Transcript Title of Project - PatientCareLink

A Rehospitalization Reduction
Program on a Geriatric Skilled
Nursing Unit
Randi Berkowitz, MD
Hebrew SeniorLife
Why decrease readmissions?
I. Excellence in care
–
–
–
errors
patient satisfaction
staff satisfaction
II. Financial
–
–
–
–
referrals
subacute beds
long-term care
census
reimbursement/patient
Learning Objectives
• Define the scope of the issue of
rehospitalization in subacute care nationally
• Describe innovative programs to reduce
transfer out to the hospital
• Show how CQI process involving transfer
can lower hospital readmission rates whole
improving patient safety and quality of care
Large geographic variation
Hospital Readmissions within 30
days from SNFs are common
 Of ~1.8 million SNF admissions in the U.S.
in 2006, 23.5% were re-admitted to an
acute hospital within 30 days
 In Massachusetts the rate is 26%
 Cost of these readmissions = $4.3 billion
Common Reasons for Transfers
 Medical instability
 Availability of:
 On-site primary care providers
 Stat tests, IVs
 Inadequate assessments to identify early changes
 Communication gaps
 Family issues/preferences
 Lack of advance directives (DNR, DNH)
Do They Have to Go?
As many as 45% of admissions of nursing
home residents to acute hospitals may
be inappropriate
Saliba et al, J Amer Geriatr Soc
48:154-163, 2000
In 2004 in NY, Medicare spent close to
$200 million on hospitalization of longstay NH residents for “ambulatory care
sensitive diagnoses”
Grabowski et al, Health Affairs
26: 1753-1761, 2007
Adverse Events Common
Coming and Going
• 46% of hospitalized patients have 1 or more
regularly taken medications omitted without
explanation. Potential for harm estimated at
39%.
– Cornish Arch Int Med 2005; 165: 424-9
• Transfers from NH to hospital have an
average of 3 med changes. 20% lead to
adverse drug events.
– Boockvar Arch Int Med 2004 (164) 545-50
Conclusion
• Rehospitalizations are going to be a prime
focus coming years
• New system paradigm will be needed to
meet the demand for prevention of
readmissions
• Focus of enhancing care in the SNF and
community treatment will take precedence
It’s a new world Obamacare!
• Center for Medicare Medicaid Innovation
• $10 billion
• Triple aim
– better health
– better care
– lower cost
• Innovation Advisors Program - Current
fellow
How?
1. Admission
2. Stay on unit
3. Discharge
Bucket #1: Problems on Admission
• Ineffective
communication of
prognosis / options
• PCP out of loop
• Inadequate care plans for
recurrent symptoms
Reduce AVOIDABLE hospital transfers
Approach to the Problem:
Admission
•
•
•
MD standardized discussions
Communication family and PCP
High risk patients
–
–
Automatic Palliative Care consult
Flag for entire team
Bucket #2: Stay on Unit:
Problems With Team Operation
• Disciplines
operating in silos
• Failure to identify
problems early
• Failure to learn
from mistakes
Reduce AVOIDABLE hospital transfers
Approach to the Problem:
Stay on the Unit
• Team Improvement for the Patient and
Safety (TIPS) conference
• Call to hospital
• Root cause analysis
Bucket #3: Problems With
Home Discharge
• Poor hand off to next team
• No teach back with
patient/HCP
• No standardized discharge
summary/ nursing process
Reduce AVOIDABLE hospital transfers
Approach to the Problem:
Home Discharge
• Project RED
– Written home care plan from electronic medical
record
– Making specific for geriatric use
• E.g. advance directives, diet, VNA, assistive devices
• Standardized discharge summaries
Target Population
• All admissions to the RSU subacute unit
• 1000 admissions a year
• 3NP/3MD- geriatric and palliative care
certified
Process and Outcome Measures
• Admission
– 90% patients have discussion with MD
• Prognosis
• Rehospitalizations past 6 months
• Communication family and PCP
– Patient/ family satisfaction survey
Advisory Committee
• Family Involvement - Daughter
• Outside institutions - Director Subacute CarePartners
• Biostatistician
• Information Technologist
• Continuum - homecare
• Senior leadership at HSL
• Rabbi from palliative care
• Staff nurse, unit coordinator, therapy, social work,
aide, administration
Process and Outcome Measures
• Middle - Stay on the unit
– Unplanned discharge rates
– Benchmarked staff safety survey for staff AHRQ
– Attendance TIPS
Process and Outcome Measures
Discharge Home
