Cost-effectiveness of Telephonic Disease Management in Heart Failure

February 15, 2008
Brad Smith, PhD
Brad Smith, PhD

,
Paul F. Hughes-Cromwick, MA
Paul F. Hughes-Cromwick, MA

,
Emma Forkner, MS
Emma Forkner, MS

,
Autumn Dawn Galbreath, MD
Autumn Dawn Galbreath, MD

Volume 14, Issue 2

This study shows that telephonic disease management was not cost-effective in a broadly representative sample of community-dwelling patients.

Objective: To evaluate the cost-effectiveness of a telephonic disease management (DM) intervention in heart failure (HF).

Study Design: Randomized controlled trial of telephonic DM among 1069 community-dwelling patients with systolic HF (SHF) and diastolic HF performed between 1999 and 2003. The enrollment period was 18 months per subject.

Methods: Bootstrap-resampled incremental costeffectiveness ratios (ICERs) were computed and compared across groups. Direct medical costs were obtained from a medical record review that collected records from 92% of patients; 66% of records requested were obtained.

Results: Disease management produced statistically significant survival advantages among all patients (17.4 days, P = .04), among patients with New York Heart Association (NYHA) class III/IV symptoms (47.7 days, P = .02), and among patients with SHF (24.2 days, P = .01). Analyses of direct medical and intervention costs showed no cost savings associated with the intervention. For all patients and considering all-cause medical care, the ICER was $146 870 per quality-adjusted life-year (QALY) gained, while for patients with NYHA class III/IV symptoms and patients with SHF, the ICERs were $67 784 and $95 721 per QALY gained, respectively. Costs per QALY gained were $101 120 for all patients, $72 501 for patients with SHF, and $41 348 for patients with NYHA class III/IV symptoms.

Conclusions: The intervention was effective but costly to implement and did not reduce utilization. It may not be cost-effective in other broadly representative samples of patients. However, with program cost reductions and proper targeting, this program may produce life-span increases at costs that are less than $100 000 per QALY gained.

(Am J Manag Care. 2008;14:106-115)

To the extent that the disease management (DM) industry can reduce program costs through economies of scale or competition, carefully targeted DM programs may produce cost-effective improvements in heart failure (HF) outcomes.

Telephonic DM produced statistically significant survival advantages among all patients, among patients with New York Heart Association class III/IV symptoms, and among patients with systolic HF.

The intervention was effective but costly to implement and did not reduce utilization. It may not be cost-effective in other broadly representative samples of patients. However, with program cost reductions and proper targeting, this program may produce life-span increases at acceptable costs.

In the United States, estimates of annual direct expenditures on heart failure (HF) range from $20 billion to $56 billion.1-3 Given the enormous effect of HF, this disease has become a primary target of disease management (DM) programs. Previous studies4-8 have shown that DM improves clinical outcomes and reduces costs in HF. However, these studies have been limited in scope and were conducted during short periods in small homogeneous commercially insured populations. Other studies9-12 call some of these findings into question, suggesting that DM in HF can be clinically effective but is unlikely to reduce costs.

The conflicting evidence on the cost-effectiveness of DM is due in large part to wide variations in methods. A Congressional Budget Office13 review of DM programs concluded that evaluations often fail to consider the staff-related and program-related costs associated with interventions. Furthermore, the focus on emergency department visits and admissions common in the literature leaves unanswered the question of what happens to other healthcare spending such as pharmaceuticals and office visits.13 If comprehensive accounting of DM programs shows that they yield superior outcomes at lower costs, then the question of whether to implement them broadly is easily answered. However, if DM offers improved outcomes at a higher cost, the question of whether to implement it large scale is more difficult. Unless the additional cost of DM is validated by the magnitude of improvement in outcomes, payers may be unlikely to support the widespread implementation of such programs.

