How to Manage Sleep Study Scheduling and Patient Follow-Ups in 2026

A sleep clinic is the only outpatient practice where the patient sleeps through the billable encounter. That single fact warps everything about scheduling. You are not booking 15-minute follow-ups. You are coordinating an eight-hour overnight in a lab bed, plus a home sleep apnea test (HSAT) device that has to physically travel to a patient's house and back, plus scoring that may take 48 hours of tech time, plus a CPAP titration, plus DME dispense, plus a 30-day adherence check that Medicare will use to decide whether to keep paying for the equipment. Miss any handoff and the patient stalls. Stall too long and they drop out — which is a clinical failure and a revenue failure simultaneously.

This guide is the procedural playbook we wish every new sleep clinic administrator had on day one. Six steps from referral to 90-day adherence, what good looks like at each handoff, the common mistakes that quietly bleed revenue, and how a modern platform like Deelo Practice collapses the workflow so a four-person team can run what used to require eight. If you are evaluating sleep clinic scheduling software in 2026, read this first — the workflow shapes the software, not the other way around.

Before we get into the steps, here is the goal state. Picture the clean version of the patient journey, the one you want every single referral to flow through with no manual intervention.

A referral arrives — fax, EHR direct message, or patient self-referral through the portal. The intake coordinator runs an eligibility check and a screening questionnaire (STOP-Bang or Berlin). Based on score and comorbidities, the patient is routed to either an HSAT or an in-lab polysomnography (PSG). If HSAT: the device ships within 48 hours, the patient wears it for one to three nights, returns it, and the raw data hits the scoring queue the same day. If PSG: the patient is scheduled into the next available lab bed — accounting for tech coverage, the right montage, and any split-night requirements — arrives that evening, sleeps, leaves by 6am, and the raw study lands in the scoring queue.

A registered polysomnographic technologist (RPSGT) scores the study within the clinic's stated turnaround window (typically 24-72 hours). The interpreting physician reviews and signs the report. The report is automatically pushed to the ordering provider and uploaded to the patient portal. If the diagnosis is obstructive sleep apnea (OSA) and CPAP is indicated, the patient is scheduled for either a separate CPAP titration study or, in a split-night protocol, the titration was already performed during the diagnostic PSG. The DME order is generated, the supplier is notified, the mask fitting is scheduled, and the patient leaves with equipment.

Thirty days later, the CPAP modem data is automatically pulled. The system flags whether the patient has met the Medicare adherence threshold (more on this in step 6). At 90 days, a follow-up visit confirms continued use and addresses any mask, pressure, or comfort issues. If the patient is non-adherent, the clinic intervenes before Medicare cuts off coverage.

That is the goal. Now here is how to actually run it.

Not every patient who snores needs a $1,200 in-lab polysomnography. The first decision in every sleep clinic workflow is whether the patient is a candidate for a home sleep apnea test or whether they need to be in a lab bed with a tech watching the signals. Get this wrong and you either burn lab capacity on patients who could have been tested at home, or you push a complex patient through an HSAT that misses their actual diagnosis.

The triage rule of thumb most clinics use: HSAT is appropriate for adult patients with a high pre-test probability of moderate-to-severe OSA and no significant comorbidities. PSG is required when the patient has comorbid heart failure, COPD, neuromuscular disease, opioid use, or a suspected non-respiratory sleep disorder (narcolepsy, REM behavior disorder, periodic limb movement disorder, parasomnias). Pediatric studies almost always require in-lab PSG. Patients who fail an HSAT — meaning the study was technically inadequate or the result does not match the clinical picture — also escalate to PSG.

Medicare coverage rules add another layer. CMS will cover HSAT for OSA evaluation in adults, but only with specific device types and only when ordered by a treating physician who has seen the patient face-to-face. Some commercial payers are more restrictive and require PSG as the first-line test in their coverage policies, even for low-comorbidity patients. The intake coordinator needs to know which payer requires which path before booking. Burning an HSAT on a patient whose payer will not reimburse it is a self-inflicted denial.

The practical workflow: at intake, run a STOP-Bang or Berlin questionnaire, document comorbidities, verify payer coverage rules for HSAT vs PSG, and route the patient to the correct study type. A good scheduling platform does this routing automatically based on the questionnaire result and the payer rule table — no front-desk judgment call required.

If the patient is routed to HSAT, the clinic now has a device-management problem on top of a clinical problem. You have a finite number of HSAT units. Each one needs to ship, get worn, come back, get cleaned and recharged, and turn around for the next patient. Lose track of one and you are out a $400-1,500 device. Let one sit in a patient's drawer for three weeks and that is one fewer study you can run that month.

