Download - Value in Health

VALUE IN HEALTH 14 (2011) S100 –S107
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/jval
Cost-Effectiveness of Supervised Exercise Therapy in Heart Failure
Patients
Eduardo M. Kühr, MD, MSc1, Rodrigo A. Ribeiro, MD, MSc2, Luis Eduardo P. Rohde, MD, ScD1,
Carisi A. Polanczyk, MD, ScD1,2,*
1
Cardiology Division, Hospital de Clínicas de Porto Alegre, Department of Medicine and Graduate Program of Cardiology, Federal University of Rio Grande do
Sul, Porto Alegre, Brazil, and National Institute for Health Technology Assessment, CNPq, Brazil; 2Graduate Program in Epidemiology, Federal University of
Rio Grande do Sul, Porto Alegre, Brazil, and National Institute for Health Technology Assessment, CNPq, Brazil
A B S T R A C T
Objective: Exercise therapy in heart failure (HF) patients is considered
safe and has demonstrated modest reduction in hospitalization rates
and death in recent trials. Previous cost-effectiveness analysis described favorable results considering long-term supervised exercise intervention and significant effectiveness of exercise therapy; however,
these evidences are now no longer supported. To evaluate the costeffectiveness of supervised exercise therapy in HF patients under the
perspective of the Brazilian Public Healthcare System. Methods: We
developed a Markov model to evaluate the incremental cost-effectiveness ratio of supervised exercise therapy compared to standard treatment in patients with New York Heart Association HF class II and III.
Effectiveness was evaluated in quality-adjusted life years in a 10-year
time horizon. We searched PUBMED for published clinical trials to estimate effectiveness, mortality, hospitalization, and utilities data.
Treatment costs were obtained from published cohort updated to 2008
values. Exercise therapy intervention costs were obtained from a reha-
Introduction
Heart failure (HF) is a common health care problem worldwide,
with elevated costs associated to its treatment [1]. During the past
20 years several effective therapies have changed HF management
and clinical outcomes and these have been formally evaluated
through economic analyses [2-4]. The decrease in HF mortality
was followed by an increase in its prevalence, with direct effect on
health care budgets resulting from the rising number of hospitalizations and therapeutic procedures [5].
HF is a complex syndrome characterized by reduced exercise
tolerance and the involvement of multiple physiopathologic
mechanisms [6]. In the past patients were often advised to limit
their efforts in daily activities; however, several studies suggest
that exercise training may reduce mortality and morbidity in HF
patients [7,8]. These studies also demonstrated that exercise training could be performed safely in appropriately evaluated cases of
patients who present in clinically compensated New York Heart
bilitation center. Model robustness was assessed through Monte Carlo
simulation and sensitivity analysis. Cost were expressed as international dollars, applying the purchasing-power-parity conversion rate.
Results: Exercise therapy showed small reduction in hospitalization
and mortality at a low cost, an incremental cost-effectiveness ratio of
Int$26,462/quality-adjusted life year. Results were more sensitive to
exercise therapy costs, standard treatment total costs, exercise therapy
effectiveness, and medications costs. Considering a willingness-to-pay
of Int$27,500, 55% of the trials fell below this value in the Monte Carlo
simulation. Conclusions: In a Brazilian scenario, exercise therapy
shows reasonable cost-effectiveness ratio, despite current evidence of
limited benefit of this intervention.
Keywords: costs, health economics, heart failure, physical therapy.
Copyright © 2011, International Society for Pharmacoeconomics and
Outcomes Research (ISPOR). Published by Elsevier Inc.
Association (NYHA) functional class II and III, as endorsed by current guidelines [9,10].
For health care managers, the decision to incorporate exercise
therapy in treatment of patients with HF should be based in several perspectives, including cost-effectiveness studies of the intervention. In 2001, Georgiou et al. [11] published a cost-effectiveness
analysis of supervised exercise intervention in HF patients showing a very favorable cost-effectiveness ratio of $1773 per life-year
saved, considering a 14-month period of supervised exercise intervention in a time horizon of 10 years applied to a North-American setting.
Recently a multicenter randomized controlled trial of 2331 HF
outpatients [12] described an exercise-based intervention being
compared with standard treatment. After 2.5 years of follow-up,
including a short training period in a facility followed by homebased exercise sessions, a benefit was observed only after adjustment for other prognostic predictors of the primary endpoint. The
authors concluded that exercise training is a safe intervention associated with a modest reduction in hospitalization and mortality,
Conflicts of interest: The authors have indicated that they have no conflicts of interest with regard to the content of this
article.
