Which red flags in a telehealth intake or subscription actually predict a bad experience?

I have now looked at seven of these services and I cannot tell the difference between marketing polish and clinical seriousness. They all use the same photography and the same phrases: "personalised", "physician-supervised", "pharmacy-grade", "clinically proven".

What I want is a list of signals with actual predictive value, ideally ones I can check before paying anything. And specifically on the subscription models: several of them price as a monthly membership that includes the medication, and I cannot work out from the checkout page how much of my payment is the consultation and how much is the drug. Is that opacity normal or is it a signal in itself?

Things I have already noticed and cannot interpret: one service will not tell me which pharmacy fills the order until after I pay; two auto-escalate the dose every four weeks unless I intervene; and one offers a twelve-month prepayment at a discount that is non-refundable.

All three of the things you noticed are real signals, and the strongest single predictor in the whole category is this: does the clinical pathway ever produce an outcome that costs the business money? Auto-escalation, undisclosed pharmacy and non-refundable prepayment are three different expressions of the answer being no.

Red flags with actual predictive value

1. Dose escalation by default rather than by assessment. Titration exists to find the highest tolerated dose, and the trial schedules pause or step back when tolerability fails. A calendar that advances unless you actively stop it has inverted the logic, and it reliably produces people at a dose they never tolerated. This is the one I would treat as disqualifying on its own.
2. The dispensing pharmacy is not disclosed before payment. The pharmacy's identity, state licences and accreditation are the only things standing between you and an unknown vial. Non-disclosure is not a competitive secret; it is a refusal to be checked.
3. No named prescriber before you commit. You are buying a clinical relationship. Anonymity on the clinician's side is not compatible with that.
4. No labs required, ever, and none offered. Undiagnosed diabetes, impaired renal function and abnormal liver enzymes are all common in this population and all change the plan.
5. No mechanism to decline. Ask directly: under what circumstances does your clinician decline to prescribe? A serious service has examples ready. A sales operation treats the question as strange.
6. Prescription is not portable. Covered in the next question, but as a signal: a service structured so that your prescription cannot leave has aligned your access to medication with your subscription payment, which is a retention mechanism sitting inside a clinical decision.
7. A combination product presented as an upgrade. Adding an ingredient with no evidence of benefit in this indication produces a compatibility and stability question nobody has answered, and historically served mainly to argue the preparation was not a copy of an approved product.
8. Fixed dose tiers marketed as personalisation. Three plans is a price ladder. Individualisation that is identical for everyone was not individualisation.
9. Non-refundable long prepayment. See the arithmetic below; this one is quantifiable.
10. No adverse-event route and no after-hours plan. The GI adverse-event burden in this class is substantial even under trial supervision [1], and someone has to answer at week three.
11. Silence on discontinuation. Weight regain after stopping is well characterised — withdrawal after a run-in led to substantial regain in the semaglutide withdrawal trial [2], and the tirzepatide withdrawal trial showed the same pattern [3]. A service with no plan for the end of treatment is selling a subscription, not managing a condition.

The prepayment arithmetic

Take the twelve-month non-refundable offer. Suppose the monthly plan is $299 and the annual prepayment is $2,988, marketed as "two months free".

• Twelve months at the monthly rate: 12 × $299 = $3,588
• Annual prepayment: $2,988
• Nominal saving: $3,588 − $2,988 = $600, i.e. 600 / 3,588 = 16.7%

Now weight it by the chance you do not complete twelve months. Adverse-event discontinuation in the pivotal trial was around 7%, and that was with study support and a protocol; real-world discontinuation over a year is considerably higher than trial figures across every published analysis, and people also stop for cost, supply, pregnancy, or because it worked. Take a deliberately conservative 30% chance of stopping at the halfway point.

• Expected cost, monthly plan: 0.7 × $3,588 + 0.3 × (6 × $299) = $2,511.60 + $538.20 = $3,049.80
• Expected cost, prepaid non-refundable: $2,988 regardless of what happens
• Difference: $61.80 in favour of prepaying

So at a 30% mid-year stop rate the entire advertised 16.7% discount has already evaporated to under 2%. At a 40% stop rate:

• 0.6 × $3,588 + 0.4 × $1,794 = $2,152.80 + $717.60 = $2,870.40, which is now cheaper than prepaying.

The break-even stop rate is where 12m × (1 − p) + 6m × p equals the prepaid figure; with these numbers that is p ≈ 33%. In other words the discount is priced to be attractive only if you are more confident of completing a year than the evidence supports. That is not fraud, it is competent pricing — but it is being sold to you as generosity, and the honest description is that you are underwriting the business's retention risk.

Green flags, since a list of negatives is hard to shop with

• Labs required or arranged, with the results actually discussed.
• Titration described as tolerability-dependent, with an explicit hold-or-reduce pathway.
• The pharmacy named, with licence and accreditation checkable before payment.
• Consultation fee and medication cost itemised separately.
• A written statement of what happens if you stop, and a route to transfer care.
• Willingness to send a prescription for an approved product to a pharmacy of your choosing.

The last one is the cleanest test in the whole list, because it costs the business its margin and only a service selling care rather than product will say yes.

On why the bundled price is opaque, since OP asked whether the opacity is itself a signal. It is, but the reason is more structural than sinister and it is worth understanding because it tells you what to ask instead.

A bundled membership fee typically covers a professional fee to the clinical entity, a technology and administration fee to the management company, the pharmacy's charge for the preparation, shipping, and margin. Itemising it creates two problems for the operator. First, once the professional fee is visible you can compare it against what a local clinician charges, and it usually does not compare well. Second, and more importantly, in states with fee-splitting restrictions the allocation between the professional entity and the lay company is legally sensitive, and publishing it invites exactly the scrutiny the structure was arranged to avoid. So the bundle is partly a legal artefact, not only a marketing one.

What to ask instead of demanding a breakdown you will not get:

• What is the cost of the consultation alone, if I bring my own prescription elsewhere? A service that cannot price that has no separable clinical product.
• What does month two cost if I need no dose change and no consultation?
• If I pause for a month for medical reasons, what happens to the fee and to my prescriber relationship?
• What is the cash price the pharmacy charges for the preparation, independent of the membership?

The answers tend to reveal the structure faster than the pricing page. In particular, a service where pausing for medical reasons ends your prescriber relationship has told you that the relationship was a subscription feature.

A counterweight, because a thread like this can leave the impression that all telehealth in this space is predatory and that is not true and not useful.

Telehealth solved a real access problem here. Obesity medicine is under-provisioned almost everywhere, waiting lists for specialist clinics run to many months, and a large share of primary-care clinicians are not comfortable initiating or titrating these agents. For someone in a rural area, or without a clinician who will engage with the topic at all, a well-run remote service is not a downgrade from in-person care — it is the difference between care and none. Several operators run genuinely careful programmes with required labs, real follow-up cadence, dietitian access, resistance-training guidance and documented discontinuation planning.

The distinction that matters is not remote versus in person. It is whether the business model puts the clinical assessment upstream of the sale as a genuine gate, or downstream of it as a formality. Every signal in the accepted answer is a proxy for that one question, and both good and bad versions exist within a few clicks of each other.

One practical note in favour of the careful operators: they are usually the ones willing to write for an approved product and to help with a prior authorisation, because their revenue is the care rather than the vial. If you have any insurance coverage at all, that is worth asking about on the first call.