Skip to main content
NCBI home page
JAMA Network logoLink to JAMA Network
. 2021 Nov 24;7(2):140–148. doi: 10.1001/jamacardio.2021.4836

Economic Outcomes of Rehabilitation Therapy in Older Patients With Acute Heart Failure in the REHAB-HF Trial

A Secondary Analysis of a Randomized Clinical Trial

Derek S Chew 1, Yanhong Li 1, Michel Zeitouni 1,2, David J Whellan 3, Dalane Kitzman 4, Robert J Mentz 1,2, Pamela Duncan 5, Amy M Pastva 6, Gordon R Reeves 7, M Benjamin Nelson 8, Haiying Chen 9, Shelby D Reed 1,10,✉
PMCID: PMC8613698  PMID: 34817542

Key Points

Question

What are the economic outcomes of the Rehabilitation Therapy in Older Acute Heart Failure Patients trial, which randomized older patients admitted for acute decompensated heart failure to a novel rehabilitation intervention vs control?

Findings

In this secondary analysis of a randomized clinical trial including 349 patients, mean medical costs were similar in both groups, surpassing $25 000 per patient by 6 months, but quality-of-life gains were greater in the rehabilitation intervention group. Lifetime cost-effectiveness ratios for the intervention varied, but most were within conventional benchmarks for good value when simulated using the validated Tools for Economic Analysis of Patient Management Interventions in Heart Failure Cost-Effectiveness Model.

Meaning

These findings suggest that initiating rehabilitation training in older patients hospitalized for acute decompensated heart failure may yield good economic value to the US health care system.

This secondary analysis of a randomized clinical trial’s economic outcomes reports the cumulative medical resource use and costs during the 6-month follow-up period for patients randomized to an intervention or control condition for acute decompensated heart failure.

Abstract

Importance

In the Rehabilitation Therapy in Older Acute Heart Failure Patients (REHAB-HF) trial, a novel 12-week rehabilitation intervention demonstrated significant improvements in validated measures of physical function, quality of life, and depression, but no significant reductions in rehospitalizations or mortality compared with a control condition during the 6-month follow up. The economic implications of these results are important given the increasing pressures for cost containment in health care.

Objective

To report the economic outcomes of the REHAB-HF trial and estimate the potential cost-effectiveness of the intervention.

Design, Setting, Participants

The multicenter REHAB-HF trial randomized 349 patients 60 years or older who were hospitalized for acute decompensated heart failure to rehabilitation intervention or a control group; patients were enrolled from September 17, 2014, through September 19, 2019. For this preplanned secondary analysis of the economic outcomes, data on medical resource use and quality of life (via the 5-level EuroQol 5-Dimension scores converted to health utilities) were collected. Medical resource use and medication costs were estimated using 2019 US Medicare payments and the Federal Supply Schedule, respectively. Cost-effectiveness was estimated using the validated Tools for Economic Analysis of Patient Management Interventions in Heart Failure Cost-Effectiveness Model, which uses an individual-patient simulation model informed by the prospectively collected trial data. Data were analyzed from March 24, 2019, to December 1, 2020.

Interventions

Rehabilitation intervention or control.

Main Outcomes and Measures

Costs, quality-adjusted life-years (QALYs), and the lifetime estimated cost per QALY gained (incremental cost-effectiveness ratio).

Results

Among the 349 patients included in the analysis (183 women [52.4%]; mean [SD] age, 72.7 [8.1] years; 176 non-White [50.4%] and 173 White [49.6%]), mean (SD) cumulative costs per patient were $26 421 ($38 955) in the intervention group (excluding intervention costs) and $27 650 ($30 712) in the control group (difference, −$1229; 95% CI, −$8159 to $6394; P = .80). The mean (SD) cost of the intervention was $4204 ($2059). Quality of life gains were significantly greater in the intervention vs control group during 6 months (mean utility difference, 0.074; P = .001) and sustained beyond the 12-week intervention. Incremental cost-effectiveness ratios were estimated at $58 409 and $35 600 per QALY gained for the full cohort and in patients with preserved ejection fraction, respectively.

Conclusions and Relevance

These analyses suggest that longer-term benefits of this novel rehabilitation intervention, particularly in the subgroup of patients with preserved ejection fraction, may yield good value to the health care system. However, long-term cost-effectiveness is currently uncertain and dependent on the assumption that benefits are sustained beyond study follow-up, which needs to be corroborated in future trials in this patient population.

Introduction

Heart failure is the leading cause of hospitalization among older individuals in the US.1 The hospital-to-home transition is a particularly critical period in the heart failure care pathway. Within the first 6 months after hospitalization for heart failure, the risk of rehospitalization or all-cause death is approximately 30% in contemporary cohorts with heart failure.2,3

Understanding the potential economic impact of interventions addressing the hospital-to-home transition period is important in the era of increasing pressures for cost containment in health care. Specifically, lower-cost interventions that improve outcomes and quality of life among patients with heart failure could help to mitigate the upward trend in health care expenditures.

Prior studies of exercise-based rehabilitation in chronic, stable heart failure have demonstrated improvements in quality of life, physical function, and clinical events.4,5 This evidence led to a coverage decision by the US Centers for Medicare & Medicaid Services to reimburse for outpatient cardiac rehabilitation in patients with stable heart failure.6 However, the Centers for Medicare & Medicaid Services do not cover cardiac rehabilitation for patients who are hospitalized or within 6 weeks of recent hospitalization or for patients with heart failure and preserved ejection fraction (HFpEF), citing lack of evidence, because prior randomized clinical studies of exercise training in heart failure4,5,7,8,9 generally excluded such patients.

