Transcript Surgery at the End of Life Kwok et al 2011

The Importance of Outcome Measurement
in Quality and Safety of Care
Alan Merry
Disclosure
Alan Merry has financial interests in
Safer Sleep LLC
Is on the Boards of
Safer Sleep LLC
NZ Health Quality and Safety Commission
Lifebox
ANZCA
and has received support for research from
ANZCA
WHO
HRC NZ
AFT Pharmaceuticals
Roche Baxter
and others
Outcomes and how
to change them
Objectives…
• The escalating cost of
healthcare
• Variation in healthcare
• The elements of quality in
healthcare
• Natural history and placebo
• Causation and association
• Drivers of treatment
• Chronic angina pectoris as
an illustrative example
“We would ask you to include
a broad perspective”
The Committee on the Quality of
Healthcare in America
“Americans now invest annually $1.1 trillion or
13.5% of the nation’s GDP in the health care
sector. This figure is expected to grow to more
than $2 trillion or 16% of GDP, by 2007 .”
IOM Crossing the Quality Chasm: A New Health System
for the 21st Century National Academy Press 2001
Since 1995 growth in health costs >
growth in national income
OECD average 18%
NZ 30%
MRG Report 2009
Davis et al 2010 The
Commonwealth Fund
USA = $7,290
Expenditures in
$US PPP
(purchasing power
parity)
Netherlands is
estimated
OECD Health Data,
2009
Australia = $3,357
NZ = $2,454
Davis et al 2010 The
Commonwealth Fund
Dartmouth Health Care Atlas
(John Wennberg)
Geographic
variations in
Medicare
resource use
across practice
intensity
www.dartmouthatlas
Variation
Profiles of
Common
Surgical
Procedures
Ratio of observed
to expected
(national average)
Medicare rates in
306 US hospital
referral regions
Ghaferi et al 2009 NEJM 361: 1368-75
Ratio highest to lowest
Radical
prostatectomy
Carotid
endarterectomy
Hip fracture
1.9
10.1 7.8
Variation
Profiles of
Common
Surgical
Procedures
“Uncertainty about
treatment
recommendations
is attributable to
disagreement
among physicians
about both facts
and values”
Ghaferi et al 2009 NEJM 361: 1368-75
Adherence to Quality
Indicators According to
Condition Adapted from
McGlynn et al 2003 NEJM
348: 2635-45
“No one method of non-operative or
operative treatment would seem
definitively to be superior to another”
“After around 35 studies have
failed to provide a definitive answer regarding the
efficacy of epidural steroid injections, it is unlikely
that future trials will do so”
Repetitive Strain Injury
“Little is known about the effectiveness of treatment
options for upper limb disorders. Strong evidence for
any intervention is scarce and the effect, if any, is
mainly short-term pain relief”
“Exercise is beneficial for non-specific
upper limb disorders”
Van Tulder 2007 The
Lancet 369: 1815-22
The Placebo Response
“Accumulated evidence suggests that the
placebo effect is a genuine psychobiological
event attributable to the overall therapeutic
context”
Finniss et al 2010 The
Lancet 375: 686-95
Drivers to Treat
• Rescue
• Death
• Paralysis
• Agony
• Professional expectations
• Targets
• $
Evidence of efficacy
The Dimensions of Quality and
Organizational Layers of Healthcare
Runciman B Merry A Walton M
Safety and Ethics In Healthcare Ashgate 2007
Surgery at the End of Life
A retrospective cohort study of 1 802 029
elderly beneficiaries of fee-for-service Medicare
who died in 2008
Kwok et al 2011 Lancet 378: 1408-13
Surgery at the End of Life
Patients who underwent an inpatient
surgical procedure during…
• Their last year of life
• Their last month of life
• Their last week of life
