Why American Health Care Keeps Producing Outcomes Nobody Wants

A couple in Tennessee discovered that one of their blood pressure medicines cost less if they paid cash at a local pharmacy than if they used their insurance. Paying cash, Charlie Joseph could get a 90-day supply for about what his CVS Caremark plan charged him for 30 days. The couple’s frustration with pharmacy benefit managers crossed party lines: she was a Republican, he was a Democrat, and both found the pricing hard to make sense of. Their story was reported by KFF Health News.

It does not tell us how every plan or drug is priced. But it captures a problem that millions of people recognize: the price at the counter can be surprising even when a person has insurance and the drug is not rare or experimental.

If the system’s purpose is to help people get the care they need, why can a patient find a cheaper price by stepping outside the insurance benefit that was supposed to help pay for it?

The answer is not simply that someone made a mistake. Nor is it enough to say that one company is greedy, one profession is self-interested, or government has too many rules. These explanations may identify real problems, but they do not explain why similar frustrations keep showing up in hospitals, insurance, prescription drugs, and public programs.

A more useful place to begin is with incentives: the rules, contracts, and payment arrangements that shape what people and organizations are rewarded for doing. Those incentives do not tell the whole story, and they do not excuse harmful behavior. They help explain how a system can keep producing results that look irrational to patients even while many of the people and institutions inside it are acting rationally by their own measures.

A system assembled over time

American health care was not designed all at once. It was assembled through successive decisions: employer benefits expanded, public insurance programs were created, private plans developed their own contracts and networks, and new rules were added to address gaps or curb costs. Each layer solved some problems and created new ones. The result is less like one machine with a single control panel and more like a city built over centuries, with old roads, new highways, tunnels, and detours sharing the same crowded map.

Employer-sponsored health insurance became especially important during the twentieth century. Tax treatment and workplace arrangements helped make insurance a common part of compensation. Medicare and Medicaid later extended coverage to groups that private employment-based insurance did not reliably cover. The Affordable Care Act added new coverage pathways and rules for private insurance. None of these changes erased the systems that came before. They were built around and on top of them.

That history matters because the organization that pays for care often differs from the person who receives it. A worker may choose from plans selected by an employer. A patient may receive care from a hospital that negotiates with multiple insurers. The insurer may hire a PBM to manage prescription benefits. The pharmacy may have separate contracts with several PBMs. A government program may set its own eligibility and payment rules. The patient sees one episode of care; the money and authority are spread across many institutions.

In ordinary shopping, people can often compare prices before deciding what to buy. Medical care rarely works that way. An emergency does not wait for a consumer to compare hospital rates. A patient may not know which clinician, facility, laboratory, or anesthesiologist will be involved until after care begins. Even when a service is planned, the final bill can depend on the details of the plan, provider network, negotiated rate, deductible, and coinsurance.

Insurance is meant to pool risk. Most people cannot predict when they will need costly treatment, so they pay premiums to share that risk with others. The pool protects a person from having to pay the full cost of a major illness on their own. But insurance also creates distance between the patient, the price, and the organization paying the bill. That distance is not necessarily a flaw; risk pooling is the point. It does mean that the patient’s experience cannot be understood by looking only at the posted price or only at the plan’s total spending.

There is no single “health care price.” A hospital may list a charge. An insurer may negotiate another amount. A public program may use a payment schedule. A patient’s portion may be based on a deductible or a percentage of a plan’s allowed charge. A manufacturer may set a drug’s list price, while rebates and other concessions change what a payer ultimately spends. Those amounts are connected, but they are not interchangeable.

This is why people can talk past one another. A hospital may say it received only a fraction of its posted charge. An insurer may say it negotiated a discount. A patient may still face a substantial bill. Each statement can be true at the same time.

Who benefits from a discount?

