Lessons Learned through Research: Do Hospitals and Ambulatory Centers Follow Guideline Recommended Care Nancy Albert, PhD, RN, CCNS, CCRN, NE-BC, FAHA,

Lessons Learned through Research: Do Hospitals and Ambulatory Centers Follow Guideline Recommended

Care

Nancy Albert, PhD, RN, CCNS, CCRN, NE-BC, FAHA, FCCM

Senior Director Nursing Research and Innovation;CNS, Kaufman Center for Heart Failure

May 2012

2

Presenter Disclosure InformationNancy M. Albert PhD, CCNS, CHFN, CCEN, FAHA, FCCMLessons Learned through Research: Do Hospitals and AmbulatoryCenters Follow Guideline Recommended Care

2

FINANCIAL DISCLOSURE:No relevant financial relationship exists

National Trends in Readmission Rates after HF Hospitalization

Ross JS, et al. Circ Heart Fail 2010;3:97-103.

RSRR, Risk-standardized 30 D all-cause readmission rate

2004 2005 2006

RS

RR

0.15

0.2

0.25

0.3

All Cause Mortality After Each Subsequent Hospitalization for HF

Setoguchi et al. Am Heart J. 2007;154:260-6.

Time Since Admission

0.0

0.2

Ka

pla

n-M

eie

r C

un

. Mo

rta

lity

0.8

1.0CHF1st Admission (n = 14,374)2nd Admission (n = 3,358)3rd admission (n = 1,123)4th Admission (n = 417)

1st hospitalization: 30 d mortality = 12%; 1 yr = 34%

0.6

0.4

0.0 0.5 1.0 1.5 2.0

Risk-Treatment Mismatch in HF: Canadian EFFECT Study

Use rates in absence of contraindications. For all drug classes, P < .001 for trend.

EFFECT, Enhanced Feedback for Effective Cardiac Treatment. Lee D. JAMA. 2005;294:1240-1247.

At Hospital Discharge 90-Day Follow-Up 1-Year Follow-Up

010

20

30

40

50

60

70

80

90

Low Risk Average Risk High Risk

ACEI ACEI or ARB

-Blocker

1-Year Mortality Rate

Pat

ien

ts, %

ACEI ACEI or ARB

-Blocker

ACADEMIC DETAILING

734 physicians surveyed

• Guidelines helped a moderate (47%) or great amount (23%) in clinical decision making

–21% respondents reported ACE-I are contraindicated if s cr. > 2.0 mg/dL

–27% thought intermittent inotropic therapy is reasonable practice per guidelines

–33.9% underestimated HF prevalence in US

–75.2% underestimated 1-year mortality for Medicare patients

Hauptman PJ, et al. Am J Medicine. 2008;121;127-135.

Timeline from Concept to Clinical Adoption of Neurohormonal Treatments in Heart Failure

1970 1980 1990 2000 2012

ACE Inhibitors

-Blockers

Concept 1st Data Definitive Approval 50% use

Adapted from M Konstam, HFSA 2000.

Aldosterone Antagonists

X

PACE of Improvement

• From 1998-1999 to 2000-2001 performance on the typical CMS inpatient/outpatient measures improved from 69.5% to 73.4%, a 12.8% relative improvement

• At this rate, by 2024, we will have 95% performance on the measures that were current in 2002

ADHERE: Variation in ACEI Use

ADHERE: Dec 2002, 206 Hospitals; 23,193 patients (subset with LVEF < .40)

ORYX Core Measure: HF 3 - LVEF < 40% prescribed ACEI at discharge

Rat

e (%

)0

2040

6080

100

ADHERE Hospitals

ADHERE: Variation in Beta Blocker Use

ADHERE: Dec 2002, 206 Hospitals (Subset with LVEF < 0.40)

ADHERE Hospitals

Rat

e (%

)0

2040

6080

100

Use of Beta blocker at Discharge for Patients with LVEF <= 40%

GWTG-HF- Aldosterone AntagonistsHospitals' use of Aldosterone Antagonist

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126 131 136 141

Hospital ID

Rate

of

Ald

oste

ron

e A

nta

go

nis

t U

se

• 34% of patients received Tx by end of 2007

Albert NM, et al JAMA; 2009;302:1658.

