Towards a Continuous Learning Ecosystem: Data Innovations and Collaborations to Improve Clinical Outcomes and

Reduce Cost of Care

Vipul Kashyap*, Ph.D. Cigna Healthcare

vipul.kashyap@cigna.com April 9, 2013

Medical Informatics World, 2013, Boston, MA

* Acknowledgments – Jeff Catteau, Knowledgent (jeff.catteau@knowledgent.com) Discussions and feedback on the Economic Decision Model

Outline

• The Healthcare Ecosystem:

– Current State Continuous Learning Ecosystem

• Interactions at the Point of Care

• “Coordinated Healthcare Intervention”

• Making it Happen: Incentivizing Research & Knowledge Sharing

– Economic Decision Model

• Conclusions and Next Steps

The Healthcare Ecosystem: Current State

Hospitals

Providers

Individuals

Provider Perspective

Pharmaceutical

Companies

Clinical Research

Organizations (CROs)

Individuals

(Populations?)

Pharma Perspective

Siloized Ecosystem – Lack of collaboration/coordination – Inefficient!

Individuals

(Populations?)

Employers Health

Advocacy

Employer/Payor Perspective

Health Plans/

Payors

The Continuous Learning Ecosystem

Patient/ Individual

Lifelong Continuous Engagement

Ability to share and exchange clinical information and knowledge is a critical enabler

Continuous Learning

Interactions at the Point of Care

Medical History? Medications? Allergies? Contraindications? Referrals? Follow Up? Reimbursement

Follow Instructions: Lab Tests, Medications, others

Continuous Learning Ecosystem: Roadblock Does the physician have time and incentive to identify insights, research new ideas and share them with other stakeholders?

Demographics Diagnosis Medication Procedure Allergy Contraindication Order Referral Result Reimb

Patient 1

Patient 2

Patient 3

Patient 4

Patient 5

Patient 6

Patient 7

Patient 8

Patient 9

Patient 10

Patient 11

Patient 12

Patient 13

Patient 14

Data, Analytics & Knowledge

Observation/Hypothesis: Physicians implicitly leverage experience to create patterns or “Care Archetypes ”

Opportunity: Leverage “Care Archetypes” as an organizational framework for improving cost/outcomes Sharing and communication of “Care Archetypes” across the Ecosystem

Interactions at the Point of Care: Populations

Cost Effective? Affordable? Good Outcomes? Alternate Treatments? Risk? Engagement/Outreach? Behavioral Archetype? Compliance? Incentivized?

Continuous Learning Ecosystem: Roadblock partially addressed Transition to Pay for Performance has begun! However: Improving Performance requires research – and sharing of insights and results Pay for Insight/Pay for Research is not yet on the agenda!

Medication/Therapy compliance Discuss Alternatives, Activation and Engagement

Interactions at the Point of Care: Research

New Interventions (Therapeutic, Incentives, Engagement)? Impact of new ongoing research? Patient Participation? Contribution of Insights: New Cost Effective approach, Drug Side Effects, New Behavioral Incentives Granular Care Delivery contexts where Interventions are effective? Genomic correlates of behavioral/activation characteristics?

Comparative Effectiveness Research Clinical Trials Clinical Studies

Aligning Research and Practice

Ideation

Opportunity Analysis

Clinical Trials/ Studies

Outcomes Measurement

Populations/ Segments

Hypothesis Generation

History & Physical

Care Archetypes

Possible Diagnoses

Differential Diagnosis/ Likelihood Estimation

Therapeutic Intervention

Lab Tests/ Longitudinal Follow Up

Do physicians have the orientation to do research in the context of clinical practice?

Physicians continue as usual – Automated Infrastructure pulls data and performs “Research Analytics”

Data, Analytics and Knowledge Population Clinical Care

Dimensions* Genome Toxicity/

Efficacy Cost Risk

Quality

Behavioral/Activation

Therapeutic Intervention

Engagement/ Outreach

Payment Incentives

Patient Incentives

Popltn 1

Patient 1

Patient 2

Patient 3

Patient 4

Patient 5

Popltn 2

Patient 6

Patient 7

Patient 8

Patient 9

Patient 10

Popltn 3

Patient 11

Patient 12

Patient 13

Patient 14

Interventions that need to be studied/evaluated

Pop

ulatio

ns

Activatio

n Segm

ents

Alignment Patient Care Archetypes Cost/Quality Populations Behavioral/Activation Segments Patient Stratification

Patient Stratificatio

n

Care A

rchetyp

es

Coordinated Healthcare Intervention

Therapeutic Intervention - Drugs, Procedures - Devices

Informational - Outreach - Wellness Apps

Motivational - Engagement - Personal Incentive

Cost and Risk Sharing - Provider Incentive

Coordinated Health Intervention

• Optimal Health Interventions will require collaborations between stakeholders across the ecosystem

• Optimizes a set of criteria: outcomes, cost, toxicity, efficacy, utilization, economic benefit

Scenario: Emergence of the Third Party RO/A

• The Healthcare Ecosystem as a “Continuous Learning System” – Driven by Research • Reduce barriers to participation: Provide common infrastructure and a set of incentives • Need to incentivize sharing data generated from on going observations and measurement • Need for a trusted “Third Party Research Organization/Aggregator”

Behavioral Data/Insights

Research/ Validation

Medication Adherence Insights

Incentives, Infrastructure Investment

Medication Compliance Insights

Licensing Model

Licensing Model

Engagement Protocols

Prescribing/ Administration Protocols

Towards an Economic Decision Model U = R – (Ic + Ir) Ir consists of one time and incremental investments: ui = ri + d – Ii

• Utilization Model for Research drives investment in research to address the information gap • Providers will rationally consume results to drive cost/outcomes improvement in clinical care • Payors and Pharma will drive investments in infrastructure and incentives to drive research • Optimization curve for market based on marginal gains for various components

U = R – (Ip + Ir) Ir consists of investments in utilization and incentives

Ui = U + (ri) – (Ic + Ir) Ir consists of investments in compliance and incentives

Um/2 = (Uprovider + Upayor + Upharma)

U = utility, R = Revenue I: Investment in Care (Ic), Research (Clinical, Utilization, Incentives) (Ir), Payment (Ip)

Conclusions • The Health Ecosystem needs to evolve into a Continuous Learning Ecosystem to achieve

cost/outcome objectives

• Need for Deep Collaborations & Information Sharing to close the Information Gap Care Archetypes Populations Activation/Behavioral Segmentations Toxicity/Efficacy Stratifications

• Need to invest in infrastructure and incentivize research and knowledge sharing – Reducing Information Gap creates value (in terms of cost/outcomes) which is consumed by

different stakeholders in the ecosystem

– Information Gap always exists (due to new knowledge and information) new economic incentives market evolution/continuous learning

• Trust – a key roadblock – may lead to the emergence of third party research organizations/aggregators

• Next steps: Develop Economic Decision Model and Work on a Roadmap of Incentives to achieve a Continuous Learning Ecosystem!

Blog

http://continuouslearningecosystem.blogspot.com

Look forward to comments, feedback, critique, suggestions!