Date post: | 27-Aug-2018 |
Category: |
Documents |
Upload: | truongphuc |
View: | 225 times |
Download: | 0 times |
Quality and Variation in Medical Practice: Why are
Doctors so Different?
Mark W. Shen, M.D.November 19, 2010M k W Sh M D
Objectives
• List 3 examples of significant variation in pediatric practice
• Describe the relationship between variation and quality of a process
• Describe one method of improving quality of care by addressing variation
A Mad Lib
1. Pick a pediatric practitioner• Generalist, pulmonologist, rheumatologist,
hematologist, neonatologist, ID, GI, ENT…2. Pick a management scenario:
• ITP, post-op T&A, HSP, bronchiolitis, protein-losing enteropathy, post-op cardiac surgery, bacterial meningitis…
3. Pick a word pair: • Given & stop OR Not given & start
Your Mad LibYou are a pediatric [insert type of practitioner] and begin
your busy Monday by seeing a patient cared for by your partner over the weekend
The patient has been receiving care for [insert disease]and was [given/not given] steroids.
You completely disagree with this approach and [stop/start] the medication.
What a frustrating start to the day. Sometimes you wonder how 2 physicians could practice such different medicine.
As you leave the room, you notice a look of puzzlement on the family’s faces…
Encountering Variation:The 5 Stages of Grief
• Denial– Is that person board-certified?
• Anger– It’s my patient, I can do what I want
• Bargaining– Let me try to use the family to get my way (I’ll tell
them my side)• Depression
– I’m an accomplice in providing poor care• Acceptance
– Just do whatever the other MD wants
The History of (the study of) Medical
Variation
Int J Epidemiol 2008;37:9–19
Variation in Incidence of Tonsillectomy: J Alison Glover
“Puzzling as is the geographical distribution, the social distribution is yet more of an enigma.Tonsillectomy is at least three times as common in the well-to-do classes.”
Sci Am 1982;246:120-34
Tonsillectomy Variation: Back Across the Pond
• 1934, American Child Health Association• 1000 New York City School Children • 40% had not yet undergone tonsillectomy
– School physicians: 45% needed an operation• Of those not selected, another group of physicians
recommended that 46% receive tonsillectomy– Of the twice-rejected children, a third group of physicians
recommended operation in 44%
After 3 exams, only 65 children remained
The Beginnings of Modern Day Variations Research
John Wennberg’sHouse
Science 1973;182:1102-1108
Extreme Variation in Tonsillectomy Rates
Sci Am 1982;246:120-34
Probability of Having Surgery in 11 Vermont Hospitals
Int J Epidemiol 2008;37:26–29
Surgical Rates for the Most Populous Hospital Areas: Maine
Int J Epidemiol 2008;37:26–29
The Surgical Signature
NEJM 1982;307:1310-14
International Differences in Surgical Rates
Sci Am 1982;246:120-34
Proof of Preference-Sensitive Care
Preference-Sensitive Careaka:
Medical Variation:The Present
Terminology
• Unwarranted Variation:– Care that is not consistent with a patient’s preference
or related to their underlying illness• Preference-Sensitive Care
– No right rate (T&A)– Misuse
• Effective Care– Evidence-based care not provided– Underuse
Evidence for Underuse of Effective Care: Adults
Evidence for Underuse of Effective Care: Children
Terminology
• Unwarranted Variation:– Care that is not consistent with a patient’s preference
or related to their underlying illness• Preference-Sensitive Care (Misuse)
– No right rate• Effective Care (Underuse)
– Evidence-based care not provided• Supply-Sensitive Care (Overuse)
– Systems supply creates demand
http://www.dartmouthatlas.org/downloads/reports/Spending_Brief_022709.pdf
The Dartmouth Atlas: Medicare Spending Varies Dramatically
http://www.ahrq.gov/about/annualconf09/brownlee.htm
63%12%