* Address correspondence to: C. A. Polanczyk, National Institute for Health Technology Assessment, Hospital de Clínicas de Porto Alegre,
Ramiro Barcelos 2350, Building 21, 5th Floor, 90035-007 - Porto Alegre, RS, Brazil.
E-mail: [email protected].
1098-3015/$36.00 – see front matter Copyright © 2011, International Society for Pharmacoeconomics and Outcomes Research (ISPOR).
Published by Elsevier Inc.
doi:10.1016/j.jval.2011.05.006
VALUE IN HEALTH 14 (2011) S100 –S107
S101
Fig. 1 – Schematic representation of the decision model.
far from the assumed estimations in previous cost-effectiveness
analysis [12].
In this study we evaluated the economic impact of a supervised
exercise intervention in a hypothetic stable outpatient HF cohort,
considering current evidence of effectiveness and costs, offering
health care professionals an updated assessment on the role of
exercise in the management of HF.
computed all-cause mortality in our model, considering evidence of exercise intervention studies. A schematic representation of our decision tree is shown in Figure 1.
The discount rate for both cost and effectiveness was 5% per
year. We used the public third-party payer perspective and a 10year time horizon.
Survival data
Methods
Target population
The target population was composed of 60-year old patients at
baseline, with clinically stable NYHA class II or III HF, intended to
reproduce the population in exercise interventions studies in HF.
Decision model structure
We developed a model based on two competing strategies: 1)
standard HF care; and 2) standard HF care plus an exercisebased intervention [13]. We constructed our decision tree model
with Markov transitional states using TreeAge Pro 2009 Suite
software (release 1.0.1, TreeAge Software Inc., Williamstown,
MA), tracking a hypothetical cohort of HF patients over time
receiving one of the strategies. During each 1-year cycle, patients could remain alive or die; patients alive could also remain
stable or be hospitalized. After having been hospitalized, these
patients could die or remain alive, with a lower survival rate,
simulating the natural history of HF. In the intervention arm we
assumed that exercise could reduce mortality and hospitalization rates, according to expected rates of effectiveness. We
Survival rates were based in data from a specialized HF outpatient clinic from a university hospital in Brazil whose patients’
characteristics are similar to the exercise intervention studies’
populations. This cohort was composed of 318 patients (68%
men), with a median age of 61 years (interquartile range 50 –71
years). Thirty-seven percent of patients had ischemic heart disease as the HF etiology; 87% were currently taking beta blockers,
and 91% were taking angiotensin converting enzyme inhibitors.
The prevalence of diabetes mellitus in this cohort was 30%, 41%
of these patients had hypertension, and 11% were tobacco users. The annual rate of hospital admission in this cohort was
16%, and patients who had been hospitalized had a diminished
survival rate compared with those who had not been hospitalized [14].
Median follow-up of this cohort was 75 months (95% confidence interval [CI] 68 – 81). To project survival during the 10-year
time horizon, we built a survival curve (Fig. 2) using a Weilbull
function. Two different curves were built based on hospitalization
status. The final equations for the survival functions were
Exp(⫺((0.0004*(_stage)^1.0715)) in nonhospitalized patients and
Exp(⫺((0.00018*(_stage)^1.3627)) in hospitalized patients.
S102
VALUE IN HEALTH 14 (2011) S100 –S107
were included, as described in the Consort statement (Fig. 3).
Thirteen trials met the eligibility criteria, resulting in a total of
3458 patients being included: 1746 in the exercise group and
1712 in the control group, as shown in Table 1 [7,12,16 –26].
When one arm of a study contained no events, we added 0.5 to
each cell to allow effect size calculation. Combined events
(death or hospitalization) showed a pooled risk ratio of 0.878
(95% CI 0.805– 0.957) using the Mantel Haenszel fixed-effects
model. Pooled risk reduction for mortality was 0.957 (95% CI
0.865–1.058) and for hospitalization was 0.90 (95% CI 0.831–
0.973).
HF treatment costs
Fig. 2 – Undiscounted survival curve projections for
standard treatment and exercise therapy plus standard
treatment, compared with actual survival curve of heart
failure (HF) patients.
Utilities
To estimate quality-adjusted life years (QALYs), we assumed a
utility index of 0.80 for HF (95% CI 0.78 – 0.82), using recommended
values proposed by Göhler et al. [15] in their decision-analytic
model, which was weighted by NYHA classes from actual prevalence on the Hospital de Clinicas de Porto Alegre cohort (66% and
33% of NYHA class II and III, respectively). We used a discount rate
of 5% (range 3%–7%).
