One of the most talked-about new ideas in health care is rewarding providers for reducing medical spending by giving them a share of the net cost savings. Driven by an interest in seeing medical homes and other providers shift to some form of performance-based payment, as well as by the Affordable Care Act’s push for "accountable care," shared-savings approaches are currently being tested by numerous payer and provider organizations across the United States.
Given uncertainties regarding an ideal shared-savings approach and how best to implement it, we examined six shared-savings pilot initiatives. For each case study, we interviewed leaders at payer and provider organizations and state agencies about their attempts to design and implement shared-savings programs.
These pilot programs, we found, vary considerably on several dimensions (Exhibit ES-1). These include the patient populations subject to shared-savings arrangements, the health care services those arrangements cover, how payers determine cost savings and payouts to providers, whether the model incorporates performance targets, and how it measures performance. These pilot projects also vary in their early impact on health care costs and payouts to providers. For example:
- One initiative measures cost savings related to preventable complications from specific procedures, and is on track to make a substantial payout to providers.
- A second initiative does not require providers to serve a minimum number of patients to participate, uses a control group and 21 quality measures to determine payouts to participating providers, but has not yet demonstrated cost savings.
- A third initiative requires providers to serve a minimum number of patients to participate, uses the average per-patient cost of health care in a metropolitan area as a benchmark, and has paid out up to 75 percent of shared savings to a provider.
Despite these variations, the case studies reveal consistent themes regarding shared-savings approaches to payment for health care services. These common elements include a willingness among most payers to absorb many of the costs entailed in setting up and sharing tools for measuring health care performance and cost savings. Overall themes also include a belief that shared-saving programs must evolve to include shared risk, and a conviction that even when pilot programs fail to achieve savings, they are moving in the right direction.
We do not yet know whether this approach is a long-term strategy for promoting better health care while lowering costs, or a transitional strategy to some other model, such as global payments for which the provider also assumes risk if spending is higher than a budget target. Exploring the organizational and environmental differences in how participants pursue shared-savings approaches, and the outcomes they achieve, will be key to determining whether they work, how to improve them, and whether and how to diffuse them.
|
EXHIBIT ES-1. KEY DESIGN FEATURES OF SHARED-SAVINGS PILOT PROGRAMS |
|||
|
MARYLAND MULTI-PAYER PATIENT-CENTERED MEDICAL HOME PROGRAM (MMPP) |
MEDICA AND FAIRVIEW HEALTH SERVICES |
HEALTH CARE INCENTIVES IMPROVEMENT INSTITUTE (PROMETHEUS PAYMENT) |
|
| Patients, Services, and Payments | |||
| Patient population(s) |
Commercial; Medicaid |
Commercial |
Commercial; Medicare Advantage |
| Methodology for attributing patients to provider groups |
12-month claims history Most frequently visited provider (most recent if tied) |
Johns Hopkins ACG software |
Not required |
| Type of payments |
Semiannual lump-sum payment Incentive payment based on |
Per-patient, per-month management fee for disease Shared savings based on comparison with a control group |
Shared savings for reducing potentially avoidable complications (PAC) associated with treating a chronic condition |
| Special Adjustments to Shared-Savings Calculations | |||
| Adjustments for patient risk |
None at the time |
ACG grouper High-cost claims truncated at $250,000 to $500,000, depending on contract |
Stop-loss provision High-cost claims (episodes exceeding two standard deviations) truncated |
|
Supports for providers |
With state funding, University of Maryland and Johns Hopkins School of Medicine provide |
Data support and analysis Some care management and health coaching may be available |
Tools and services to help payers and providers use the Prometheus Payment software and share best practices |
| Inputs for Calculating Savings | |||
|
Minimum number of patients per practice group |
None |
15,000–20,000 member-months |
Minimum number of patients with certain conditions and treatments |
|
Retention of initial percent of savings by payer |
None |
None |
Payer takes upfront share (discounted PAC allowance) |
| Method for Calculating Savings | |||
| Comparison with: |
External benchmark (TBD) |
Control group |
Budget |
| Method for Distributing Savings | |||
| Percent cost savings eligible to providers |
30%–50%, depending |
Up to 75%, split among providers: one-third to hospital; one-third to care management; one-third to physician groups |
Varies by payer, but PAC allowances |
|
Performance targets |
21 quality measures; and reductions in use of high-cost services, such as emergency department and hospital readmissions |
Minimum quality gate, then confidential algorithm |
Must achieve net 6% reduction in cost of PAC to receive payout Quality scorecard available for use |
|
EXHIBIT ES-1. KEY DESIGN FEATURES OF SHARED-SAVINGS PILOT PROGRAMS (CONTINUED) |
|||
|
BLUE CROSS BLUE SHIELD |
HEALTHPARTNERS |
HARVARD PILGRIM HEALTH CARE |
|
| Patients, Services, and Payments | |||
| Patient population(s) |
Commercial |
Commercial; Medicaid (two of the latter providers opted to participate, but with separate calculations) |
Commercial |
| Methodology for attributing patients to provider groups |
Jointly developed confidential methodology with a two-year look-back and tie-breaking algorithm |
Retrospectively attributed using internally developed algorithm |
Prospectively assigned using internally developed algorithm |
| Type of payments | Shared savings based on comparison with a control group | Shared savings based on comparison with a negotiated target | Shared savings compared with a budget |
| Special Adjustments to Shared-Savings Calculations | |||
| Adjustments for patient risk |
DxCG software Services excluded: transplantation |
ACG software High-cost claims truncated |
DxCG software High-cost claims (>$50,000) truncated Services excluded: transplantation, behavioral health, and out-of-area services |
| Supports for providers |
Reports, data feeds, and software Analytical support for targeted |
Detailed analysis of performance, including cost and use of services, location of services, medical conditions, and specialty Some grant money to help build |
Basic reporting tools for tracking performance; access to best-practice examples Care management tools, hospital utilization review, and in-depth reports available for extra fee |
| Inputs for Calculating Savings | |||
| Minimum number of patients per practice group | None, but only provider groups with >100,000 members are participating | None, but only large provider groups are now participating | 3,000–5,000 patients |
| Retention of initial percent of savings by payer |
Yes (amount confidential) |
None |
2% |
| Method for Calculating Savings | |||
| Comparison with: |
Control group: |
Jointly negotiated target |
Budget |
| Method for Distributing Savings | |||
| Percent cost savings eligible to providers |
Up to 50% |
50% (commercial and Medicaid savings calculated separately) |
50/50 split of savings up to 6% of total budget, after plan takes 2% (3% max for provider group) |
| Performance targets |
Minimum quality gate includes |
None |
No, but separate pay-for-performance program |