There is a second way to offer therapeutic neurotoxin that does not require you to solve the credentialing problem first, and a great many aesthetic practices are already running it without having thought it through properly. Off-label Botox for migraines, delivered on a cash-pay basis, tailored to the individual patient rather than to a fixed protocol, is a legitimate model. It is also one where the compliance work moves from the billing department to the consent form, and where the thing most likely to cause you a problem is a sentence you did not say.
We can do this in our practices as medical providers. We are not billing insurance. We tell the patient plainly that this is an off-label treatment, and we escalate and tailor the dose according to that patient's response over time. That is the model. This article is about doing it deliberately.
What "off-label" means here, precisely
FDA's position is straightforward, and worth quoting to yourself before you quote it to a patient: "once the FDA approves a drug, healthcare providers generally may prescribe the drug for an unapproved use when they judge that it is medically appropriate for their patient." The agency also states the flip side directly — if you and your provider use an approved drug for an unapproved use, "remember that FDA has not determined that the drug is safe and effective for the unapproved use" (FDA, Understanding Unapproved Use of Approved Drugs "Off Label").
FDA describes three ways a use can be off-label: the drug is used for a condition it is not approved to treat, given in a different way, or given in a different dose.
That last pair is the one that matters here, and it is the distinction most injectors get muddled. In a typical cash-pay migraine case you are not necessarily treating an unapproved condition — the patient may genuinely meet the on-label definition of chronic migraine. What is off-label is the regimen: fewer units than 155, a different distribution of sites, sometimes unilateral treatment, escalated over time according to response. Different dose, different way. Off-label by the second and third routes rather than the first.
The other scenario — treating a patient with fewer than 15 headache days a month — is off-label by the first route, and it is a materially weaker position, because the label records that safety and effectiveness were not established for episodic migraine across seven placebo-controlled studies, and a Cochrane review remained uncertain whether it works in that population at all (Herd CP, Tomlinson CL, Rick C, et al. Cochrane Database Syst Rev. 2018;6:CD011616). That is not an evidence gap. It is a repeated negative result, and a patient is entitled to hear so.
The model, stated plainly
Many of us in aesthetic practice begin by treating the anterior fields we already know well — the corrugator and frontalis region and the temporalis — because those sit close to areas that are on-label for migraine injection and they are territory we inject every day. Where a patient has a reliably unilateral headache, we may treat only that side. If they get resolution of their headache with that, and they are a cash-paying patient, we do not have to go further. If they do not, we escalate.
What many of us have found is that in aesthetic practices you can inject in a very similar fashion and get adequate control of migraines without using the same number of units as the on-label protocol. Over time many patients will say that their headaches are much better as long as they are getting their treatment.
Two things must be said alongside that, and I say both when I teach it.
First, the honest framing. This is a clinical observation from practice, not a trial result. The published efficacy data belong to the fixed 155-unit, 31-site regimen delivered every 12 weeks. A tailored lower-unit approach is not that regimen, which means you cannot quote the trial numbers as though they describe what you are selling. You can describe your own results. You cannot borrow someone else's.
Second, the counterweight. If a patient is not getting a response — if they are having migraines, if they are being treated by their neurologist — then consider that in aesthetics we are often under-treating the on-label way that this has traditionally been used. The anterior fields are the ones we are comfortable with, and comfort is not a clinical rationale. A partial response in a patient who has only ever had partial coverage is not evidence that the treatment failed.
Why the model works, commercially and clinically
It removes the credentialing wall. Participation status becomes irrelevant, which for a practice with no medical-plan relationships is the difference between offering this and not.
It uses the skill set you have. The anterior fields are your daily work. Extending into the posterior fields is a training investment; starting in the anterior ones is not.
It permits genuine tailoring. In a covered, on-label service you deliver the fixed paradigm because that is what is approved and that is what generated the data. In a cash-pay off-label setting you can respond to a unilateral pattern, escalate by response, and stop at the dose that works. That is real clinical latitude and it is the model's main advantage.
Pricing is transparent to the patient. There is no prior authorisation, no explanation of benefits, no surprise. Set your fee structure for the assessment period, not per-visit, so that a patient who needs three cycles is not making a fresh financial decision at each one.
