What Happens After the Incentives Change?

Much of the discussion around health care reform eventually arrives at one of two propositions:

“We need to align the incentives.” or “This changes the system.”

Infusion site-of-care economics provides a useful way to examine both.

The first two supplements focused on the purchaser. A purchaser can identify a higher-priced treatment setting, determine what portion of the opportunity is clinically and operationally movable, estimate the expected price reduction, and later measure whether the projected savings were realized.

That analysis can be entirely valid.

But once the purchaser changes the economic incentives, the analysis is no longer limited to the purchaser. Hospitals, physician groups, alternative infusion sites, health plans, drug manufacturers and other participants operate within the same economic environment. Changing the value of one activity changes the incentives facing the other participants.

What economic result remains after the other participants respond?

Changing incentives changes the game

The hospital-outpatient infusion differential does not exist independently of the market around it.

Hospital systems may have greater negotiating leverage than independent physician practices because of consolidation and market concentration. Hospital outpatient services can also carry facility reimbursement that is unavailable in physician offices.

The RHV discussion describes another part of the same economic structure. Providers may resist releasing patients from an existing treatment relationship, and the guests describe plan design, network structure and alternative infusion arrangements as ways of overcoming that resistance. Some of what appears operational is therefore also economic.

If an intervention removes profitable infusion volume from a hospital or physician group, the intervention has changed that counterparty's payoff. The relevant response may be to accept the lost revenue. But it may also be to change pricing, contracting, ownership, service mix or other behavior.

The sources establish that these incentives exist. They do not establish which response will dominate or how large it will be.

Purchaser savings and system savings are different propositions

Suppose an employer successfully moves an infusion from a hospital outpatient department to a lower-priced setting.

The employer may have produced a genuine purchaser savings result.

That result does not by itself tell us what happened to total economic cost.

At least three different types of economic change are possible.

Cost removal - Fewer economic resources are required to produce the care. Lower acquisition costs, less expensive infrastructure, reduced administrative intensity or a more efficient delivery model can reduce the underlying cost base.

Cost transfer - The financial burden shifts among participants without an equivalent reduction in the resources consumed. A purchaser may pay less while the provider absorbs the reduction, or some portion of the economic burden may ultimately be borne elsewhere.

Resource reallocation - Resources remain in the system but are used differently. Infusion volume may shift from hospital facilities to ambulatory centers or the home. Nursing labor, pharmacy capacity, capital and margin may move with it rather than disappear.

These outcomes are not mutually exclusive.

A site-of-care intervention could remove some costs, transfer others and cause still other resources to be redeployed. The phrase “savings” can refer to several different economic objects. A purchaser can save money even if the broader system has primarily experienced a transfer. Conversely, a delivery model can become more resource-efficient without every purchaser immediately capturing the resulting economic benefit.

Other participants respond to changed incentives

Changing an incentive can make a desired behavior more likely, but it does not tell us what the affected participants will do after the incentive changes.

A purchaser may make hospital outpatient infusion less attractive by restricting the network, changing authorization rules, offering member incentives or steering treatment toward another site. Providers may then respond to the changed economics.

• An alternative infusion site that gains volume may expand capacity.

• Its prices may also rise.

• A hospital system may purchase or affiliate with an alternative site.

• Contract negotiations may change.

• Providers may alter which therapies they administer, how they source drugs or where they direct patients.

None of those responses is established as the inevitable consequence of site-of-care optimization. But they illustrate why “align the incentives” is the beginning of an economic analysis rather than its conclusion.

When incentives are changed, the new incentive structure may or may not produce the economic result the purchaser expected after the other participants have had an opportunity to respond.

The lower cost alternative can change

The prior supplement showed that persistence depends partly on whether the alternative site remains economically favorable.

EBRI provides a useful empirical example. For a consistent group of physician-administered drugs, the median HOPD markup declined from 98% in 2019 to 70% in 2024. But the narrowing was driven substantially by rising physician-office reimbursement rather than falling HOPD reimbursement.

The observation establishes only that the economic relationship between sites can change. It provides no basis for attributing the increase in physician-office reimbursement to purchaser steering.

A purchaser that underwrites a savings opportunity from a point-in-time differential should therefore be cautious about treating the alternative price as fixed.

The market response question goes beyond simply observing that the differential narrowed.

It asks why it changed, who responded, and whether those responses were related to changing volume, ownership, contracting leverage or some other market force.

The available sources do not answer those causal questions.

Local optimization and system change operate at different scales

This is where the claim that an intervention “changes the system” requires another evidentiary step.

An individual employer may identify a site-of-care opportunity and capture legitimate savings. That does not necessarily change the underlying market. The intervention may simply be too small relative to the revenue of the affected hospital system to alter its behavior materially.

A larger purchaser, health plan, coalition or coordinated group of employers may have more economic gravity.

The scale of change matters with regard to impacts.

Local purchaser optimization - whether one purchaser can improve its own economic position.

Coordinated market change - whether enough purchasing behavior changes to alter the incentives facing counterparties.

System change - whether the resulting responses materially change prices, capacity, ownership, resource use or total economic cost.

Those propositions require progressively broader evidence.

A collection of successful local interventions may eventually contribute to systemic change. But the transition should not be assumed merely because the interventions point in the same direction.

Treatment and product changes complicate the analysis

Site of care is only one source of change in infusion economics.

Biosimilars can alter drug acquisition cost and provider margin without changing the treatment site. EBRI's prior work found that biosimilars generally reduced acquisition costs while higher HOPD reimbursement could offset part of that advantage.

New formulations or movement from infused to oral therapies can change the relevant patient population more fundamentally.

These changes can alter the value of the original site-of-care opportunity and the incentives facing providers.

A hospital responding to reduced infusion volume in a market with rapid biosimilar adoption may face a different economic problem from a hospital responding in a market where the drug mix is stable.

That is another reason the system should not be treated as a fixed background against which the purchaser acts.

The intervention and the environment can change at the same time.

What can a purchaser reasonably observe?

This is a less mature analytical space than the purchaser underwriting and measurement questions addressed in the first two supplements.

Claims data can show that an infusion moved and what the purchaser paid before and after.

They are much less capable of establishing whether a hospital later changed another negotiated rate because of that intervention, whether an alternative site repriced because demand increased, or whether resources were removed rather than reallocated elsewhere.

A purchaser interested in those broader questions could examine changes over time in alternative-site reimbursement, provider ownership and affiliation, contracting terms, utilization across sites, and total spending relationships with important systems.

But even then, causal attribution would remain difficult.

A later hospital price increase, by itself, provides insufficient basis for attributing the increase to site-of-care optimization.

The evidentiary burden rises as the claim becomes broader.

Evidentiary ceiling

Site-of-care optimization can produce a real purchaser savings result while the broader economic effect remains unresolved. Some of the effect may represent cost removal, some cost transfer, and some resource reallocation. Hospitals, physicians, alternative sites and other participants may also respond in ways that alter the original price relationship or economic opportunity.

Changing incentives sets those responses in motion. Understanding the broader effect requires observing what remains after the affected participants adjust.

EBRI Issue Brief Source: Fronstin, Paul, and M. Christopher Roebuck, “Location, Location, Location: Spending Differences for Physician-Administered Outpatient Medications by Site-of-Treatment,” EBRI Issue Brief, no. 666 (Employee Benefit Research Institute, September 24, 2026)