Exercise therapy intervention
The model assumed a supervised exercise therapy session occurring in a facility center, with direct supervision of a qualified
exercise training professional (eg, a physiotherapist). These sessions consist of an initial warm-up, followed by 15 to 30 minutes
of aerobic training on a treadmill or performing stationary cycling in individually prescribed program. All patients should be
monitored during the exercise session and each physiotherapist could supervise up to four patients each session. In initial
phases of rehabilitation (until 12 weeks), every patient in the
intervention group should participate in at least three weekly
1-hour supervised group-based exercise sessions.
After 36 sessions we determined that patients should return
once a week to the rehabilitation center for additional supervised
sessions during a 9-month period, totaling 72 sessions per year in
the first intervention stage. In subsequent stages we considered a
maintenance program of 50 yearly sessions (weekly supervised
sessions).
Annual HF treatment costs in the Public Healthcare System
(PHS) were extracted from a Brazilian cohort described by
Araujo et al. [27]. These authors described treatment costs categorized in outpatients and hospital costs, in 2002 values. We
updated published values to 2008 costs according to the official
Brazilian inflation index [28]. Hospital costs were adapted from
our university hospital billing system, which includes in standard admission payments some high-complexity diagnostic
procedures, including special drugs and devices. For outpatient
cost estimates for stable conditions we included drug costs,
complementary exams, and ambulatory costs (Table 2). All
costs are expressed in international dollars (Int$), using the purchasing power parity conversion rate. According to a 2008 report of the World Bank regarding conversion rates [29], Int$1 ⫽
R$1.357. We used an annual discount rate of 5% in all described
costs, expressed in internation dollars.
Intervention costs (Supervised Exercise)
In Brazil outpatient cardiac rehabilitation procedures are not
reimbursed by the PHS. To establish yearly intervention costs
we used data from a rehabilitation center in a private facility of
an insurance plan, located in the Brazilian state of Santa Catarina.
Salaries of fitness center personnel accounted for 47% of the
total annual cost of rehabilitation. Maintenance costs (including
rental costs, licenses, and repairs) and equipment costs (e.g.,
cycle ergometers and pulse oximeters) accounted for 33% and
20%, respectively. This structure is able to provide 2000 annual
Intervention effectiveness
We performed a literature search in PUBMED using “Heart Failure” and “Exercise Therapy” MeSH terms, selecting randomized
controlled trials published from 1980 to 2009. We included trials
with the following characteristics: 1) randomized parallel group
controlled trials; 2) exercise programs during at least 8 weeks or
more initially based in facility centers; and 3) clinically relevant
endpoints (death and/or hospitalization) described in the results. Initial search identified 169 articles that were subsequently scrutinized and those that fulfilled inclusion criteria
Fig. 3 – Flow diagram of study selection for exercise
effectiveness.
S103
VALUE IN HEALTH 14 (2011) S100 –S107
Table 1 – Studies meeting inclusion criteria to estimate exercise effectiveness.
Author, year (ref)
Austin, 2005 [16]
Belardinelli, 1999 [7]
Gianuzzi, 2003 [17]
Jolly, 2009 [18]
Jonsdottir, 2006 [19]
Kiilavuori, 2000 [20]
Kulcu, 2007 [21]
Hambrecht, 2000 [22]
McKelvie, 2002 [23]
Nilsson, 2008 [24]
O’ Connor, 2009 [13]
Wielenga, 2000 [25]
Willenheimer, 1998 [26]
Total
Sample size
(exercise controls)
Events (exercise, controls)
Combined
Deaths
Hospitalization
13
23
14
34
2
2
23
25
9
13
0
1
0
2
5
4
11
10
2
1
918
958
2
4
0
3
999
1082
5
4
9
20
0
1
7
5
2
2
0
0
0
1
3
2
11
10
2
1
189
198
1
4
0
0
229
248
9
19
5
14
2
1
16
20
7
11
0
1
0
2
2
2
0
0
0
0
729
760
1
1
0
3
771
834
Exercise: 86
Controls: 98
Exercise: 50
Controls: 49
Exercise: 45
Controls: 45
Exercise: 84
Controls: 85
Exercise: 21
Controls: 22
Exercise: 12
Controls: 15
Exercise: 27
Controls: 26
Exercise: 35
Controls: 34
Exercise: 90
Controls: 91
Exercise: 40
Controls: 40
Exercise: 1159
Controls: 1172
Exercise: 41
Controls: 39
Exercise: 22
Controls: 30
Exercise: 1712
Controls: 1746
training hours to four patient groups, with a total cost of
Int$66,633, resulting in a single session cost of Int$8.33 per patient. The proposed cost of each session is similar to the estimated costs per session used by a previously published model
[11], which would cost Int$7.14 in the Brazilian scenario. Values
used in the model are lower than reimbursement values in Brazil (Int$17.65 in the first 12-week program and Int$5.88 in the
maintenance program).