The Rehabilitation Therapy in Older Acute Heart Failure Patients (REHAB-HF) trial was designed to address this important gap in the management of acute decompensated heart failure in older hospitalized patients, regardless of ejection fraction. The REHAB-HF trial was a multicenter, randomized, attention-controlled, single-blind trial that randomized 349 older participants who had been hospitalized for acute decompensated heart failure to a novel, early, transitional, tailored, progressive, multidomain, 12-week physical rehabilitation intervention or a control condition.10,11,12 In the rehabilitation intervention group, significant improvements in physical function, quality of life, and depression were found at 3 months, but there were no significant reductions in rehospitalizations or mortality at 6 months.12

In a prespecified economic analysis of the REHAB-HF trial, we evaluated the cumulative medical resource use and costs during the 6-month follow-up period. A secondary objective was to estimate the cost-effectiveness of the rehabilitation intervention compared with a control condition.

Methods

REHAB-HF Trial Design

The design of the REHAB-HF trial has been described previously,10,13 and the trial protocol is provided in Supplement 1. Briefly, the study randomized 349 participants 60 years or older who had been admitted for acute decompensated heart failure to either a rehabilitation intervention group or a control group, stratified by reduced or preserved ejection fraction (eMethods 1 and eFigure 1 in Supplement 2). The institutional review boards at each enrolling site approved the study. All patients provided written informed consent. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.

The rehabilitation intervention was initiated during the acute hospitalization and continued for 12 weeks after hospital discharge in the outpatient setting. One-on-one outpatient sessions were 60 minutes, provided 3 days per week for 12 weeks, for a total of 36 sessions. Participants in the control arm received biweekly telephone contacts from study personnel. Although participants were encouraged to adhere to prescribed therapy and follow-up appointments, they received no specific recommendations regarding exercise.

Data Collection

Detailed data on medical resource use were collected from patients and caregivers at follow-up visits and supplemented by available medical records. Medical resource data included hospitalizations, emergency department visits, outpatient clinic visits, cardiac procedures, non–study-related physical or occupational therapy visits, skilled nursing facility or acute rehabilitation center visits, and medications (eMethods 2 in Supplement 2). For inpatient care, admission and discharge dates were collected along with discharge diagnoses (eTable 1 in Supplement 2).

The 5-level EuroQol 5-Dimension (EQ-5D-5L) instrument, a preference-based measure of health-related quality of life, was administered to participants at baseline and 1, 3, and 6 months. Responses on the EQ-5D-5L were then converted to US health utility weights, which is a weighting scale wherein 1 corresponds to perfect health and 0 corresponds to death (eMethods 2 in Supplement 2).14

A time survey was administered to obtain information on the additional time clinicians spent on activities associated with the supervised training sessions that would not be reflected in the exercise logs designed for the REHAB-HF trial, including time spent before and after each session on activities with patients (eg, scheduling, warm-up) and without patients (eg, cleaning equipment, documentation). In addition, patients were asked to report time spent traveling to and from supervised training sessions, nonstudy cardiac rehabilitation, and exercising at home.

Cost Assignment

Costs were reported in 2019 US dollars from the perspective of the US health care system and were adjusted for inflation using the US Medical Care Consumer Price Index where appropriate.15 Physician fees were assigned using the 2019 Medicare Physician Fee Schedule based on Current Procedural Terminology (CPT) codes.16 Medication costs were based on the Federal Supply Schedule for Pharmaceuticals and Drugs.17 Hospitalization costs were assigned using diagnosis-related group–based 2019 US Medicare reimbursement rates and calculated based on hospital length of stay.18 Inpatient days were calculated based on days in the hospital since the date of study randomization. The cost of an emergency department visit was based on the mean expenditure for all-cause emergency department visits (ie, $1877 in 2019 US dollars).19 We did not assign costs to study-related visits because activities during these visits were limited to research-related activities.

Costs associated with physical rehabilitation intervention were estimated for a 1-hour session of direct patient time using the 2019 Medicare Physician Fee Schedule and CPT code 97530.16 We conducted sensitivity analyses using different methods (ie, bottom-up and top-down costing approaches) to estimate the rehabilitation intervention costs.20 These costs were estimated using the Tools for Economic Analysis of Patient Management Interventions in Heart Failure (TEAM-HF) cost-estimation tool, an Excel-based tool designed to assist users in comprehensively estimating costs required to provide patient-centered interventions.21 Because we adopted the perspective of the US health care system, this economic evaluation did not include patients’ out-of-pocket costs or productivity losses. However, given that time dedicated to rehabilitation sessions, associated travel time, and home exercise is important from the patient perspective, we reported indirect costs (eMethods 3 in Supplement 2).

Cost-effectiveness Analysis

Concordant with economic evaluation best practices,22 we conducted an analysis to project costs and quality-adjusted life-years (QALYs) during a (remaining) lifetime horizon to better reflect the potential long-term consequences of the rehabilitation intervention. To extrapolate lifetime costs and QALYs, we used the previously validated TEAM-HF cost-effectiveness model, which uses the Seattle Heart Failure Model as the primary means of generating estimates of projected survival based on sets of demographic and clinical covariates representing patients in intervention and control groups at the end of study follow-up (eMethods 4 in Supplement 2).23

Clinical, cost, and utility model inputs were obtained directly from the REHAB-HF trial, supplemented with default laboratory measurements from the TEAM-HF model. The web-based model generates 10 000 Monte Carlo simulations, each representing predicted outcomes for 2 cohorts representing the rehabilitation intervention and control groups. Base-case parameters are provided in eTables 2 and 3 in Supplement 2. Consistent with guidelines for conducting cost-effectiveness analyses, both costs and QALYs were discounted at 3% per year.22