• In 140 hospitals w 10/more patients meeting criteria, use was highly variable: median, 28.3%

Albert NM, et al. JAMA 2009; 302:1658

GWTG-HF Aldosterone Ant. Tx

0%

20%

40%

60%

80%

100%

2005, Q 1-2 2005, Q 3-4 2006, Q 1-2 2006, Q 3-4 2007, Q 1-2 2007, Q 3-4

Use over time, N = 12,565; P = 0.0005

28.0%34.5%

GWTG-HF Aldosterone Antagonist Tx- Trends in Adherence Over Time


Aldosterone Ant. Users (n / N)

Jan-Jun 05

%

Jul-Dec 05

%

Jan-Jun 06

%

Jul-Dec 06

%

Jan-Jun 07

%

Jul-Dec 07

%P

value*

Inappropriate use overall (n= 640; 7.43%)

7.61 8.25 7.05 7.61 7.91 6.31 0.49

Serum K+ >5.5 & <6.0 mEq/L (n=18; 0.21%)

0.18 0.09 0.07 0.44 0.29 0.48 0.09

Serum Cr. ≥2.5 & < 3.0 mg/dL (n= 233; 2.71%)

3.61 3.53 2.57 2.64 3.63 1.45 0.66

EF >40% and without HTN (n= 396; 4.60%)

4.60 5.51 5.16 5.07 4.92 4.88 0.75


GWTG-HF Aldosterone Ant. Tx

50

60

70

80

90

100

DischargeInstructions

LV Function ACE-I/ARB in LVSD Beta-blocker inLVSD

Warfarin in AF

Yes, Aldosterone AntNo, Aldosterone Ant.

%,

Per

form

ance

Mea

sure

Co

nfo

rmit

y

GWTG-HF: Warfarin at Hospital Discharge Among Pts Admitted for HF w Atrial Fibrillation• 72,534 pts from 01/2005 – 03/2008

–255 hospitals

• Results:

–20.5%, atrial fib on admission (n=14,901)

–13.7%, prior Hx atrial fib but SR at adm. (n=9,918)

–Contraindications to warfarin Tx were documented in 9.2%

–Median prevalence of warfarin Tx, 64.9%Piccini JP et al. JACC 2009;54:1280-1289

Warfarin at Hospital Discharge Among Pts Admitted for HF w AF

• Median prevalence of warfarin Tx, 64.9%• Trend over time, P = 0.146

Piccini JP et al. JACC 2009;54:1280-1289

0%

20%

40%

60%

80%

100%

2005 2006 2007 2008, 1st QTR

Warfarin Use Over Time


GWTG-HF: Warfarin at Hospital Discharge Among Pts Admitted for HF w Atrial Fib

Site1 8 15

War

fari

n D

isch

arg

e (%

)

0

20

40

60

80

100

22 29 36 43 50 57 64 71 78 85 92 99 106 113 120 127 134 141 148 155 162 169 176

GWTG-HF: Warfarin at Hospital Discharge Among Pts Admitted for HF w Atrial Fib


CHADS2 score: CHF, HTN, age > 75, DM, prior stroke/TIA

War

fari

n D

isch

arg

e (%

)

01

20

40

60

80

100

2 3 4 5 6Chads2 Score

70.966.5 63.6 64.3 65.8

59.5

P<.0001 for trend

Are hospitals delivering

optimal evidence-based

recommendations?

Hernandez, A. F. et al. JAMA 2007;298:1525-1532.

