The New York State Behavioral
Health Readmissions Quality
Collaborative
Molly Finnerty, MD
Edith Kealey, PhD
Kate M. Sherman, LCSW
New York State Office of Mental Health
June 26, 2014
Participants:1-866-639-0744, no code needed
Minnesota RARE Campaign
Monthly Call, June 26, 2014
The New York State
Behavioral Health
Readmissions Quality
Collaborative
Molly Finnerty, MD
Edith Kealey, PhD
Kate M. Sherman, LCSW
New York State Office of Mental Health
Outline

Overview of the Project
Participants and activities
 Project metrics and data


Lessons Learned and Recommendations

Interventions
Emergency Department
 Inpatient
 Aftercare



Managing the Project
Future Plans
Collaborative
Participants, Activities
and Time Line
Readmissions Collaborative
Project Context and Focus

Statewide behavioral health systems transformation

Previous successful learning collaborative to reduce
use of antipsychotics with higher risk for metabolic
disturbance for individuals with existing metabolic
conditions

Focus on behavioral health readmissions
(individuals discharged from behavioral health
inpatient services who are readmitted to behavioral
health inpatient services within 30 days of
discharge)
Readmissions Collaborative
Sponsors and Participants

Sponsors


NYS Office of Mental Health
The 2 major hospital associations in NYS



Steering Committee



Greater NY Hospital Association (NYC)
Healthcare Association of NYS
Sponsors plus 8 hospitals / systems
Specified project focus and requirements
Participants: 45 hospitals statewide



Invited all hospital association members with inpatient
behavioral health services
Participation not required, no direct financial incentives
24% of eligible hospitals participated (some attrition)
Project Options: Participation

Select services to participate



Select settings to participate




Psychiatry and/or
Substance abuse services
Inpatient
Outpatient
Emergency departments
Multiple services encouraged to participate

Inpatient strongly encouraged but not required
Project Options: Strategies

Project focus: menu of options in 3 domains

Medication strategies




Outpatient engagement




Increase use of Long-Acting Injectables / Clozapine
Medication fill at discharge
Counseling for medication adherence
Referrals to ACT / case management / health homes
Counseling for adherence to treatment
Peer services
Integrated dual diagnosis treatment

Enhanced discharged planning required in Inpatient
and Emergency Services (defined by hospitals)

Target population defined by hospitals according to the
intervention selected
Collaborative Activities

Conferences




Kick-Off
Mid-point, share successful strategies
Concluding
Monthly Learning Collaborative Calls

Interactive, report on progress

Strategies Calls: Training on specific strategies

Site Visits (selected hospitals)


Technical assistance
Identify best practices
Resources and
Technical Assistance

PSYCKES Application



NYS Medicaid claims / encounter data
Behavioral health population (4.6 million)
Track performance and identify clients with quality
concerns

Project Website

Clinical tools (e.g., Readmission Risk Assessment)



Developed for the collaborative
Shared by participants
Identified from outside sources
Time Line

6/2012 - 12/2012: Kick-off and Planning




Begin monthly calls
Project Planning form due 10/2012
Note: Superstorm Sandy 10/2012
1/2013 - 6/2013: Begin delivering and tracking
interventions (monthly reporting), Midpoint Conference

Decision to extend Collaborative through 6/2014

7/2013 - 6/2014: Site Visits (n=15) and Calls (n=3)

11/2013: Midpoint Survey

6/2014: End / Concluding Conference
Project Data and
Measurement
Data Sources

NYS Medicaid Claims/Encounter Data



Hospital Self-Report



PSYCKES application
Data Analysis Team
Reported monthly by each hospital
Aggregated and distributed to hospitals monthly
Surveys


Prescriber Survey on LAI and Clozapine
Midpoint Survey on project interventions (value,
feasibility) and lessons learned
Key Project Metrics

Inpatient (primary indicator)