Statistical Analysis

Data were analyzed from March 24, 2019, to December 1, 2020. For baseline characteristics, categorical variables are presented as frequencies with percentages and continuous variables as means with SDs. Using generalized linear modeling, we compared medical resource use and costs incurred during the follow-up period between study groups. The models were specified with negative binomial error distributions and log links for comparisons of medical resource use and gamma error distributions and log links for comparisons of costs.24,25,26 Comparisons of health utilities by study group were based on multilevel mixed-effect linear regression models. In these models, individual patients were modeled as random effects. Time, treatment assignment, and an interaction between time and study group were modeled as fixed effects. The model also included the baseline EQ-5D-5L utility weight, study site, age, sex, systolic blood pressure at baseline, and type of heart failure (ie, HFpEF or reduced ejection fraction [HFrEF]) as fixed-effect covariates. Nonparametric bootstrapping was used to calculate bias-corrected 95% CIs for differences in costs and QALYs and their joint distributions. SAS, version 9.4 (SAS Institute Inc) was used for all statistical analyses. Two-sided P < .05 indicated statistical significance.

Results

From September 17, 2014, through September 19, 2019, 349 patients were enrolled at 7 sites in the US. The mean age at baseline was 72.7 (8.1) years; 183 patients (52.4%) were women and 166 (47.6%) were men. A racially diverse sample of patients was recruited: 7 (2.0%) were American Indian or Alaska Native, 4 were Asian (1.1%), 154 (44.1%) were Black or African American, 173 (49.6%) were White, and 11 (3.2%) reported more than 1 race. Seven patients (2.0%) reported Hispanic or Latino ethnicity. The relatively large proportion of Black or African American patients strengthened the generalizability of the study sample because of the higher prevalence of heart failure among non-Hispanic Black adults in the US.27 Heart failure with preserved ejection fraction was slightly more common at baseline than HFrEF (185 [53.0%] vs 164 [47.0%]) (eTable 4 in Supplement 2).

Medical Resource Use

A total of 406 rehospitalizations occurred during the 6-month follow-up period. Of these rehospitalizations, 204 (50.2%) were due to heart failure, 60 (14.8%) were due to cardiovascular causes other than heart failure, and the remainder were due to noncardiovascular causes (142 [35.0%]).

There was no statistically significant difference in the mean number of all-cause rehospitalizations per patient among the rehabilitation intervention group (1.10 [1.37]) and the control group (1.22 [1.44]; P = .63) (Table 1). The mean number of inpatient days was also similar in both groups (8.35 [13.39] in the rehabilitation intervention group and 8.84 [12.61] in the control group; P = .97).

Table 1. Within-Trial Medical Resource Use and Costs for the Overall Cohort.

Treatment group, mean (SD) Difference (95% CI) Adjusted relative means ratio (95% CI)a P value
Rehabilitation intervention (n = 175) Control (n = 174)
No. of medical resources used
All-cause rehospitalizations 1.10 (1.37) 1.22 (1.44) −0.12 (−0.41 to 0.15) 0.94 (0.72 to 1.23) .63
Rehospitalization inpatient days 8.35 (13.39) 8.84 (12.61) −0.50 (−3.18 to 2.24) 0.99 (0.66 to 1.50) .97
Total inpatient days 10.68 (13.83) 11.55 (15.69) −0.80 (−3.60 to 2.15) 0.97 (0.75 to 1.24) .78
Emergency department visits 0.54 (0.93) 0.57 (1.04) −0.03 (−0.24 to 0.17) 0.95 (0.66 to 1.36) .77
Outpatient care visits, total 6.94 (5.17) 6.62 (5.21) 0.32 (−0.79 to 1.38) 1.05 (0.89 to 1.24) .59
Primary care physician 2.20 (2.28) 2.05 (2.2) 0.15 (−0.27 to 0.65) 1.08 (0.88 to 1.34) .99
Cardiologist 2.21 (2.1) 2.13 (2.1) 0.08 (−0.33 to 0.52) 1.03 (0.84 to 1.27) .78
Other specialist physician 1.90 (3.15) 1.92 (3.05) −0.02 (−0.70 to 0.58) 1.00 (0.75 to 1.35) .99
NP or PA 0.51 (1.23) 0.37 (0.96) 0.14 (−0.08 to 0.38) 1.37 (0.82 to 2.31) .23
Other clinician 0.13 (0.7) 0.15 (0.69) −0.02 (−0.18 to 0.12) 0.90 (0.30 to 2.69) .84
Non–study-related rehabilitation visits 2.99 (6.50) 6.71 (11.54) −3.72 (−5.93 to −1.97) 0.42 (0.24 to 0.73) .002
Direct medical costs, 2019 US $
Inpatient care 23 972 (38 556) 24 837 (30 510) −864 (−7639 to 6813) 0.93 (0.67 to 1.28) .64
Outpatient care
Total 754 (659) 696 (586) 57 (−63 to 190) 1.09 (0.81 to 1.47) .57
Visits 657 (505) 634 (521) 23 (−84 to 130) 1.03 (0.77 to 1.39) .82
Procedure 97 (346) 62 (197) 35 (−15 to 101) 1.54 (0.87 to 2.72) .14
Emergency department visits 1019 (1751) 1068 (1950) −49 (−454 to 313) 0.92 (0.48 to 1.77) .81
Non–study-related rehab visits 325 (717) 708 (1246) −383 (−621 to −187) 0.42 (0.22 to 0.77) .006
Medications 350 (289) 339 (260) 10 (−48 to 62) 1.03 (0.84 to 1.28) .76
Total, excluding intervention 26 421 (38 955) 27 650 (30 712) −1229 (−8159 to 6394) 0.97 (0.78 to 1.21) .80
Intervention 4204 (2059) NA NA NA NA
Total 30 625 (38 442) 27 650 (30 712) 2976 (−3853 to 10 552) 1.13 (0.92 to 1.38) .25
Open in a new tab

Abbreviations: NA, not applicable; NP, nurse practitioner; PA, physician assistant.

a

Baseline covariates included age, sex, clinical site, and baseline Short Physical Performance Battery score.