• 30 day readmission rates after discharge
from SNF
• Satisfaction survey of discharge
preparedness
Perceived Facilitators/Barriers
•
•
•
•
•
•
Pt acceptance of less aggressive approaches
Increased liability
Increase cost keeping sicker patients
Difficulty obtaining information from hospital
Time needed to engage primary care
Lack of practitioner access to computer systems in
key referral sites
• Limited IT resources for Project RED
RSU Acute Transfer/Total Discharges
Data Unplanned Transfers
• January 2008- June 2009 compared with
post TIPS July 2009-November 2009
• Massachusetts 30 day 22-28%
Pre-intervention
16.5%
Post-intervention 13.3%
Rate Reduction
-20%
Pre and Post Discharge
Dispositions
•
•
•
•
•
Pre N=862
Community 68.6%
Died
1.2%
LTC
13.8%
Hospital 16.5%
•
•
•
•
•
Post N=8863
73%
2.2%
11.6%
13.3%
When staff report something that could harm a
resident, someone takes care of it
Agree and Strongly Agree
100
90
80
70
60
50
40
30
20
10
0
PRE TIPS
6 Months
One Year
On this unit, we talk about ways to keep incidents
from happening again
Agree and Strongly Agree
100
90
80
70
60
50
40
30
20
10
0
Pre TIPS
6 Months
One Year
Staff ideas and suggestions are valued on this unit
Agree and Strongly Agree
70
60
50
40
30
20
10
0
Pre TIPS
6 Months
One Year
It is easy for staff to speak up about problems on this unit
Agree and Strongly Agree
60
50
40
30
20
10
0
Pre TIPS
6 Months
One Year
Staff feel like they are part of a team
Agree and Strongly Agree
80
70
60
50
Series1
40
30
20
10
0
Pre TIPS
6 Months
One Year
Staff are blamed when a resident is harmed
Disagree and Strongly Disagree
60
50
40
Series1
30
20
10
0
Pre TIPS
6 Months
One Year
Implications for HSL
• Family/patient involvement
• Create culture of system management rather than
blame
• Share knowledge learned across sites/teams
• True multidisciplinary team- swarm the problem
and front line solutions which can be used
organizational wide
• Use of run/control charts to guide CQI into
frontlines and understand common cause
variability
Why take on this pain financially?
Census, census, census
• Hospital care!
• CMS demonstration project
• Preferred provider network
RED
• Computerized After Hospital Care Plan
– Code status, meds, VNA info, PCP info, speech
and therapy directions
– Is Meditech good for something?
– Phone number to call with questions with
picture care coordinator and name
– Give at first care plan meeting and on discharge
update
RED
• Change culture patient/family
empowerment
• Involvement of front line staff- NASA
comparison
• Culture of QA and monthly feedback
• Clear numerical goals for entire team
Project RED
Empowering the Patient
Setting goals of January, 2011
How are we doing?
How good EXACTLY do we want to
be?
Respondents Reached
• 305 patients
• 96%
• 30 days after discharge RSU
Rehospitalization Once Home
• 56/302 patients
• 18.5%
• GOAL- We will reduce this to 15% or 2.7%
How many see PCP in 30 days?
• 171/282
• 60.6%
• GOAL - We will increase this to 75%.
Understood Medications Very Well
or Extremely Well
• 216/279
• 77.4%
• GOAL- We will increase this to 80%
Understood Medications Very Well
or Extremely Well
• 216/279
• 77.4%
• GOAL- We will increase this to 80%
How would you rate HSL?
• Respondents 263
• Mean number scale 1-10 (10 the best)
• 8.55
• GOAL- We will increase this to 9
The life of a RED packet
• Given to patient approximately 1 week after
admission (“rough draft”)
• Nurses use RED as a tool to help educate
patients about their illness, meds, etc.
• “Final draft” is given to patient on discharge
RED Data – How are we doing?
• 90% of patients have been reached
• 13% of the intervention patients have been
readmitted to a hospital or had a visit to the
ER compared to 17.4% of the nonintervention patients
Data (cont.)
• 73.2% or RED patients saw their PCP
within 30 days compared with 45.8% nonRED
• 92.5% understood their medications
compared to 60.5%
Data (cont.)
• 56.5% of patients were told of side effects
for new medications, compared with 16.6%
RED Problems
• Original plan was to for social workers give
RED to patients
• Current plan is to have nurses deliver and
teach RED to each patient
What are you all doing?
• Identify high risk patients
• Assessment versus actual interventions- all
assessed and no where to go
• Sharing between systems- STAAR et al
• Communicating across continuum
– Talking, data, funding