Rethinking Copay Accumulators: Focus on Days-on-Therapy Over Dollars Spent

Copay accumulators are argued about in dollars, which is why the argument never ends. Every party to it can produce a number showing they lost, and none of those numbers describes what happened to the patient.

An accumulator does not reduce what a drug costs. It relocates when the patient pays, moving the full weight of the deductible to the moment manufacturer assistance runs out. The dollars eventually net out somewhere in a ledger. What does not net out is the interval in the middle, when the patient is asked to absorb a cost they had been shielded from and some of them stop filling. Measured in dollars, that year is a dispute. Measured in days on therapy, it is a fact.

Why the dollar frame produces a stalemate

Run the same accumulator year past three parties and you get three defensible grievances. The manufacturer spent assistance that never reduced the patient’s deductible. The plan collected cost sharing against the same prescription twice, once from assistance and once from the patient, which is the design working as intended. The patient reaches the point of exhaustion still owing the entire deductible, and now owes it unassisted.

Each of those is true. None of them is a measure of the outcome, because all three are denominated in money and the thing that actually changed was continuity.

What the accumulator actually moves

The mechanism is worth stating plainly, because it is often described as a cost-control tool and it is not. Total spend on the drug is unchanged. Assistance is still spent, the plan still pays its share, the patient still owes their deductible. What changes is the sequence: assistance covers the early fills without reducing the patient’s accumulated cost sharing, so the deductible arrives intact at the moment the assistance is gone.

That is a timing decision, and timing is exactly what determines abandonment. A patient who would have absorbed a deductible spread across a year encounters it as a single wall in month four or five. The clinical decision at the counter is not made against an annual total. It is made against one price, on one day.

The regulatory answer is not coming on schedule

It is tempting to treat this as a compliance question that policy will eventually settle. The current record argues otherwise. Twenty-five states plus the District of Columbia and Puerto Rico now require plans to count copay assistance toward cost-sharing limits, and Avalere’s analysis put that at no less than 16.8 percent of the United States commercial market, roughly 34.3 million lives.

Those laws reach state-regulated plans. Self-insured employer plans are governed federally and sit outside them, which is where a large share of commercial coverage actually lives. At the federal level, a district court vacated the 2021 provision permitting accumulators in September 2023, returning the standard to the 2020 rule, and the agency then said it would not enforce pending new rulemaking. That rulemaking has not arrived.

So the practical situation is a benefit design that varies by state, by funding type, and by whether a medically appropriate generic exists. That is not a condition anyone should be waiting out.

Days on therapy is the measure both sides can audit

The reason to move to a days-based measure is not that dollars do not matter. It is that days are the only variable in this system that every party can observe the same way and none of them can reframe.

1. Days covered before the cliff. How long assistance sustained continuous therapy before the patient’s exposure changed. This is the part of the program that worked.
2. Gap length at exhaustion. The interval between the last assisted fill and the next fill of any kind. A gap of zero means the patient absorbed it. A long gap means they did not.
3. Restart rate. The share of patients who resume after a gap, and how long they take. Non-restart is abandonment with a delay in front of it.
4. Persistence at twelve months against a matched population. The comparison that isolates the accumulator from everything else happening to the patient.

Those four are auditable from data that already exists in hub and pharmacy reporting. None requires agreement about who bore which dollar.

What this changes about program design

Once the measure is days rather than dollars, the design questions change with it. A support program calibrated to a dollar cap per patient per year is optimizing the wrong constraint, because the cap does not know where the cliff falls. A program calibrated to continuity spends differently: it front-loads less, it monitors the approach to exhaustion, and it treats the transition itself as the event to manage rather than an accounting boundary.

It also changes what gets escalated. A patient at 80 percent of assistance with a deductible still unmet is a predictable gap, visible weeks before it happens, and nothing in a dollars-based program is built to notice.

The same discipline, one step earlier

This sits in the same place as every other access question. The enterprise develops, manufactures, prices, contracts, distributes and supports. It does not decide what a plan counts toward a deductible. What it can decide is what it measures, and whether its own support programs are instrumented to see a patient stop.

The dollar argument about accumulators will continue, in legislatures and in courts, on a timeline nobody controls. The days argument can be settled internally, with data already in hand, and it is the one that tells you whether the patient is still on therapy.

Market and enrollment figures are the published findings of Avalere Health. Regulatory status is described as of publication and is not legal advice. The operating patterns described reflect my own experience and are offered as commentary, not as legal, regulatory, clinical or accounting advice. The views expressed are my own and are not made on behalf of any current or former employer. All trademarks belong to their respective owners.