A Catalyst to Drive Behavior and Engagement in a Value-Based World

17
©2016 SCIO Health Analytics ® . Confidential and Proprietary. All rights reserved. | 1 Predictive Analytics A Catalyst to Drive Behavior and Engagement in a Value-based World July 22, 2016 Rose Higgins President, North America

Transcript of A Catalyst to Drive Behavior and Engagement in a Value-Based World

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |1

Predictive Analytics

A Catalyst to Drive Behavior and Engagement in a Value-based World

July 22, 2016

Rose HigginsPresident, North America

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |2

Market Driver: Connecting and Coordinating CareD

RIV

ING

FO

RC

ES

HEALTHCARE REGULATIONS

AFFORDABLE CARE ACT (ACA)

MEDICARE

Data needs are expanding beyond paid claims.Acquiring, aggregating, and managing heterogeneous

data is a core competence.Utilize data for actionable business value

PARADIGM SHIFT

Fee for Service Fee for Value

Sick Care Well Care

Fragmented Episodes Coordinated Care

Acute Care Focused Cross Continuum

Patient Consumer

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |3

Increased scrutiny, audits,

rate cuts, STAR ratings and

Quality/Satisfaction

Measures

Value-based care

• Rewards outcomes and

effectiveness

• Financial accountability for

care

• Requires precise resource use

• Patient Satisfaction and

Engagement

Fee for service

• Illusion of provider sovereignty

• Financial conflicts for patients

• Rewards increased intensity

and volume

2000 2020

Intensity of Care

Labor Costs

Supply Costs

Whatis Reimbursed

$

Unsustainable

Costs

Changing

Payment

Models

Pressure from Payers

(Employers,

Govt, Plans)

Market Driver: Cost Pressures

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |4

Emerging Trends

Increased mobile usage

Payment tied to VBH

Consumers as new $

managers

Precision Medicine

Predictive/Anticipatory

Cloud

Wearables become fully

integrated Medical devices

data

ACO & IDNs become the

standard care delivery

models

Willingness to participate in

telehealth

Payers demand evidence of

“patient success”

Participatory

Real time

Social Networks

MACRA / MIPS / APS

Shift to lowest level care

setting- home will become

where most care is delivered

Pharma patient centricity

Preventive

Big Data

Collaborative and social BI

increases

Rising drug cost impact

Consumers choose

nontraditional providers

Pharma shifts to drive

better pt. outcomes.

Personalized

Multi data sources

Technology Enabled Behavior

Change

Volume to Value

Consumer Engagement

Personalized Care Delivery

Actionable Analytics

Data Handling

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |5

Increasingly Complex Patient Populations

Older patients with multiple chronic conditions consume an outsized amount of resources

Longer Life Span

Boomer Generation

Two thirds of the Medicare +beneficiaries have two or more chronic conditions.

(2010 Center for Disease Control Study)

There are

49,432,610 Medicare beneficiaries in the US

(Kaiser Family Foundation)

Most Common Chronic Conditions High Blood Pressure High Cholesterol Heart Disease Arthritis Diabetes

Average health costs for someone who has one or more chronic conditions is greater than for someone without any chronic conditions

55X

2 2 3 4 2 3 1 0 1

50% 45% 31% 29% 28%

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |6

Using Analytics to Better Understand Your Populations

Increasingly Complex

Patient Populations

Accelerating Shift to

Value-Based Care

≥1 Chronic Condition = the Cost5x

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |7

The Foundation of Population Health: Robust Profiles

Member, Patient or

Provider Profile

Clinical FactorsRisk, Cost,

Quality, Utilization, & Attribution

Demographic Behaviors Attitudes

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |8

View Populations Through the Lens of Impactability

Population

100%

Impactability Prospective Risk

Moderate Impactability

12% of Members

Low Impactability

75% of Members

High Impactability

12% of Members

High

Low

Op

po

rtun

ity

Goal

Close Gaps

and Steerage

to Managed

Networks

Close Gaps

and Steerage

to Managed

Networks

Manage

High Costs

and Risk

Factors

Manage

High Costs

High Risk

10%

Moderate Risk

1.5%

Low Risk

0.5%

High Risk

8%

Moderate Risk

3%

Low Risk

1%

High Risk

13.5%

Moderate Risk

27%

Low Risk

34.5%

High Cost

1% of Members

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |9

Allocate Resources Towards Impactable Conditions

Diabetes, $326,515,914

COPD, $56,262,780

Seizures, $20,743,606

Obesity, $53,401,848

Rheumatoid Arthritis, $47,819,586

Inflammatory Bowel, $13,258,925

Back Pain, $303,270,098

Depression, $158,480,248

Hyperlipidemia, $326,077,730

Asthma, $119,587,531 CHF, $69,839,071

Maternity, $69,955,753

CKD, $48,898,470

Parkinson Disease, $1,602,486

CAD, $112,816,271

Hypertension, $497,076,823

-10

-8

-6

-4

-2

0

2

4

6

8

10

-10.0 -8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0 10.0

Condition Intervention Summary

Harder to Impact

More Complex Interventions

Less Complex Interventions

Easier to Impact

High Volume Silent Diseases

High Volume Symptomatic Diseases

Maternity

Symptomatic Chronic Low Numbers

Diabetes, $497,076,823

Hypertension,

$326,515,914

Optimal Interventions for Diabetes

1 Eye Exam

2 HbA1c

3 Lipid Test

4 Medication Regimen Compliance

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |10

Track Compliance and Progress

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |11

Provider & Patient Engagement

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |12

Segmenting Patients by Consumer Type

Healthy & Affluent

BalancedAdults

High Utilizers Quality Driven Cost ConsciousChronic older

AdultsHigh Cost Baby

Boomers

No.of chronic

conditions

ER Paid PMPM

IP Paid PMPM

ER Utilization

IP Utilization

0.540.70 0.71

0.86 0.82

1.021.13

Median Risk Prospective Score

0.6 0.7 0.8 1.2 1.2 1.3 1.6

0.09 0.05 0.10 0.04 0.07 0.08 0.09

0.25 0.22 0.34 0.23 0.18 0.21 0.23

$75 $73 $147 $54 $75 $118 $248

$10 $9 $14 $9 $7 $10 $11

Healthy & Affluent Balanced Adults Value Driven Quality Driven Cost ConsciousChronic older

