Chronic heart failure (CHF) is an extremely common, high-mortality and high-cost disease among older veterans. Several technologies, including devices such as implantable cardioverter defibrillators (ICDs) and cardiac resynchronization therapy-defibrillators (CRT-Ds), as well as pharmacotherapies such as the beta-blockers carvedilol and extended-release metoprolol, have been demonstrated to reduce mortality among patients with CHF. While some technologies such as ACE-inhibitors currently are used in almost all veterans with CHF, devices and beta-blockers are not yet used in all clinically eligible patients. The costs of these technologies are substantial, and thus it is important to quantify the benefits that these technologies have yielded among veterans with CHF during the past ten years relative to their costs, and it is also critical to identify existing opportunities to improve CHF health care quality at reasonable cost.
The goals of this project were: (1) to examine national trends across VA from 2001-2010 in the use of ICDs/CRTs and carvedilol/metoprolol and to determine if changes in technology use at the VAMC/VISN level were correlated with changes in CHF outcomes; (2) to measure the changes in costs of care for veterans with CHF resulting from the increased use of devices and carvedilol/metoprolol; (3) to identify opportunities for improvement in VA CHF care through greater use of these therapies, estimate the magnitude of the veteran CHF population health benefit that would result from greater technology use, and compare this benefit to the increase in costs to VA that would be necessary for full dissemination of these technologies.
This study used multiple sources of VA data describing health care utilization and costs among Veterans with CHF, including the VA's Medical SAS datasets at the Austin Information Technology Center, VA Decision Support System data, the VA Vital Status File, and the VA-Centers for Medicare and Medicaid Services datasets (managed by the VA Information Resource Center-VIReC) that provides information on Veterans dually enrolled in VA and Medicare. We identified annual cohorts of CHF patients within each VAMC and VISN from 2001-2010, and we used VA's medical procedure, surgery, and pharmacy databases to measure longitudinal trends in technology use rates, outcomes, and costs among these cohorts via hierarchical linear regression models using a "difference-in-difference" approach.
We found that VA health care costs for heart failure care related to new technologies greatly increased from 2001-2010. An illustration of this is the cost of implantable cardioverter defibrilaltors, which cost VA $42 million in 2001, but rose to $166 million in 2010. There was a dramatic increase in use of high-cost technologies (both devices and pharmaceuticals) for heart failure care between 2001-2010, and a wide variance in the rate of uptake across the VA's 140 medical centers. Using multivariable statistical models to control for differences across VAMC hospital populations, we observed that VAMCs with higher utilization rates of implantable defibrillators and evidence-based beta blockers (e.g., carvedilol and metoprolol succinate) had significantly higher costs per average veteran with heart failure than those VAMCs with lower technology use rates. However we did not observe improved heart failure outcomes among those VAMCs that used the highest amounts of new technologies. We concluded that although evidence-based heart failure therapies have been widely adopted across VA, it is possible that those VAMCs that are most "aggressive" in using new technologies may be (on the margin) selecting patients for treatment who are less likely to benefit.
This project investigated how increasing use of evidence-based pharmaceutical and device therapies from 2001-2010 among Veterans with CHF has affected clinical outcomes, and we quantified the substantial rise in VA costs associated with increasing use of these therapies. Important potential impacts to our project are:
(1) VA has generally done very well in adopting evidence-based therapies, thus providing favorable "feedback" evidence on the quality processes encouraging their use.
(2) However, it is possible that on the margin, VAMCs with very high utilization rates of these therapies may be selecting candidates for treatment who, on the margin, are less likely to benefit. This would be the most likely reason that we saw no correlation between a VA hospital's utilization rate of high-cost heart failure technologies and the VAMC's heart failure mortality rate.
We believe these findings can inform VA policymakers with information on how high-cost technologies impacted the outcomes and costs of CHF care in the recent past. As new high-cost technologies are perpetually in the pipeline and will constantly be introduced into VA clinical care, our findings provide a cautionary note on how best to manage the diffusion and use of these technologies to maximize beneficial outcomes for veterans without excessive (and potentially unnecessary) health care spending.
