Collaborative Care in Primary Care and Behavioral Health: Are We at the

Tipping Point?Henry Chung, M.D.

Vice President and Chief Medical OfficerCare Management Company of Montefiore Medical Center

AndClinical Associate Professor of Psychiatry New York University School of Medicine

Agenda

Why does Integration Matter for the Health of Our Patients?

Is Integrated Care the Same as Collaborative Care?

What are the key ingredients of Collaborative Care?

Quality Outcomes and Measurement

Health Homes and ACO’s – the Promised Land?

Health and Behavioral Health Challenges Basic medical and behavioral health services must be available in all clinical

settings –most people with behavioral health disorders are seen in medical settings NOT specialty settings

Specialty behavioral health settings must have excellent access to medical services - most people with severe and persistent mental illness or severe addiction have excess morbidity and mortality due to premature CV disease

All health settings must have care coordination capability in the appropriate continuum depending on case mix and severity – episodic point of care treatment is INEFFECTIVE resulting in major gaps in treatment; in high cost complex safety net patients, medical and behavioral health readmissions are a HUGE excess cost to the system

Care management for patients with multiple disorders as well as specialty care management must be used in a stepped care manner – locus of patient care and costs should be accounted for in the segmentation process

Primary Care and Behavioral Health Integration Models (Mechanic D)

Enhanced screening, treatment and referral - Trained primary care providers screen, identify, treat and /or refer to mental health specialists (usually off site) for treatment

Co-location of services –behavioral health clinicians provide consultation and/or short term treatment

Systematic integration with shared protocols, health information, and quality metrics and outcomes

Key Principles of Depression Integration in Primary Care

Training of PCP to Screen and Diagnose

Starting Treatment (almost always medication)

Patient Education

Follow Up and Stepped Care

Easy Access to Specialty Care

BUT?

What are the Limitations?

Organizational Commitment

Cultural Tailoring necessary

Payment Alignment Problems

Sustainability Problems

“Quest for Cost Offset!”

Working Through the Cultural Values of Primary Care and Behavioral Health

Landmark Randomized Controlled Trials in Depression

IMPACT – Older Adults and Depression

RESPECT-D – Primary Care Patients and Depression

? PRISMe – Older adults with depression in primary care or at-risk drinking randomized to collaborative care vs enhanced specialty care

RCT for geriatric patients presenting in primary care settings (VA, FQHC, academic) with depression randomized to Integrated care or enhanced specialty care with more realistic resource allocations (ie limited!)

Rationale : assess real world outcomes with greater generalizability for integrated care compared to improved specialty mental health; ie no algorithms, no treat to target interventions

Main hypotheses: Improved Engagement in Integrated Settings; Equivalent 6 month outcomes;

Why PRISM-E?

Appointment available within 2-4 weeks of PCP dx of depression/ at risk alcohol

Licensed MH/SA professionals with some indication of geriatric expertise

Coordinated followup contacts with PCP and patient if they missed first visit

Concrete services incl transport if necessary

Availability of urgent and emergency appointments

PRISM-E Enhanced Specialty Mental Health Model Requirements

Engagement and treatment of depression in PRISM-E

0

10

20

30

40

50

60

70

80

Integrated Care Specialty

Engaged Response

N=599 N=621

71%

29%

49%

37%

Very high refusal rates in PC settings, except for VA PC settings

Higher engagement in PC than in specialty settings

No difference in outcomes at 3 or 6 months

Patients with both depression and SA did slightly better in specialty settings (but very small sample)

PRISM-E Outcomes for Hazardous Drinking

“TRYING HARDER WILL NOT WORK. CHANGING SYSTEMS

OF CARE WILL.”

Don BerwickInstitute for Healthcare Improvement

What is Collaborative Care?

Collaborative (Chronic) Care Model

Effective chronic illness care requires a re-organization of the health care system to manage chronic or recurring illnesses more effectively

Application of evidence at point of care

Patient activation to promote effective self-management and patient-provider collaboration

Systematic outreach to ensure appropriate monitoring and follow-up care

Care Management/Coordination

Holistic and coordinated approach for conditions often put into separate silos in the delivery system

Offers flexibility in patient-centric treatment modalities and settings sensitive to cultural and economic barriers to treatment

Supports adherence through self- management and active care coordination for multiple chronic illnesses

Self ManagementGoal directed patient behaviors that enhance clinical &

functional outcomes:Medication management and adherencePsychotherapy previewing and adherenceSelf-monitoring of symptoms, treatment status Managing effects of illness on social role functionReducing health risks (alcohol misuse, smoking)

Preventive maintenance (e.g., exercise, screening

check-ups)Working with health care professionals

Organizing Framework for Team Based Care

Peer Based Learning with clinical application (rapid piloting)

Patient Centered Care Coordination as a Foundational Concept = Population Based Health

QI principles tied to Specific Measures = Accountable Care

Ability to Incorporate Latest Evidence Base into The Model

Patient Self Management = Empowerment

Sustainability is paramount

Why is the model effective?

