Post on 20-Jun-2015
description
transcript
Health IT: The Critical Tool for Managing Clinical Care
Brent C. James, M.D., M.Stat.Executive Director, Institute for Health Care Delivery ResearchIntermountain HealthcareSalt Lake City, Utah, USA
iHT2
The Health IT Summit in Beverly HillsIntercontinental Los Angeles Hotel, Beverly Hills, California
Wednesday, 7 November 2012 -- 11:25a - 12:10p
Disclosures
Neither I, Brent C. James, nor any family members, have any relevant financial relationships to be discussed, directly or indirectly, referred to or illustrated with or without recognition within the presentation.
I have no financial relationships beyond my employment at Intermountain Healthcare.
Quality, Utilization, & Efficiency (QUE)Six clinical areas studied over 2 years:- transurethral prostatectomy (TURP)- open cholecystectomy- total hip arthroplasty- coronary artery bypass graft surgery (CABG)- permanent pacemaker implantation- community-acquired pneumoniapulled all patients treated over a defined time period
across all Intermountain inpatient facilities - typically 1 yearidentified and staged (relative to changes in expected utilization)- severity of presenting primary condition- all comorbidities on admission- every complication- measures of long term outcomescompared physicians with meaningful # of cases
(low volume physicians included in parallel analysis, as a group)
IHC TURP QUE StudyMedian Surgery Minutes vs Median Grams Tissue
M L K J P B C O N A I D H E G F0
20
40
60
80
100
0
20
40
60
80
100
Attending Physician
Median surgical time Median grams tissue removed
Gra
ms
tissu
e / S
urge
ry m
inut
es
IHC TURP QUE Study
1500 1549 1568 16181543
1697
1913
22332140 2156
1598
12691164
1552 15561662
A B C D E F G H I J K L M N O PAttending Physician
0
500
1000
1500
2000
2500
Dol
lars
0
500
1000
1500
2000
2500
Average Hospital Cost
1. Well-documented, massive, variation in practices (beyond the level where it is even remotely possible that all patients are receiving good care)
2. High rates of inappropriate care
3. Unacceptable rates of preventable care- associated patient injury and death
4. A striking inability to "do what we know works"
5. Huge amounts of waste leading to spiraling prices that limit access (46.6 million uninsured Americans)
The opportunity (care falls short of its theoretic potential)
50+% of all resource expenditures in hospitals is
quality-associated waste:recovering from preventable foul-upsbuilding unusable productsproviding unnecessary treatmentssimple inefficiency
Andersen, C. 1991James BC et al., 2006
60,001.8
2009
0
10
20
30
40
50
60
Trill
ion
$
Total U.S. fiscal exposuresBy layering on future obligations, the total net prevent value (PV) of debt rises
to over $60 trillion -- about $195,000 for every man, woman and child in the U.S. More than two-thirds of the shortfall arises from health care delivery.)
Source: GAO. Financial Reports of the United States Government for the Years Ended September 30, 2009 and 2008.
Federal employee and veteran benefits ($5,283.7 B)
Federal debt securities ($7,582.7 B)
PV of Social Security shortfall ($7,677.0 B)
PV of Medicare Part A shortfall ($13,770.0 B)
PV of Medicare Part B shortfall ($17,165.0 B)
PV of Medicare Part D shortfall ($7,172.0 B)
Other explicit liabilities
($1,257.4 B)
The Fiscal Gap (unfunded federal obligations - 2009)
Social Security$7.7 trillion
Medicare$38.1 trillion
TotalNational Debt
$14.1 trillion
Stimulus$862
billion
NationalDefense$714
billion
TARP$700
billion
Unfunded obligations
Health care payments will be cut
NIH-funded randomized controlled trialassessing an "artifical lung" vs. standard ventilator managementfor acute respiratory distress syndrome (ARDS)
discovered large variations in ventilator settings across and within expert pulmonologists
created a protocol for ventilator settings in the control arm of the trial
Implemented the protocol using Lean principles (Womack et al., 1990 - The Machine That Changed the World)- built into clinical workflows - automatic unless modified- clinicians encouraged to vary based on patient need- variances and patient outcomes fed back in a learning loop
Dr. Alan Morris, LDS Hospital, 1991:We have found proven solutions
Challenges building guidelines
Lack of evidence for best practice- Level 1, 2, or 3 evidence available only about 15-20% of the time
Expert consensus is unreliable- experts can't accurately estimate rates using subjective recall
(produce guesses that range from 0 to 100%, with no discernable pattern of response)- what you get depends on whom you invite (specialty level, individual level)
Guidelines don't guide practice- systems that rely on human memory execute correctly ~50% of the time (McGlynn: 55% for adults, 46% for children)
Results:survival (for ECMO entry criteria patients) improved from 9.5% to 44%costs fell by ~25% (from $160k to $120k)physician time fell by ~50%
we generalized the concept: Shared Baseline protocols ("bundles") to standardize care whileencouraging clinicians to vary based on individual patient needs;then feeding back variation and patient outcome data in a "learning system"
Dr. Alan Morris, LDS Hospital, 1991
07 Ja
nMarMay Ju
lSep Nov
08 Ja
nMar May Ju
lSep Nov
09 Ja
nMarMay Ju
lSep Nov
10 Ja
nMar
Month
0
20
40
60
80
100
% c
ompl
ianc
e
0
20
40
60
80
100
ER bundle ICU bundle All components
Sepsis bundle compliance
04 Ja
nMay Sep
05 Ja
nMay Sep
06 Ja
nMay Sep
07 Ja
nMay Sep
08 Ja
nMay Sep
09 Ja
nMay Sep
10 Ja
n
Month
0
0.1
0.2
0.3
0.4
0.5
Mor
talit
y ra
te
0
0.1
0.2
0.3
0.4
0.5
Sepsis mortality - ER-ICU transfers
20.2%
8.0%
125+ fewer inpatient deaths per year
2832
4437
4542
4223
3429
4133
4553
3850
4739
3130
3424
4041
3528
2722
2827
2432
4436
3952
5170
6560
4757
5250
6151
4377
7377
6571
6948
5259
4663
6868
6370
9490
7581
6979
8178
8270
7484
91n=
We count our successes in lives ...