GWTG-HF: Race and Gender Disparities in ICD Use at Discharge Among Eligible Patients With HF

Black female

White female

Black male

0 0.4 0.80.2 0.6 1.0

43.633.4 29.8 28.2

0102030405060708090

100

White Male

Black Male

WhiteFemale

BlackFemale

N= 13,034 pts

GWTG-HF: Hospital Variation and Characteristics of ICD Use

• January 2005 – June 2007

• New or Discharge prescription for ICD Tx in patients with EF ≤ 30% without documented contraindication

–54,750 pts from 234 hospitals

–Of 12,693 pts, 2545 had prior ICD (20% use)

–Of 10,148 (134 hospitals):

–Overall Use/Planned implementation = 20%

Shah B, et al. JACC 2009;53:416-22.

Hospital Variation in ICD Use

Patient Factors

High Use

N=48

Med. Use

N=42

Low Use

N=44

P Value

Female 34% 38% 36% <0.03

Race-AA 22% 33% 32% <0.001

High Chol 38% 32% 30% <0.001

Hx MI 15% 9% 11% <0.001

Hx HTN 63% 66% 64% 0.004

Shah B, et al. JACC 2009;53:416-22.

Are disparities in care present

by patient features?

New or Planned ICD Use-GWTG-HF

Shah B, et al. JACC 2009;53:416-22.

GWTG - HF Hospital Site

1 11 21 31 41 51 61 71 81 91 101 111 121 131

Ho

spit

al I

CD

rat

e (%

)

0

20

40

60

80

100

Hospital ICD Use Associations with Hospital Characteristics

Shah B, et al. JACC 2009;53:416-22.

Aca

dem

ic

Adj

uste

d IC

D r

ate

(%)

0

20

40

60

80

100N

on-a

cade

mic

Hea

rt tra

nspl

ants

No

hear

t tra

nspl

ants

PC

I cap

able

No

PC

IC

AB

G c

apab

leN

o C

AB

GN

orth

east

Mid

wes

tS

outh

Wes

tB

eds

<100

Bed

s 10

0-19

9B

eds

200-

299

Bed

s 30

0-39

9B

eds

400-

499

>50

0 B

eds

Are disparities in care present

by hospital features?

Median, 49.1Mean, 50.7

IMPROVE-HF: Variation in OPD HF Care

Fonarow GC, et al. Circ Heart Fail. 2008;1:98–106.

ADHERE-HF: Rates of Conformity by Practice Setting and Differences in Hospital Level Outcomes; > 80,000 Hosp. Admissions

Fonarow GC, et al. Arch Intern Med 2005; 165:1469–1477










Are cardiologist practices

delivering optimal

evidence-based

recommendations?

IMPROVE-HF: Improvement in Quality Measures at 24 Months (Pt. Level Analysis)

80%

86%

34%

69%

38%

49%

62%

84%

93%

51%

69%

58%

71%69%

87%

94%

62%

69% 69%

79%

71%

0%

20%

40%

60%

80%

100%

Baseline 12 months 24 months

ACEI/ARB ß-blocker Aldosterone Antagonist

Anticoagulant for AF

CRT ICD HF Education

*

*

*

*

*

**

*

*

*

**

Eli

gib

le P

atie

nts

Tre

ate

dE

lig

ible

Pat

ien

ts T

rea

ted

Fonarow GC, et al. Circulation. 2010;122:585-96.

P-values are for relative change; *, P <0.001 vs. baselineBaseline N= 15,177; 24 Mo. N= 11,621; 167 practices

IMPROVE-HF: Baseline Measure Conformity: Alive vs. Dead at 24-Months

The baseline process measure conformity was significantly lower among patients who died compared with those who survived for 5 of 7 individual measures.

The baseline process measure conformity was significantly lower among patients who died compared with those who survived for 5 of 7 individual measures.

Fonarow GC, et al. Circulation. 2011;123(15):1601-1610.

