Welcome to the OPQC NAS July Action Period Call
• Thank you for joining; our webinar will start shortly!
• In the mean time; please
sign in the chat box with the names of all webinar participants and hospital affiliation.
Neonatal Abstinence Syndrome Project
July Action Period Call
Ohio Perinatal Quality Collaborative July 2015
Welcome!
The line will be placed on Group Mute
To ask a question: Click on the Raised Hand icon
You can type your question
into the Chat Box You can use *6 to come off of GROUP
MUTE (and *6 to go back on).
Time Topic Presenter
3:00 pm Welcome & Agenda Review Susan Ford, RN
3:05 pm Data Overview – June Results
Rick McClead, MD
3:15 pm NAS Orchestrated Testing Lloyd Provost, MS Heather Kaplan, MD
3:45 pm Team Discussion – Q&A All teams
3:55 pm Next Steps •Save the Date Fall Learning Session •Data Submission Reminder •MPR/PDSA Reminder
Susan Ford
Agenda
Promedica Toledo Children’s
Miami Valley
Mercy Anderson
Aultman
Mt. Carmel East OSU
UH Rainbow Babies & Children’s
Bethesda North Hospital
Nationwide Dublin Methodist
Akron Children’s Summa
Cincinnati Children’s
Hillcrest Hospital Fairview Hospital
Cleveland Clinic
Dayton Children’s
Nationwide Riverside Methodist
Nationwide Grant
Nationwide Mt. Carmel St. Ann’s
UH Cincinnati
Good Samaritan Hospital
MetroHealth
Mt. Carmel West Nationwide Doctor’s
Akron Children’s
Nationwide Children’s
Mercy Children’s Hospital
Atrium Medical Center
Fort Hamilton
Mercy Hospital Fairfield
Mercy Medical Center Canton
The Christ Hospital
St. Rita’s Medical Center
Southview Medical Center
Good Samaritan Hospital Dayton
Kettering
Mercy Health West
Southern Ohio Medical Center
Genesis Healthcare System
OhioHealth MedCentral Mansfield
Marion General
Elyria Medical Center -UH
Mercy Regional Medical Center Lorain ProMedica
Bay Park
Lima Memorial Health System
Springfield Regional Medical Center
Adena Regional
Medical Center
Soin Medical Center
Upper Valley Medical Center
Licking Memorial Health System
NAS Participating Sites 2014
1/2014 start Level 3 and Level 2 teams
Akron Children’s
St. Elizabeth Health
Center/Mahoning Valley
Trumbull Memorial
4/2014 start Level 2 teams
Key Driver Diagram Project Name: OPQC Neonatal NAS Leader: Walsh
SMART AIM
KEY DRIVERS INTERVENTIONS
By increasing identification of and
compassionate withdrawal treatment for full-term infants born with
Neonatal Abstinence Syndrome (NAS), we will reduce length of stay by 20% across participating sites by June 30, 2015.
Improve recognition and non-judgmental support for Narcotic
addicted women and infants
Connect with outpatient support and treatment program prior to
discharge
Standardize NAS Treatment Protocol
Optimize Non-Pharmacologic Rx Bundle
Initiate Rx If NAS score > 8 twice. Stabilization/ Escalation Phase Wean when stable for 48 hrs by 10%
daily.
Swaddling, low stimulation. Encourage kangaroo care Feed on demand- MBM if appropriate
or lactose free, 22 cal formula
All MD and RN staff to view “Nurture the Mother- Nurture the Child”
Monthly education on addiction care
Attain high reliability in NAS scoring by nursing staff
Partner with Families to Establish Safety Plan for Infant
• Fulltime RN staff at Level 2 and 3 to complete D’Apolito NAS scoring training video and achieve 90% reliability.
Establish agreement with outpatient program and/or Mental Health
Utilize Early Intervention Services
Collaborate with DHS/ CPS to ensure infant safety.