What it is not is a cheaper version of the same product. It is a different product, and the patient should know that.
The consent conversation is the compliance work
Here is the thing to be clear-eyed about: there is no federal statute or FDA rule imposing a special informed-consent requirement for off-label prescribing. Consent standards in this area come from state law and professional guidance. That is a reason to do it thoroughly, not a reason to relax.
The most useful published framework is the AMA's opinion on innovative practice, which — while it does not use the words "off-label" — sets out exactly the disclosure a nonstandard therapeutic recommendation requires. It says physicians must base recommendations on patients' medical needs, refrain from offering such services until they have acquired appropriate knowledge and skills, and recognise that informed decision making requires the physician to disclose:
- how the recommended service differs from the standard therapeutic approach, if one exists
- why the physician is recommending this modality
- the known or anticipated risks, benefits and burdens of the recommended therapy and the alternatives
- what experience the professional community in general, and the physician individually, has had with it to date
- what conflicts of interest the physician may have
It also instructs physicians to discontinue innovative therapies that are not benefiting the patient (AMA Code of Medical Ethics, Opinion 1.2.11, Ethically Sound Innovation in Medical Practice). The AMA's general informed consent opinion adds the baseline expectations: present the diagnosis, the nature and purpose of the recommended intervention, and the burdens, risks and expected benefits of all options including forgoing treatment — and document the conversation in the record (Opinion 2.1.1, Informed Consent).
Map that onto this treatment and your consent needs to name, explicitly:
Ready to put this into practice?
Explore Empire's hands-on, CME-accredited Practice Development & Business Strategies courses — live patients, expert faculty, and ongoing mentorship.
- That this is a preventive, not a rescue treatment. It does nothing for an attack in progress.
- That the regimen you are proposing is off-label — and in what way. Different dose and different site distribution from the approved protocol, or an unapproved indication, or both. Name which.
- That FDA has not determined the drug is safe and effective for this use. Use FDA's own phrasing.
- What the approved alternative is, and that it exists. A patient choosing a tailored cash-pay regimen should know that a fixed 155-unit protocol with published trial data is the standard approach, that it may be available to them through a credentialed provider, and why you are recommending this instead.
- The realistic magnitude of benefit. In chronic migraine, the systematic-review estimate for the on-label regimen is about 1.9 fewer headache days per month versus placebo. Say a number. Never promise a result, and never offer a guarantee of any kind.
- The timeline. Response commonly takes more than one cycle; roughly one in five eventual responders in the pivotal programme did not declare until cycle two or three.
- The adverse event profile, including that all botulinum toxin products carry a boxed warning for distant spread of toxin effect, with symptoms including muscle weakness, diplopia, ptosis, dysphagia, dysphonia, dysarthria and breathing difficulties reported hours to weeks after injection.
- Your stopping rule, in both directions — what constitutes success, and at what point you will tell them this is not working and stop taking their money.
Empire's guidance on what to include in Botox consent forms is the right starting structure; the eight items above are the therapeutic-specific additions on top of it.
Document the reasoning, not just the signature
A signed form proves a form was signed. What protects a clinical decision is the reasoning in the note.
It is worth knowing what a reviewer's expectation looks like when one is written down. In the Medicare context, contractor coverage policies for botulinum toxin state that dosing must be used in accordance with FDA-approved labelling, and that for off-label use without an approved dosing indication, the clinician "must provide robust published clinical evidence to support the dosing use." You are not billing Medicare in this model, so that policy does not govern you — but it is a useful calibration of the standard a reviewer applies to off-label dosing generally. "I gave less because the patient preferred it" does not meet it. "I treated the unilateral field on the symptomatic side and escalated by documented diary response over three cycles" is a clinical rationale.
So for each patient, the record should hold: the confirmed diagnosis and who made it, the red-flag screen, the prospective headache diary establishing the baseline, the reason this regimen rather than the approved one, the consent conversation itself, and the response data that justified each subsequent adjustment.
The lines you do not cross
Do not bill insurance for a tailored regimen. If you have chosen the cash-pay model, run it as a cash-pay model. Submitting a claim for a service you delivered differently from the approved protocol is a different category of problem from anything discussed above.
Do not let a patient infer coverage. Silence reads as "probably yes." Say the words.