IMPROVE-HF Nested Case-Control Analysis: Baseline Use of Guideline Recommended Therapies in Cases (N=1376; Dead) and Controls (N= 2752; Alive) at 24-months (matched at 1:2 ratio)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

ACEi/ARB

B-Block

ers

Aldo. Antag

onist

Anticoag fo

r AF

ICD/CRT-D

CRT-P/C

RT-D

HF Educa

tion

CasesControls

P<0.0001

P=0.4046

P=0.0094

P<0.0001

P=0.0019

P<0.0001

P<0.0001

Fonarow GC, et al. J Am Heart Assoc 2012;1:16-26.

IMPROVE-HF Nested Case-Control Analysis: Mortality Reduction Based on Number of Guideline-Recommended Therapies at Baseline

24 Month MortalityAdjusted Odds Ratios (95% CI)

24 Month MortalityAdjusted Odds Ratios (95% CI)

Number of Therapies (vs. 0 or 1 therapy)

2 therapies

3 therapies

4 therapies

5, 6, or 7 therapies

Odds Ratio(95% confidence interval)

0.63 (0.47-0.85)(P =0.0026)

0.38 (0.29-0.51)(P <0.0001)

0.30 (0.23-0.41)(P <0.0001)

0.31 (0.23-0.42)(P <0.0001)

0 0.5 1 1.5 2


IMPROVE-HF: Incremental Benefits with HF Therapies(Cumulative % Reduction in Odds of Death at 24 Months)

-39%

-63%

-76%-81% -83% -81%

-100%

-90%

-80%

-70%

-60%

-50%

-40%

-30%

-20%

-10%

0%Beta Blocker

Beta Blocker +ACEI/ARB

Beta Blocker +ACEI/ARB + ICD

Beta Blocker +ACEI/ARB + ICD + HF

Education

Beta Blocker +ACEI/ARB + ICD + HF

Education +anticoagulants for AF

Beta Blocker +ACEIARB + ICD + HF

Education +Anticoagulant for AF

+ CRT

-28% to -49%P<0.0001

-54% to -71%P<0.0001

-68% to -81%P<0.0001

-75% to -86%P<0.0001

-77% to -88%P<0.0001

-72% to -87%P<0.0001


IMPROVE-HF: Incremental Benefit with HF Therapies(Cumulative % Reduction in Odds of Death at 24 Months Associated with Sequential Treatments)

-38%

-77%-90%

-100%

-90%

-80%

-70%

-60%

-50%

-40%

-30%

-20%

-10%

0%ACEI/ARB ACEI/ARB + BB ACEI/ARB + BB + CRT + ICD

+20% to -68%P=0.1566

-43% to -91%P<0.0001

-70% to -96%P<0.0001


Hospitalization for HF : CV / Medical Conditions

Klapholz, et al. JACC 2004;43:1423-8.

Reasons for Clinical Decompensation were Identified in only ½ of Patients

Syst. BP >200 mmHg

Non-Compliance

MR / AR >3+

ACS

Renal Insuff.

Afib / Flutter / SVT

Sev. COPD / Asthma

Pneumonia

AS / MS <1.0 cm2

Sepsis

Patients (%)0 3 6 9 12 15

Did we assess patient

knowledge, psychosocial,

behavioral, economic

needs/issues?

Primary endpoint was:Mortality or Readmission for heart failure

COACH Study: Self-Care Compliance

COACH study: Self-Care ComplianceCOACH study: Self-Care Compliance

Van der Wal MH, et al. Eur Heart J 2010;31:1486-1493Van der Wal MH, et al. Eur Heart J 2010;31:1486-1493

Compliance = scoring “mostly” or “always” in following 3 recommendations: 1) Sodium-restricted diet2) Fluid restriction3) Exercise

Compliance = daily weight monitoring three times per week to daily

Compliance measured 1 month after hospital discharge & followed for 18 months

48% (N=830 patients)

