27% of Employers Now Point Members to Cash-Pay GLP-1 Platforms. The Spending Doesn't Disappear: Your Ability to See It Does.

A new pulse survey from the International Foundation of Employee Benefit Plans, fielded June 2–12 of this year, found that 27% of employers now encourage employees to buy weight-loss GLP-1s through direct-to-consumer platforms like LillyDirect and NovoCare, and 21% encourage them to pay for those drugs with HSA or FSA dollars. On a spreadsheet, this looks like a clean win: the pharmacy claim vanishes from the plan, trend flattens, and the renewal conversation gets easier. But moving a drug outside the pharmacy benefit does not remove the drug from your population. It removes the drug from your data. And in a benefit where nearly every management decision you make is downstream of claims data, losing visibility is not a neutral event, it is a decision with a price tag, and most plan sponsors never see the invoice.

The Insight Most Employers Miss

There is a version of cost control that reduces spend, and a version that relocates it. Pointing members to a cash-pay channel is almost always the second kind. The member still takes the drug. The member still has obesity, or sleep apnea, or elevated cardiovascular risk. What changes is that the transaction now happens in a place your PBM cannot see, your care management vendor cannot see, your stop-loss carrier cannot see, and your own reporting cannot see.

Think of it the way a CFO thinks about an operating lease that never touched the balance sheet. The obligation was always real. The statement just got quieter. Quiet financials feel like progress right up until the moment you need to explain a number, and you discover the underlying activity was never captured.

That is precisely the position a plan sponsor is in two renewals from now, when the medical trend moves and nobody can say whether GLP-1 use in the population went up, went down, or collapsed entirely.

Why This Gets Overlooked

Three reasons, and none of them involve carelessness. First, the pricing looks compelling in isolation. LillyDirect has recently listed Zepbound at roughly $299 to $449 per month, and NovoCare has listed Wegovy at roughly $149 to $349 per month, though cash-pay pricing in this category has moved repeatedly and is worth re-verifying before it goes into a model. Compared with a plan-paid brand claim, those are real numbers, and it is reasonable for a benefits team to notice them.

Second, the decision usually gets framed as a coverage question rather than a data question. The meeting is about whether to cover weight-loss indications. It is rarely about what happens to accumulators, drug utilization review, and measurement when the answer is "not through the plan."

Third, the recommendation frequently arrives through the advisory channel, which carries a lot of weight. In that same IFEBP survey, broker, consultant, and PBM recommendations were the single most-cited factor shaping GLP-1 coverage decisions at 58%, more influential than obesity's role as a chronic disease risk factor (46%) or the effectiveness of cost-control mechanisms on premiums (44%). When the recommendation comes from the people you trust to model cost, it does not feel like a strategy that needs stress-testing.

Five Questions To Ask Before You Route Members Off-Plan

1. Does our PBM agreement actually permit this?

Many PBM contracts contain exclusivity provisions requiring prescription drug coverage, and often the associated clinical programs, to flow through the PBM. Removing a class from the formulary can, depending on language, jeopardize pricing guarantees across the entire book, not just the class you carved out. There is a second trap here that gets missed: if your rebate guarantee is stated as dollars per brand script, pulling a high-rebate class out of the mix changes the denominator, and most PBMs reserve the right to reprice guarantees when plan design changes materially. A carve-out that saves $400,000 in drug spend and costs you a repriced guarantee across a $9 million pharmacy book is not a savings strategy.

Related trap: if your PBM offers its own weight management program and you use a different vendor, that alone may breach the agreement or void guarantees.

2. What happens to the member's accumulators, and to HSA eligibility?

Cash-pay purchases do not flow through the pharmacy benefit, which means they do not count toward the deductible or the out-of-pocket maximum. For a member with comorbidities and other claims during the year, that quietly raises total household health spending even though the plan's reported per-member cost went down. The members absorbed the difference.

If you go a step further and reimburse the drug, through an HRA-style arrangement or manual claims process, you are now operating a health reimbursement arrangement whether or not you called it one, and it needs to be documented, integrated with a minimum-value plan, and tested for discrimination in favor of highly compensated individuals. If you offer an HDHP, be especially careful: reimbursing non-preventive drug costs before the statutory deductible is satisfied can disqualify the employee from contributing to an HSA for the year. Excepted-benefit HRAs are capped at $2,200 for 2026, which does not cover a year of therapy at $300 to $450 a month. This is legal-counsel territory, not a benefits-committee improvisation.

