Report Back from Breakout Sessions Moderated by Penny Mohr, MA - - PowerPoint PPT Presentation

Moderated by Penny Mohr, MA

Senior Program Officer Improving Healthcare Systems

Report Back from Breakout Sessions

1

Breakout Group 1: ACO Structures and Risk Sharing Arrangements

2

Of the questions being considered, which patient-centered comparative effectiveness questions PCORI might fund?

(Consider: populations/subpopulations that should be target; interventions and comparators; relevant patient-centered outcomes.)

What are the characteristics of ACOs* that encourage packaged population health innovations^, and what are the impacts of these on patient engagement and patient-centered outcomes? What types of data information feedback is optimal for encouraging team behavior and patient engagement?

• *Characteristics of ACOs include ownership, external risk and payment

arrangements, patient and community engagement in governance, internal payment arrangements, and local and national market considerations, etc.

• ^Packaged, interdependent population health innovations: access

(responsiveness to patient needs), team care (strategic distribution of work), risk-stratified care management and care coordination, integrated/seamless behavioral health integration, internal incentives for patients and providers, HIT functionality and interoperability

3

Which CER question is most compelling and why?

Question 1 was agreed upon by the group

• Felt we could fold question two into the first
• We can use current systems to capture data and currently

available measures

• Sufficiently broad to capture important characteristics of

ACOs and patient outcomes

• Incorporates idea of interdependent, packaged care

modules/processes

4

Why or why not are these questions particularly well suited for PCORI to fund?

Big interest in this right now, most good can come in complex population and those at risk of developing multiple chronic conditions Level of knowledge is advanced to the point where we know enough about the complex innovation package (not starting at zero) Addresses the Triple Aim Will allow us to see what will do the most good in the next 4-5 years

5

What are the challenges raised in conducting research on these questions, and how might those challenges be addressed?

Methodological rigor may be difficult, randomization in particular is tricky if not impossible here

• Designs that take advantage of large systems may be

advantageous (cluster random, stepped wedge designs within a large system/ACO)

• Proving causality in study design may not be an absolute

requirement (numerous less rigorous studies may provide sufficient evidence for systems)

Data availability Cooperation of organizations Involving patients in meaningful ways (in governance and in practices) Involving providers in meaningful ways

6

Breakout Group 2: Patient and Provider Activation

7

Of the questions being considered, which patient-centered comparative effectiveness questions PCORI might fund?

(Consider: populations/subpopulations that should be target; interventions and comparators; relevant patient-centered outcomes.)

Are different approaches (models, intensity, relationship, incentives, proximity) of patient and/or provider engagement better at improving patient-centered outcomes for different subpopulations than others? Are different approaches of incorporation of patient input into program strategy and program design and operations better at improving patient and/or providers activation and patient- centered outcomes than others.

8

Populations/subpopulations and Patient- centered Outcomes

Populations

• High risk (high cost?) / multiple comorbidities
• Medically complex
• Care management
• Low SES
• Mental/behavioral health
• Generalizability to commercial/Medicaid
• Procedures vs. chronic illness

Patient-centered outcomes:

• Process (treatment adherence) and outcomes on chronic disease management
• Quality of life
• Level of activation or confidence
• Clinical outcomes relevant to condition
• Decision quality
• Avoidable admissions
• Readmissions

9

Which CER question is most compelling and why?

Are different approaches (models, intensity, relationship, incentives, proximity) of patient and/or provider engagement better at improving patient- centered outcomes for different subpopulations than

• thers?

10

Why or why not are these questions particularly well suited for PCORI to fund?

Very focused on patient-centeredness of care and

• utcomes

Rapid adoption of ACOs A lot of experimentation without evidence base to compare approaches

11

What are the challenges raised in conducting research on these questions, and how might those challenges be addressed?

This is organizational research and each organization is an N of 1

• Contextual issues (leadership, culture, IT)

Defining and how to measure:

• Intensity (high touch, low touch; technology vs. human

interaction)

• Models
• Proximity (to patient/clinician interaction)

12

Breakout Group 3: Delivery Services

13

Of the questions being considered, which patient-centered comparative effectiveness questions PCORI might fund?

(Consider: populations/subpopulations that should be target; interventions and comparators; relevant patient-centered outcomes.)