With few exceptions,8,14 evaluations of DM in HF have not provided formal assessments of cost-effectiveness. The present analysis explores the cost-effectiveness of a telephonic DM intervention in HF previously shown to have positive effects on clinical outcomes but not on healthcare costs.11

METHODS

The data for this analysis come from a randomized controlled clinical trial, the details of which have been previously published.11 The study was approved by all institutional review boards, and written informed consent was obtained. One thousand sixty-nine adult subjects with documented systolic HF (SHF) or diastolic HF were enrolled for 18 months per subject. Participants were randomized into 1 of the following 3 study groups: usual care, DM, and augmented DM. Subjects in the intervention arms were assigned a disease manager, a registered nurse who performed patient education and medication management with the patient’s primary care provider for the full 18-month enrollment period. Subjects in the augmented DM group also received in-home devices for enhanced self-monitoring. The devices consisted of an electronic blood pressure monitor, a finger pulse oximeter, and a wristwatch activity monitor. Blood pressure and oximetry data were transmitted to the disease manager but were not forwarded to the primary care provider. Activity monitoring data were not transmitted to the disease manager or to the primary care provider. No differences in outcomes were noted between the 2 intervention groups; therefore, the present analysis uses a pooled treatment group consisting of patients from the 2 intervention arms. Three hundred twenty-five subjects did not complete all 4 planned study visits; 93 died, and 232 were missing 1 or more visits, withdrew, or were lost to follow-up. Data analysis was by intent-to-treat. All cases were included in the estimation of cost and survival differences. Subjects who withdrew from the trial were considered censored after the date of withdrawal.

Utilization Data

Intervention cost data were based on fees paid to the DM subcontractor of the project, with additional detail on cost structure provided by the subcontractor. Based on the total cost of labor, benefits, and office and operating expenses, as well as the number of patient-months for which services were provided, the mean cost of DM services per patient per month was $246 (67% labor and benefits, 29% overhead, and 4% travel, protocol licensing, and equipment). The devices issued to half of the intervention group patients are not included in the monthly cost.

Economic Analysis

Because the intervention was expensive, we computed a separate set of ICERs using 50% ($123) of the actual monthly cost of the intervention to examine the price sensitivity of our findings. We also computed ICERs for varying utility factors and for a series of hypothetical cases in which the DM intervention reduces congestive HF–related emergency department visits and admissions.

RESULTS

Except for a statistically significant but small (approximately 2 mm Hg) difference in diastolic blood pressure, no differences were apparent between the control and experimental groups across an array of clinical and demographic variables at baseline (Table 1). Results from the analysis of the primary study end point are summarized in Table 2. Disease management produced statistically significant survival advantages among all patients (17.4 days, P = .04), among patients with NYHA class III/IV symptoms at baseline (47.7 days, P = .02), among patients with SHF (24.2 days, P = .01), and in the combined group (79.4 days, P = .04).

To define whether DM was cost-effective, we assessed the cost of life extension, quantified as the ICER. Table 4 gives the ICER data for all patients and for the subgroups for allcause and HF-specific expenditures. For all patients and considering all costs, the ICER was $146 870 per quality-adjusted life-year (QALY) gained, exceeding the standard of $100 000 considered the upper limit of a societally acceptable expenditure. 22 Of note, the ICER for HF-related costs alone was $139 133. For patients with NYHA class III/IV symptoms, the ICER was well under $100 000 in all cases, and it was just under the $100 000 mark for patients with SHF. In patients with SHF and NYHA class III/IV symptoms, the ICER for allcause care was $43 733. To explore the effect of the cost of delivery of DM, we recalculated the ICER assuming a 50% reduction and found that with that level of efficiency the ICER for all-cause expenditure for all patients dropped markedly to $79 511.

We also ran a series of analyses in which the utility factor was fixed across all NYHA classes to values between 0.6 and 1.0 inclusive (Table 4). The estimates in which the utility factor was set at 1.0 give an estimate of cost-effectiveness per unadjusted QALY gained. As the utility factor increases, the intervention is cost-effective at the $100 000 per QALY mark across all but 1 of the subsamples.

DISCUSSION

In a sample of community-dwelling patients with HF during an 18-month follow-up period, DM improved outcomes but with costs per QALY gained well in excess of $100 000. In contrast, in subgroups of patients with advanced HF (NYHA class III/IV symptoms), with SHF, or with both, our analysis shows that the intervention may produce life-span increases at costs that are below or close to the amount considered societally acceptable as cost-effective.

This is one of the first studies to assess the cost-effectiveness of DM in HF in a large sample with a follow-up period exceeding 12 months. The variation in DM interventions, cost-accounting methods, and study designs in previous studies5,6,24,25 confounds precise comparisons between the present study and existing work in this area. However, telephonic DM in this heterogeneous sample of community-based patients with HF seems to be substantially more costly than in the few studies5,6,24,25 of HF DM programs in which cost-effectiveness has been formally assessed or may be inferred.