Table 2 – Model variables, values, and sources.
Variable
Median age of cohort (y)
Annual risk of hospitalization (%)
Exercise variables
Annual hospitalization risk reduction (%)
Annual mortality risk reduction (%)
Costs
Annual exercise intervention costs in the first year (Int$)
Base
60
16
0.90
0.957
Sensitivity
analysis variation
50–70
10–22
0.831–0.973
0.865–1.058
375
187–562
Annual exercise intervention cost in 2- to 10-y interval (Int$)
100
50–150
Total annual conventional heart failure treatment costs (Int$)
Annual hospitalization costs (Int$)
Annual ambulatory costs (Int$)
Annual complementary exams costs (Int$)
Annual medication costs (Int$)
Total annual exercise-group heart failure treatment costs
(conventional heart failure treatment cost plus exercise
intervention costs) (Int$)
Utilities
Heart failure patient utility (New York Heart Association
Class II-III)
Discount rate (%)
3752
1798
54
363
1538
4187
1,876–5,628
899–2,697
27–81
182–545
769–2,306
2,093–6,280
0.80
5
0.78–0.82
3–7
Source
15
15
Meta-analysis
Meta-analysis
Health care insurance
plan, primary data
Health care insurance
plan, Primary data
17
17
17
17
17
Estimated
16
Estimated
S104
VALUE IN HEALTH 14 (2011) S100 –S107
Table 3 – Model predicted cost, effectiveness, and cost-effectiveness of competing strategies.
Total cost (Int$)
Conventional treatment
Conventional treatment ⫹ Exercise therapy
12,720
15,331
Effectiveness
Incremental
cost-effectiveness
Mean life years
Mean QALYs
Int$/LYS
Int$/QALY
5.45*
5.58*
4.36*
4.46*
–
21,169
–
26,461
Int$, international dollars; LYS, life year saved; QALY, quality-adjusted life year.
* All values are discounted.
Sensitivity analysis
We performed one-way sensitivity analysis in all model parameters described in Table 2. Model effectiveness variables (mortality
and hospitalization reduction with exercise) were varied between
the boundaries of meta-analysis confidence intervals. Costs were
varied ⫾50% of their original values; utilities varied according to
previously described values. Discount varied between 3% and 7%.
Two-way sensitivity analysis was performed in the most important model variables.
Model robustness was tested in a Monte Carlo simulation, with
a generation of 1000 trials and variation in the range described
above. Gamma distribution was chosen for cost variables, lognormal distribution for effectiveness, and beta distribution for utility
variables.
Results
The model predicted a mean survival of 5.58 years in the exercise
group and 5.45 years in the control group. When adjusted for quality of life, we found a mean survival of 4.46 and 4.36 QALYs, respectively, as shown in Table 3. The exercise intervention increased 0.13 life-years and 0.10 QALYs. The total costs in the
intervention group was Int$15,331 and Int$12,720 in the standard
care group. The incremental cost-effectiveness ratio (ICER) was
Int$21,169 per life-year and Int$26,462 per QALY. Monte Carlo simulation with 1000 trials is represented in Figure 4.
Results from one-way sensitivity analyses are summarized
in Table 4, and our evaluation was adequate to most one-way
analyses. At the lowest values of established intervals for variables related to HF standard costs of treatment—such as ambulatory costs, complementary exam costs, hospitalization costs,
and drugs costs—there was small influence on the ICERs. Utili-
ties and discount rates variations also produce discrete modifications on the ICERs, varying between Int$25,816 and Int$27,140
and Int$25,155 and Int$27,832, respectively. More pronounced
effect occurred when varying probability of hospitalization,
with an ICER of Int$24,499 at a 22% annual rate and of Int$30,350
at a 10% annual hospitalization rate.