Do not call it "the migraine protocol" if it is not the protocol. The word carries the trial data with it. If you are delivering a tailored regimen, describe it as a tailored regimen.
Do not treat an unconfirmed headache. Cash-pay changes who is paying. It does not change the requirement that the diagnosis has been made by someone qualified, with the appropriate workup, and that you have screened for secondary causes. Injecting is not a substitute for a workup, and a cash-pay setting makes the temptation to skip that step stronger rather than weaker.
Do not keep treating a non-responder. The AMA opinion is explicit about discontinuing innovative therapies that are not benefiting the patient, and your own stopping rule should have said so before you started.
Do not escalate a patient who was controlled and has stopped being controlled. That is a change in pattern, which is a red flag for a secondary cause. They go back to their neurologist for evaluation before anyone touches the dose.
What to do about it on Monday
- Decide explicitly which model you are running, and write it down. Drifting between covered and cash-pay is how practices end up doing both badly.
- Rewrite your consent for this indication rather than adapting the cosmetic one. The eight items above are the minimum.
- Price the assessment period, not the appointment.
- Put the diary in the workflow before the first injection, because without a baseline you cannot justify escalation or defend a stopping decision.
- Say "this is off-label, and FDA has not determined it is safe and effective for this use" out loud, every time, and note that you did.
- Book three cycles and state the stopping rule at the start.
For adjacent therapeutic applications that raise similar tailoring and consent questions in an aesthetic practice, see Empire's material on Botox for jaw clenching and management of bruxism.
Injection technique, dilution and neurotoxin handling are taught hands-on in Empire's Complete Botox Training and Cosmetic Neurotoxins Training workshops.
The tailored cash-pay approach described here reflects Dr. Chris Croley's clinical practice as taught in Empire Medical Training's hands-on curriculum, and is presented as his practice rather than as standard of care. It is not supported by the trial data that support the fixed on-label regimen, and this article says so. Technique is learned under supervision; this article is educational and is not a substitute for training, and is not legal or compliance advice.
Frequently Asked Questions
Is it legal to treat migraines off-label with neurotoxin?
FDA states that once a drug is approved, healthcare providers generally may prescribe it for an unapproved use when they judge it medically appropriate. FDA also states plainly that it has not determined the drug is safe and effective for that unapproved use. The decision, and the responsibility for it, sits with the prescriber — and with state law and professional standards.
If the patient has genuine chronic migraine, what makes it off-label?
The regimen, not the condition. FDA describes off-label use as including a drug given in a different way or in a different dose. A tailored approach using fewer units, a different site distribution, or unilateral treatment is off-label on dose and route even where the indication itself is on-label. Name which form applies in your consent.
Do I need a special consent form for off-label treatment?
There is no federal off-label consent mandate; requirements come from state law and professional guidance. Practically, yes — write one. AMA guidance on nonstandard therapy requires disclosing how the treatment differs from the standard approach, why you recommend it, the risks and benefits of it and the alternatives, the experience behind it, and any conflicts of interest.
Can I quote the PREEMPT trial results for a tailored regimen?
No. Those results belong to the fixed 155-unit, 31-site regimen delivered every 12 weeks in a defined chronic migraine population. A tailored lower-dose approach is a different intervention. You may describe your own outcomes; you may not attach another regimen's published numbers to yours, and doing so in marketing is a substantiation problem as well as a clinical one.
What should I charge for a cash-pay migraine course?
That is your commercial decision, but structure it around the assessment period rather than the visit. Response commonly takes more than one cycle, so a patient asked to make a fresh financial decision at twelve weeks with no improvement yet will usually stop — including patients who would have responded at cycle two or three.
Disclaimer
This article reflects the clinical opinions and experience of Dr. Chris Croley, Chief Medical Officer, Empire Medical Training, an independent faculty member contributing to Empire Medical Training's curriculum. The views expressed are the author's own and do not necessarily represent those of Empire Medical Training.
It is professional education, not medical advice, and is no substitute for hands-on training or independent clinical judgment. Licensed clinicians remain responsible for their own patient selection, technique and outcomes, for verifying current product labelling, and for practising within their scope and applicable law. Empire Medical Training accepts no liability for reliance on this content.