AdultsHigh Cost Baby

Boomers

44% 24% 19% 29% 8%12% 11%

67% 63% 32% 54% 14% 40% 54%

58% 30% 14% 33% 2% 13% 12%

Net Worth

> $100K

Estimated HH Income

> $100K

Estimated HH Debt

> $15K

Median Home Value

0.540.70 0.71

0.86 0.821.02

1.13

Median Risk Prospective Score

$333K $285K $216K $165K $151K $146K $175K

Healthy & Affluent Balanced Adults Value Driven Quality Driven Cost ConsciousChronic older

AdultsHigh Cost Baby

Boomers

Habitual Frequent Frequent Frequent Occasional Sporadic Sporadic

Phone MailCell

Text

E-mail Web

Text

E-mail E-mail

Channel Preference

Spending Pattern

No. of Automobiles

0.540.70 0.71

0.86 0.821.02

1.13

Median Risk Prospective Score

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |13

Prioritize Care Gap Closure at an Individual Level

Member ID Risk Score Impactability Score Gap1 Gap2 Gap3

000000010506 0.89 1.68 Diabetes - Consider Foot Exam HbA1c Less Than 7 Target

000000010331 0.83 1.51 Lipid Panel Spirometry

000000010043 0.81 1.64 Consider Pulmonary Rehabilitation AST Test Physical Therapy

000000010154 0.73 1.39 Lipid Panel Spirometry Alpha-Glucosidase

000000010539 0.73 1.04Diabetes and Macroalbuminuria - Consider

Adding an ACE Inhibitor or ARB

Diabetics 50 years and Older - Consider

Screening for Peripheral Arterial Disease

Member ID

In Last 12 Months Cost Incurred in Last 12 Months Probability

of ER

Admission

Predicted

Probability of ER

Admit IF all the

gaps are closed

DifferenceImpactability

Score#

Hospitalization

# ER

Visits

InPatient

(PMPM)

ER

(PMPM)

OutPatient

(PMPM)

Professional

(PMPM)

Pharmacy

(PMPM)

000000010506 1 1 $2,999 $302 $209 $201 $130 93% 22% 71% 1.68

000000010331 0 0 $237 $158 $147 90% 27% 64% 1.51

000000010043 0 2 $287 $231 $225 $133 91% 22% 69% 1.64

000000010154 0 0 $231 $178 $103 74% 16% 58% 1.39

000000010539 0 0 $340 $181 $96 70% 27% 44% 1.04

000000010507 0 0 $333 $208 $134 73% 24% 49% 1.15

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |14

Analytics Driving Engagement Outreach

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |15

Small Improvements Carry Significant Revenue Implications

Calculation: $800 average pmpm payment x RAF

A 45,000 member health plan purchases Medicare risk adjustment

and HEDIS/Star/P4P monitoring analytics

Within 90 days their systems are online to support new suspecting and provider collaboration programs:

• Identification: A prioritized list of all patients that need to be seen by 12/31 to ensure care gaps are closed

and revenue streams remain constant

• Provider Collaboration: Each morning physicians receive patient-specific pre-populated forms containing

previously diagnosed conditions and any outstanding Stars measure assessments needed for that member.

PMPM Revenue

In two months, the health plan increases their average Medicare RAF score from .95 to .98 and sees significant

improvement on a number of clinical Star measures

$760

$784

$0 $100 $200 $300 $400 $500 $600 $700 $800 $900

.98 RAF

.95 RAF

Pre-Solution Revenue Post-Solution Revenue

$24

Provider

Reimbursement for many providers is based on % of revenue/premium

At 35% of premium, this example generates an additional provider revenue of

$4,500,000

Health Plan

An increase of just 0.03 to the RAF score generated an additional $24/member/month.

For a 45,000 member plan, this equates to an annual revenue increase of

$12,960,000

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |16

Case Study Diabetes Management at a Medicaid ACO

Member Data Analytics Engine Actionable Insights

Challenges Solution Value

Analysis of the program

demonstrated the following outcomes

for participants:

• 5% Total Cost Savings

• Patient compliance to LDL &

HbA1c both increased by 9%

• Visit to PCP: Increased by 29%

• Acute Utilization: Decreased

by 25% & 20% in IP & ER

respectively

• Eligible population stratified as

medium or high risk

• Patient cohorts receiving

multiple incentives identified.

• Tracked ‘non-compliance’

members for 6 months for gap

closure

• Dashboards to measure quality

outcomes & utilization

• Understand the impact of

Diabetes Disease Management

Program:

1. Incur less cost

2. Utilize less ER and Inpatient

services

3. Utilize more physician office

services

4. Have higher compliance

EHR Data

Rx & DxClaims

Patient Medical History

Cost Data

Identify

Engage

Measure

©2016 SCIO Health Analytics®. Confidential and Proprietary. All rights reserved. |17

In Summary

Leveraging Predictive Analytics as a Catalyst to Drive Behavior and Engagement

Multi-source persona data including behavioral, clinical factors, demographic, attitudes

Impact and receptiveness to treatment

Focus on the population and individual level

Optimizing incentives

Measure the outcome and ongoing sustainability

Improved quality and effectiveness

Better care for lower cost