Emergency department visits were similar between the study groups. Similarly, there was no significant difference in the mean number of outpatient care visits overall or outpatient visits by clinician care type. Patients randomized to the control group had a higher mean number of non–study-related rehabilitation visits (6.71 [11.54]) compared with patients randomized to the rehabilitation intervention (2.99 [6.50]; P = .002).

Rehabilitation Intervention Costs

Of the 28 clinicians who facilitated the rehabilitation intervention across the 7 study sites, 27 (96.4%) completed a survey that collected information on time spent on activities associated with the rehabilitation session in addition to direct exercise time. In the outpatient setting, clinicians reported spending a mean total of 31.2 (27.8) minutes on nonexercise activities for the initial session and 23.3 (23.3) minutes for subsequent sessions. Patients completed a mean of 23.7 (12.4) supervised intervention sessions representing a mean of 24.3 (12.8) hours of exercise. The mean cost for intervention (including non–exercise- and exercise-related activities) was an estimated $4204 ($2059) per patient using CPT-based Medicare reimbursement fees. Intervention costs were similar or lower in our sensitivity analyses using alternate bottom-up and top-down approaches to cost estimation (eTable 5 in Supplement 2).

Direct Medical Costs and Indirect Patient Time Costs

Table 1 shows the direct medical costs by medical resource use category. Mean direct medical costs, excluding the cost of the intervention, were not significantly different between the rehabilitation intervention and control groups (mean difference, −$1229; 95% CI for the difference, −$8159 to $6394; P = .80).

Total mean direct medical costs, including intervention costs, were estimated to be $30 625 ($38 442) per patient in the rehabilitation intervention group (excluding intervention costs, $26 421 [$38 955]) and $27 650 ($30 712) per patient in the control group. Thus, the rehabilitation intervention was associated with $2976 higher costs (95% CI, −$3853 to $10 552) compared with the control group, though this difference was not statistically significant (P = .25). Based on the numbers of intervention sessions completed and patient-reported non–study rehabilitation sessions and time for travel and home exercise, mean indirect costs for patient time were estimated at $2323 ($1995) for the rehabilitation intervention group and $1093 ($1629) for the control group.

Health Utilities

At baseline, the mean EQ-5D-5L utility score was 0.57 (0.29) in the rehabilitation intervention group and 0.62 (0.28) in the control group (P = .07) (Figure). By 1 month, mean utilities increased by 0.12 units in the rehabilitation intervention group and 0.05 units in the control group. Mean utility scores were consistently higher in the rehabilitation intervention group at all follow-up visits through 6 months. By 6 months, the mean utilities were 0.74 (0.26) and 0.68 (0.31) in the rehabilitation intervention and control groups, respectively (P = .007). In the adjusted multilevel mixed-effects linear regression model, the mean difference in utilities during the follow-up period was 0.074 (95% CI, 0.029-0.118; P = .001). When we imputed 0 utility scores to account for deaths during follow-up, mean utility scores remained higher in the rehabilitation intervention group at all follow-up visits through 6 months. In the adjusted model, there remained a statistically significant difference in utilities during the trial follow-up period (0.054; 95% CI, 0.003-0.105; P = .04).

Figure. Unadjusted Mean 5-Level EuroQol 5-Dimension Health Utility Scores at Trial Follow-up, by Study Group.

Figure.

Open in a new tab

Using generalized linear modeling (adjusted for age, sex, clinical site, heart failure type [reduced vs preserved ejection fraction], and baseline Short Physical Performance Battery score), the rehabilitation intervention group had higher utility scores compared with the control group after 1 month of follow-up (baseline, P = .07; month 1, P = .01; month 3, P < .001; and month 6, P = .007). Treatment group, month, and interaction terms between treatment and months 1, 3, and 6 were included as independent variables. Error bars indicate SD.

Within-Trial Cost-effectiveness

During the 6-month follow up, mean undiscounted QALYs were 0.310 (0.140) for the intervention group and 0.300 (0.140) for the control group. The within-trial difference in QALYs was 0.010 (95% CI, −0.022 to 0.041). The within-trial incremental cost-effectiveness ratio (ICER) was estimated to be $292 521 (95% CI, −$159 650 to $293 190 515).

HFpEF Subgroup

As reported elsewhere, improvements on clinical outcomes with the rehabilitation intervention were greater among patients with HFpEF vs patients with HFrEF.28 In this prespecified subgroup analysis, there were no significant between-group differences in total direct medical costs (Table 2). Mean within-trial QALYs were 0.301 (0.142) for the intervention group and 0.259 (0.145) for the attention-control group, a gain of 0.042 (95% CI, −0.0003 to 0.081) QALYs. The within-trial ICER was an estimated $72 933 per QALY gained (95% CI, −$165 916 to $1 019 114) in the HFpEF subgroup.

Table 2. Within-Trial Medical Resource Use and Costs for Patients With HFpEF.