Prescription drugs provide a particularly clear view of these competing prices. A manufacturer’s list price is the starting point for some negotiations and can affect what patients pay when their cost sharing is based on that price. But list price is not the same as the net amount received by the manufacturer after rebates and discounts. Nor is it the same as what a pharmacy pays a wholesaler or what an insurer pays under a contract.

46brooklyn Research has spent years examining public Medicaid and Medicare Part D data to show how drug prices vary across products, plans, and payment arrangements. Its work on benchmarks such as Average Wholesale Price, Wholesale Acquisition Cost, and the National Average Drug Acquisition Cost makes an important point: each benchmark measures something different. AWP, despite its name, is not a survey of the average price pharmacies pay. NADAC uses pharmacy acquisition-cost survey data and offers a different view of what pharmacies spend to obtain drugs. A benchmark can still shape reimbursement even when it does not closely track a pharmacy’s actual cost.

For patients, the distinction between gross and net prices can matter immediately. Some people pay a copayment, a fixed amount for a covered drug. Others pay coinsurance, a percentage of the drug’s price, or the full negotiated amount until they reach a deductible. If a rebate is paid later to a plan or another intermediary, it may reduce the payer’s overall costs without reducing the amount a patient pays at the counter for that prescription. The effect depends on the benefit design and how the rebate is used.

Drug Channels estimated that gross-to-net reductions for brand-name medicines totaled about $334 billion in 2023. That number does not mean PBMs pocketed $334 billion. The estimate includes different kinds of manufacturer rebates and discounts, Medicaid rebates, and 340B discounts. The point of the estimate is that the difference between list prices and net prices is large and distributed among many parties. The more the system depends on confidential rebates and fees, the harder it can be for patients and employers to see who received which discount and whether it lowered the cost that matters to them.

This structure can create a strange incentive. A plan may prefer a drug with a high list price and a large rebate if the rebate makes the plan’s net spending look lower. Yet a patient whose coinsurance is based on the pre-rebate price may pay more out of pocket. In that situation, the payer can benefit from a discount that does not reach the patient at the moment of purchase.

Not every rebate arrangement works this way, and rebates can reduce overall premiums or plan costs. They can also help manufacturers compete for preferred placement on a formulary. The key question is who receives the benefit and when. A discount that lowers total spending for a plan is not automatically a discount at the pharmacy counter.

PBMs, or pharmacy benefit managers, sit in the middle of many of these transactions. They administer drug benefits for employers, insurers, and public programs. Their work can include negotiating with manufacturers, designing formularies, processing claims, and setting network terms with pharmacies. By combining the purchasing power of many covered people, PBMs may help payers negotiate lower costs and manage complex benefits.

At the same time, the largest PBMs are connected to insurance companies and pharmacies, and some own or are affiliated with the pharmacies where their members fill prescriptions. Vertical integration can make it easier to coordinate services and reduce administrative friction. It can also create conflicts: the same corporate family may decide which drugs are preferred, what a competing pharmacy is paid, and where a patient is steered to fill a prescription.

That does not prove every decision is designed to favor an affiliated business. It does mean the incentives deserve scrutiny, especially when plan sponsors and patients cannot easily inspect the underlying contracts. A contract can guarantee a rebate or charge a spread between what the PBM bills the plan and what it reimburses the pharmacy. The employer may see a final price without seeing every transaction that produced it. Policy debates that focus only on rebate pass-through can miss other sources of PBM revenue or other ways a contract can shift costs.

A 2023 survey of employer benefit arrangements analyzed by Drug Channels found that employers reported different approaches to rebates and spread pricing. The survey was limited and does not represent every employer or PBM contract, but it illustrates a crucial point: plan sponsors themselves do not all buy the same arrangement. A reform that forces one payment design may help some employers while creating costs or tradeoffs for others. It may also fail to ensure that savings reach the patient whose prescription generated the rebate.