The core tracking unit is the device itself, not the patient. Every HSAT unit should have an ID, a current status (in inventory, shipped, with patient, returned, in scoring, in cleaning), a current location, and a chain-of-custody log. When the unit is shipped, the system should generate a tracking number, send the patient a setup video and instructions, and start a return-by deadline. If the deadline passes without a return, an automated outreach kicks in — text, email, then a phone call.

Three distribution models work depending on clinic size. Pickup model: patient comes to the clinic, gets fitted with the device, and returns it the next morning. Lowest device loss rate, highest staff time. Mail-out model: device ships to the patient's home with prepaid return. Highest patient convenience, requires careful tracking and longer turnaround. Hybrid: high-comorbidity or first-time patients pick up in person, repeat or low-risk patients mail out. Most growing clinics end up here.

The scoring handoff matters too. The moment the device is returned and the data is offloaded, the raw study should hit the scoring queue with a stamped turnaround clock. RPSGTs work from a queue with priorities (rush, routine, recheck) and the queue should be visible to the front desk so they can give honest answers when patients call asking for results.

In-lab polysomnography is where scheduling complexity peaks. You are matching three resources at the same time: a lab bed that is available overnight, a registered technologist credentialed to run the study, and the right montage and equipment for the suspected diagnosis. A standard diagnostic PSG, a split-night protocol, a CPAP titration, an MSLT (multiple sleep latency test), and a pediatric study all use the same bed but require different setup, different equipment, different tech skills, and different billing codes.

The most common failure mode is the empty lab bed. A patient cancels at 4pm for an 8pm study and the clinic does not have a same-day waitlist process. That bed represents $1,000-2,500 in lost revenue plus a tech who is paid to sit there. Every sleep lab needs a standby waitlist that can be activated within hours: patients who have completed all paperwork, are insurance-cleared, and have agreed to take a slot if it opens. A text blast to the standby list when a cancellation hits should be the default response, not a manual decision.

The second failure mode is room turnover. Beds need to be cleaned, linens changed, electrodes prepped, and equipment recalibrated between studies. If your lab runs more than one study per bed per night (rare but possible during MSLT scheduling), turnover time is the difference between making the next study on time and pushing into overtime. Build it explicitly into the schedule — do not assume techs will absorb it.

The third failure mode is montage drift. The wrong sensors get applied because the order did not specify the diagnosis clearly. The patient sleeps for eight hours and the resulting study cannot answer the clinical question. A good scheduling platform attaches the suspected diagnosis and the required montage to the appointment so the tech setting up the room is not guessing.

After the study comes the scoring bottleneck. A typical PSG generates eight hours of multi-channel signal data that an RPSGT manually scores epoch by epoch — staging sleep, identifying respiratory events, scoring leg movements, calculating an AHI. Most clinics use scoring software like Polysmith, Compumedics ProFusion, Natus SleepWorks, or RemLogic. The clinic's sleep medicine physician then reviews the scored study, interprets the findings, and signs the final report.

Turnaround is what patients judge you on. A 24-72 hour turnaround from study completion to signed report is the modern standard. Anything past a week and patients call your competitor for the next study. The bottleneck is almost never the equipment — it is the handoff between the scoring tech and the interpreting physician, and the report distribution back to the ordering provider.

Three practices that cut turnaround in half. First, batch scoring by physician — give each interpreter a queue that batches their patients together so they review in one focused session rather than one chart at a time. Second, automate report distribution. The moment the physician signs the report, it should fax to the ordering provider, upload to the patient portal, and trigger the next-step workflow (CPAP titration scheduling, follow-up appointment, DME order) without anyone clicking a button. Third, use templated interpretations for the routine cases — moderate OSA on CPAP recommendation is roughly the same paragraph for ninety percent of patients. Templates with smart fields collapse a 20-minute report into 5 minutes.

Integration with the scoring software matters. If your scheduling platform and your scoring software are separate and uncommunicative, every study requires manual re-entry of patient data, manual report upload, and manual status updates. A properly integrated stack pulls the scored data back into the chart automatically and updates the patient's status from 'in scoring' to 'awaiting interpretation' to 'report signed' without staff intervention.

If the diagnostic study confirms OSA and CPAP is indicated, the next step is determining the right pressure. There are two paths. Split-night protocol: during the original diagnostic PSG, if a clinically significant AHI is detected in the first portion of the night, the tech transitions the patient to CPAP and titrates pressure during the second half. This is efficient — one bed-night, one bill, one patient trip — but only works for clear-cut moderate-to-severe cases. Separate titration study: the patient returns for a second night dedicated to CPAP titration. Higher cost, two trips, but more reliable for borderline cases or patients who could not tolerate split-night.