Treatment group, mean (SD) Difference (95% CI) Adjusted relative means ratio (95% CI)a P value
Rehabilitation intervention (n = 93) Control (n = 92)
No. of medical resources used
All-cause rehospitalizations 1.15 (1.52) 1.32 (1.39) −0.16 (−0.62 to 0.23) 0.87 (0.61 to 1.25) .46
Rehospitalization inpatient days 8.62 (13.55) 8.98 (11.77) −0.35 (−4.14 to 3.22) 1.00 (0.58 to 1.75) .99
Total inpatient days 10.80 (13.56) 11.53 (11.97) −0.74 (−4.34 to 2.90) 0.92 (0.66 to 1.28) .60
Emergency department visits 0.46 (0.83) 0.67 (1.22) −0.21 (−0.52 to 0.08) 0.68 (0.41 to 1.12) .13
Outpatient care visits, total 7.45 (5.30) 7.07 (5.81) 0.39 (−1.32 to 2.02) 1.05 (0.85 to 1.30) .63
Primary care physician 2.37 (2.10) 2.25 (2.18) 0.12 (−0.45 to 0.74) 1.06 (0.83 to 1.36) .63
Cardiologist 1.89 (1.76) 1.92 (2.24) −0.03 (−0.60 to 0.52) 0.97 (0.72 to 1.32) .86
Other specialist physician 2.43 (3.86) 2.51 (3.84) −0.08 (−1.24 to 1.05) 0.99 (0.68 to 1.45) .97
NP or PA 0.63 (1.40) 0.28 (0.67) 0.35 (0.07 to 0.70) 2.28 (1.15 to 4.55) .02
Other clinician 0.13 (0.70) 0.10 (0.36) 0.03 (−0.10 to 0.23) 2.01 (0.43 to 9.45) .38
Non–study-related rehabilitation visits 3.92 (6.89) 7.51 (12.68) −3.59 (−6.77 to −0.82) 0.50 (0.24 to 1.03) .06
Direct medical costs, 2019 US $
Inpatient care 23 551 (31 492) 24 156 (26 070) −605 (−8242 to 7599) 0.82 (0.53 to 1.27) .38
Outpatient care
Care, total 746 (577) 737 (648) 9 (−168 to 187) 1.02 (0.71 to 1.46) .92
Visits 704 (543) 683 (598) 21 (−155 to 188) 1.03 (0.72 to 1.48) .86
Procedure 42 (140) 54 (187) −12 (−64 to 36) 1.03 (0.45 to 2.37) .95
Emergency department visits 867 (1555) 1265 (2296) - 397 (−977 to 155) 0.68 (0.28 to 1.69) .41
Non–study-related rehabilitation visits 445 (839) 789 (1419) −344 (−706 to −21) 0.54 (0.23 to 1.27) .16
Medications 373 (308) 337 (250) 35 (−50 to 117) 1.09 (0.83 to 1.43) .53
Total, excluding intervention 25 983 (32 061) 27 286 (26 479) −1302 (−9064 to 7027) 0.96 (0.72 to 1.28) .80
Intervention 4363 (1950) NA NA NA NA
Total costs 30 346 (31 541) 27 286 (26 479) 3061 (−4653 to 11 246) 1.13 (0.87 to 1.46) .35
Open in a new tab

Abbreviations: HFpEF, heart failure with preserved ejection fraction; NA, not applicable; NP, nurse practitioner; PA, physician assistant.

a

Baseline covariates included age, sex, clinical site, and baseline Short Physical Performance Battery score.

Lifetime Extrapolation of Costs and Outcomes

For the overall cohort, remaining lifetime costs and QALYs projected using the TEAM-HF model resulted in an ICER of $58 409 per QALY with the rehabilitation intervention compared with the control groups (Table 3). Using probabilistic sensitivity analysis to vary all model parameters simultaneously, 45% and 54% of simulated ICERs were at or below conventionally accepted thresholds of $50 000 and $100 000 per QALY gained, respectively. Uncertainty associated with the estimated ICERs is represented in the cost-effectiveness acceptability curve (eFigure 2 in Supplement 2).

Table 3. Lifetime Cost-Effectiveness of the Rehabilitation Intervention vs Controla.

Overall cohort (95% CI) Patients with HFpEF (95% CI)
Total costs, $ Total LYs Total QALYs Total costs, $ Total LYs Total QALYs
Rehabilitation intervention group 194 866 (177 946-212 278) 5.73 (5.23-6.22) 4.58 (4.18-4.98) 180 237 (158 830-202-318) 5.61 (4.96-6.28) 4.49 (3.96-5.02)
Control group 190 683 (173 634-208 437) 5.66 (5.16-6.16) 4.52 (4.12-4.93) 176 210 (155 044-198 866) 5.50 (4.83-6.19) 4.39 (3.86-4.94)
ICER, US $
Per life-year gained 69 421 NA NA 41 871 NA NA
Per QALY gained 58 409 NA NA 35 600 NA NA
Open in a new tab

Abbreviations: HFpEF, heart failure with preserved ejection fraction; ICER, incremental cost-effectiveness ratio; LY, life-year; NA, not applicable; QALY, quality-adjusted life-year.

a

Costs, life-years, and QALYs were discounted at 3% per year.

We also estimated the lifetime cost-effectiveness of the rehabilitation intervention vs control groups in the subgroup with HFpEF at $35 600 per QALY gained. In the probabilistic sensitivity analysis, the rehabilitation intervention was considered cost-effective in 53% of simulations and 58% of simulations at willingness-to-pay thresholds of $50 000 and $100 000 per QALY gained, respectively, for an additional 0.10 QALYs gained.

Discussion

In this prespecified economic analysis of the REHAB-HF trial, the mean total direct medical costs in both study groups surpassed $25 000 per patient for 6 months, reflecting the frequent interaction between older patients with heart failure and the health care system. Although slightly lower mean costs in the rehabilitation intervention group did not offset the cost of the rehabilitation intervention estimated at $4204 per patient, significant improvements in physical functioning and quality of life afforded by the intervention contribute to its overall value to the health care system in an older, frail, at-risk patient group. Prespecified exploratory analyses found that incremental gains in within-trial QALYs with the rehabilitation intervention were larger among patients with HFpEF, thus further enhancing expected value of the novel patient-centered program in this patient subgroup.