SBIRT for Alcohol

Copyright © 2011 The National Center on Addiction and Substance Abuse at Columbia University

SBIRT Components

Screening Brief strategy to identify at-risk population

Brief Intervention

One or more discussions with clinician (10-15 min each):

1.Assessment & feedback on drinking

2.Advice, goal setting, agree on plan

3. Follow-up contact

Referral to Specialty Treatment

Patients with more severe problems require more than a brief intervention

SBI for Alcohol in Primary Care

Effectiveness and Cost-Effectiveness

Most effective intervention for alcohol problems based on clinical trials research

Solberg et al. (2008):SBI for alcohol ranked among top of 25 USPSTF-

recommended screening practices based on effectiveness and cost-effectiveness

Similar in ranking to screening for hypertension or colorectal cancer

Copyright © 2011 The National Center on Addiction and Substance Abuse at Columbia University

Life-Years Saved, Costs, And Savings From 20 Evidence-Based Clinical Preventive Services (2006 Dollars)

Clinical preventive service

Life-years saved per 10,000 people per year of intervention

Medical cost of service per person per year

Medical savings of service per person per year

Annual net medical costs per person per year

Childhood immunizations 1,233.1 $306 $573 −$267

Influenza immunization 23.8 28 20 8

Pneumococcal immunization 6.4 46 113 −67

Discuss daily aspirin use 63.0 21 87 −66

Smoking cessation adviceand assistance

97.5 10 50 −40

Alcohol screening andbrief counseling

7.0 9 20 −11

Breast cancer screening 45.0 64 3 61

Cholesterol screening 27.8 128 24 104

Colorectal cancer screening 40.8 46 31 15

Depression screening 0.0 42 0 42

Hypertension screening 10.7 79 50 29

• Source Authors’ analyses; sources for data used in each model are available from the authors.

• Health Affairs September 2010 29:9

Other Promising Collaborative Care Models in the Arsenal

GAD and Panic Disorder

Bipolar Disorder

PTSD

Opioid Abuse and Dependence using Buprenorphine

Others?

Behavioral Health and Measurement: A Quality Imperative

Why Measurement?Improve individual outcomes by assisting in

treatment planningGroup level outcomes can serve as benchmarks

and goals that can be used as critical information to confirm or address effectiveness of service model changes

Creates a common language across disciplines and providers to promote effective collaboration

Translating PHQ-9 Depression Scores into Initial Planning

Score Description Actions

1-4 Community Norms No further action

5-9 Mild Symptoms Watchful waiting, periodic re-screening, education, patient activation and evaluation

10 – 14 Moderate Symptoms Develop treatment plan, consider counseling, education, assertive follow-up and evaluation, pharmacotherapy

15 – 19 Moderate -Severe Immediate institution of treatment including medication and/or counseling

≥ 20 Severe Pharmacotherapy, counseling & referral to mental health specialist

Using the PHQ-9 to Monitor & Adjust Treatment at 4-6 Weeks

PHQ-9 Treatment Response Treatment Plan

Drop of 5 points from baseline Adequate No treatment change neededFollow-up in four weeks

Drop of 2-4 points from baseline

Possibly Inadequate May warrant an increase in antidepressant dose

Drop of 1 point, no change or increase

Inadequate Increase dose; Augmentation; Informal or formal psychiatric consultation; Add psychotherapy

APA NDLMI Psychiatry (Psychiatrist and Non-Physician Co-Leader)

Depression Measurement Informed Care Project

PHQ9 was helpful in Tx decisions 93% (n=6,096 Patient Contacts)

For contacts where PHQ9 was helpful, how did PHQ9 influence Tx? (n=5,578 Patient Contacts)

- Change Tx 40%

- Confirm Tx 60%

Impact of Measurement on Psychotherapy Treatment Outcomes*

* Harmon, S. Cory, Lambert, Michael J., et al. (2007). Enhancing outcome for potential treatment failures: Therapist-client feedback and clinical support tools. Psychotherapy Research, 17(4), 388

Health Homes and Accountable Care Organizations

Superb Access to Care

• Patients can easily make appointments and select the day and time.

• Waiting times are short.• eMail and telephone

consultations are offered.• Off-hour service is available.

• Patients have the option of being informed and engaged partners in their care.

• Practices provide information on treatment plans, preventative and follow-up care reminders, access to medical records, assistance with self-care, and counseling.

• These systems support high-quality care, practice-based learning, and quality improvement.

• Practices maintain patient registries; monitor adherence to treatment; have easy access to lab and test results; and receive reminders, decision support, and information on recommended treatments.

Care Coordination

• Specialist care is coordinated, and systems are in place to prevent errors that occur when multiple physicians are involved.

• Follow-up and support is provided.

Team Care •Integrated and coordinated team care depends on a free flow of communication among physicians, nurses, case managers and other health professionals (including BH specialists). •Duplication of tests and procedures is avoided.

Patient Feedback

•Patients routinely provide feedback to doctors; practices take advantage of low-cost, internet-based patient surveys to learn from patients and inform treatment plans.

• Publically available informatio

n

•Patients have accurate, standardized information on physicians to help them choose a practice that will meet their needs.

