💊 The Prescription Middlemen

Editors’ Note: Our website features a healthspan-focused job board, an archive of past newsletters, a growing dictionary of terms we’ve highlighted in previous editions and more to come!

Visit us at: https://stacked-health.com/

Or copy and paste this link to others: {{rp_refer_url}}

A patient in Ohio gets a prescription for Zepbound. Before the first dose reaches her, it will pass through a manufacturer, a drug wholesaler, a pharmacy, a pharmacy benefit manager, that manager's rebate-negotiating arm, an insurer, and the employer that hired the insurer. Seven organizations touch one weekly injection, and each one takes a cut.

The popular instinct is that all of them are skimming and the fix is to cut them out. That instinct is about half right. Some of these companies do work that would have to be reinvented the day they disappeared. Others earn more when the drug costs more, which is a strange thing to be paid for. 

How we got here

American healthcare has so many middlemen because the person receiving care is almost never the person paying for it. During World War II, wage controls stopped employers from raising pay, so they competed for workers with health benefits instead. In 1954 the tax code made those benefits tax-free, and employer-sponsored insurance became the default for most working Americans.

Once a third party pays, a chain of new jobs appears. Someone has to process the payment, decide what is covered, and negotiate the price. Each intermediary in the chain was created to solve a problem the previous one introduced. Pharmacy benefit managers, for example, began in the late 1960s as claims processors for insurers drowning in paper prescription receipts.

The better question is which of these middlemen are still solving the problem they were created for?

A simple test

Here is the test to apply at each step. When the system gets cheaper, does this company make more money, less money, or about the same?

A middleman paid for work performed, like moving boxes or processing claims, is roughly indifferent to price. A middleman paid a percentage of the price it helps set has a reason to prefer that the price go up. The first kind is usually creating value. The second kind might be, but its incentives are pointed the wrong way.

The warehouse nobody thinks about

After the drug leaves the factory, it goes to a wholesaler. Three companies, McKesson, Cencora, and Cardinal Health, move more than 90% of prescription drugs in the United States. They hold thousands of products in inventory, deliver to tens of thousands of pharmacies (often daily), and float the working capital in between, paying manufacturers before pharmacies pay them.

Their operating margins sit around 1%. They are paid largely on volume and are mostly indifferent to what an individual drug costs. By our test, this is what a value-creating middleman looks like. It is boring, concentrated, and thin-margined, and if it vanished tomorrow, someone would have to rebuild it.

The switchboard

When the pharmacist enters the Zepbound prescription, a pharmacy benefit manager (PBM) checks within seconds whether the patient is covered, whether the drug is on her plan's formulary, and what she owes at the counter. The three largest PBMs, Express Scripts, CVS Caremark, and Optum Rx, handle about 80% of the prescription claims in the country.

This first job is real and valuable. Adjudicating billions of claims a year in real time is hard infrastructure, and nothing about it requires the PBM to care what the drug costs.

The PBM's second job is negotiation. When two drugs compete, as Wegovy and Zepbound do, the PBM can threaten to leave one off the formulary and extract a discount from the manufacturer in exchange for preferred placement. That discount is paid as a rebate after the fact. Bargaining power on behalf of millions of patients is also a genuine service.

The trouble is how PBMs were paid for it. Compensation was historically tied to the size of the rebate or to the drug's list price. Consider a drug with a $1,000 list price and a 50% rebate, and a competitor with a $500 list price and no rebate. The net cost is identical, but the first drug generates far more money for the PBM. Manufacturers learned to raise list prices and rebates together, and the gap between list price and what anyone actually paid ballooned.

The patient is the one who absorbs that gap. Deductibles and coinsurance are calculated on the list price, so a patient paying 20% of a $1,000 drug pays $200 even if her plan's true cost was $500. The people who use the most medication end up subsidizing everyone else's premiums.

Regulators spent 2026 codifying this distinction. The Federal Trade Commission sued the three big PBMs in 2024 over insulin pricing, and all three settled this year. The settlements require them to stop tying compensation to list prices, stop preferring high-list-price drugs over cheaper equivalents, and offer plans where patient costs are based on the net price. Express Scripts also agreed to move its rebate-negotiating subsidiary back to the United States from Switzerland. In February, Congress required the same delinking in Medicare Part D, with commercial-market provisions phasing in by 2029.

Read together, the policy answer maps cleanly onto our test. Claims processing is fine. Being paid a cut of a price you help inflate is not.

The insurer's odd job

Insurance does one thing no other institution does as well, which is pooling risk. Most people won’t get cancer or be in a serious car accident, but everyone pays a little so that the unlucky few are not ruined. That is valuable in the most basic economic sense.

GLP-1s for obesity strain that logic. Insurance is built for events that are expensive and unpredictable. A weekly obesity drug is expensive and predictable, taken indefinitely by a population that includes a large share of American adults. Covering it is less like pooling risk and more like prepaying a known bill with an administrative layer on top.

There is also a structural wrinkle. The Affordable Care Act requires insurers to spend 80% to 85% of premiums on medical care. That caps overhead as a share of spending, which also means the most reliable way for an insurer to grow its absolute profit is for total spending to grow. It is not the same distortion as the rebate system, but it points in the same direction.

Employers see this clearly because most large ones self-insure, meaning they pay claims directly and hire insurers and PBMs only to administer them. Among the largest employers, GLP-1 coverage for weight loss fell from 72% in 2025 to 60% in 2026, and 14% have dropped it or plan to by 2027. When the payer realizes it is not buying protection against risk, it stops paying.

The newest middleman

When coverage disappears, patients route around it. Novo Nordisk reported that in one April week, roughly 100,000 of about 270,000 injectable Wegovy prescriptions were filled through the self-pay channel. The Wegovy pill sells for $149 to $299 a month direct, and Lilly sells Zepbound vials for $299 to $449 through its own pharmacy.

That new chain is shorter. The manufacturer sells to a telehealth prescriber's patient and ships from its own pharmacy, cutting out the PBM, the rebate, and the insurer entirely. Telehealth platforms are middlemen too, but they pass our test more comfortably than most. They are paid for access to a prescriber in days rather than weeks, and subscription models reward them when patients stay on treatment rather than when the drug costs more.

The shorter chain has costs of its own. Cash purchases usually do not count toward a patient's deductible or out-of-pocket maximum. And nobody is negotiating on the patient's behalf. The price is low because two manufacturers decided it should be, and they can decide otherwise.

Why you should care

Middlemen that are paid for work performed, like moving product, processing claims, or pooling unpredictable risk, tend to earn their cut. Middlemen that are paid a percentage of a price they influence tend to push that price up, and in 2026 both Congress and the FTC effectively agreed.

For metabolic health, the bigger shift is that the most important drugs in the category are migrating out of insurance and into cash. That is roughly what the test predicts for a predictable, chronic expense. But the migration leaves a gap. For decades, the PBM was supposed to compare prices, steer patients to cheaper options, and vouch for the pharmacy on the other end. The cash-pay patient now does all of that alone, and that job does not have an owner yet.

Or copy and paste this link to others: {{rp_refer_url}}

Current Referral Count: {{rp_num_referrals}}