission from Optum.
Measurable impact of SDOH
Increasing in-office assessments for MA plans

© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 32

Do not distribute or reproduce without express permission from Optum.
Measurable lift in annual care visits through focused targeting

• A and B markets
had both Analytics
and Outreach
efforts. All other
markets only reflect
Analytics
• For the all-other
markets the visits
per member were
1.11% for the
Control group
compared to 4.26%
for the Optum group
• For A and B the
visits per members
was 6.04% for the
Optum group and
1.40% for the
Control group
© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 33
Do not distribute or reproduce without express permission from Optum.
Lessons Learned & Best Practices

© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 34

Do not distribute or reproduce without express permission from Optum.
Roadmap for continued payer/provider
collaboration
Goal: Leverage SDOH data to
quickly and effectively
engage with providers and
Q1 Member Outreach: Leverage SDOH data to improve
call success rate
members

New Membership: Share with providers to

Goal: Share insight around
how best to connect with Q2 understand how new members want to engage to
ensure they are quickly seen & assessed
members to get the most
critical members seen ASAP

Goal: Work with health plan
to map out other areas of
opportunity for 2022
Q3-
Q4
Q2- Drive ACVs: Share with providers to drive ACV completion
rates, focus on MWOV + multiple HCCs
Q3
Improve CAHPS/HOS Scores: Leverage SDoH data to
address patient needs and build
© 2021 Optum, Inc. All rights reserved. Confidential property of Optum. 35
Do not distribute or reproduce without express permission from Optum.
Questions and Answers
Thank You!
 HFMA On-line Evaluation

URL: http://api.hfma.org/Site/events/atc_evaluation.cfm

Enter your member ID # (can be found in your confirmation email when you
registered)

Enter Meeting Code: 21AT28

Your comments are very important and enable us to bring you the
highest-quality programs!

38 Healthcare Financial Management Association | hfma.org