Prenatal Identification of Mom Implement Optimal Med Rx Program
Engage families in Safety Planning. Partner with other stakeholders to influence policy and primary
prevention. Provide primary prevention materials to sites.
To reduce the number of moms and babies with narcotic exposure, and
reduce the need for treatment of NAS.
GLOBAL AIM
Site Specific Graphs • Site specific data, as well as regional aggregate charts
are now in your hospital’s folder on SharePoint.
• This month we are excited to introduce enhancements to the monthly charts which provide more information in an easy to read table below your graphs.
• The data on your graphs has not changed.
• If you have any questions or comments about this new format please feel free to contact us!
Factorial Design and
Orchestrated Testing
Lloyd P. Provost, MS Associates in Process Improvement
OPQC NAS Project July 21 2015
API - 2015 17
18
Quality Improvement using the Model for Improvement
Hunches Theories
Ideas
Changes That Result in
Improvement
A P S D
A P S D
Very Small Scale Test
Follow-up Tests
Wide-Scale Tests of Change
Implementation of Change
What are we trying to accomplish?
How will we know that a change is an improvement?
What change can we make that will result in improvement?
Model for Improvement
Sequential building of knowledge under a wide
range of conditions API - 2015 18
Experimental Designs for QI Research
Speroff and O’Connor, Study Designs for PDSA Quality Improvement Research, Q .Manage Health Care, Vol 13, No.1, 2004
API - 2015 19
Why Experiments evaluating more than 1 Change?
PE book, p. 109-111
Two major deficiencies with studying one factor at a time: 1. The first is that there can be interactions between the factors
under study that produce the best results. • An interaction means that the effect a factor has on the response may
depend on the levels of some other factors.
2. The second deficiency in studying one factor at a time is inefficiency. As each factor is changed in turn, the data previously collected to study other factors are set aside and new data are collected. • Each set of data supplies information on only one factor. • Factorial designs provide an effective alternative to studying one factor at a
time.
API - 2015 20
Factorial Designs 1. They allow experimenters to accomplish two functions:
• determine what works best • understand the cause-and-effect system that produced the results.
2. They are most useful to determine the effects of multiple changes (2, 3, or 4 factors) on an outcome variable.
3. In addition to the effects of each individual change (factor), interactions between the factors can be studied.
4. Studying factors at two levels (with and without the change present) requires relatively few runs, leads to a simple analysis, and meets most of the needs of experiments
5. The use of graphical displays (run charts, cubes, paired comparisons, and response plots) can be used with factorial designs to effectively communicate the results of the experiment to others.
API - 2015 21
R. A. Fisher’s Tools for Experimentation
• Experimental pattern • Planned grouping • Randomization • Replication
API - 2015 22
Change A Change B Infection Rate*
Test 1 (baseline) no no 8.0
Factorial Designs: Testing Changes to Reduce an Infection Rate
Which change caused the improvement?
Response Variable – Infection Rate Factors – Change A, Change B Blocks – none identified Levels – no (not used) and yes for each factor
Test 2 yes no 7.0
Test 3 no yes 8.0
Test 4 yes yes 2.0
* 2% change is significant
API - 2015 23
“Orchestrated” Testing: Coordinate PDSA testing in a network to evaluate
Ideas for improvement
Using Factorial Designs in an Improvement Network
API - 2015 24
ImproveCareNow Orchestrated Testing Matrix ImproveCare
Now Teams
Enrollment & Data Quality
Pre-visit Planning
Population Management
Self-Management
Support Current #
Sites # Pts
(9/20/12)
A, B, C, D + - - - 3 84+*
E, F, G, H, I + + - - 5 1865
J, K, L, M + - + - 4 929
N, O + - - + 2 n/a
P, Q, R, S + + - + 4 702
T, U, V + - + + 3 308 W, X, Y, Z, AA, AB + + + - 6 689
AC, AD, Ae + + + + 3 903
Current enrollment data, not target for September 30, 2012
API - 2015 25
Why call this “Orchestrated Testing?”
Sections: the wind section, brass section, string section, percussion section, etc.