Outpatient



Among clients discharged from your hospital’s inpatient
service (psychiatry or substance abuse)
Percentage readmitted to the same service at any hospital
within 30 days
Among clients seen in your outpatient service who had a
behavioral health hospitalization at any hospital
Percentage readmitted to behavioral health inpatient at any
hospital within 30 days
Emergency


Among clients who come to ED within 30 days of discharge
from psychiatric inpatient at any hospital
Percentage readmitted by your ED
Readmissions within 30 Days of
Discharge from Inpatient Psychiatry
Length of Stay 4+ Days
25%
20%
Average Annual
Percent Change
6/2012 to 9/2013
15%
Participating:
-0.5 (ns.)
Start of Project
NonParticipating:
-3.8 (sig.)
10%
5%
Participating Observed Readmission Rate
NonParticipating Observed Readmission Rate
Sep-13
Aug-13
Jul-13
Jun-13
May-13
Apr-13
Mar-13
Feb-13
Jan-13
Dec-12
Oct-12
Nov-12
Sep-12
Aug-12
Jul-12
Jun-12
May-12
Apr-12
Mar-12
Feb-12
Jan-12
Dec-11
Nov-11
Oct-11
Sep-11
Aug-11
Jul-11
Jun-11
May-11
Apr-11
0%
Mar-11
Includes age 18+
Excludes SUD
Participating Modeled Readmission Rate
NonParticipating Modeled Readmission Rate
Readmissions within 30 Days of Discharge
from Inpatient Psychiatry
Length of Stay 4+ Days
Average Annual Percent Change, 6/2012 – 9/2013
by Hospital
60
50
40
30
20
10
0
-10
-20
-30
-40
-50
Green = Significant improvement
Yellow = Strong trend toward improvement
Red = Significant increase
Psychiatric ED visits by Individuals with a
Psychiatric Inpatient Stay in the Prior 30
days, and Disposition:
Aggregate data for all Participating Hospitals
Baseline
(June
2012)
ED visits
Baseline rate of
Readmissions
in ED
Most
recent
(Sep
2013)
ED visits
Most recent
rate of
Readmissions
in ED
(N)
(n)
%
(N)
(n)
%
ED visits with Psychiatric
Inpatient stay at any
hospital in prior 30 days
824
444
54%
914
454
50%
ED visits with Psychiatric
Inpatient stay at the same
hospital in prior 30 days
413
223
54%
473
220
47%
Average
Annual
Percent
Change
AAPC
95% CI
Statistically
Significant
Trend?
(P-Value
<0.05)
Low
High
-4.2
-8.3
0.1
No
-7.6
-12.9
-1.9
Yes
30-Day BH Readmissions (Any Hospital)
among Mental Health Outpatients
PSYCKES Indicator: 12-month look-back
Average Annual
Percent Change
7/1/2013 to
4/1/2014
Participating:
-10.4
Project Start
NonParticipating:
-7.5
Both statistically
significant
Includes individuals
of all ages
Measurement Challenges:
Defining Readmissions

What is a hospitalization?



Service types


Separate psychiatry and substance abuse
What is a readmission?




Any length of stay?
Exclude short term observation?
Same service type vs.
Any behavioral health vs.
Any service type including medical
Time frame


15 / 30 / 45 day
Readmission vs. high utilization over time
Measurement Challenges:
Other Issues

Data maturity: need to wait 6 months to see both index
admission and readmission appear in claims/encounter data

Observation periods: Monthly data vs. longer intervals

Confounding trends and variation



Seasonal fluctuations
Super-storm Sandy
Health Home and other systems transformation initiatives

Limited baseline data

Exploring alternative statistical methods

Exploring other related outcome and process measures
Recommendations and
Lessons Learned
Methods:
Review of Models and Initiatives

RQC: Behavioral Health Readmissions Quality Collaborative

Clinic CQI: OMH Continuous Quality Improvement Initiative
for Health Promotion and Care Coordination