• 8• Source: Health2 Resources 9.30.08

Defining the Medical Home

Health Homes OverviewBeneficiary criteria - At least two chronic conditions, one chronic

condition and at risk for another, or one serious and persistent mental health condition. Chronic conditions include mental health condition, substance abuse disorder, asthma, diabetes, heart disease, being overweight (BMI over 25).

Designated Providers -Physicians, clinical group practices, rural health clinics, community health centers, community mental health centers, home health agencies; interdisciplinary health teams.

Payment - flexibility in designing the payment methodology including structuring a tiered payment methodology that adjusts for severity of illness and the “capabilities” of the designated provider.

• 31

All Medicaid by Clinical Risk Groups (CY 2008)

Entire Medicaid by CRG Recips Total Member

Months

Pct Total Member Months Avg MM

Pct of Total Member Months

Sum Total Claim Expenditures

Pct of Total Claim Expenditures

Total Claim PMPM    

1) Healthy

2,891,590

26,366,511 53.65% 9.12 53.65% $ 6,256,508,376 14.30% $ 237    

2) Maternity/Delivery

227,175

2,060,955 4.19% 9.07 4.19% $ 1,501,611,849 3.43% $ 729    

3&4) Minor Conditions

333,807

3,656,225 7.44%

10.95 7.44% $ 1,801,754,388 4.12% $ 493    

5) Single Chronic

741,860

7,915,447 16.10%

10.67 16.10% $ 8,925,736,961 20.40% $ 1,128 Pairs & Triples

6) Pairs Chronic 628,772

7,132,396 14.51% 11.34 14.51% $ 16,778,912,692 38.34% $ 2,352 % of Memb

Mos % of $

7) Triples Chronic 76,427

870,377 1.77%

11.39 1.77% $ 3,457,392,177 7.90% $ 3,972 16.28% 46.24%

8) Malignancies 25,157

255,422 0.52%

10.15 0.52% $ 839,000,518 1.92% $ 3,285

9)Catastrophic 31,219

342,664 0.70%

10.98 0.70% $ 2,236,148,131 5.11% $ 6,526    

10) HIV / AIDS 49,589

549,384 1.12%

11.08 1.12% $ 1,964,903,822 4.49% $ 3,577    

Total

5,005,596

49,149,381 100% 9.82 100.00% $ 43,761,968,915 100.00% $ 890    

• 32

Goal: Improve quality of medical care for SPMI patients at one CMHC

Method: 12 month RCT of Patients in one CMHC randomized to nursing care management (MI, coaching, navigation, followup with appts) versus encouragement and PCP list

Demo: 85% of patients had schizophrenia, depression and bipolar. 25% had co-occurring SA disorders. Most common CMI were: HTN, arthritis, dental, diabetes.

Primary Care Access Referral and Eval (PCARE) Study - Proof of Concept for Health Home

PCARE Improvement in Medical Care

Druss, Von Esenwein, Am J Psychiatry 167:2, February 2010

Health Care Reform: ACOs The Accountable Care Act provides for a shared savings program:

that promotes accountability for a patient population and encourages investment in infrastructure and redesigned care processes for high quality and efficient service delivery. Under such program:

– groups of providers meeting pre-determined criteria may work together to manage and coordinate care for Medicare fee-for-service beneficiaries through an accountable care organization (referred to in this section as an ‘ACO’); and

– ACOs that meet quality performance standards are eligible to receive payments for shared savings

PCMH as Foundation for Accountable Care Organizations

• 38• Source: Premier Healthcare Alliance

• ACOs are defined as a group of providers that has the legal structure to receive and distribute incentive payments to participating providers.

65 Quality Measures in 5 Domains in : Patient/Caregiver Experience (7)

Care Coordination (12)Patient Safety (2) Preventive Health (9)At Risk Pop/Frail Elderly in Chronic

Conditions (20)Specific Behavioral Health Measures: ONLY 1? HUH!?

ACO Quality Measures

So Are We at a Tipping Point? (Gladwell)

Point at which a movement/event/intention becomes inevitable, inescapable, when everything changes all at once

Transformation occurs through initial infection (idea) then contagiousness (others latch on, tell others), little actions having big effects (multiple pilot testing, things begin to work, new application seems to work), change is often sudden (then everybody gets it)

Examples – flu epidemic, internet usage, Google.

NCDP © New York University

Key Ingredients to Reach a Tipping Point

1. Law of the Few

a.Connectors – People who know everyone and create intersections and opportunities

b.Mavens – People who are credible information hounds and validate ideas and methods

c. Salesmen – People who are able to use information and people to spread idea to the skeptical

Case example: “The Midnight Ride of Paul Revere”

NCDP © New York University

Key Ingredients to Reach a Tipping Point

2. The Stickiness Factor

a.Memorable – evokes strong feelings

b.Minimal barriers to entry – can’t avoid seeing it or seeing others do it – “why not me?”

c. “Cool” factor – in the know, the in crowd, identification with status

d.Little things make a difference in making big changes, example – PDSA’s

NCDP © New York University

So Are We There Yet?