A randomized trial of a comprehensive multidisciplinary intervention by Rich and colleagues6 in a population of recently discharged patients with HF produced notable clinical improvements and dramatic reductions in the rate of rehospitalization. Even after accounting for the cost of the intervention, DM exhibited clear dominance over usual care, yielding a net cost savings of $460 per patient during a 90-day period. Similar findings emerged from the Trans-European Network-Home-Care Management System study.26 Although a complete cost-effectiveness analysis has not yet been published (to our knowledge), results suggest that the home telemonitoring arm of this trial improved survival and reduced hospital days per patient enough to be cost saving, even after accounting for program expenses.26,27 The cost per QALY gained in our study is also well in excess of the cost reported in the analysis by Capomolla and colleagues8 of a day hospital intervention. Similarly, the Specialized Primary and Networked Care in Heart Failure trial was shown to reduce utilization, but program costs and increases in non–HF-related utilization offset the cost savings over time.14

There are several possible explanations for the contrast between the present findings and those of previous studies of DM interventions. First, these findings confirm that DM interventions have their optimum effect when targeted toward the highest-risk patients.28,29 The lower risk for inpatient events in the large group of patients with NYHA class I/II HF meant that DM was unlikely to produce notable reductions in utilization costs in the overall sample. Second, dramatic mortality reductions were unlikely to result purely as a result of bringing patients’ drug regimens into compliance with treatment guidelines given the widespread use of angiotensin-converting enzyme inhibitors and angiotensin receptor blockers in our sample.11 In this environment, the high labor costs of the intervention make achieving a cost-effective result difficult. Third, it is possible that the DM intervention tested herein is less effective than previous interventions. However, this particular intervention was chosen because it had the strongest evidence-based support at the time the trial was initiated,7,30 suggesting improper targeting rather than wholesale ineffectiveness. Moreover, the effectiveness of the intervention in the present case is supported by the clinical improvements observed in the DM group, including a mortality benefit, NYHA class improvement, and the use of guideline- based therapies (for patients with SHF).11

The intervention studied herein is also more expensive than several other therapies for HF. The DM intervention produced QALY gains at a higher cost than enalapril maleate therapy ($9700 per year of life saved),31 B-type natriuretic peptide–guided management ($7787 per QALY gained),32 implantable cardioverter-defibrillators ($34 000-$88 700 per QALY gained),33,34 and cardiac resynchronization therapy (≥$10 200).35

Our study must be interpreted in light of several limitations. First, the cost data are based in part on patient self-report and possibly underrepresent the true costs of caring for these patients. Although studies conducted among patients drawn from a single-payer or healthcare system may allow collection of more complete cost data, the fact that these patients are drawn from a wide spectrum of health plans and healthcare systems allows us to assess the effect of broad application of a DM program in a real-world setting and constitutes a strength of the study. Second, although our trial is one of the longest-duration studies conducted to date, it only followed up patients for 18 months, during which mortality remained low. Because extensive censoring, as occurred in our study, may lead to underestimation of the true mean survival time, the values we report may be shorter than the true extension of survival with HF in the experimental group. The application of decision analytic modeling techniques or a longer period of observation might produce notable improvements in the accuracy of the findings reported herein. Third, the rate of inpatient admissions for HF among the present sample is substantially lower than that reported in some previous trials.6,10,36 The distinctions of this sample with respect to inpatient admission rates and healthcare utilization may limit the generalizability of our findings to populations with substantially different patterns of inpatient utilization.

In conclusion, these data show that telephonic DM did not reduce costs and was not cost-effective in this sample. However, when targeted properly, DM seems capable of producing life-span increases at costs that are less than $100 000 per QALY gained. If the DM industry can reduce program labor costs through technological innovation, economies of scale, or competition, carefully targeted DM programs may produce cost-effective improvements in HF outcomes.

Author Affiliations: From the Disease Management Center, University of Texas, San Antonio (BS); Altarum Institute, San Antonio, Tex (BS, EF); Altarum Institute, Ann Arbor, Mich (PFH-C); and Texas Transplant Institute, San Antonio (ADG).

Funding Source: This study was supported by contract DAMD17-99- C9099 from the US Army Medical Research and Materiel Command. The disease management vendor employed in the trial had no involvement in the design, analytic, or editorial stages of the project and provided no promotional pricing consideration for its services.