Three variables had a more expressive effect on the base
case estimates: exercise intervention costs, estimated mortality
reduction, and estimated hospitalization reduction with rehabilitation program. When the relative risk (RR) for hospitalization with exercise was 0.83, the ICER would be Int$20,856; with a
RR of 0.97, the ICER increased to Int$36,245. Regarding exercise
costs, a cost of Int$217 per year of supervised exercise produces
an ICER of Int$13,501, and a cost of Int$899 per year increased
the ICER to Int$52,056 (Table 4).
The variable with greater impact on the ICER in the sensitivity
analysis was mortality reduction with exercise: a RR of 0.86 increased 0.24 in total QALY, resulting in an ICER of Int$12,738/
QALY; a RR of 0.96 yielded 0.09 additional QALY and an ICER of
Int$28,141/QALY. Finally, considering no effect on mortality, exercise intervention results in additional 0.04 QALY and an ICER of
Int$66,576/QALY.
Two-way sensitivity analysis altering hospitalization reduction effect and exercise cost is illustrated in Figure 5. Assuming a
mortality RR of 0.96 with exercise and a willingness-to-pay of
Int$27,495, the exercise intervention should cost below Int$440
yearly to become cost-effective, assuming a RR of 0.90 in hospitalization with exercise.
In the Monte Carlo simulation, we established a threshold of
three times the Brazilian gross domestic product, which represents Int$27,495 (R$37,311) in 2008 and evaluated how many simulations fell below this value. As shown in Figure 6, 55% of trials
were below this value. In 34% of trials exercise was more costly
Table 4 – One-way sensitivity analysis.
Variables
Mortality reduction with exercise
Hospitalization reduction with
exercise
Exercise intervention costs
Annual rate of hospitalization
Utility of heart failure
Discount rate
Ambulatory costs
Hospitalization costs
Complementary exam costs
Medication costs
Fig. 4 – Monte Carlo 1000 trials scatter plot. The number of
points below wiliness-to-pay threshold line are 55.2%
(Int$27,500/quality-adjusted life year—three times Brazil’s
gross domestic product per capita).
Lower ICER
(Int$/QALY)
Higher ICER
(Int$/QALY)
12,738
20,856
Dominated
36,245
13,501
24,499
25,816
25,155
26,430
25,945
26,251
25,571
52,056
30,350
27,140
27,832
26,493
26,977
26,672
27,353
Note: Range values are those presented in Table 2.
ICER, international cost-effectiveness ratio; Int$, international dollars; QALY, quality-adjusted life year.
VALUE IN HEALTH 14 (2011) S100 –S107
Fig. 5 – Three-way sensitivity analysis considering
exercise costs (Int$) and hospitalization risk reduction,
assuming mortality risk reduction with exercise = 0.96 and
a willingness-to-pay of Int$ 27,500.
and above the wiliness-to-pay and in 10.8% the exercise intervention was inferior (dominated).
Discussion
The results of this model show that exercise therapy in HF patients has a modest but favorable incremental cost-effectiveness ratio of Int$26,462 per QALY and Int$21,169 per life-year in
a Brazilian PHS scenario. The results were consistent considering sensitivity analyses performed and assumptions described.
Our results show that this intervention has a reasonable costeffectiveness ratio when compared to other incoming therapies,
such as implantable cardioverter defibrillator devices [30], but
closer to proposed willingness-to-pay for Brazil, as demonstrated in the acceptability curve. The ICER of this intervention
is also higher than the estimated Brazilian hemodialysis cost
per life-year gained (US$10,065) [31].
During the past three decades the exercise training approach
concerning HF patients moved from absolute restraint to enthusiastically prescribed, based on a growing body of evidence suggesting physiologic and clinical benefit from exercise. Randomized controlled trials showed significant reductions in the
composite endpoint of hospitalization and mortality in HF patients submitted to supervised sessions of exercise interventions, reaching almost 50% in one report [7], and averaging a
35% reduction in mortality and 28% in hospitalization rates in a
meta-analysis published in 2004 [8]. Nonetheless, a recently
published trial failed to prove superiority of an exercise therapy
intervention in a multicenter-based strategy [12]. The HF-Action trial [12] was designed to test the hypothesis that exercise
training prescribed to HF patients would reduce mortality, and
although overall intention-to-treat results were of small magnitude, protocol-driven results and those adjusted for prognostic variables indeed suggest some benefit on mortality beyond
reduction of hospitalization and improvement on quality of life,
functional capacity, and other markers of well-being [32,33].