3. Where does the clinical visibility go?

When fills bypass the PBM, they typically fall outside standard pharmacy and provider workflows. The dispensing pharmacy your member normally uses may not know they are on a GLP-1. Neither may the prescriber manage their other conditions. Concurrent drug utilization review, interaction screening, duplicate-therapy checks, and the safety net of a single medication profile all weaken at once. For a drug class with meaningful gastrointestinal effects, dose titration schedules, and important perioperative and anesthesia considerations, that is not a trivial gap. It is also a gap the plan sponsor created, which matters when you are the one holding fiduciary responsibility for how the benefit is administered.

4. What does discontinuation cost us, and would we even know it happened?

This is where economics gets uncomfortable. Pharmaceutical Strategies Group's 2026 drug benefit design research found that nearly two-thirds of patients without type 2 diabetes discontinue GLP-1 therapy within a year. Meanwhile, Aon's analysis of roughly 192,000 GLP-1 users drawn from commercial medical and pharmacy claims found that members who stayed on therapy for 18 months showed a pattern of slower medical cost growth and fewer hospitalizations for major adverse cardiovascular events.

Read those two findings together and the strategic question becomes clear: the value of this drug class appears to depend heavily on persistence, and persistence is exactly what an unmanaged cash-pay channel is worst at supporting. There is no refill reminder tied to your plan, no adherence outreach, no clinical program touchpoint, no data feed telling anyone the member stopped in month five.

Illustrative scenario, using round numbers: a 3,000-life employer with 150 members obtaining Zepbound through a DTC platform at $449 per month. If persistence follows the pattern above and average therapy duration lands near six months, that population spends roughly $400,000 out of pocket in a year, largely for a partial course of treatment, with weight regain likely for most who stop. Nothing about that shows up in your pharmacy report. Some of it shows up two years later in your medical claims, where you cannot attribute it to anything.

5. How will we measure whether this worked?

If the answer is "our pharmacy trend will be lower," you have designed a measurement system that cannot fail. Removing claims from the denominator guarantees the result. Ask instead: what specific data will tell us whether members got appropriate therapy, stayed on it, and generated the downstream benefit that justifies obesity treatment in the first place? If the honest answer is "none," you have not adopted a strategy. You have adopted a blind spot with a favorable-looking report attached.

The Version Of This That Actually Works

There is a legitimate design here, and it looks different from simply pointing members at a website. The emerging direct-to-employer models, including arrangements from vendors such as Waltz Health and 9am Health partnering with Novo Nordisk and Eli Lilly, attempt to preserve fixed, pre-negotiated pricing while keeping a clinical management wrapper around the member. Employers pursuing that route typically carve GLP-1s out of the pharmacy benefit but retain reimbursement through the plan, which means the spend stays visible, accumulators can be applied manually, and utilization management still exists. It costs more in vendor fees and administrative complexity, and it requires PBM contract accommodation. It also produces something you can actually manage.

The distinction is simple: a strategy that keeps the claim inside your plan while lowering the price is cost control. A strategy that pushes the claim outside your plan is cost transfer, with reporting improvement as a side effect.

What To Do In The Next Thirty Days

Pull your PBM agreement and locate the exclusivity provision, the formulary-deviation language, and the guarantee-reset clause. Ask your account team, in writing, what happens to every pricing and rebate guarantee if GLP-1s are removed for weight-loss indications.

Ask your PBM for a two-year GLP-1 utilization report broken out by indication, showing new starts, 6-month persistence, and 12-month persistence. If your plan currently covers these drugs, this is your baseline. If you are considering a cash-pay steer, this is the last clean data you will ever have.

Ask your broker or consultant a direct question: "If we do this, what will I be able to see next year that I can see today?" Then ask them to put the answer in the recommendation document.

If you already reimburse any portion of DTC purchases, have counsel confirm whether you are operating an undocumented HRA, and whether your HDHP population's HSA eligibility is intact.

Key Takeaways

Moving GLP-1s to a cash-pay channel reduces reported pharmacy spend without reducing the underlying clinical and financial exposure. It shifts cost to members in a way that bypasses deductibles and out-of-pocket maximums. It weakens medication safety oversight. It removes the persistence data that determines whether this drug class generates value at all. And depending on your contract, it may put pricing guarantees across your entire pharmacy book at risk.

None of that means a carve-out is wrong. It means the question is not "should we cover this?" The question is "if we move it, what are we giving up, and did anyone price that?"