A similar question appears in pharmacy reimbursement. Pharmacies have to buy medicines before they dispense them. Their final payment may depend on acquisition costs, negotiated formulas, dispensing fees, and later adjustments. Before a CMS policy change took effect in 2024, the term Direct and Indirect Remuneration, or DIR, became associated with price concessions assessed after a pharmacy had filled a Medicare Part D prescription. Some adjustments were tied to performance measures intended to encourage quality. But if a pharmacy could not predict its final reimbursement when it accepted a claim, it could be difficult to know whether the transaction would cover the drug and the work required to dispense it. Since 2024, Part D plans have been required to apply pharmacy price concessions to the negotiated price at the point of sale, changing that part of the payment design.

That pressure can affect a pharmacy’s finances and its willingness to participate in networks. It is not a complete explanation for why pharmacies close. Labor costs, rent, competition, retail sales, corporate strategy, and changing patient behavior also matter. Still, payment rules have real consequences. A contract term that looks like a small adjustment to a payer can become a large cash-flow issue to a pharmacy operating on thin margins.

Hospitals are not simple businesses either

The same incentives show up in hospitals, where a patient’s need for care meets an institution’s need to remain financially viable. Hospitals have to keep staff and equipment available around the clock, maintain facilities, meet regulatory requirements, and provide services that may not be profitable. Emergency departments and trauma services cannot be turned on only when demand is high. Hospitals also treat patients who cannot pay the full cost of care.

The way hospitals are paid can reward some activities more clearly than others. Under fee-for-service, hospitals and clinicians are generally paid for delivering specific services. This makes it possible to bill for a defined procedure or visit, but it can also reward volume more directly than prevention or coordination. A payment model that pays for each service may encourage more services, even when no one sets out to provide unnecessary care. Efforts to pay for quality or outcomes try to change that signal, but measuring quality is difficult. If payment depends on a small set of reported measures, organizations may invest heavily in documenting those measures or avoid patients whose care is harder to predict.

Hospital mergers are another example of incentives with more than one side. Consolidation may allow a health system to share technology, administrative functions, and staffing across sites. In a struggling rural area, a merger might keep services open or provide access to resources a small hospital could not afford alone. Hospital associations often make this case, and it deserves to be considered when judging an individual transaction.

But consolidation can also reduce competition. Research reviewed by Penn’s Leonard Davis Institute and KFF has found that hospital mergers often lead to higher prices, without clear evidence of better quality. When a health system gains market power, insurers may have fewer alternatives during negotiations. Higher negotiated prices can raise premiums and public spending. The effect can be felt well beyond the hospitals involved, because employers and families share the cost through insurance premiums and taxes.

The policy question is not simply whether mergers are good or bad. It is what problem a merger claims to solve, whether the expected benefits are likely to reach patients, and what happens to prices and access when local competition disappears. A merger might preserve a service in one market and raise prices in another. Evidence should be examined market by market, while the incentives of both the merging institutions and their critics remain visible.

Coverage changes the choices people can make

Insurance coverage also changes whether people can seek care before a problem becomes urgent. Medicaid expansion under the Affordable Care Act offers one example of how national rules and state decisions shape access. States made different choices about expansion, and those choices affected who could qualify for coverage. KFF’s state-by-state reporting documents how the map of eligibility and coverage has differed across the country.

When coverage is available, people may use more outpatient and preventive services and have a better chance of managing conditions before they become emergencies. Coverage does not automatically create enough local clinicians, shorten appointment waits, or make every treatment affordable. Insurance is a financial doorway; the care system still needs staff, facilities, and payment arrangements on the other side.

This helps explain why debates about coverage can become frustrating. One person may focus on the moral and financial harm of being uninsured. Another may focus on taxes, payment rates, or wait times. These are not necessarily arguments about the same question. A policy can expand eligibility and improve financial security while leaving gaps in provider availability or continuity of care. It can also produce different effects in urban and rural communities, in states with different labor markets, and among people with different medical needs.

People respond to the way care is paid for, too. If a medicine has a high copayment, a patient may delay filling it, split doses, or decide it is not affordable. If a plan requires prior authorization, a clinician or pharmacist may spend time assembling documentation before a patient can start treatment. These rules can be designed to prevent waste or ensure that a treatment is appropriate. They can also shift administrative work and uncertainty onto patients and care teams.