25% 63%1212%12%12%12%21
25%
Preference Sensitive Care
Effective Care
Supply Sensitive Care
Unwarranted Variationin Medicare Spending
Source: John E. Wennberg and Dartmouth Atlas
www.nejm.org
A Trending Topic
www.pediatrics.org
Trending in Pediatrics
Examples of Data Sources in Pediatrics
• Databases– PHIS (Pediatric Health Information Systems)
• Propietary administrative database• Maintained by Child Health Corporation of America
(CHCA)– Business Alliance of 42 children’s hospitals
• Collaborative Networks– VIP (Value in Inpatient Pediatrics)
• Grassroots collaborative improvement network• Data: administrative and chart review
Pediatrics 2009;123;636-642
Osteomyelitis: Variability in Early Conversion to Oral Therapy
0
10
20
30
40
50
60
70
80
90
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Hospital
Con
vert
ed to
ora
l the
rapy
(%)
Pediatrics 2010;126;196-203
UTI in Infants: Variability in Length of IV Therapy
Circulation. 2010 Nov 8. [Epub ahead of print]
Corticosteroid Use After Congenital Heart Surgery
Source: VIP Network
Bronchodilator Doses Per Patient in Acute Bronchiolitis
Center
I See Variation
Isn’t Variety the Spice of Life?
Unwarranted Variation:The Losers
• Patients• Learners• System (everyone loses)
Preference-Sensitive Losers:The Patients
• Patients lose when not involved– Recent Dell Children’s patient comments:
• “Doctors, deliver a consistent message. We heard different plans from different doctors.”
• “Lack of communication between doctors”• “I was given conflicting info, on which I had to
make a judgment call.” I didn’t know who to talk to.• “Too many doctors involved”
Preference-Sensitive Losers:The Learners
“What do you want to do?”(everyone does things differently so just tell me
what you want to do)
“I don’t care”
Adverse Effects of Unmeasured Variation
Adverse Systems Effects of Unwarranted Variation
Medicare: Cost vs Quality The Value Equation
Quality• Value = ---------------------
Cost
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
$- $10,000 $20,000 $30,000 $40,000 $50,000 $60,000 $70,000 $80,000 $90,000
Average Annual Reimbursement per Beneficiary (Wage-Index Adjusted)
Aver
age
Qua
lity
of C
are
Scor
e
* Based on percent of beneficiaries with three conditions (diabetes, chronic obstructive pulmonary disease, and congestive heart failure) who had a doctor’s visit four weeks after hospitalization, a doctor’s visit every six months, annual cholesterol test, annual flu shot, annual eye exam, annual HbA1C test, and annual nephrology test. Source: G. Anderson and R. Herbert for The Commonwealth Fund, Medicare Standard Analytical File 5% 2001 data.
Best Practice CurveA
Greenville, NC
B C
DDNewark, NJ
CCMelrose Park, IL
BSaginaw, MI
Manhattan, NY
Orange County, CAEast Long Island, NY
t Practice CEaEE
Ft. Lauderdale, FL
Boston, MA
Variation in Annual Total Cost and Quality for Chronic Disease Beyond Just The Numbers
McAllen vs El Paso: Medicare Spending
http://www.whitehouse.gov/omb/blog/09/06/04/McAllenRedux/ http://www.newyorker.com/reporting/2009/06/01/090601fa_fact_gawande?currentPage=all
To Vary is Human
http://www.managedcaremag.com/archives/0311/0311.variation.html
CABG Rates in California
Public Perception:CABG Rates, Redding vs CA
Doctors’ Decisions and Impact on Medical Care
Lenses Under Which to Analyze Doctors’ Decisions
• Uncertainty & limits of the human brain– Medical decision-making, clinical problem-
solving• Different Disciplines
– Clinical, economic, sociological, psychological• Components
– Patient, physician, system
Health Affairs 1984;3:74-89
We Are Surrounded By Uncertainty
• Defining a Disease• Making a Diagnosis• Selecting a Procedure (e.g., test or
intervention)• Observing Outcomes• Assessing Preferences
Health Affairs 1984;3:74
Colorectal Experts:Consensus???