Cu

m s

urv

ival

Time to primary endpoint

0 100 200 300 400 500 6000.0

0.2

0.4

0.6

0.8

1.0

Overall non-compliantOverall compliantHR 1.40 (1.08-182); P=0.01

Van der Wal MH, et al. Eur Heart J 2010;31:1486-1493

Compliance with advice on weight monitoring, low sodium diet, fluid restriction and exercise

*, P< 0.01; **P<0.05


0

20

40

60

% p

rim

ary

end

po

int

Total compliance score (0-4)

4 3 2 1 or 0

25%

38% 36%

45%*

**

*

Non-compliance adjusted HR (95% CI)Primary endpoint: 1.48 (1.15-1.91) P=0.002Time to:-- HF readmission: 1.55 (1.13-2.13) P=0.007-- Death: 1.24 (0.89-1.72) P=0.20

Non-compliance adjusted HR (95% CI)Primary endpoint: 1.20 (0.87-1.65) P=0.27Time to:-- HF readmission: .93 (0.60-1.44) P=0.76-- Death: 1.58 (1.08-2.27) P=0.02


COACH study: Self-Care ComplianceCOACH study: Self-Care ComplianceC

um

su

rviv

al

Time to primary endpoint0 100 200 300 400 500 600

0.0

0.2

0.4

0.6

0.8

1.0

Non-compliant exerciseCompliant exercise


Non-compliant weighingCompliant weighing

Compliant with exercise Compliant with daily weighing

0.0

0.2

0.4

0.6

0.8

1.0


Conclusion: Advice and compliance in all 4 self-care behaviors are important


Non-compliant exerciseCompliant exercise


Non-compliant fluidCompliant fluid

Compliant with diet

Cu

m s

urv

ival

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

Compliant with fluid restriction


Do we assess patient’s

adherence to self-care and

learn reasons for

non-adherence?

Performance Measures & Quality

• Based on nationally established guidelines• Include outcomes and processes of care

known to positively influence overall outcomes of care–Institution and system level

–Hospital–Healthcare practitioners/care providers

–Incorporate risk-adjustment methodology to account for significant differences in patient populations among institutions

Performance Measures & Quality• Often, we question the reliability of

performance measures in specific settings–Documentation of patient education

–Checking a box does not reflect delivery of quality care

1. Koelling TM, et al. Circulation 2005;111:179-185.2. Krumholz HM et al. JACC 2002;39:83-89.

• 1:1 RN-delivered comprehensive HF education during a 1-hour hospital session decreased 6-month days hospitalized or days dead1

• RN delivered, 1 hour, comprehensive hospital education + ongoing 1 year telephone-based support decreased 1-year hospital readmission rate and hospital costs2

Performance Measures & Quality• Public release of performance data in changing

the behavior of healthcare consumers, professionals or organizations –> 35,000 consumers & 1560 hospitals

• Conclusions: –“The small body of evidence available

provides no consistent evidence that the public release of performance data changes consumer behavior or improves care.

–Evidence that the public release of performance data may have an impact on the behavior of healthcare professionals or organizations is lacking.”

Ketelaar NA, et al. Cochrane Database Syst Rev. 2011;11:CD004538.

PROGRAMS Developed to Improve Performance Measures• Usually designed to enhance quality of patient care

that promotes adoption of evidence-based, guideline-recommended therapies

–Force a deeper look into actions and practices

–Requires:

–Leadership

–Evidence-based clinical decision support (algorithms; pocket cards, customizable order sets...)

–Patient education resources

–Regular review of data; benchmarking

–Process improvement model of change

Lessons Learned through Research: Do Hospitals and Ambulatory Centers Follow Guideline Recommended Care Nancy Albert, PhD, RN, CCNS, CCRN, NE-BC, FAHA,

Documents

hf setoguchi

underestimated hf prevalence

opd hf care fonarow

hf w atrial fibrillation

hf w atrial fib site

users n n

hf w atrial fib piccini

hf hospitalization ross