The true effect of exercise on mortality in this population is
a matter of debate, although exercise is still considered a key
aspect in the management of HF [9,34]. Our model was sensitive
to this parameter; assuming even a small effect of 4% reduction
in mortality, modest reduction of combined risk of death or
HF-related hospitalization, exercise would still have a favorable
cost-effectiveness ratio. Unfortunately we do not have large
Brazilian studies describing the actual result of such interven-
S105
tions. There are several short-term randomized studies evaluating the effect of exercise among Brazilian HF patients, mostly
limited to physiologic parameters, functional class, or quality of
life [35,36]. We could not identify studies that have evaluated
hard clinical endpoints such as death or hospitalization. Indirect evidence from these studies; however, support the concept
that the results obtained locally are similar to the ones described by other international research groups [35,36]. Based on
these assumptions we believe our findings could be a reasonable estimate of the cost-effectiveness ratio for the Brazilian
PHS scenario, although the effect of the health care system itself
and cultural and socioeconomic characteristics on the main result might be unpredictable.
In another published cost-effectiveness analysis [11], the impressive absolute reduction of 19% fewer hospitalizations in the
exercise group compared to standard treatment group and additional survival of 1.82 years was the determinant factor in
effectiveness estimation, with a cost-effectiveness ratio of
$1773 per life-year saved. In our model, we considered a less
pronounced increase in survival (0.13 years) produced by exercise intervention, reflecting a more realistic ICER taking into
account current data. Prospective economic evaluation from
HF-Action [33] demonstrated that along the trial the exercise
training program was of little systematic benefit in terms of
overall resource use, but individual data indicate that most estimates were consistent with a decrease in costs (89.9%) and an
increase in QALYs (76.5%), and that most of the bootstrap replications were either associated with cost-saving result or with
ICER below $50,000 per QALY.
Usual exercise intervention in HF requires at least 12 weeks
of training, because during a period of this length patients can
receive adequate care and perceive functional improvement,
increasing adherence to the intervention. In our model, we determined that the 36 sessions should be supplemented with
additional weekly supervised sessions to reinforce compliance
to the intervention in an attempt to reach the effectiveness
demonstrated in clinical trials. Perhaps these additional costs
could be counterbalanced with increased effectiveness, but we
opted to consider a conservative strategy in light of current
evidence. Assuming that including different professionals and
activities in every program is proportional to increasing costs of
the intervention, reducing the number of sessions in a facility
center is a valuable attempt to reduce treatment costs and increase cost-effectiveness ratio.
In the real world exercise therapy is not considered the core
of the standard care for HF patients. Commonly HF patients are
Fig. 6 – Acceptability curve with a willingness-to-pay of
Int$27,500/quality-adjusted life year.
S106
VALUE IN HEALTH 14 (2011) S100 –S107
aged and have other diseases (e.g., osteoarticular limitations)
that restrict their mobility; plus, successful exercise therapy
demands motivation and both family and patient engagement
time. In today’s health care system model, which is based in
hospital procedures, there are few examples of rehabilitation
centers to promote significant changes in patients’ habits and
minimize usual risk factors associated with cardiovascular diseases, such as lack of exercise, tobacco use, inappropriate food
ingestion, and depression. Despite all these points, there is evidence—including that produced in our study—that committed
patients with cardiac disease who have access to facilities with
trained professionals and appropriate equipment could benefit
from exercise therapy with reasonable cost-effectiveness ratio
[37]. Common to other interventions that also rely on human
behavior, long-term adherence to exercise in patients with HF
remains a challenge and requires additional research to determine strategies aimed at improving compliance, which might
be associated with increased effectiveness. Areas of needed research include identifying the subgroups of patients who benefit the most from this intervention, as well as the determining
the optimal intensity, duration, and frequency of exercise
needed to maximize clinical benefits.
Some limitations in our modeling should be mentioned.
First, we established a constant effectiveness. In the real world
a considerable number of patients oscillate their attendance in
rehabilitation programs. Another limitation is that true cost of
HF treatment could be higher than assumed costs, although we
used primary data from a cohort of HF patients, with values
updated to 2008. As we assumed a third-party payer (i.e., government) perspective, we did not considered patient or family
displacement cost in regard to the rehabilitation facility; indirect cost associated with productivity losses were also not included, although HF in Brazil is a retirement cause insured by
the PHS. The main caveat regarding the results of our study is
the broad variance in exercise effectiveness, even though in
two-way sensitivity analysis we found that every scenario with
effectiveness greater than 5% (considering mortality reduction)
the exercise intervention brought benefits to HF patients.