31.9%
18.3%
8.0%
Kwok et al 2011 Lancet 378: 1408-13
Surgery at the End of Life
• Substantial regional variation
• Rate of surgery related to
total Medicare spending
no of hospital beds
age (inversely)
• Not related to
no of surgeons
High surgical
intensity
regions also
had higher
rates of death
Kwok et al 2011 Lancet 378: 1408-13
Surgery at the End of Life
Also more often men and non-white
Kwok et al 2011 Lancet 378: 1408-13
Risk of Cardiac Surgery the STS
Database and EuroSCORE
Nashef et al 2002 European Journal of Cardio-thoracic
Surgery 22: 101-5 (and see 2012 41: 734-45)
Bypass and the Brain
T2-WI
Intracardiac surgery
• 43% - new ischemic
lesions
• Cognitive decline –
100% vs 35%
Barber et al 2008 Stroke 1427-33
CABG vs PCI
ACCF Registry & STS Database
65yrs + 2-3 vessels, no AMI
(“ASCERT”)
At 4 years
• RR 79%
• ARR 4.4%
Mean age: 74 years Weintraub et al 2012 NEJM 366: 1467-76
“It is useful to begin by examining the trials comparing CABG
surgery with medical treatment of coronary artery disease . A
meta-analysis.. ..demonstrated a significant survival
advantage for surgery over medical treatment for patients with
three-vessel and left main stem coronary artery disease.1”
Results
“2649 patients entered the seven trials between 1972
and 1984….”
Yusef et al 1994 Lancet
Alison et al 2007 NZMJ 1255: U2557
Medicine, Angioplasty, or
Surgery Study (MASS II)
(N = 611: 5 Year Follow Up)
“Background—Despite routine use of coronary
artery bypass graft (CABG) and percutaneous
coronary intervention (PCI), no conclusive
evidence exists that either modality is superior to
medical therapy (MT) alone for treating
multivessel coronary artery disease with stable
angina and preserved ventricular function”
Hueb W et al
Circulation 2007 115: 1082-89
Primary endpoint: Survival free of
• total mortality
• unstable angina 
revascularization
• Q-wave MI
Hueb W et al
Circulation 2007 115: 1082-89
Hueb W et al
Circulation 2007 115: 1082-89
Natural History
“A ticking bomb in my chest”
“Brought back from the perilous precipice
of paralysis”
(Photograph used with permission)
SCS for Angina
20 patients  paced to angina  rest  SCS 
paced to same rate  rest  paced to angina
SCS:• Time to angina time to recovery
• Magnitude of ST depression 
• Lactate production  extraction
• Angina at higher rate (still felt)
Eliasson et al 1996 Pain 65: 169-79
The ESBY Study
• 104 patients randomised: SCS 53 CABG 51
• Symptomatic indication - not prognostic
• High risk for CABG
• peripheral disease
• diabetes mellitus
• ejection fraction < 40%
• peripheral vascular disease
• renal dysfunction
Mannheimer 1998 Circulation 97: 1157-63
The ESBY Study
Angina (attacks/week)
Pre
Post
CABG
16.2(12.6)
5.2(10.3)

SCS
14.6(13.5)
4.4(7.4)

<0.001
 ns
<0.001
Mannheimer 1998 Circulation 97: 1157-63
The ESBY Study
Mortality
CABG*
7
 0.02
SCS**
1
*6 cardiac, 1 cardiac + stroke, 3 preop
**MI at 3/12
Mannheimer 1998 Circulation 97: 1157-63
The ESBY Study
Cerebrovascular Morbidity
CABG
8
 0.03
SCS
2
Mannheimer 1998 Circulation 97: 1157-63
The ESBY Study
5-year (4.8) Follow-up
Survival at 5 years
 SCS: 75.5% CABG: 68.6% (n.s.)
 Comparable with similar groups of CABG
and PCI
 66% deaths cardiac
Ekre et al 2002 European Heart Journal 23:
1938-45
SCS for Angina
Risk
• Retrospective multicentre study 1987-97
• 21 centres:
• 517 pts - range of risk factors…
“...characterised by end stage CAD, severely
disabling angina not responsive to conventional
treatment, and a moderately reduced LVEF.”
