The amount a provider receives for an episode of care results from the combination of several factors: the contracted price (the topic of this post), the chosen risk adjustment methodology (discussed previously here and here), quality or outcomes-based incentives (the subject of our next post), and overage thresholds and gainshare caps (to be discussed in future posts).
A common starting point for payers seeking to establish an appropriate contract price is to determine the enterprise-level average risk-adjusted price for a given episode definition. A payer might never share this price with the providers it approaches about participating in a bundled payment program, but the information helps payers anchor future pricing in their historical experience.
Before calculating the average enterprise-level price, most bundled payment models eliminate high-cost outlier episodes. PROMETHEUS, for example, excludes the highest- and lowest-cost 1% of bundles, but I generally recommend payers further narrow the range of payments, such as by eliminating the least and most costly 10% of episodes when calculating the anchor price. Once the enterprise average is calculated, I recommend selecting one of three pricing philosophies when contracting with specific providers (selection will vary based on local market dynamics):
- Market-wide reference pricing
- Provider-specific pricing
- Combined or weighted market- and provider-specific pricing
The first option, market-wide reference pricing, uses data for a particular region or another specific geographic area, like a metropolitan statistical area (MSA), to calculate a common risk-adjusted price. This can be a difficult starting point for payers’ conversations with potential bundle providers if the providers do not fully understand or trust the chosen risk adjustment framework. Frequently, the reference price approach is met by providers’ refrain, “my patients are sicker.” However, if well executed, this strategy has can be tremendously powerful in reducing variation across the entire market.
A much easier approach to opening the pricing conversation is to use a provider’s own historical performance. This method of pricing can be a great way to get providers on board with a bundled payment program because it makes the case that a provider’s future patients will have similar risk levels to their past patients. Furthermore, there is always room for improvement in a provider’s performance, and by sharing data down to the claim line detail, a payer can both build trust with their providers and help them understand how they can improve.
One of the primary limitations of the provider-specific model is that in order to effectively employ it, payers need to limit their contracting efforts to providers who have completed a minimum number of episodes. This threshold will differ based on a variety of market factors, but I recommend that payers exclude providers with fewer than 30 episodes per year as a general rule of thumb, to ensure pricing is informed by sufficient historical experience.
Finally, a useful compromise between the previous two strategies outlined is hybrid market- and provider-specific pricing. This is the model that is used by Medicare in its Comprehensive Care for Joint Replacement (CJR) program that began in April 2016. It works by weighting the provider's experience and the market average as follows: two-thirds of the hospital’s own spending and one-third regional spending for years 1 and 2, one-third hospital-specific and two-thirds regional spending for year 3, and regional historical spending only for years 4 and 5. This model creates a glide path for providers as they move to market-based reference pricing.
At this point, the payer can start to identify where costs can potentially be reduced. This next section describes a number of options that payers and providers can consider together to generate savings without sacrificing quality:
Annual Inflation Targets — Many fee-for-service (FFS) contracts contain a pre-negotiated medical cost inflator based on historical trends. For example, a provider’s fee schedule might call for a 2-5% annual increase to keep up with the rising costs of medical technologies and wages. Because episode pricing is based on historical FFS claims, one of the easiest ways to generate tangible savings is to establish the episode rate with no, or a reduced, inflation adjustment.
For example, a pricing analysis may only include episodes triggered between January 1, 2016 and December 31, 2016. If contracts are priced based on this analysis and are effective as of July 1, 2017, running for three years (i.e., June 30, 2020), savings would be realized annually equal to the avoided inflation rate. In this case, if the expected inflation rate was 3% each year, by 2020, this contract would yield 9-10% savings compared to the baseline.
Potentially Avoidable Complications (PACs) — As defined under the PROMETHEUS payment model, a PAC is “any event that negatively impacts the patient and is controllable by providers.” When using a PROMETHEUS episode definition or a custom episode design based on a PROMETHEUS definition, PACs are already identified in the model. However, payers and providers may also negotiate the specific complications that are under a provider’s control and should be avoided.
Complications have real, sometimes life-and-death consequences for patients, and starting the pricing conversation here allows both payer and provider to focus on improving care to benefit patient health and outcomes. To relate this back to price setting, Aver analyzes a provider’s historical PAC spending rate, and the payer can then model a reduction in that spending, subtracting that amount from the provider’s historical costs.
For example, assume that a provider’s historical, risk-adjusted cost for a knee replacement, including care provided up to 30 days before and 90 days after the surgery, is $30,000. Further analysis finds that, on average, $4,000 of that cost is attributable to avoidable complications. Using Aver’s analysis, the payer and provider identify likely causes of those complications and determine ways to avoid them in the future. To incentivize this complication avoidance, the payer could begin by negotiating a 25% reduction in payment for avoidable complications. This would reduce the provider’s bundled payment rate by $1,000, to $29,000.
Readmission rates and spending — Similar to analysis conducted to identify PAC spending, Aver can also determine historical spending on readmissions: those related to the episode, all-cause readmissions, or both. The payer and provider can then negotiate a reduction in spending on readmissions and apply that reduction to the historical risk-adjusted rate for that provider.
Care Pathway Utilization and Efficiency — Aver’s solutions identify the care pathways used by individual providers, highlighting the average amount and percent of total episode spending attributable to different post-acute settings. For example, a knee replacement patient may be discharged to a skilled nursing facility, inpatient rehab, outpatient rehab, or to their home with home health services.
Here, a comparison of the provider’s performance to their region might be useful. If a provider uses more skilled nursing care than other providers in their region and less home health services, this could contribute to higher costs. Similarly, this type of pathway analysis can be used to identify cases that can move from inpatient to outpatient settings of care for certain procedures. In both of these cases, the payer and provider are identifying a portion of care that could be moved from higher cost to lower cost settings and reduce the overall episode costs.
Durable Medical Equipment — It is common for providers to be unaware of the variation in prices for durable medical equipment (DME). For example, a hospital may provide patients with a walker and commode at a cost to the payer of $400. The provider, however, can work with DME suppliers to obtain the same equipment at a cost of $200. The payer and provider may negotiate a reduction in DME spending of $100, and the provider keeps the other $100.
Applying some (or all) of these factors to the historical average price eventually leads payers and providers to a risk-adjusted contracted price. And they might stop there. It’s common to include only upside risk in the first bundled payment contract. In this scenario, if the cost of care is less than the contracted price, the provider retains the difference. The payer is responsible for any costs that exceed the contracted price.
In the next post, we will consider how a payer might offer a provider additional incentives for outcomes and quality.
Want fresh insights on Value-Based Care delivered right to your Inbox? Become a member of the Aver Insights Community.