Skip To Content

Voices On Value RSS Feed

CBO report: Silver Lining for controlling health care costs

This week, the Congressional Budget Office (CBO) released an analysis of 10 Medicare demonstration projects undertaken over the last 20 years. All were designed to save the program money, but only one succeeded in doing so. Do these findings mean we should abandon efforts to redesign our country’s health care payment and delivery systems?

Not at all. In fact, when you look below the surface of the CBO report, you reach precisely the opposite conclusion. The reason most of these pilots did not achieve their desired goals is because they were built upon our existing fragmented delivery and fee-for-service/pay-for-piecework system—a system that incents more, not better care, pays a second time for avoidable complications and provides no and incentive for care coordination and better outcomes.

Indeed, CBO says explicitly:

Demonstrations aimed at reducing spending and increasing quality of care face significant challenges in overcoming the incentives inherent in Medicare’s fee-for-service payment system, which rewards providers for delivering more care but does not pay them for coordinating with other providers, and the nation’s decentralized health care delivery system, which does not facilitate communication or coordination among providers.

It is instructive to note that, of the four value-based-purchasing demonstrations studied, the only one to demonstrate cost savings was the one that didn’t use fee-for-service: bundled payments for heart bypass surgery. The beauty of bundled payments is that they align incentives between physicians, hospitals and health insurers by making a lump sum payment to cover all services related to a given procedure (e.g.  pre-surgery, surgery hospital and follow-up care); they also include quality targets and cover costs of complications for at least 30 days. Because physicians and hospitals must meet both cost and quality targets, hospitals negotiate lower prices for supplies, physicians coordinate more closely on inpatient and follow-up care. And ultimately, it is the patients who benefit.

While CBO does not examine quality outcomes explicitly, it’s important to recognize that the bundling demonstration showed important quality improvements: both mortality rates and average length of hospital stay declined for patients in the pilot.

Medicare isn’t the only entity to demonstrate important quality improvements and cost reductions from bundled payments. A similar program at the Geisinger Health System in Pennsylvania reduced complications, infections, length of stay and hospital costs by meaningful amounts and provided a “warranty” for any avoidable complications. The incentive is to get it right the first time. Bundled payments for orthopedic surgeries at Dr. Ingham and Johnson Medical Center in Texas reduced potentially avoidable complications (e.g., post-surgical infections) and reoperations—again, leading to reduced costs.

While it is true that some other bundling initiatives (e.g., a Medicare cataract surgery pilot) haven’t demonstrated similar success, the improvements noted above illustrate the value this approach can add to our health care system. That is why CIVHC is actively engaged with health plans, hospitals and physicians in facilitating bundling pilots in Colorado.

The CBO analysis also points to lessons learned from other demonstration projects, which back up the need to fundamentally reshape our health care system:

  • Gather timely data on the use of care, especially hospital admissions.
    • Providers can’t manage care effectively without robust, timely data from health insurers showing them the volume and cost of services they’re providing and mapping those against quality indicators. That’s why tools such as the Colorado Regional Health Information Organization and Quality Health Network (information highways allowing health care providers to share data with each other) and the All Payer Claims Database (a secure warehouse of aggregated, encrypted, deidentified health claims data that will let us see cost and utilization patterns) are so essential, and why CIVHC’s vision for Colorado is based on using data to improve health care, health outcomes and control costs.
  • Focus on transitions between care settings.
    • Improving the hand-offs of patients from one care setting to the next work is crucial to minimizing avoidable hospital readmissions—and, when we do that, we reduce costs and improve patients’ outcomes. Colorado is a national leader on this issue, with effective initiatives in place under the leadership of the Colorado Foundation for Medical Care and its Northwest Care Transitions project and its , the Colorado Hospital Association the and University of Colorado’s Dr. Eric Coleman and his team. CIVHC is coordinating efforts among these and other organizations working on improving care transitions and handoffs.
  • Use team-based care.
    • As a patient, don’t you want to know that your doctors and nurses are working as a team? That means not only coordinating care between office-based clinicians and hospitals, but also between primary care physicians and specialists, and using teams that combine different skill sets and licensure levels to ensure patients receive timely, high-quality care from appropriate providers. It also means deploying nurse care coordinators to work closely with physicians and patients, following up with patients to monitor medication compliance, track symptoms, etc.—a team-based approach that improves patients’ health care experience and outcomes. As the CBO report notes, the prevailing fee-for-service model doesn’t pay for that coordination. But here again, Colorado is already a leader, with health plans, Medicaid, HealthTeamWorks and others  jointly piloting a patient-centered medical home initiative that provides care coordination payments, and applications from those same public and private payers for a new Comprehensive Primary Care Initiative that will expand use of that model.

Change isn’t easy. Medicare blazed the trail with their pilots, and we can learn valuable lessons from what worked—and what didn’t—in their demonstration projects. We know our current system is broken. It is extravagantly expensive, fragmented and ill-equipped to provide the quality care we all want. We know we need new approaches to better integrate care and pay our providers for delivery quality, not simply volume. Let’s use the CBO findings to accelerate change for the better.

Share on: DiggFacebookRedditDeliciousTwitter

Comments

Blog post currently doesn't have any comments.