When a payer is ready to set the price of an episode or bundle, they’ve already defined the episode—maybe they’re using Bundled Payments for Care Improvement or PROMETHEUS, or even a custom episode design—and they’ve given some initial thought about how to handle risk adjustment within the bundle design. At this point, it is important to realize how every decision leads to a price for that episode of care. Whether to include a pre-trigger window, what codes to include or exclude, what is the length of the post-trigger window? All these decisions are building blocks in determining the price of the episode.
But there isn’t just one price.
When helping payer clients determine a pricing structure for an episode, Aver uses the following framework:
Working our way from the bottom to the top of this framework, a gainsharing cap is an optional part of a bundle pricing structure that can protect the interests of both patients and payers. In order to ensure that providers do not reduce costs too aggressively, a payer might establish a cap on the amount of gainsharing a provider may earn. By establishing this cap, a payer ensures that providers aren’t incented to drastically cut costs, and by extension, cut services that patients may need. A gainsharing cap also provides payers with an upper limit on their provider gainsharing budget.
Next up, the contracted price is the total budget for an episode of care, and this is usually the first element of the pricing framework to be negotiated. A contract price may be paid prospectively or retrospectively. If actual costs are lower than the contracted price, the episode initiating provider pockets the difference. If actual costs are above the contracted price and the contract includes downside risk, the episode initiating provider is responsible for the overage. The contracted price receives the most attention in any pricing structure, and we’ll take a look at how Aver helps payers determine a contracted price in a future post.
Even providers who accept a contract with downside risk are unlikely to accept responsibility for all costs above the contracted price, which is where the overage threshold comes in. Once total costs for an episode reach the overage threshold, the payer re-assumes responsibility. In this way, the overage threshold caps the provider’s risk, providing them some assurance of their total potential responsibility for each episode.
Between the contracted price and the overage threshold is the warranty. Under a warranty, an episode initiating provider agrees to accept responsibility for a more limited set of items or services included in the bundles for a period of time beyond the end of the episode. For example, if a knee replacement episode extends for 90 days post-trigger event, the surgeon may agree to warranty the knee implant for 365 days post-trigger event for an additional payment. In this case, most of their responsibility for a patient ends 90 days after the surgery, but the surgeon would be responsible for any problems with the actual implant for a year after surgery. For a surgeon who has done hundreds, if not thousands, of knee replacement surgeries and never had a problem with an implant, an additional payment may well be worth accepting some incremental risk.
The final element in Aver’s pricing framework is the stop loss level. Although a large health insurer may not need to consider reinsurance for individual episodes, given their available reserves, a self-funded employer may wish to purchase a reinsurance policy that caps their liability exposure in the event of an episode with catastrophic costs.
At this point, Aver processes historical claims data using the chosen episode model to help payers identify a potential contracted price, warranty (if any), overage threshold, and stop loss level for each type of episode. By looking at comparative statistics—mean, median, standard deviation, and more —Aver’s first step is typically to provide payers with an enterprise-level average episode cost for all the episodes of that type.
While an enterprise-level average episode cost is a useful metric, it is important to acknowledge the wide range in episode costs, even among episodes of the same type. This is where risk adjustment comes in. Adjusting an episode’s price based on a patient’s risk isn’t necessarily required, but acknowledging the effects of clinical risk factors on the expected cost of an episode of care and modifying bundle payments accordingly will go a long way towards securing their participation.
Payers committed to using risk adjustment can consider several different models. The first is to incorporate risk adjustment into the price of the bundle, so that the budget varies based on the impact of the risk factors attributable to each patient. The PROMETHEUS risk adjustment methodology identifies risk factors that may impact the cost of each patient’s episode, using historical claims data to assign a coefficient to each factor, and then applying those coefficients to the cost of care for an average patient to compute their expected cost of care. In addition to informing expected episode costs that can be used to develop a prospective budget, this risk adjustment methodology is very useful in giving a fair picture of provider performance as payers identify potential provider partners for an episodic payment program.
Incorporating complex risk adjustment into payment rates, however, can be tricky. Because there are so many factors that contribute to an episode’s risk level, it can be difficult to communicate how each factor might affect a provider's reimbursement for each new patient. More specifically, retrospective risk adjustment makes it nearly impossible for providers to predict the budget he/she will receive for a particular patient because the adjustment process happens after the episode completes.
The risk adjustment approach that Aver is seeing increasingly used is one that classifies patients into one of three tiers— low, medium, and high risk—with an episode budget for each tier. Using the risk factors identified by PROMETHEUS, review of the clinical literature, and Aver’s data analysis, payers and providers can negotiate how patients are assigned to various risk tiers. For example, patients undergoing a knee replacement may be classified based on the presence of chronic conditions and/or their BMI. Individuals with lower BMI are assigned to the low-risk tier while individuals with higher BMI are assigned to the medium- or high-risk tier. Using this tiered method increases transparency, allowing for more dynamic negotiation between payers and providers.
Regardless of methodology, however, Aver recommends using a risk adjustment mechanism that sets an episode’s budget in advance of its triggering event. This enables providers to tailor care pathways to meet the patient’s risk level, thus improving the likelihood of a successful outcome.
In addition, using historical claims analysis, payers can share with providers their typical patient risk levels, allowing providers to make investments in staff or infrastructure to help mitigate patient risk. Finally, payers and providers may choose to incorporate outcomes and quality measurement to determine a final provider payment. We will discuss outcome and quality considerations for a bundled payment program in a future post.
Over the next several posts, other Aver experts and I will share additional insights on negotiating a budget with a provider or group, incorporating quality in episode pricing, and more. Be sure to subscribe to these insights delivered to your inbox.
Want fresh insights on Value-Based Care delivered right to your Inbox? Become a member of the Aver Insights Community.