Multiple test sites that are all different
Led by a conductor who sets and maintains the tempo (speed), dynamics, and interpretation, including articulating style.
QI “Conductor”
All sections cooperate to create symphony. That means sometimes just resting and not playing
Each site has it’s own assigned test
All members of the orchestra agree to standardize and tune
All sites agree to the standard changes and the test plan
All orchestra rehearse and perform Preparation, execution, and reporting by all sites
All sections contribute to the whole; there can be no performance if any section is missing.
All tests must be completed to have an experiment. Each site gets more out of the experiment than they put in.
26
Orchestra Orchestrated Test
API - 2015
Proposed 24-1 Factorial for Hospital Readmissions
Standard Teaching Enhanced Teaching and Learning
Handoff Handoff
Current Handoffs
Enhanced Communication
Current Handoffs
Enhanced Communication
Current Follow-up Practices
Current Assessment Unit 3 Unit 7
Early Discharge Needs
Assessment Unit 8 Unit 4
Post-Hospital Follow-up
Current Assessment Unit 1 Unit 6
Early Discharge Needs
Assessment Unit 5 Unit 2
API - 2015 27
Each test with a different hospital or ward, IHI STAAR project, 2012
Potential Benefits and Challenges of Orchestrated Testing
1. Increase power for learning (larger sample size) from multiple sites.
2. Factorial designs allows more than one change to be tested at once (including synergy/antagonism)
3. Better design (standardization and replication) than current before & after testing each change independently
4. Bottom line: Potential to learn more, with less resources, and faster.
5. Down side? More complex to set up and manage
API - 2015 28
API - 2015 29
Children’s Hospital Neonatal Consortium (CHNC) in partnership with Children’s Hospital Association, 2013
Orchestrated testing in planned experimentation allowed simultaneous testing of multiple factors and the interaction of these factors on CLABSI rates.
The effect of four factors were studied 1. – Monitoring of hub care compliance 2. – Use of clean or sterile tubing change technique
3. – Monitoring unit policy on limitation of central venous catheter access 4. – Assessment of need for central line - The 17 Children Hospitals NICUs were allowed to choose the factors that fit within their practice culture
– Centers divided into eight groups based on their current practices and willingness to change
API - 2015 30
Design of the Orchestrated Test
API - 2015 31
CHNC Orchestrated Test
24-1 Fractional Factorial Design for Orchestrated Test
API - 2015 32
Shewhart U Chart for CHNC Orchestrated Test
API - 2015 33
API - 2015 34
CHNC Orchestrated Test
API - 2015 35
CHNC Orchestrated Test
Confirmation and Sustainability (2014)
API - 2015 36
CHNC Follow-up to Orchestrated Test
Results of Confirmation Study – 3 Groups of NICU’s
API - 2015 37
“Orchestrated” Testing: Coordinate PDSA testing in a network to evaluate
Ideas for improvement
Using Factorial Designs in an Improvement Network
API - 2015 38
“Orchestrated” Testing: Why Should OPQC Do this?
Top 5 Reasons We Should Pursue OT…
5. Neonatal Abstinence Syndrome (NAS) is still a relatively new disease, and is not yet fully understood
4. There is still variation in how different hospitals treat infants with NAS and, therefore, an opportunity to learn
Morphine Methadone Clonidine Phenobarbital Buprenorphine
3. We are doing well, but we can do even better
Length of Treatment 13.9 days Length of Stay 20.6 days
47.8% of Exposed Infants Treated
2. Together we can tease out the most effective components of care to further refine our current evidence-based, standardized care protocol
1. …and further improve the outcomes and reduce hospital length of stay for babies with NAS.
“Orchestrated” Testing: What is Our Proposed Plan?