Do ACOs which have access to and use information that is timely and actionable to inform care coordination strategies have better results on improving patient- centered outcomes? Do ACOS that incorporate community and social services into their care coordination perform better on patient centered outcomes than ACOs that do not? Do patients in an ACO that report having a team taking care of them have better patient centered outcomes. Do these results correlate with structural measures of team-based care? Do ACOs that have a collaborative payer/provider arrangement do better on patient centered outcomes vs solely provider driven arrangements on patient centered outcomes? Patient Centered Outcomes: improved patient satisfaction, pt activation, pt engagement, clinical quality outcomes, absenteeism (from work/school/other), avoidable hospitalization/ER visits, medication adherence Targeted Populations: General: at risk, for which interventions are likely to prove beneficial ‘influenceable’

14

Which CER question is most compelling and why?

Do ACOs which have access to and use information that is timely and actionable to inform care coordination strategies have better results on improving patient-centered

• utcomes?

15

Why or why not are these questions particularly well suited for PCORI to fund? Research gap on patient

• utcomes/experience (instead of cost focus)

Suggested scalable strategies Information as to how to deliver interventions

16

What are the challenges raised in conducting research on these questions, and how might those challenges be addressed?

Willingness of ACOs to participate Selection bias of data among research participants Research challenges; definitional challenges How to isolate impact of specific interventions

17

Breakout Group 4: Medicaid

18

Of the questions being considered, which patient-centered comparative effectiveness questions PCORI might fund?

(Consider: populations/subpopulations that should be target; interventions and comparators; relevant patient-centered outcomes.)

• 1. Do certain Medicaid populations (complex medical needs, SPMI, kids,

etc.) benefit more than others from ACO models, in terms of outcomes such as: achieving patient-defined goals, health outcomes, TCOC, etc.?

• 2. Do (Medicaid) ACOs that incorporate behavioral health (incl. substance

abuse) into the program perform better than those who do not on

• utcomes such as: achieving patient-defined goals, care coordination,

increased primary care utilization, population health indicators, social service utilization, etc..

• 3. Do Medicaid-only ACOs vs. Medicaid + other payer ACOs perform

better on population health outcomes, TCOC, patient satisfaction and

• ther patient-centered outcomes?

19

Which CER question is most compelling and why?

#2 because it is relevant to vulnerable populations in any coverage type, has more research clarity, and is more actionable from the policy maker perspective.

20

Why or why not are these questions particularly well suited for PCORI to fund?

Populations with behavioral health care needs

• ften not well represented in health care services
• research. This question looks at the impact of a

system-level intervention on a particular population

• f interest.

21

What are the challenges raised in conducting research on these questions, and how might those challenges be addressed?

Still a little early in terms of number of entities incorporating behavioral health but there is growing interest, integration of BH is a priority issue, and there is enough action to support study; study can inform further action Can be difficult to measure outcome of achieving patient-defined goals

22

BREAK

3:30 – 3:45 p.m.

23

Moderated by Penny Mohr, MA

Senior Program Officer Improving Healthcare Systems

Priority Research Questions for PCORI and Justification

24

Priority Questions

25

Criteria to Keep in Mind

Patient-Centeredness: is the comparison relevant to patients, their caregivers, clinicians or other key stakeholders and are the outcomes relevant to patients? Impact of the Condition on the Health of Individuals and Populations: Is the condition or disease associated with a significant burden in the US population, in terms of disease prevalence, costs to society, loss of productivity or individual suffering? Assessment of Current Options: Does the topic reflect an important evidence gap related to current options that is not being addressed by

• ngoing research.

Likelihood of Implementation in Practice: Would new information generated by research be likely to have an impact in practice? (e.g. do

• ne or more major stakeholder groups endorse the question?)

Durability of Information: Would new information on this topic remain current for several years, or would it be rendered obsolete quickly by new technologies or subsequent studies?

26

Voting Sheet

1 (low) 2 3 (modest) 4 5 (high) Patient- Centeredness Impact on Health and Populations Assessment of Current Options Likelihood of Implementation Durability of Information Overall Importance

Are different models of patient engagement in ACOs better at improving patient-centered outcomes than others? Are different models of patient engagement more effective for different subpopulations (e.g., children versus adults, socioeconomic status)? Score this topic from (1 – low / does not or barely meets the criterion) to (5 – High scoring / fully addresses the criteria). Please reference the scoring criteria guide as reference. The total score will measure how highly this topic is prioritized.

Given your consideration of all CER questions, which one would you recommend that PCORI should pursue and why?

27

Steven Clauser, PhD, MPA

Program Director, Improving Healthcare Systems

Closing Remarks

Bryan Luce, PhD, MS, MBA

Chief Science Officer, Office of the Chief Science Officer

28

We Still Want to Hear from You

We welcome your input on today’s discussions. We are accepting comments and questions for consideration on this topic through January 23rd, 2015 via email (info@pcori.org) We will take all feedback into consideration.

29

Thank You for Your Participation

30