Question: What is the effect of screening annual fecal occult blood and flexible scope on colorectal cancer?
Doctors Decisions and the Cost of Medical Care. Michigan, 1986
Eisenberg: Determinants of Medical Decision-Making
• Physician as a self-fulfilling practitioner
• Physician as patient’s agent
• Physician as guarantor of social good
Doctors Decisions and the Cost of Medical Care. Michigan, 1986
Eisenberg: Determinants of Medical Decision-Making
• Physician as a self-fulfilling practitioner1. Desire for income2. Desire for a style of practice3. Personal characteristics4. Practice setting5. Standards established by clinical leadership
Medical Care 1981;19:297-309
Older Doctors Use Fewer Laboratory Tests
Doctors Decisions and the Cost of Medical Care. Michigan, 1986
Eisenberg: Determinants of Medical Decision-Making
• Physician as a self-fulfilling practitioner1. Desire for income2. Desire for a style of practice3. Personal characteristics4. Practice setting5. Standards established by clinical leadership
Health Affairs 1984;3:74
David Eddy on Practice Setting’s Impact on Variation
This tendency to follow the pack is the most important single explanation of regional variations in medical practice.
If uncertainty caused individual physicians to practice at random, or to follow their personal interpretations and values, without any attempts to match the actions of their neighbors, the variations in practice patterns would average out, and no significant differences would be observed at the regional level.
Differences between regions are observed because individual physicians tend to follow what is considered standard and accepted in the community.
Doctors Decisions and the Cost of Medical Care. Michigan, 1986
Eisenberg: Determinants of Medical Decision-Making
• Physician as patient’s agent1. Economic agent2. Clinical agent3. Patient demand4. Defensive medicine5. Patient characteristics6. Convenience
Doctors Decisions and the Cost of Medical Care. Michigan, 1986
Eisenberg: Determinants of Medical Decision-Making
• Physician as guarantor of social good– Duty to the patient vs steward of resources– Tension between “the prisoner’s dilemma”
and “the tragedy of the commons”• Classic scenario: end-of-life care (flat of the
curve medicine)
Does Genotype Determine Medical Decision-Making? Knowledge is Paralyzing
How Do We Improve?
Learn from Patients
• Paternalistic Craft-based Silos are archaic
Learn from Improvement Science
1. Measure the process2. Analyze the data3. Intervene: Control the process
(Research is a Slightly Different Order)
1. Measure the process2. Control everything3. Intervene4. Analyze the data
Learn from Improvement Science
1. Measure the process2. Analyze the data3. Intervene: Control the process
• Control unwarranted variation through standardization
• Continue to measure and analyze
Quality Improvement in Action: Control Chart
0%10%20%30%40%50%60%70%80%90%
100%
11/15
/2009
11/22
/2009
11/29
/2009
12/6/
2009
12/13
/2009
12/20
/2009
12/27
/2009
1/3/20
10
1/10/2
010
1/17/2
010
1/24/2
010
1/31/2
010
2/7/20
10
2/14/2
010
2/21/2
010
2/28/2
010
3/7/20
10
3/14/2
010
3/21/2
010
3/28/2
010
4/4/20
10
%
Step 1: Measure
Step 3: Intervene - Standardize
Step 2: Analyze
Learn fromHigh Performers
• Pediatric Oncology– Minimal unwarranted variation– All variation is measured & patient-level– Enormous success
• Pediatric cancer transformed from uniformly fatal disease in 1950s to 78% five-year survival for all types
– Better outcomes than adult groups for adolescents and young adults (AYA)
Blood 2008;112:1646-1654
Pediatric vs Adult Trials in AYA with ALL
• Reasons– Better compliance on
pediatric protocols– Better enrollment in
pediatric trials• Next Steps
– Enrolling adults in trials with pediatric protocols
If to vary is human,then only through
collaboration will we truly divine