CTI: Critical Time Interventions

Transitions: ACT Transitions Project

RED: Project RED (Re-Engineered Discharge)

STAAR: State Action on Avoidable Readmissions

AHRQ: Agency for Healthcare Research and Quality
(AHRQ) Reducing Medicaid Readmissions Project

RARE: Reducing Avoidable Readmissions Effectively
Note: all quotations are from RQC Midpoint Survey
Emergency Department
Prevent avoidable readmissions in ED

Identify high utilizers and potential readmissions

Consult/ approval by last inpatient team (they
come to ED to evaluate) before determining
disposition


Is the client’s status the same as last discharge?

Is another admission likely to be helpful?

Are there safe alternatives that could be tried?
Identify and contact community-based supports
before disposition/admission
Source(s): RQC
On Admission /
During Inpatient Stay
Assessment

Identify readmissions / high utilizers

Conduct in-depth review or case conference




What was the last discharge plan? How well did it work?
Why were they readmitted (root causes)?
What can we do differently this time?
Review in treatment team meeting, cross department
meetings (ER, inpatient, case workers, outpatient)
“Engaging the patient in reasons why the prior discharge
failed can help staff gain insight.”
Source(s): STAAR, AHRQ, RQC
After Hospital Care Plan

Develop and use After Hospital Care Plan (e.g.
Project RED format), including




Clear medication instructions
Follow-up appointments (arranged before discharge)
Name and phone number to call with any problems
Educate client and family using teach-back
method throughout inpatient stay
Source(s): Project RED (key intervention), STAAR, RARE
Access to Medication
Ensure access to medication post discharge!
Verify
insurance formulary for meds before initiating
Obtain
and verify pre-authorization for meds before
discharge
Fill
prescriptions at discharge: patients leave with meds
in hand (or are walked to the pharmacy by staff)
Check
Medicaid status - enroll in Medicaid if eligible
“Make sure that the patient can afford the medications they
are discharged on.”
Source(s): RARE, RQC
Involve Family / Natural Supports

Support evaluation

Assess family needs

Provide crisis intervention

Psychoeducation and skill-building
“Family involvement is key to a patient's recovery.”
“Family support makes a tremendous difference with patient
compliance.”
Source(s): RQC, CTI, STAAR, RED, RARE
Bridging and “Warm Hand-offs”

Face to face meeting with receiving outpatient
provider during inpatient stay or immediately
upon discharge. Ideally:

Discharge planning meeting: outpatient provider,
client, family, and inpatient team; and

Individual meeting/session: outpatient provider
and client
Source(s): STAAR, RARE, RQC, Transitions Project, CTI
Co-Occurring Mental Health and
Substance Use Disorders


Provide Integrated Dual Diagnosis Treatment, e.g.:

Screening at intake

4-quadrant model of assessment

Motivational interviewing
Refer to providers of integrated treatment for
aftercare
Source(s): RQC, EBP for co-occurring disorders
Post Discharge /
Outpatient
Aftercare

Follow-up appointment with aftercare mental health
provider within 3 days of discharge (5 at most)

Use higher-intensity outpatient services for hospital
diversion and hospital step-down

Partial Hospitalization Program (PHP)

Some clinics developing Intensive Outpatient (IOP)
level of care

Identification of and coordination with existing services
such as ACT
Source(s): RARE, RQC, Transitions
Follow-Up Phone Calls

Follow-up phone call to client/family






Within 72 hours
Clinical intervention, intensive (not just a reminder call)
Use teach-back method (don’t read the med list)
Ideally by staff known to client
Not “discharged” until attends first outpatient
appointment
Follow-up phone call to provider
“Follow-up phone calls are very important, to make sure that
discharged patients continue to take their meds and keep their
follow-up appointments.”
Source(s): Project RED (key component), RARE, RQC, Transitions
Follow-Up Phone Call to Client:
Project RED Key Components
1.
Assess clinical status
2.
Review and confirm each medication
3.
Review follow-up appointments
4.
Assess for barriers, problem-solve, and review
what to do if a problem arises
5.
After call: take any needed follow-up actions /
inform treatment team of any issues
Short-Term Case Management

Services may be provided by case manager,
bridger, peer, etc.