No one policy rule acts alone. A patient may face a high deductible, a narrow network, an authorization delay, a provider shortage, and a complicated billing process at the same time. Each obstacle may have a separate rationale. The combined experience can make care feel inaccessible, even for a person who technically has insurance.

Why the easiest explanation is rarely enough

When the system produces a bad outcome, it is tempting to identify one villain and stop there. That can be emotionally satisfying. It is also a poor guide to reform if the behavior is rewarded by the rules in place.

A pharmaceutical company can seek a high price while pointing to the cost and risk of research. A PBM can negotiate rebates while also earning revenue from complicated contracts. An insurer can restrict coverage to control spending while shifting the burden of paperwork to clinicians and patients. A hospital can merge to survive and gain bargaining leverage at the same time. A public program can protect taxpayers by limiting payment and leave providers reluctant to participate. These are not equivalent actions, and they do not all carry the same consequences. But they show why a system can produce harmful outcomes without a single command center directing every result.

Some people in health care do act in bad faith. Some firms exploit market power. Some organizations hide information that patients and purchasers need. The incentive lens does not erase responsibility. It helps separate deliberate misconduct from predictable responses to a poorly designed rule—and tells us where a policy change might matter.

This lens also makes it harder to promise that one reform will solve everything. More transparency may help employers compare contracts, but the data must be understandable and comparable. Ending rebates may change list prices and the way manufacturers compete, but it may not guarantee lower costs for every patient. Directly setting payment rates can restrain spending but may put pressure on hospitals or pharmacies that already operate close to the margin. Encouraging competition can improve choices where real competitors exist; it cannot create a second hospital or pharmacy overnight in every community.

Reform is not only about choosing between markets and government. Every health system makes decisions about what is covered, how providers are paid, who bears financial risk, and what happens when resources are limited. Some countries set prices publicly; others rely more on private negotiation. None can avoid tradeoffs. The American system’s challenge is that its decisions are often dispersed across contracts and programs, making it harder for patients to see who made a choice or how to appeal it.

A better debate starts with concrete questions. What behavior does this payment reward? Who receives the savings? Who takes on the risk? What happens to the patient who needs the service but does not fit the average case? Does the reform change the underlying incentive, or only move the money to a different part of the chain? How will we know if access, quality, and affordability improved?

These questions are less satisfying than a slogan. They are more likely to lead to a reform that survives contact with the real world.

Follow the money, then follow the patient

The Tennessee couple’s pharmacy experience is a useful place to begin, not because it proves that all insurance is bad or that every PBM is acting improperly. It shows that the word “price” can hide a complicated set of decisions. The patient compared two ways to pay and found an outcome that did not match the promise he associated with his insurance benefit.

That kind of surprise is worth investigating. It asks us to follow the transaction: who set the list price, what the plan negotiated, whether a rebate was paid, how the pharmacy was reimbursed, and what the patient owed at the counter. Then we can ask what each party was trying to accomplish and what the rules rewarded.

The same approach works for a hospital bill, a denied claim, a pharmacy closure, or a prescription that costs more under insurance than it does in cash. Start with what happened to the patient. Then trace the decisions and incentives that shaped the result. Keep the evidence close. Check whether the explanation applies beyond one case. Look at who benefits from a proposed fix as well as who bears its costs.

Health care will always require judgments about risk, value, and limited resources. But patients should not have to reverse-engineer a maze of contracts just to understand why a bill or prescription costs what it does. Better rules should make those decisions clearer, put incentives closer to the outcomes we value, and make it possible to tell whether care is working for the people who depend on it.

The system is not irrational because every person inside it has stopped thinking rationally. It is complicated because many people and institutions are responding to different rules at once. To change what patients experience, we have to understand those rules—and change the incentives that keep producing the wrong results.