External Links for this Project
Grant Number: I01HX000523-01
- Kini V, Soufi MK, Deo R, Epstein AE, Bala R, Riley M, Groeneveld PW, Shalaby A, Dixit S. Appropriateness of primary prevention implantable cardioverter-defibrillators at the time of generator replacement: are indications still met? Journal of the American College of Cardiology. 2014 Jun 10; 63(22):2388-94. [view]
- Groeneveld PW. Can advanced healthcare technology save money? Circulation. Cardiovascular quality and outcomes. 2013 Sep 10; 6(5):509-10. [view]
- Kini V, McCarthy FH, Rajaei S, Epstein AJ, Heidenreich PA, Groeneveld PW. Variation in use of echocardiography among veterans who use the Veterans Health Administration vs Medicare. American heart journal. 2015 Oct 1; 170(4):805-11. [view]
- Groeneveld PW. How drug-eluting stents illustrate our health system's flawed relationship with technology: comment on "use of drug-eluting stents as a function of predicted benefit". Archives of internal medicine. 2012 Aug 13; 172(15):1152-3. [view]
- Rajaei S, Groeneveld P, Teerlink J, Whooley M, Turakhia M, Richman IB, Sahay A, Kini V, Heidenreich PA. Racial Disparities in echocardiography use within the Veterans Administration and Medicare systems. [Abstract]. Circulation. Cardiovascular quality and outcomes. 2015 May 1; 2015(8):A217. [view]
- Groeneveld PW. Implantable cardioverter-defibrillators in the VHA and their impact on heart failure costs and outcomes [Presentation to CHF QUERI Executive Committee]. 2014 Apr 1. [view]
- Groeneveld PW. Implantable cardioverter-defibrillators in VHA and healthcare cost growth: 2001-2010 [HERC Health Economics Seminar]. [Cyberseminar]. 2012 Jun 20. [view]
- Groeneveld PW. Cardiovascular devices: comparative effectiveness research and critical economic realities. Paper presented at: Duke University Clinical Research Institute Annual Clinical Medicine Series; 2010 Nov 1; Durham, NC. [view]
- Groeneveld PW, Richardson DM. Cost consequences of increasing use of implantable cardioverter-defibrillators in the Veterans Health Administration: 2001-2010. Poster session presented at: American Heart Association Quality of Care and Outcomes Research Council Annual Scientific Session; 2011 May 12; Washington, DC. [view]
- Epps KC, Holper EM, Gualano SK, Vlachos HA, Selzer F, Abbott JD, Jacobs AK, Marroquin OC, Naidu SS, Srinivas VS, Groeneveld PW, Wilensky RL. Higher rates of repeat coronary revascularizations in young women. Poster session presented at: Transcatheter Cardiovascular Therapeutics Annual Scientific Conference; 2013 Oct 29; San Francisco, CA. [view]
- Rajaei S, Groeneveld P, Teerlink J, Whooley M, Turakhia M, Richman IB, Sahay A, Kini V, Heidenreich PA. Racial disparities in echocardiography use within the Veterans Administration and Medicare systems. Poster session presented at: Quality of Care and Outcomes Research in Cardiovascular Disease and Stroke Annual Scientific Forum; 2015 Apr 29; Baltimore, MD. [view]
- Rajaei S, Groeneveld P, Teerlink J, Whooley M, Turakhia M, Richman IB, Sahay A, Kini V, Heidenreich PA. Use of echocardiography and nuclear scintigraphy for patients with heart failure in the Veterans Administration and Medicare systems. Poster session presented at: Quality of Care and Outcomes Research in Cardiovascular Disease and Stroke Annual Scientific Forum; 2015 Apr 29; Baltimore, MD. [view]
- Kini V, Heidenreich PA, McCarthy FH, Rajaei S, Epstein AJ, Groeneveld PW. Variation and volume of cardiac imaging in the Veterans Health Administration versus Medicare. Paper presented at: Quality of Care and Outcomes Research in Cardiovascular Disease and Stroke Annual Scientific Forum; 2015 Apr 30; Baltimore, MD. [view]