Lesson 1
Very often,
better care is cheaper care ...
Lesson 2
Aligning financial incentivesNeonates > 33 weeks gestational age
who develop respiratory distress syndromeTreat at birth hospital with nasal CPAP (prevents
alveolar collapse), oxygen, +/- surfactantTransport to NICU declines from 78% to 18%.Financial impact (NOI; ~110 patients per year; raw $):
Birth hospitalTransport (staff only)
Tertiary (NICU) hospitalDelivery system total
Integrated health planMedicaid
Other commerical payersPayer total
Before 84,24422,199
958,4671,064,910
900,599652,103
429,1011,981,803
After 553,479
- 27,222 209,829736,086
512,120373,735
223,2151,109,070
Net 469,235
- 49,421 -748,638-328,824
388,479278,368
205,886872,733
Current payment mechanisms
Actively incent overutilization: do more, get paid more - even when there is no health benefit
I am paid to harm my patients (paid more for complications)
Actively disincents innovation that reduces costs through better quality (a key success factor for the rest of the U.S. economy)
Very strong, deep, wide evidence showing exactly this effect throughout U.S. healthcare
Bending the cost curve
1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
2010
2015
0
5
10
15
20
25
% G
ross
Dom
estic
Pro
duct
0
5
10
15
20
Tota
l $ p
er U
S ci
tizen
(tho
usan
ds)
2,281
4,729
3,762
6,683
9,173
12,357
148 357 1,106
1. ACOs, AMHs, bundled payment, shared savings, pay for value: sophisticated forms of capitation- provider at (financial) risk ... but with far better data systems for
(1) quality measurement and (2) risk adjustment
2. Represent "managed care at the bedside"- ask clinical teams at the bedside to manage the care, not distant
and disengaged insurance companies
3. More than 80% of cost saving opportunities live on the clinical side; 70+% of clinical improvement activities reduce costs by freeing up care delivery capacity (technically, "fixed cost leverage").
Capitation makes a comeback
Our answer:
A Shared Accountability Organization:
Physicians,hospitals,payers, and
patients
with aligned professional and financial incentivesto seek
the best medical resultat the lowest necessary cost
Some key elements:Pay first dollar, not last dollar
(defined contribution, not defined benefit; reference payment)
Whoever makes the consumption decision bears the (appropriate) financial consequences (patients and physicians have skin in the game)
No incentive to risk-select patients (community-rated premiums, but risk-adjusted capitation payments)
Levers: No incentives to overtreat or undertreatPayments targeted at break-even, most efficient cost of operations;
all upside $$ contained in shared savingsHitting measured quality thresholds a prerequisite
to participate in shared savingsInvolve employed and affiliated physician groups
via partner health plans
Process management is the keyhigher quality drives lower costsunder capitation, all of the savings come
back to clinical process managersmore than half of all cost savings will
take the form of unused capacity (fixed costs:empty hospital beds, empty clinic patient appointments, and reduced procedure, imaging, and testing rates)
balanced by increasing demand(Baby Boom; obesity; community growth; technological advances; may still require some capacity management / reduction)
major financial model shift, from revenue enhancement to cost control
key difference: it takes a team
1. Identify a high-priority clinical process (key process analysis)
2. Build an evidence-based best practice protocol(always imperfect: poor evidence, unreliable consensus)
3. Blend it into clinical workflow (= clinical decision support; don't rely on human memory; make "best care" the lowest energy state, default choice that happens automatically unless someone must modify)
4. Embed data systems to track (1) protocol variations and (2) short and long term patient results (intermediate and final clinical, cost, and satisfaction outcomes)
5. Feed those data back (variations, outcomes) in a learning loop - constantly update and improve the protocol- provide true transparency to front-line clinicians- generate formal knowledge (peer-reviewed publications)
Process management means health IT
Better has no limit ...
an old Yiddish proverb