Business Process and Data Analytics

Report
Data + Process: The
Peanut butter cup of HIT!
Howard Landa, M.D.,
Chief Medical
Information Officer
Alameda Health System
Oakland, CA
Quoting smart people
makes you seem smarter
• Data is like salt water. It can drown you in its
native state, but filtered and processed it sustains
life. “The goal is not to drown in data while
thirsting for information”-Reid Coleman
• Using data to impact processes
drives value from the data
• Dashboards are not very useful
when standing still
2
By any other name…
Beyond Meaningful Use
MU2 : Meaningful use of
Meaningful Use
Why we all started doing this:
To use technology to improve
the process of Healthcare
3
Healthcare Information Technology Truisms
1) The EHR shines a bright spotlight on all your bad
procedures, processes and actors
2) Workflow is everything (Implementation)
3) Computers make smart people
do stupid stuff
4) Where Workflow isn’t everything,
Data is everything
5) Data impacting Workflow is really everything
(Optimization)
4
Scenario #1
• Situation
– Well resourced Accountable Care Organization
– Mature EHR implemented and optimized
– Reliable data and clinical decisions support
• Goal:
– Improve a specific aspect of care
– One that Passes the “Champagne test”
5
Goal: Improving Mammogram Screening
• Improve ordering of Mammography
– Identify
Alerting
– Ease ordering
• Improve cycle time
• Improve patient satisfaction with process
• Earlier detection  Improved outcome
Initial attempt
• Provide the structure
– Alerts and Reminders (Staff and Patients)
– Process support
• Mature ubiquitous EHR
• Incentives
– Analytic data
• Screening rates, Treatment benefit
• Monitor alerts and orders
7
Modest improvement
• Modest increased screening rates
• PCP most impacted
• Data on resistance of other providers
– Educational issue?
– Ineffective notification?
– Ease of ordering issue?
– Result workflow?
NO
NO
NO
BINGO!
8
Mitigating Reluctance of non-PCP
Result-o-phobia
The irrational fear of looking stupid when an otherwise competent
professional does not know how to deal with a test result. e.g.:
• Pediatric urologist sees patient for a circumcision, sees
alert and orders a mammogram for his patient’s mother
• Result comes back “BIRADS 3, Borderline abnormal”
• Pediatric urologist develops sweating and palpations when
he realized he has to contact what has inadvertently
become “his patient!”
Mitigating Reluctance of non-PCP
The cure for Result-o-phobia
• Ordering provider legally must get the test result
Ordering provider gets “Courtesy” copy
• APN intercepts ALL results and is accountable for negative
study communication and follow-up
• Communicates with Breast team non-negative
• Breast team accountable for all subsequent care
• Resistance is futile!
Now we’re talking
• Resistance is now mitigated
• Kaiser Permanente Hawaii in NCQA 2009
– Highest breast CA screening rate (42 – 69)
~72%  85% in 2009
– EHR can alert and facilitate
– Incentives and accountability
– Operations to support workflow based on sequential
evaluation of analytic data
Scenario #2
• Situation
–
–
–
–
County Hospital/Inner city trauma center
Poorly resourced, Resident run
Minimal Electronic systems (ED EMR)
“The best thing about standards is that there are so many to
choose from!”
• Goal: To improve the quality of care
– Optimize and Standardize disparate paper order sets,
leverage evidence and facilitate workflow
• Recommendation: CPOE!...when we have $$$
12
Is there a stepping stone to CPOE?
• Electronic Order EntryPaper Output
– Increase the number of order sets
– Minimize workflow change
– Resource “light” implementation
– Expedite creation and maintenance of order
set content
– “Test drive” order sets to support future
enterprise EHR
13
Stepping stones to CPOE
• Electronic Order Entry with Paper Output
– “Electronify” starter/current paper forms
– Add “low impact” Clinical Decision support
– Expedite creation and maintenance of
order set content
– Expand functional and conditional order sets
– “Test drive” order sets
– Iterate!
Immediate Clinical Content Committee:
Rapid cycle content management
Deploy new
Content
ICCC:
 Pharmacy
 HIM
User
feedback
ICCC Design
 Quality
 Safety
 Informatics
Order use
analytics
Care quality improvement
• Improved compliance with ordering of complete
insulin coverage
• DVT prophylaxis compliance has increased from 50%
to 97%
• Compliance with required orders
– E.g. Admit, Code status, Telemetry criteria,
Resident/Attending/Service identification
• 28% reduction in time from decision to admit to
creation of inpatient orders
16
Why it worked
• Workflow tools at the Point-of-Care
• Clinical Decision Support
– Leading and guiding vs. Interrupting
• Minimal change to manage
• Ordering analytics to support rapid, iterative
improvement cycles
• Make the “Right” thing to do
the “Easy” thing to do
17
Taking it to the next level
• Goal: Reduce Sepsis Deaths by 15%
• Challenge to get baseline data
– Bad data  Denial, Anger, Bargaining
3/5 Kubler-Ross
• Challenge to impact process with bad data
• Change Management a major problem
• Need for standardized protocol and procedure
for physicians and nurses
Overcoming
Barriers
• Multidisciplinary education
• Standardized Protocol and
Procedures
– Physicians and nurses
– Sepsis education
• Paper based pilot
Sepsis the Highland Way
Youtube video
Initial results promising but limited
• Data evaluation and process iteration
– Lactate not coming back quickly 
POCT testing
– Antibiotic Delivery: Involve Pharmacy
 Change Management: Making resistance futile
– Paper to Electronic
• ED order sets in EMR
• Admission in SmartOrders
• Provider and nurse focus and education
Way better!
• From 2009 to 2011 56% reduction in
sepsis mortality! But…
– Bundle compliance still not close to 100%
– Significant paper processes remained
– Data collection was still a challenge
– Change is hard, maintaining change is harder
Let’s implement an EHR!
• Mid-2012 through Early-2013 we implemented
ambulatory and inpatient EHR
• Better order processing
• Better decision support
• Lots of distraction (Quality, Clinicians, IT)
• Rule #2: Workflow is everything
(during Implementation)
Let’s optimize and EHR!
• Don’t drown in data while thirsting for info
– Too much data, not enough information
– Too much distraction and too much change
– Meaningful Use <> Using Meaningfully
 Rule #5: Data impacting Workflow is really everything
(Optimization)

Identify and track key metrics
 Implement process improvements and iterate!
 Still in early phase
24
Success factors
• It is NOT just the data or just the workflow!
• Use good data to drive great change
• Manage the change:
– Make the change feel do-able: small bites
– Try to pass the Champagne test
– Make the “Right” thing to do the “Easy” thing to do
• Metrics are tools, not destinations. The process of
Healthcare is just that, a process.
26

similar documents