Based on surveys of health plans by Catalyst for Payment Reform, less than 3 percent of payments in the commercial market are “bundled.” One can argue that there have been more conferences and webinars about bundled payment than actual bundled payments. In the public sector, however, there is much more experimentation with and usage of bundled payment through Medicare and Medicaid.

There is good reason to pursue bundled payment. At least in theory, a bundle represents the best thinking on how much it costs to treat the course of an identifiable illness, disease, or injury. It also addresses the tendency to inflate costs with excessive, and often unnecessary, medical intervention. But progress has been slow, in part, because of a wide array of variables that go into calculating a bundled payment, including:

  • Which costs does the bundle include?
  • Which providers get which portion of the bundle?
  • What timeframe does a bundle cover?
  • What happens with outlier cases that come in below cost or, more likely, above?

These and other questions have stunted the growth of what could be a breakout idea.

Is The Focus On Prospective Payment A Barrier?

Some believe that the holy grail of bundled payments is prospective bundles—setting a price for a course of care that has not yet occurred and paying only that amount at the start of the episode of care. The thinking behind prospective payment is that the incentives for cost containment will be stronger because providers will be loath to overspend when they know there’s no chance of additional payment. They also sound administratively simpler for the payer—just cut one check, and don’t worry about figuring out who did what. (It’s strange, however, that there doesn’t seem to be an equal push for prospective global payments. Total cost of care contracts with accountable care organizations or other health systems are almost always settled retrospectively against a target decided in advance.)

However, prospective bundled payments can be difficult to implement because they require deciding which provider receives the funds and can be trusted to distribute them properly among all providers involved in an episode. Legal and turf issues can stymie the effort.

The difference between retrospective and prospective payments is not significant enough to delay implementation of bundled payment in the hopes of perfecting prospective bundling. Payers shouldn’t go into a retrospective bundled payment arrangement without having agreed upon a budget for a procedure or other episode of care in advance, even though they may settle debts and credits on the back end. In addition, providers, like most people, have an aversion to loss, and if the payment arrangement includes downside risk, they will work hard to avoid any losses. This means that waiting for the means to implement prospective bundles on a broad scale is not sufficient cause for delaying the use of bundled payment.

Focus On The Real Problems

There are real challenges that need to be addressed in many of the alternative payment models being discussed today, including bundled payments, whether paid prospectively or retrospectively. These include:

Prospectively Setting The Price And Budget

Coming to agreement on a reasonable budget (and therefore payment amount) takes analysis and negotiation on the part of the providers and payers. Ideally, the arrangement includes a separate payment for evaluation, so providers are paid partially for helping to determine when patients do not need the care the bundle covers. And bundles won’t save payers money if the price for the bundle is too high.

Risk Adjustment

In setting a bundled payment amount, providers will want payers to consider that some patients are sicker and more complicated than others. Identifying these patients and deciding how much to adjust the bundled payment upward can be a subject of debate. The idea is to hold providers responsible for the costs they can control, not the ones they can’t.


While medical professionals might accept an occasional loss from a patient whose care costs more than the bundle payment, in the aggregate they must operate with a margin to remain viable and continue participation in the payment arrangement. Preventing intolerable losses is integral to the bundled payment strategy and establishing stop-loss insurance can give providers the confidence to participate.

Percent of Upside And Downside Risk

The amount a provider can earn on any bundle is limited by the costs of caring for a given patient. However, the amount a provider can lose is significantly higher, even with a stop loss. For example, assume a target price of $2,000 for a colonoscopy. The most a provider could earn is between $500 and $1,000, depending on the type of anesthesia used (Note 1). But the total cost of a colonoscopy episode can easily be $5,000 if there’s a redo, leading to a $3,000 loss. As such, should the risk-sharing formula be the same (for example, 50 percent of the gains and 50 percent of the losses), or should the upside share be greater than the downside share?

Quality Measures

Without a set of meaningful quality measures, bundled payments could motivate providers to cut corners to increase profits. The bundled payment arrangement can include an adjustment to the final payment amount based on whether the provider meets certain quality standards agreed to in advance. This could also help providers mitigate their losses when they have a bad financial year but good quality results.

A Reasonable Path Forward

The answers to these issues may already be out there. Public purchasers, including Medicare and several Medicaid agencies, already have large implementation efforts underway. Some vendors have also stepped up to provide pre-packaged solutions to employers and third-party administrators, providing nearly turn-key programs for implementing bundled payment arrangements.

Whether a bundled payment is made in advance or after the fact is not that important. That is an administrative decision that should be based on how well integrated the delivery system is and the ability of providers to agree on which of them takes the lump sum and delegates it to others.

What matters is the incentives being created. Setting a prospective target budget, instituting reasonable downside financial risk, and tying final payment amounts to performance on a set of outcome measures all provide powerful incentives to manage patients well while adhering to a budget.

Let’s put aside the discussion of prospective versus retrospective and continue to experiment with various ways to implement bundled payment—a method that has real potential to improve quality and reduce spending.

Note 1

Internal analysis of nationally representative commercial insurance plan data, 2016, Health Care Incentives Improvement Institute.