Most of the overall costs (>75%) were attributed to inpatient care, which is consistent with previous studies.29,30 Based on the point estimates of inpatient costs, mean unadjusted direct medicals costs were $1229 lower in the rehabilitation intervention group compared with the control group (P = .80), and there was no significant reduction in the rates of rehospitalization or mortality between groups. It is worth noting, however, that the trial was insufficiently powered for rehospitalization events, because the observed effect size (−10%) was smaller than that assumed in the power calculation (−25%).

Under the conservative assumption that the clinical benefit of the rehabilitation intervention is limited to improvements in quality of life and physical functioning, these benefits alone support the value proposition of the rehabilitation intervention. Specifically, we observed similar within-trial direct medical costs and lifetime extrapolated costs. Even when not accounting for potential reductions in inpatient costs or fewer nonstudy rehabilitation visits, at an intervention cost of $4204, the intervention would be considered economically attractive at traditionally accepted willingness-to-pay thresholds of $50 000 or $100 000 per QALY gained as long as lifetime QALY gains surpassed 0.084 or 0.042, respectively. These QALY gains may be achievable with sustained improvements in quality of life beyond the trial follow-up. In the present study, we estimated an additional 0.010 QALYs in the intervention group during just 6 months. Indeed, under the conservative assumption that mortality was similar in both treatment groups, our lifetime cost-effectiveness model estimated an additional 0.06 QALYs gained and the lifetime cost per QALY gained for the rehabilitation intervention was an estimated $58 409 per QALY compared with the control group, which is within conventional thresholds for value within the US31 and similar to other HF interventions such as combined sacubitril-valsartan or sodium-glucose contransporter 2 inhibitors.32,33 However, as shown in eFigure 2 in Supplement 2, approximately 45% of simulated ICERs surpass conventional value thresholds. Any future adequately powered, randomized studies demonstrating clinical benefits of the rehabilitation intervention, either in terms of reduced mortality or hospitalization rates, would improve and reduce uncertainty about its cost-effectiveness.

Finally, we found that the value proposition of the rehabilitation intervention may be particularly attractive among patients with HFpEF. Notably, the within-trial ICER of $72 933 per QALY gained (ie, rehabilitation intervention vs control) would be considered economically attractive at just 6 months of follow-up. Improved cost-effectiveness in the HFpEF subgroup was driven by the greater magnitude of quality-of-life improvement associated with the rehabilitation intervention vs control—that is, there was a greater observed incremental gain in QALYs in the HFpEF subgroup compared with the overall cohort (0.042 vs 0.010 QALYs gained). These results should be interpreted with caution until the magnitude of benefit observed in the HFpEF subgroup in the REHAB-HF trial can be corroborated in a larger, adequately powered randomized trial.

Limitations

The results of this economic evaluation need to be interpreted in the context of several limitations. First, because cost data were not directly collected in the REHAB-HF trial, medical resource use was valued with externally derived unit costs. However, unit cost estimates were based on costs of Medicare patients with HF treated at geographically diverse US hospitals, uncertainty of estimates was incorporated in analyses, and estimated inpatient costs reflect the range of case mix complexity reported in the trial. Second, we estimated the costs associated with the rehabilitation intervention from CPT-based reimbursement from Medicare, which may not be generalizable to settings where care is provided by nonparticipating clinicians or patients participating in Medicare Advantage plans. However, our base-case approach was conservative. In our sensitivity analyses using bottom-up and top-down costing approaches, the cost of the rehabilitation intervention was estimated to be lower than CPT-based Medicare reimbursement. Third, the duration of follow-up in the REHAB-HF trial was relatively short, and there is uncertainty about whether quality-of-life gains are sustained beyond 6 months. An additional concern may pertain to the application of the TEAM-HF Cost-Effectiveness model to the HFpEF subgroup. The TEAM-HF model relies on the Seattle Heart Failure Model, which was originally developed and externally validated in several cohorts of patients with HFrEF, to predict survival beyond a study’s follow-up period. However, a recent analysis34 revealed good calibration of the Seattle Heart Failure Model on predicting mortality in patients with HFpEF.

Conclusions

In this economic analysis of the REHAB-HF trial, although there was minimal impact of the rehabilitation intervention on medical resource use and costs compared with the control condition, our findings suggest that longer-term effects of this novel rehabilitation intervention on QALYs attributable to improvements in physical functioning may yield good economic value to the US health care system, particularly in the subgroup of patients with HFpEF. When interpreting these findings, however, uncertainty regarding the duration of the intervention’s effect should be considered, because benefits of behavioral interventions tend to wane over time. Evidence from a larger-scale trial with extended surveillance using secondary data will be needed to corroborate or support adjustments to our findings about the long-term value of the REHAB-HF rehabilitation program for patients with HFpEF who currently have few effective treatment options.

Supplement 1.

Trial Protocol

Click here for additional data file. (1.1MB, pdf)
Supplement 2.

eMethods 1. REHAB-HF Trial Design

eMethods 2. Data Collection

eMethods 3. Indirect Cost Assignment

eMethods 4. TEAM-HF Cost-Effectiveness Model

eTable 1. Diagnosis Related Group (DRG) Codes Assigned to Prespecified Discharge Diagnoses

eTable 2. TEAM-HF Cost-Effectiveness Model Input Parameters for the Overall REHAB-HF Cohort

eTable 3. TEAM-HF Cost-Effectiveness Model Input Parameters for the Heart Failure With Preserved Ejection Fraction Subgroup

eTable 4. Baseline Characteristics

eTable 5. Sensitivity Analysis Using Different Methods for Estimation of Rehabilitation Intervention Cost

eFigure 1. CONSORT Diagram

eFigure 2. Cost-Effectiveness Acceptability Curves for the Overall Cohort and the HFpEF Subgroup

eReferences

Click here for additional data file. (345.2KB, pdf)

