Bariatric Telehealth Scheduling: Which Visits Can Be Video

Bariatric telehealth scheduling looks like a convenience question and behaves like a billing question. A patient in a months-long weight-management sequence asks to do the next visit by video. Somebody says yes. Nothing breaks that day, and nothing breaks that month.

It breaks when the insurance packet gets assembled and one visit in the sequence does not count. By then the patient has been in the program for half a year, the surgery date has been discussed out loud, and the fix is another month of waiting. The person who said yes to video was a scheduler with no way to know which visits in a bariatric program are modality-restricted and which are not. That rule is real, it varies, and it lives nowhere near the schedule.

Modality is a booking decision before it is a clinical one

Coverage rules are written with the modality baked in. Medicare’s national coverage determination for intensive behavioral therapy for obesity specifies a visit cadence in explicit terms: one face-to-face visit every week for the first month, then one face-to-face visit every other week for months 2-6, then monthly for months 7-12 if the beneficiary meets the weight-loss threshold at six months.

Read that as an operator rather than a clinician. The cadence is a schedule, the eligibility test at month six is a date, and the phrase doing the work is “face-to-face.” None of that is a clinical call. It requires somebody to know, at the moment a patient asks for a video slot, whether this particular visit is one of the visits where modality is fixed.

Commercial payers write their own supervised-period rules, and they do not agree with each other or with Medicare. That is the actual problem. There is no single rule to memorize, so the front desk defaults to whatever the patient asked for.

Where a video slot quietly costs the program

Bariatric programs are one of the few places in an outpatient practice where a scheduling mistake has a delay measured in months instead of days. The supervised sequence has to be continuous. A visit that does not count is not a visit that gets redone next week, it is a gap, and a gap can send the patient back to the start of the sequence.

The same shape shows up on the pre-operative side. Clearances arrive as faxes from outside psychology, cardiology, pulmonary, and sleep providers, and each item carries its own expiry window. A workup that stalls invalidates testing already sitting in the chart. Modality errors stall workups.

What makes this hard to catch is that the failure is silent at the point of booking. The slot fills, the patient shows up, the visit happens, and the documentation is fine in every respect except the one that matters. Nobody finds it until the packet is assembled.

What the automation checks before it offers a video slot

This is administrative work, and it is the kind of administrative work automation is actually good at, because it is a lookup that a human has to do under time pressure while a patient waits on the phone.

Working inside athenaOne, the AI resolves the request against the appointment type rather than against the patient’s preference. A bariatric program typically carries separate types for the seminar, the surgical consult, the monthly supervised visit, the pre-operative visit, and each post-operative interval. Modality eligibility is a property of the type, not of the patient. So the sequence is: identify which visit in the program this is, look up whether that type permits a telehealth slot for this patient’s plan, then offer only slots that are valid on both counts.

The complication that breaks naive versions of this: the appointment-type catalog changes without notice. Practices retire types, fold several into one, or add a virtual variant partway through a quarter. An automation that hard-codes the mapping is wrong within a month. The mapping has to be read from the practice’s live configuration, and when a type appears that the mapping does not recognize, the correct behavior is to stop offering slots for it, not to guess.

When the request is valid, the rest is logistics the automation carries: sending the visit link, re-sending it when the patient cannot find it, and making the tech-readiness call before a first video visit so the slot is not lost to a camera that was never turned on.

The handoff, stated plainly

The automation does not decide whether a patient should be seen virtually. It decides whether the visit type the patient is asking for can be booked as a telehealth slot under their plan, which is a coverage and configuration question.

Everything past that goes to a person. If the patient wants video for a visit that has to be in person, the AI books the in-person slot or, when the patient cannot come in, routes the request to the program coordinator with the reason attached. If the plan’s rule is ambiguous, it does not resolve the ambiguity. It hands the coordinator a request that already has the visit type, the plan, and the sequence position filled in, which is most of the work of answering it.

That boundary is not a limitation to apologize for. It is what keeps the automation useful, because a coordinator who trusts the handoff stops re-checking the bookings that went through cleanly.

The visit series that no scheduling template holds

Post-operative follow-up in a bariatric program runs for years, not weeks, and program accreditation depends on follow-up rates. No scheduling template expresses a multi-year interval series, so in most practices it is a coordinator with a spreadsheet and a recurring reminder.

Much of that series is exactly where telehealth belongs, and it is also where patients drift. A patient two years out has lost the habit of the practice. The outreach that brings them back is a call, a booked slot at the right interval, and a link that works, repeated at intervals nobody has hours to track manually.

The eligibility side stays bounded and checkable. Medicare’s bariatric surgery coverage determination sets its own thresholds, covering the listed procedures for beneficiaries with a body-mass index of 35 or higher who have at least one obesity-related co-morbidity and have been previously unsuccessful with medical treatment. Those are facts about a chart and a plan, not clinical assessments, and they are the kind of thing a front-office system can be held to.

Key takeaways

Treat modality as a property of the appointment type, not a patient preference. If your athenaOne types do not distinguish video-eligible visits from face-to-face-required visits, that is the first configuration to fix.
Audit the supervised-period visits specifically. Those are the ones where a modality error costs months rather than a reschedule.
Read the appointment-type mapping from live configuration instead of hard-coding it. Catalogs change mid-quarter, and a stale mapping fails silently.
Define the handoff before you automate. The automation should book what is unambiguous and route what is not, with the visit type and plan already attached.
Put post-operative interval outreach on the automation rather than on a coordinator’s spreadsheet. It is a multi-year series and accreditation follow-up rates depend on it.

Bariatric programs are complicated in a way that rewards specificity. The visit sequence is long, the rules attach to visit types rather than to patients, and the cost of a wrong booking shows up a quarter later. That is a configuration problem sitting inside a scheduling problem, and it is worth solving before it becomes a packet problem.

Sources: