On September 3, the FDA acknowledged that some patients were having trouble finding their usual estradiol patches. The agency attributed the problem to increased demand, even though the supply of FDA-approved patches had nearly doubled over the previous year.
That combination is the story. More product entered the market, but demand moved faster.
In my work, I spend a lot of time thinking about whether a healthcare message is clear enough to change behavior. The estradiol patch situation points to the question that comes next: what happens if it works?
Last November, the FDA asked manufacturers of menopausal hormone therapy products to remove boxed-warning language related to cardiovascular disease, breast cancer and probable dementia. The agency said decades of evidence required a more specific benefit-risk discussion, particularly around age, time since menopause and differences between systemic and local products. In February, it approved the first six revised labels.
Now the FDA says demand for these therapies increased following its November action. The message did more than clarify risk. It changed the market around the treatment.
Demand was already growing
It would be convenient to tell this as a clean before-and-after story: the FDA changed the warning, women felt comfortable seeking treatment and pharmacies ran out of patches. The evidence does not support something that simple.
Interest in menopause care had been growing for years. A preliminary analysis by Truveta found that estrogen-based hormone therapy prescribing more than doubled from January 2018 to February 2026. Patch dispensing more than tripled over the same period, making patches the most common route of administration in its dataset by the end of the study.
The growth continued around the FDA action. Truveta found a 19.1% increase in estrogen-based hormone therapy prescribing between July 2025, when the FDA convened an expert panel, and February 2026. The analysis was based on a subset of Truveta data and was not peer reviewed, so it should be treated as a directional view rather than a national count.
The FDA announcement did not create demand on its own. It entered a market already being reshaped by specialist advocacy, changing clinical conversations, media attention and broader investment in menopause care. But the agency now explicitly says demand increased after the November action. An authoritative public message added momentum to a trend that the supply chain was already struggling to match.
The timing also shows how quickly health communication can travel. The FDA announced its request in November, then approved the first six revised labels in February. Patients and clinicians did not need every package insert to be updated before the larger message reached them.
A changed warning still needs nuance
The warning change does not mean hormone therapy is appropriate for everyone.
The Menopause Society supported removing the boxed warning from low-dose vaginal estrogen, which has very low systemic absorption. It was more cautious about systemic estrogen, the category that includes estradiol patches. The organization continues to recommend individualized decisions based on factors such as age, time since menopause, medical history, symptoms and personal preferences.
The FDA also retained boxed-warning language about endometrial cancer for systemic estrogen-alone products. Its stated goal was to make the benefit-risk discussion more specific, not to replace one universal message with another.
That distinction matters because a boxed warning is blunt, but the public shorthand around its removal can be blunt too. If the old message became “hormones are dangerous,” the new one cannot simply become “hormones are safe.” Better risk communication should create more informed conversations, not demand for one treatment regardless of the patient.
Operational readiness therefore means more than producing additional patches. It also means making sure clinicians have the time and information to help patients interpret a changing message.
Available nationally can still be inaccessible
The FDA says estradiol patches remain available, although a particular product or brand may vary by pharmacy or location. The American Society of Health-System Pharmacists lists multiple estradiol patch products as affected by shortage, with some doses on back order, others on allocation and several without an estimated release date.
Those statements are not necessarily contradictory. The FDA determines shortage status at the national drug level using supply and projected demand. ASHP tracks availability at the product and presentation level. One can show adequate aggregate supply while the other captures the specific dose or manufacturer a pharmacy cannot obtain.
The distinction makes sense to a regulator. It is much less useful to a patient holding a prescription she cannot fill.
A patch can be available somewhere in the United States and still be inaccessible to a woman who needs a particular dose, refill schedule and local pharmacy. Finding an alternative may require calls to other pharmacies, a substitution by the pharmacist when appropriate or another conversation with the prescriber about a different product. Even an apparently simple dose workaround may be unsafe to improvise. ASHP cautions that cutting a patch can lead to uneven drug release and variable therapy.
The FDA’s own advice reflects this chain. It tells patients to contact their pharmacist, talk with their healthcare provider if their usual patch is unavailable and avoid stopping or changing medication without clinical guidance. Each step is reasonable. Together, they show how a supply problem becomes work for patients, pharmacists and clinicians.
This is why the official question of whether the country is in a shortage does not fully capture access. National availability is a supply measure. Filling the prescription a clinician and patient selected is the care experience.
Plan for what happens if people listen
The FDA is not ignoring the supply side. It says it is working with all six manufacturers, offering to expedite reviews related to production increases and supporting measures such as additional manufacturing shifts, larger batches, specialized equipment and new production lines. Some changes can help quickly. Others take longer because transdermal patches require specialized manufacturing processes.
The agency should not have waited to update risk information until supply was perfect. Preserving an overbroad or outdated message would create its own harm, and a regulator cannot guarantee that every dose will be available at every pharmacy. The takeaway is that risk communication should be planned as a demand intervention.
Before a major public reframe, the planning cannot end with the evidence review, regulatory language, press release and patient education. It should also ask what will happen if patients and clinicians act on the new information. Which products are likely to see the largest change in demand? How quickly can manufacturers respond? What will pharmacists be able to substitute? What information will prescribers need when a patient’s first choice is unavailable? How will the system distinguish a national supply assessment from the access problem an individual patient is actually experiencing?
The measurements have to extend beyond message reach and prescription growth too. Time to fill, product switches, repeat pharmacy contacts, treatment interruptions and abandoned prescriptions would reveal whether greater awareness is producing access or simply moving the burden downstream.
Communication teams do not own manufacturing capacity. They do have a responsibility to make the likely behavioral consequences visible early enough for operations, clinical teams and supply partners to prepare.
If the goal is informed choice, success is not that more women ask about hormone therapy or that more prescriptions are written. It is that a woman can have an individualized conversation, obtain the treatment she and her clinician choose and continue it without being sent through a pharmacy search every month.
Changing the warning changed more than the label. It changed the system expected to respond.
When a piece rests on my own data, I say so. When it rests on someone else’s, I say whose, and whether they funded it.