References

  • 1.Roger VL. Epidemiology of heart failure. Circ Res. 2013;113(6):646-659. doi: 10.1161/CIRCRESAHA.113.300268 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Chun S, Tu JV, Wijeysundera HC, et al. Lifetime analysis of hospitalizations and survival of patients newly admitted with heart failure. Circ Heart Fail. 2012;5(4):414-421. doi: 10.1161/CIRCHEARTFAILURE.111.964791 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Kozhuharov N, Goudev A, Flores D, et al. ; GALACTIC Investigators . Effect of a strategy of comprehensive vasodilation vs usual care on mortality and heart failure rehospitalization among patients with acute heart failure: the GALACTIC randomized clinical trial. JAMA. 2019;322(23):2292-2302. doi: 10.1001/jama.2019.18598 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Bozkurt B, Fonarow GC, Goldberg LR, et al. ; ACC’s Heart Failure and Transplant Section and Leadership Council . Cardiac rehabilitation for patients with heart failure: JACC Expert Panel. J Am Coll Cardiol. 2021;77(11):1454-1469. doi: 10.1016/j.jacc.2021.01.030 [DOI] [PubMed] [Google Scholar]
  • 5.O’Connor CM, Whellan DJ, Lee KL, et al. ; HF-ACTION Investigators . Efficacy and safety of exercise training in patients with chronic heart failure: HF-ACTION randomized controlled trial. JAMA. 2009;301(14):1439-1450. doi: 10.1001/jama.2009.454 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Centers for Medicare & Medicaid Services . Decision Memo for Cardiac Rehabilitation (CR) Programs—Chronic Heart Failure (CAG-00437N). February 18, 2014. Accessed April 22, 2021. https://www.cms.gov/medicare-coverage-database/view/ncacal-decision-memo.aspx?proposed=N&NCAId=270
  • 7.Fleg JL, Cooper LS, Borlaug BA, et al. ; National Heart, Lung, and Blood Institute Working Group . Exercise training as therapy for heart failure: current status and future directions. Circ Heart Fail. 2015;8(1):209-220. doi: 10.1161/CIRCHEARTFAILURE.113.001420 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Tucker WJ, Beaudry RI, Liang Y, et al. Meta-analysis of exercise training on left ventricular ejection fraction in heart failure with reduced ejection fraction: a 10-year update. Prog Cardiovasc Dis. 2019;62(2):163-171. doi: 10.1016/j.pcad.2018.08.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Fukuta H, Goto T, Wakami K, Kamiya T, Ohte N. Effects of exercise training on cardiac function, exercise capacity, and quality of life in heart failure with preserved ejection fraction: a meta-analysis of randomized controlled trials. Heart Fail Rev. 2019;24(4):535-547. doi: 10.1007/s10741-019-09774-5 [DOI] [PubMed] [Google Scholar]
  • 10.Reeves GR, Whellan DJ, Duncan P, et al. ; REHAB-HF Trial Investigators . Rehabilitation Therapy in Older Acute Heart Failure Patients (REHAB-HF) trial: design and rationale. Am Heart J. 2017;185:130-139. doi: 10.1016/j.ahj.2016.12.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Warraich HJ, Kitzman DW, Whellan DJ, et al. Physical function, frailty, cognition, depression, and quality of life in hospitalized adults ≥60 years with acute decompensated heart failure with preserved versus reduced ejection fraction. Circ Heart Fail. 2018;11(11):e005254. doi: 10.1161/CIRCHEARTFAILURE.118.005254 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Kitzman DW, Whellan DJ, Duncan P, et al. Physical rehabilitation for older patients hospitalized for heart failure. N Engl J Med. 2021;385(3):203-216. doi: 10.1056/NEJMoa2026141 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Pastva AM, Duncan PW, Reeves GR, et al. Strategies for supporting intervention fidelity in the Rehabilitation Therapy in Older Acute Heart Failure Patients (REHAB-HF) trial. Contemp Clin Trials. 2018;64:118-127. doi: 10.1016/j.cct.2017.10.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Pickard AS, Law EH, Jiang R, et al. United States valuation of EQ-5D-5L health states using an international protocol. Value Health. 2019;22(8):931-941. doi: 10.1016/j.jval.2019.02.009 [DOI] [PubMed] [Google Scholar]
  • 15.US Department of Labor : Bureau of Labor Statistics. Consumer Price Index–all urban consumers. 2020. Accessed December 10, 2019. https://www.bls.gov/cpi/
  • 16.Centers for Medicare and Medicaid Services . National physician fee schedule payment amount file. Updated July 1, 2021. Accessed January 20, 2020. https://www.cms.gov/medicare/physician-fee-schedule/search/overview
  • 17.US Department of Veterans Affairs . Office of Procurement, Acquisitions and Logistics (OPAL): pharmaceutical prices. Updated October 5, 2021. Accessed February 5, 2020. https://www.va.gov/opal/nac/fss/pharmPrices.asp
  • 18.Centers for Medicare and Medicaid Services . Acute inpatient PPS [prospective payment system]. Modified February 20, 2020. Accessed March 11, 2020. https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS
  • 19.Galarraga JE, Pines JM. Costs of ED episodes of care in the United States. Am J Emerg Med. 2016;34(3):357-365. doi: 10.1016/j.ajem.2015.06.001 [DOI] [PubMed] [Google Scholar]
  • 20.Chapko MK, Liu CF, Perkins M, Li YF, Fortney JC, Maciejewski ML. Equivalence of two healthcare costing methods: bottom-up and top-down. Health Econ. 2009;18(10):1188-1201. doi: 10.1002/hec.1422 [DOI] [PubMed] [Google Scholar]