Exercise therapy seems to be safe and should be considered
in stable HF patients as part of treatment regardless of individual beliefs. Although effectiveness of exercise therapy in HF
patients seems to be lower than initially expected, its cost-effectiveness ratio remains acceptable for health policy decision
makers to incorporate this intervention in the care of patients
with HF.
Source of financial support: Dr. Rohde and Dr. Polanczyk received
research sponsorship from Conselho Nacional de Desenvolvimento
Científico e Tecnológico, Brazil. This study was supported by the Brazilian Institute for Health Technology Assessment.
REFERENCES
[1] Lee WC, Chavez YE, Baker T, Luce BR. Economic burden of heart
failure: a summary of recent literature. Heart Lung 2004;33:362–71.
[2] Gregory D, Udelson JE, Konstam MA. Economic impact of beta
blockade in heart failure. Am J Med 2001;110(Suppl. 7A):74S– 80S.
[3] Andersson F, Cline C, Ryden-Bergsten T, Erhardt L. Angiotensin
converting enzyme (ACE) inhibitors and heart failure. The consequences
of underprescribing. Pharmacoeconomics 1999;15:535–50.
[4] McAlister FA, Ezekowitz J, Dryden DM, et al. Cardiac resynchronization
therapy and implantable cardiac defibrillators in left ventricular
systolic dysfunction. Evid Rep Technol Assess (Full Rep) 2007):1–199.
[5] Mosterd A, Hoes AW. Clinical epidemiology of heart failure. Heart
2007;93:1137– 46.
[6] Executive summary: HFSA 2006 Comprehensive Heart Failure Practice
Guideline. J Card Fail 2006;12:10 –38.
[7] Belardinelli R, Georgiou D, Cianci G, Purcaro A. Randomized,
controlled trial of long-term moderate exercise training in chronic
heart failure: effects on functional capacity, quality of life, and clinical
outcome. Circulation 1999;99:1173– 82.
[8] Piepoli MF, Davos C, Francis DP, Coats AJ. Exercise training metaanalysis of trials in patients with chronic heart failure (ExTraMATCH).
BMJ 2004;328:189.
[9] Dickstein K, Cohen–Solal A, Filippatos G, et al. ESC Guidelines for the
diagnosis and treatment of acute and chronic heart failure 2008: The
Task Force for the Diagnosis and Treatment of Acute and Chronic
Heart Failure 2008 of the European Society of Cardiology. Developed in
collaboration with the Heart Failure Association of the ESC (HFA) and
endorsed by the European Society of Intensive Care Medicine (ESICM).
Eur J Heart Fail 2008;10:933– 89.
[10] Wenger NK. Current status of cardiac rehabilitation. J Am Coll Cardiol
2008;51:1619 –31.
[11] Georgiou D, Chen Y, Appadoo S, et al. Cost– effectiveness analysis of
long–term moderate exercise training in chronic heart failure. Am J
Cardiol 2001;87:984 – 8, A4.
[12] O’Connor CM, Whellan DJ, Lee KL, et al. Efficacy and safety of exercise
training in patients with chronic heart failure: HF-ACTION
randomized controlled trial. JAMA 2009;301:1439 –50.
[13] Sonnenberg FA, Beck JR. Markov models in medical decision making: a
practical guide. Med Decis Making 1993;13:322–38.
[14] Beck-da Silva L, Goldraich L, Bonzanini L, et al. Pulse pressure and QRS
width evaluation as an inexpensive tool for heart failure assessment.
Congest Heart Fail 2009;15:222–7.
[15] Gohler A, Geisler BP, Manne JM, et al. Utility estimates for decisionanalytic modeling in chronic heart failure-health states based on New
York Heart Association classes and number of rehospitalizations.
Value Health 2009;12:185–7.
[16] Austin J, Williams R, Ross L, et al. Randomised controlled trial of
cardiac rehabilitation in elderly patients with heart failure. Eur J Heart
Fail 2005;7:411–7.
[17] Gianuzzi P, Temporelli PL, Corra U, Tavazzi L. Antiremodeling effect
of long-term exercise training in patients with stable chronic heart
failure: results of the Exercise in Left Ventricular Dysfunction and
Chronic Heart Failure (ELVD-CHF) Trial. Circulation 2003;108:
554 –9.
[18] Jolly K, Taylor RS, Lip GY, et al. A randomized trial of the addition of
home-based exercise to specialist heart failure nurse care: the
Birmingham Rehabilitation Uptake Maximisation study for patients with
Congestive Heart Failure (BRUM–CHF) study. Eur J Heart Fail 2009;11:
205–13.