• Follow up
• mean 23 months
• range 0 -128 months
Ten Vaarwerk et al 1999 Heart 82: 82-88
SCS for Angina
Risk
• Annual mortality
• all causes = 7-8%
• cardiovascular = 3.5-5%
• 78% survival at 3 years - similar to CASS
“...does not exceed mortality in more or less
comparable groups of patients”
• NYHA functional class 3.5  2.1 (P < 0.01)
Ten Vaarwerk et al 1999 Heart 82: 82-88
Revenge Effects
The outcome of any recommended
action to improve safety should be
measured because
• It may not be effective
• “Revenge effects may occur”
Example of a revenge effect: screening
for prostatic cancer means many men
have treatment in whom the cancer
would not have manifest clinically – net
benefit is not proven (Harris 2002
Annals of Internal Medicine 131 917-29)
(Photograph used with permission)
271 patients – education, demystification, CBT
• Hospital admissions per year from 2.4 to 1.8
• Bed occupancy days per year from 15.5 to 10.3
• Overall mortality lower than comparable CABG
Moore et al 2007 JPSM 33: 310-16
Are French Red Wines More
Cardioprotective?
• 1786 Heberden noted relief of angina pectoris by
“spirituous cordials”
• Confirmed by almost 100 geographic, case-control,
cohort, and epidemiological studies
• J curve
Risk
Dose
Vogel et al 2003 JACC
41: 479–81
Are French Red Wines More
Cardioprotective?
• Women 1, men 2-3 glasses per day
• Regular (binges are bad)
• ?Better with meals
• Benefit confined to middle aged and older people
Risk
Dose
Vogel RA JACC 41 2003
479–81
Are French Red Wines More
Cardioprotective?
• Phenols and flavinoids may be relevant  NO
•  HDL
•  LDL
Risk
Dose
Vogel RA JACC 41 2003
479–81
WHAT THIS STUDY ADDS
The results confirm the beneficial effects of moderate
alcohol consumption and the need to elucidate the
underlying pathophysiological mechanisms
All cause mortality for drinkers vs non drinkers
RR = 0.87 (0.83 to 0.92)
1 drink a day is good
4 is bad!
Ronskley et al 2011
BMJ 342: d671
WHAT THIS STUDY ADDS
• Moderate alcohol consumption had favourable effects on
levels of high density lipoprotein cholesterol,
apolipoprotein A1, adiponectin, and fibrinogen
• These results strengthen the case for a causal link
between alcohol intake and reduced risk of coronary
heart disease
Brien et al 2011
BMJ 342: d636
Alcohol and Ischaemic Heart Disease:
Probably no Free Lunch
• Raises HDL-cholesterol
• Aspirin-like thrombolytic effect
• Moderate to heavy drinking  clean
coronary arteries at (early!) PM
Jackson et al 2005 Lancet 366 1911-2
Alcohol and Ischaemic Heart Disease:
Probably no Free Lunch
• No RCTs – uncontrolled confounding
• Cardiovascular associated risk factors significantly more
prevalent in non drinkers than in light to moderate
drinkers
Jackson et al 2005 Lancet 366 1911-2
Wallerath et al 2003 J
Am Coll Cardiol 41:471-78
Effect of red wines on endothelial-type nitric oxide synthase
(eNOS) messenger ribonucleic acid (mRNA) expression in
human endothelial EA.hy
The Importance of Outcome
Measurement
USA = $7,290
Expenditures in
$US PPP
(purchasing power
parity)
Netherlands is
estimated
OECD Health Data,
2009
Australia = $3,357
NZ = $2,454
Davis et al 2010 The
Commonwealth Fund
(Photograph used with permission)
The Most Important Aim?
“Patient-centered – providing care that is
respectful of and responsive to individual patient
preferences, needs and values and ensuring that
patient values guide all clinical decisions”
IOM Crossing the Quality Chasm: A New Health System
for the 21st Century National Academy Press 2001