More clinics now use auto-titrating CPAP (APAP) at home in lieu of an in-lab titration for straightforward cases. The patient receives an APAP machine, uses it for 1-2 weeks, and the resulting data is reviewed to set the final fixed pressure (or to leave the patient on APAP). This shifts cost from the lab to the DME process, which is generally a positive trade for both clinic and payer.

DME order fulfillment is the next handoff that breaks. The patient is diagnosed with OSA, CPAP is prescribed, and now a DME supplier — sometimes the clinic itself, sometimes an external supplier — needs to deliver the machine, the mask, the humidifier, and the supplies, plus do a setup visit and a mask fitting. If the clinic dispenses in-house, this is a billing and inventory workflow inside the practice management system. If a third-party DME, this is a referral with status tracking and a follow-up loop. Either way, the patient should not be the messenger — the system should track the DME order from prescription to delivery to fitting confirmation.

Mask fitting is underrated. The most common reason CPAP patients become non-adherent is mask discomfort. A 20-minute fitting visit with a respiratory therapist or trained DME tech, with the patient trying multiple mask types, is the highest-leverage intervention you can make for downstream adherence. Schedule it. Bill it. Document it.

This is the step that most sleep clinics underbuild and that Medicare will use to decide whether to keep paying for the equipment. Under Medicare's CPAP coverage policy, the patient must demonstrate adherence during a 90-day initial trial — defined as use of CPAP for at least 4 hours per night on 70% of nights during a consecutive 30-day window within the trial period. Fail that threshold and Medicare can pull coverage, leaving the patient with the device and no reimbursement to the supplier. Many commercial payers have adopted similar adherence rules.

Modern CPAP machines transmit usage data via cellular modem to a cloud platform (ResMed AirView, Philips Care Orchestrator, etc.). The clinic should be reviewing this data weekly during the trial, not at the end. A patient using CPAP 2 hours per night in week one is going to fail the 90-day check unless something changes. Catching that in week one is a phone call. Catching it in week 12 is a coverage denial.

The practical workflow: at 7 days post-dispense, automatically pull modem data and run an adherence projection. Flag patients projected to fail the 70%/4hr threshold. The flagged patients get a respiratory therapist outreach — usually a phone call, sometimes a virtual visit — to troubleshoot mask, pressure, leak, or sleep position issues. At 30 days, run the formal adherence check. At 90 days, the patient comes in for a billable follow-up visit, the clinician reviews the data and the patient's subjective response, the prescription is finalized, and the long-term DME billing is established.

Clinics that build this workflow into their software see adherence rates of 70-80% at 90 days. Clinics that wait until the patient calls with a complaint see 40-50%. The difference is not clinical skill — it is a workflow and reminder system that catches the problem early.

Deelo Practice is the practice management foundation. The Scheduling app handles the multi-resource booking — lab beds, technologists, and equipment — with rules for split-night, MSLT, pediatric, and titration protocols. Standby waitlists with automated text blasts on cancellation are built in, so an empty bed becomes a backfilled bed within minutes rather than lost revenue.

Referral intake runs through a digital intake flow with STOP-Bang and Berlin questionnaires, automatic routing to HSAT or PSG based on score and comorbidities, and payer-rule checks that flag when a payer requires PSG over HSAT before the appointment is booked. HSAT inventory tracking gives every device an ID, a status, a current location, and a return deadline with automated outreach when patients are late returning.

For scoring and reporting, Deelo integrates with major scoring software vendors so the scored study lands back in the patient chart automatically and the report distribution to ordering providers happens the moment the physician signs. CPAP titration scheduling, DME order generation, and mask fitting appointments are linked workflows triggered by the diagnosis, not separate manual steps.

The adherence reporting layer pulls CPAP modem data on a weekly cadence, projects whether the patient is on track to meet Medicare's 4hr/70% threshold, and surfaces flagged patients to a respiratory therapist work queue. Compliant patients flow to their 90-day follow-up automatically. Non-compliant patients get scheduled outreach before coverage is at risk.

Deelo is HIPAA-compliant by default with role-based access, encrypted PHI, audit logs, and a signed BAA. Pricing is $19/seat/month for Starter, $39 for Business, $69 for Enterprise — with annual volume discounts for clinics running 4+ beds. There is a free tier so clinics can test the workflow before committing.