Key principles

Assess client risk/needs, adjust intensity and time
frame accordingly

Include home visits if needed

Actively follow up on non-adherence to the plan, e.g.:
make another appointment if missed
Source(s): CTI, RARE, RQC, Transitions
Community Functioning / Support


Build, practice and test self-management skills

Examples: filling pill boxes, keeping appointments

Skill-building at each level of care to prepare for next
Refer to intensive community supports, e.g.:

ACT

Health Home / other care management
“Very helpful to establish referral links to Health Homes for
care coordination services and ACT Teams.”
Source(s): RQC
Outpatient Crisis Management

Outpatient programs develop strategies for crisis
management, e.g.:





relapse prevention plans
monitoring for early warning signs
urgent care / walk-in appointments
on call availability
Educate clients (and staff) not to use the ED for
urgent care
Source(s): Clinic CQI
Managing the Project
Continuous Improvement
Across All Settings

No single solution



Portfolio of mutually reinforcing interventions
Ongoing incremental changes
All relevant services within the hospital should
participate and collaborate on the project
“There is definitely a need for increased collaboration between
the inpatient and outpatient staff. Though we are one agency,
and consider ourselves seamless, reviewing our internal
referral process has demonstrated a disconnect in identifying
and following up with patients deemed high-risk for
readmission.”
Source(s): RED, STAAR, RARE, RQC, Transition
Data-Driven Decision Making:
Project Level and Client Level

Start with a root cause analysis of a sample of
readmissions, including:
 client/caregiver interviews
 quantitative analysis
 input from hospital staff and other providers

Track interventions and outcomes over time
“Reducing behavioral health re-hospitalizations requires
developing a system for close monitoring and tracking of
patients identified as at-risk for re-hospitalization.”
Source(s): RED, STAAR, AHRQ, RQC
Collaboration across the
Continuum of Care

Know and engage your community partners
 Standardize communication
 Develop protocols for expedited referrals
 Collaboration on treatment and discharge planning
 Must include: BH, medical, housing

Develop a relationship with at least one pharmacy

Improved, real-time communication between inpatient
and outpatient behavioral health providers and
primary care physician
Source(s):STAAR, AHRQ, RQC, RED, RARE
Importance of Leadership

Buy-in / Motivation

Education

Resource Allocation
“Behavioral health re-admissions can be reduced when
providers use the proper, evidence-based treatments for
serious mental health problems….”
“When administration plans a project without staff buy-in or
support, it is doomed to be less successful than if staff had
themselves designed the interventions/strategies. Any
future collaborative project needs to incorporate more
representation from front line staff.”
Future Plans:
Readmissions Collaborative
Phase II
Expanded Focus

Discharge from behavioral health inpatient



Psychiatry
Substance abuse
Readmission to any inpatient service within 30
days



Psychiatry
Substance abuse
Medical
Project Structure and Strategies

All behavioral health services in the hospital
participate and work collaboratively

Focus on processes and care transitions

Timeline:

Summer 2014: Planning with Steering Committee

Fall 2014: Learning Collaborative Kick-Off
Question and Answer
Upcoming RARE Events….
Stay tuned for the next RARE Mental Health Webinar:
July 23, 2014 (12-1pm)
Care transitions for the homeless
Minnesota Department of Human Services
Future webinars…
To suggest future topics for this series,
MH - Reducing Avoidable
Readmissions Effectively “RARE”
Networking Webinars, contact:
Kathy Cummings, kcummings@icsi.org
Jill Kemper, jkemper@icsi.org