Focus on Non-Pharmacologic Care • Areas of the care bundle where we are still
unsure…
Focus on Non-Pharmacologic Care Variation Exists
Focus on Non-Pharmacologic Care Some early evidence that 22 kcal feeds may
be beneficial in reducing LOS
Focus on Non-Pharmacologic Care Some early evidence that low lactose feeds
may NOT be an important part of care
OPQC Factorial Design (22)
Group 22 kcal/oz Standard
Low Lactose Standard
A No Yes
B Yes Yes
C No No
D Yes No
OPQC Factorial Design (22)
Wind Section (A) Low Lactose or BM
Standard 20 kcal/oz Standard
String Section (B) Low Lactose or BM
Standard 22 kcal/oz Standard
Horn Section (C) Regular Formula or BM
Standard 20 kcal/oz Standard
Percussion Section (D) Regular Formula or BM
Standard 22 kcal/oz Standard
Important Factors (But Not Under Study)
• Location of Treatment – Inborn vs. Outcome
• Availability of and approach to the use of breast milk
• Drug Used for Treatment – Methadone or Morphine
• Protocol Used for Treatment – OCHA vs. Cincinnati vs. Other
“Orchestrated” Testing: Will You Participate?
Ask Yourself These Questions… • Is this exciting to you?
• Are you willing to sail unchartered waters with
OPQC as we try out this method of learning?
• Do you want to help tease out the most effective components of care to develop and further refine our current evidence-based, standardized care protocol?
• Do you want to be part of continuing to improve outcomes and reduce LOS?
Ask Yourself These Questions…
• Are you willing to work with OPQC to select one of 4 groups to participate in?
• Will you agree to continue the practices of group and not to implement factors not assigned to their group for the duration of the project?
• Will you agree to keep your pharmacologic care practices the same for the duration of the project?
Ask Yourself These Questions…
• Will you be able to collect and submit data regularly?
• Can you collect data on all babies in whom Finnegan scores are initiated (not JUST those being treated) at your hospital whether in normal newborn or NICU?
• Do you have buy in for OT from your entire team (physicians, nurses, social workers)?
Ask Yourself These Questions…
• IF YES – We are excited to have you as part of
the core OPQC OT group
• IF NO – You are still part of OPQC and can
continue to learn with us, participate in AP calls, collect data, attend learning sessions
“Orchestrated” Testing: Next Steps…
What to Expect… • OPQC staff will be in touch to confirm your interest in participate and
discuss which group you will join
• We will provide you with a letter to circulate explaining our plan for pursing an OT design for the next 9 months including an “elevator speech” to help explain to your colleagues
• We will be asking for volunteers to help pilot test a new data collect form and draft clearer instructions – Thanks to those that already volunteered via the MPR
• We will be soliciting interested people to serve on a steering committee
and/or help lead action period calls for the 4 groups (“learning labs”) – Thanks to those that already volunteered via the MPR
Timeline
July Aug Sept October-June July
Hospitals commit to participate OPQC “assigns” sites to 1 of 4 groups Test & finalize Data Collection Form Test & finalize Data Collection Instructions Call with Lead Physicians (Aug)
Learning Session—PROJECT KICK OFF
Run OT Testing “Learning Lab” AP Calls
Every other Month Group Calls Data Collection & Submission
Continue work on other Key Drivers
Analyze Results Update &
Implement Best Care Practices
Questions and Discussion Please Tell Us…
• Does your team have any concerns about participating in orchestrated testing, such as: – Data collection logistics – Focusing on testing a specific piece of the NAS protocol – Continuing to engage team members throughout the next year – Sustainability for ongoing NAS work
• Would anyone from your team be willing to: – Be on a steering committee for orchestrated testing? – Work on improving the data collection form? – Help lead monthly “learning labs” for small group Action Period calls?
Next Steps… • Let us know if you are interested in
participating on a Steering Committee and/or helping to design Orchestrated Testing
• Please submit NAS Data by July 30th.
Remember to please submit and check “No Eligible Babies for the Month” if there were no NAS patients at your site.
• Monthly Progress Report was sent to Key
Contacts July 20th, due July 30th.
The OPQC NAS Project is funded by The Ohio
Department of Medicaid