[19] Jónsdóttir S, Andersen KK, Sigurosson AF, Sigurosson SB. The effect
of physical training in chronic heart failure. Eur Heart J 2006;8:97–
101.
[20] Kiilavuori K, Naveri H, Salmi T, Harkonen M. The effect of physical
training on skeletal muscle in patients with chronic heart failure. Eur
J Heart Fail 2000;2:53– 63.
[21] Kulcu DG, Kurtais Y, Tur BS, Gulec S, Seckin B. The effect of cardiac
rehabilitation on quality of life, anxiety and depression in patients
with congestive heart failure. A randomized controlled trial, shortterm results. Eura Medicophys 2007;43:489 –97.
[22] Hambrecht R, Gielen S, Linke A, et al. Effects of exercise training on
left ventricular function and peripheral resistance in patients with
chronic heart failure: a randomized trial. JAMA 2000;283:
3095–101.
[23] McKelvie RS, Teo KK, Roberts R, et al. Effects of exercise training in
patients with heart failure: the Exercise Rehabilitation Trial (EXERT).
Am Heart J 2002;144:23–30.
[24] Nilsson BB, Westheim A, Risberg MA. Long-Term effects of a groupbased high-intensity aerobic interval-training program in patients
with chronic heart failure. Am J Cardiol 2008;102:1220 –24.
[25] Wielenga RP, Huisveld IA, Bol E, et al. Safety and effects of physical
training in chronic heart failure. Results of the Chronic Heart
Failure and Graded Exercise study (CHANGE). Eur Heart J 1999;20:
872–9.
[26] Willenheimer R, Erhardt L, Cline C, Rydberg E, Israelsson B. Exercise
training in heart failure improves quality of life and exercise capacity.
Eur Heart J 1998;19:774 – 81.
[27] Araujo DV, Tavares LR, Verissimo R, et al. Cost of heart failure in the
Unified Health System. Arq Bras Cardiol 2005;84:422–7.
[28] Instituto Brasileiro de Geografia e Estatística. Available at: http://www
.ibge.gov.br/home/estatistica/indicadores/precos/inpc_ipca/ipca-inpc_
200908_3.shtm. [Accessed September 9, 2009].
[29] Word development indicators. Available at: http://data.un.org/Data
.aspx?q⫽purchase⫹power⫹parity&d⫽CDB&f⫽srID%3a29947. [Accessed
April 7, 2009].
VALUE IN HEALTH 14 (2011) S100 –S107
[30] Ribeiro RA, Stella SF, Camey SA, et al. Cost-effectiveness of
implantable cardioverter-defibrillators in Brazil: primary prevention
analysis in the public sector. Value Health 2010;13:160 – 8.
[31] Sesso R, da Silva CB, Kowalski SC, et al. Dialysis care, cardiovascular
disease, and costs in end-stage renal disease in Brazil. Int J Technol
Assess Health Care 2007;23:126 –30, 132.
[32] Flynn KE, Piña IL, Whellan DJ, et al. Effects of exercise training on
health status in patients with chronic heart failure: HF-ACTION
randomized controlled trial. JAMA 2009;301:1451–9, 1533.
[33] Reed SD, Whellan DJ, Li Y, et al; HF-ACTION Investigators. Economic
evaluation of the HF-ACTION (Heart Failure: A Controlled Trial
Investigating Outcomes of Exercise Training) randomized controlled
trial: an exercise training study of patients with chronic heart failure.
Circ Cardiovasc Qual Outcomes 2010;3:374 – 81.
S107
[34] Keteyian SJ, Piña IL, Hibner BA, Fleg JL. Clinical role of exercise
training in the management of patients with chronic heart failure.
J Cardiopulm Rehabil Prev 2010;30:67–76.
[35] Winkelmann ER, Chiappa GR, Lima CO, et al. Addition of inspiratory
muscle training to aerobic training improves cardiorespiratory
responses to exercise in patients with heart failure and inspiratory
muscle weakness. Am Heart J 2009;158:e1–7.
[36] Lima MM, Rocha MO, Nunes MC, et al. A randomized trial of the
effects of exercise training in Chagas cardiomyopathy. Eur J Heart Fail
2010;12:866 –73.
[37] Papadakis S, Reid RD, Coyle D, et al. Cost-effectiveness of cardiac
rehabilitation program delivery models in patients at varying cardiac
risk, reason for referral, and sex. Eur J Cardiovasc Prev Rehabil 2008;15:
347–53.