Author Disclosure: The authors (BS, PFH-C, EF, ADG) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (BS, PFH-C, EF, ADG); acquisition of data (BS, PFH-C, EF, ADG); analysis and interpretation of data (BS, PFH-C, EF, ADG); drafting of the manuscript (BS, ADG); critical revision of the manuscript for important intellectual content (BS, ADG); statistical analysis (BS); provision of study materials or patients (ADG); obtaining funding (ADG); administrative, technical, or logistic support (ADG); and supervision (ADG).

Address correspondence to: Brad Smith, PhD; Altarum Institute, 3737 Broadway, Ste 205, San Antonio, TX 78209. E-mail: brad.smith@altarum.org.

1. Massie BM, Shah NB. Evolving trends in the epidemiologic factors of heart failure: rationale for preventive strategies and comprehensive disease management. Am Heart J. 1997;133(6):703-712.2. O’Connell JB. The economic burden of heart failure. Clin cardiol. 2000;23(3)(suppl):III6-III10.

4. Hershberger RE, Ni H, Nauman DJ, et al. Prospective evaluation of an outpatient heart failure management program. J Card Fail. 2001;7(1):64-74.

6. Rich MW, Beckham V, Wittenberg C, Leven CL, Freedland KE, Carney RM. A multidisciplinary intervention to prevent the readmission of elderly patients with congestive heart failure. N Engl J Med. 1995;333(18):1190-1195.

8. Capomolla S, Febo O, Ceresa M, et al. Cost/utility ratio in chronic heart failure: comparison between heart failure management program delivered by day-hospital and usual care. J Am Coll Cardiol. 2002;40(7):1259-1266.

10. Laramee A, Levinsky S, Sargent J, Ross R, Callas P. Case management in a heterogeneous congestive heart failure population: a randomized controlled trial. Arch Intern Med. 2003;163(7):809-817.

12. Mejhert M, Kahan T, Persson H, Edner M. Limited long term effects of a management programme for heart failure. Heart. 2004;90(9):1010-1015.

14. Gregory D, Kimmelstiel C, Perry K, Parikh A, Konstam V, Konstam MA. Hospital cost effect of a heart failure disease management program: the Specialized Primary and Networked Care in Heart Failure (SPAN-CHF) trial. Am Heart J. 2006;151(5):1013-1018.

16. 2003 Drug Topics Red Book. Montvale, NJ: Thomson PDR; 2003.

18. Brazier J, Roberts J, Deverill M. The estimation of a preferencebased measure of health from the SF-36. J Health Econ. 2002;21:271-292.

20. Briggs AH,Wonderling DE, Mooney CZ. Pulling cost-effectiveness analysis up by its bootstraps: a non-parametric approach to confidence interval estimation. Health Econ. 1997;6:327-340.

22. Ubel P, Hirth R, Chernew M, Fendrick A. What is the price of life and why doesn’t it increase at the rate of inflation? Arch Intern Med. 2003;163(14):1537-1541.

24. Philbin EF, Rocco TA, Lindenmuth NW, Ulrich K, McCall M, Jenkins PL; MISCHF Study Investigators. The results of a randomized trial of a quality improvement intervention in the care of patients with heart failure. Am J Med. 2000;109(6):443-449.

26. Cleland JG, Louis AA, Rigby AS, Janssens U, Balk AH;TEN-HMS Investigators. Noninvasive home telemonitoring for patients with heart failure at high risk of recurrent admission and death: the Trans European Network-Home-Care Management System (TEN-HMS) study. J Am Coll Cardiol. 2005;45(10):1654-1664.

28. DeBusk RF, Miller NH, Parker KM, et al. Care management for low-risk patients with heart failure. Ann Intern Med. 2004;141(8):606-613.

30. DeBusk RF. MULTIFIT: a new approach to risk factor modification. Cardiol Clin. 1996;14(1):143-157.

32. Morimoto T, Hayashino Y, Shimbo T, Izumi T, Fukui T. Is B-type natriuretic

33. Sanders GD, Hlatky MA, Owens DK. Cost-effectiveness of implantable cardioverter-defibrillators. N Engl J Med. 2005;353(14):1471-1480.

of a microvolt T-wave alternans screening strategy for implantable cardioverter-defibrillator placement in the MADIT-II–eligible population. J Am Coll Cardiol. 2006;48(1):112-121.

36. Kimmelstiel C, Levine D, Perry K, et al. Randomized, controlled evaluation of short- and long-term benefits of heart failure disease management within a diverse provider network: the SPAN-CHF trial. Circulation. 2004;110(11):1450-1455.