  • 21.Reed SD, Li Y, Kamble S, et al. Introduction of the tools for economic analysis of patient management interventions in heart failure costing tool: a user-friendly spreadsheet program to estimate costs of providing patient-centered interventions. Circ Cardiovasc Qual Outcomes. 2012;5(1):113-119. doi: 10.1161/CIRCOUTCOMES.111.962977 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Sanders GD, Neumann PJ, Basu A, et al. Recommendations for conduct, methodological practices, and reporting of cost-effectiveness analyses: Second Panel on Cost-Effectiveness in Health and Medicine. JAMA. 2016;316(10):1093-1103. doi: 10.1001/jama.2016.12195 [DOI] [PubMed] [Google Scholar]
  • 23.Reed SD, Neilson MP, Gardner M, et al. Tools for Economic analysis of patient management interventions in heart failure cost-effectiveness model: a Web-based program designed to evaluate the cost-effectiveness of disease management programs in heart failure. Am Heart J. 2015;170(5):951-960. doi: 10.1016/j.ahj.2015.08.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Grieve R, Nixon R, Thompson SG, Normand C. Using multilevel models for assessing the variability of multinational resource use and cost data. Health Econ. 2005;14(2):185-196. doi: 10.1002/hec.916 [DOI] [PubMed] [Google Scholar]
  • 25.Manca A, Rice N, Sculpher MJ, Briggs AH. Assessing generalisability by location in trial-based cost-effectiveness analysis: the use of multilevel models. Health Econ. 2005;14(5):471-485. doi: 10.1002/hec.914 [DOI] [PubMed] [Google Scholar]
  • 26.Thompson SG, Nixon RM, Grieve R. Addressing the issues that arise in analysing multicentre cost data, with application to a multinational study. J Health Econ. 2006;25(6):1015-1028. doi: 10.1016/j.jhealeco.2006.02.001 [DOI] [PubMed] [Google Scholar]
  • 27.Rethy L, Petito LC, Vu THT, et al. Trends in the prevalence of self-reported heart failure by race/ethnicity and age from 2001 to 2016. JAMA Cardiol. 2020;5(12):1425-1429. doi: 10.1001/jamacardio.2020.3654 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Mentz RJ, Whellan DJ, Reeves GR, et al. Rehabilitation intervention in older patients with acute heart failure with preserved versus reduced ejection fraction. JACC Heart Fail. 2021;9(10):747-757. doi: 10.1016/j.jchf.2021.05.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Dunlay SM, Shah ND, Shi Q, et al. Lifetime costs of medical care after heart failure diagnosis. Circ Cardiovasc Qual Outcomes. 2011;4(1):68-75. doi: 10.1161/CIRCOUTCOMES.110.957225 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.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(4):374-381. doi: 10.1161/CIRCOUTCOMES.109.907287 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Anderson JL, Heidenreich PA, Barnett PG, et al. ; ACC/AHA Task Force on Performance Measures; ACC/AHA Task Force on Practice Guidelines . ACC/AHA statement on cost/value methodology in clinical practice guidelines and performance measures: a report of the American College of Cardiology/American Heart Association Task Force on Performance Measures and Task Force on Practice Guidelines. Circulation. 2014;129(22):2329-2345. doi: 10.1161/CIR.0000000000000042 [DOI] [PubMed] [Google Scholar]
  • 32.McEwan P, Bennett H, Khunti K, et al. Assessing the cost-effectiveness of sodium-glucose cotransporter-2 inhibitors in type 2 diabetes mellitus: a comprehensive economic evaluation using clinical trial and real-world evidence. Diabetes Obes Metab. 2020;22(12):2364-2374. doi: 10.1111/dom.14162 [DOI] [PubMed] [Google Scholar]
  • 33.Gaziano TA, Fonarow GC, Claggett B, et al. Cost-effectiveness analysis of sacubitril/valsartan vs enalapril in patients with heart failure and reduced ejection fraction. JAMA Cardiol. 2016;1(6):666-672. doi: 10.1001/jamacardio.2016.1747 [DOI] [PubMed] [Google Scholar]
  • 34.Rich JD, Burns J, Freed BH, Maurer MS, Burkhoff D, Shah SJ. Meta-analysis Global Group in Chronic (MAGGIC) Heart Failure Risk Score: validation of a simple tool for the prediction of morbidity and mortality in heart failure with preserved ejection fraction. J Am Heart Assoc. 2018;7(20):e009594. doi: 10.1161/JAHA.118.009594 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

Trial Protocol

Click here for additional data file. (1.1MB, pdf)
Supplement 2.

eMethods 1. REHAB-HF Trial Design

eMethods 2. Data Collection

eMethods 3. Indirect Cost Assignment

eMethods 4. TEAM-HF Cost-Effectiveness Model

eTable 1. Diagnosis Related Group (DRG) Codes Assigned to Prespecified Discharge Diagnoses

eTable 2. TEAM-HF Cost-Effectiveness Model Input Parameters for the Overall REHAB-HF Cohort

eTable 3. TEAM-HF Cost-Effectiveness Model Input Parameters for the Heart Failure With Preserved Ejection Fraction Subgroup

eTable 4. Baseline Characteristics

eTable 5. Sensitivity Analysis Using Different Methods for Estimation of Rehabilitation Intervention Cost

eFigure 1. CONSORT Diagram

eFigure 2. Cost-Effectiveness Acceptability Curves for the Overall Cohort and the HFpEF Subgroup

eReferences

Click here for additional data file. (345.2KB, pdf)

Articles from JAMA Cardiology are provided here courtesy of American Medical Association

RESOURCES