Pulmonology After-Hours Calls: How AI Triages Breathing Complaints

Pulmonology after-hours calls sit on a fault line. A COPD patient calls at midnight because they feel more short of breath than usual. That could be a normal bad night with a chronic condition, or it could be an exacerbation heading toward respiratory failure. The words the patient uses – “I can’t catch my breath” – sound the same either way. The difference is in the details, and sorting those details after hours is exactly where most practices struggle.

The call volume is dominated by manageable questions: CPAP problems, inhaler technique, medication timing, oxygen equipment. But breathing is the one symptom where under-triage can turn deadly fast. Any after-hours system for pulmonology has to handle the routine majority efficiently while never missing the patient who is actually decompensating.

Why pulmonology after-hours volume is different

Respiratory conditions are chronic, symptomatic, and anxiety-provoking at night. Lying flat makes breathing harder. Cough worsens in the dark. CPAP masks leak, machines alarm, and sleep apnea patients wake up frustrated at 2 a.m. COPD and asthma patients live with a baseline level of breathlessness and have to constantly judge whether tonight is worse than usual.

That judgment is the problem. Patients are not equipped to distinguish a manageable flare from an exacerbation that needs the ED. So they call. And without triage, the call about a leaking CPAP mask lands in the same queue as the COPD patient whose oxygen saturation is dropping.

The clinical stakes are high. Respiratory distress can escalate quickly. That single fact means a pulmonology after-hours system cannot lean toward reassurance by default. It has to be built to catch the decompensating patient every time, while still handling the equipment questions without waking the on-call physician.

What pulmonology after-hours calls actually look like

The call mix breaks into four groups.

Equipment and device calls are the largest and most routine. CPAP and BiPAP mask fit, leaks, pressure discomfort, machine alarms, humidifier issues, and supply reorders. Home oxygen concentrator questions. Nebulizer troubleshooting. These are logistical, well-documented, and almost never require a physician after hours.

Medication and inhaler questions are the second group. Inhaler technique, rescue-versus-maintenance confusion, missed doses of controller medications, questions about prednisone tapers, or whether it is time to start a rescue course. Some need clinical input; many are answerable from the patient’s documented care plan.

Symptom-check calls are where the judgment lives. “I’m more short of breath than usual.” “My cough is worse.” “My chest feels tight.” These require structured questioning to place them on the spectrum from stable-chronic to acute-exacerbation. This is the group that cannot be handled generically.

Calls that need urgent escalation are the minority by count but the reason the system exists. Severe or rapidly worsening breathlessness, chest pain, blue lips or fingertips, confusion, oxygen saturation dropping on a home monitor, inability to speak in full sentences. These need a clinician immediately – and the risk is that they arrive undifferentiated alongside the CPAP question.

Why answering services fail respiratory patients

Most practices cover after hours with an answering service or direct on-call physician routing. Answering services fail respiratory patients because they have no chart access and no pulmonary context.

When a COPD patient calls about worsening shortness of breath, the answering service does not know their baseline, their oxygen requirement, whether they have a rescue action plan, or how many exacerbations they have had this year. Without that, “more short of breath than usual” is impossible to interpret. So the service either escalates everything, burying the on-call physician in CPAP questions, or falls back on generic guidance that under-serves the patient who is genuinely decompensating.

For a symptom where minutes matter, generic triage is not just inefficient. It is a safety gap. And none of it gets documented where the care team will see it the next day.

What AI can actually handle

AI voice agents integrated with a pulmonology EHR change the equation because they know the patient before the call starts.

When a patient calls at midnight short of breath, the AI pulls their chart. It knows they have moderate-to-severe COPD, are on home oxygen at 2 liters, had an exacerbation three months ago, and have a documented rescue action plan. That context turns a vague complaint into a structured, protocol-driven triage.

“You mentioned you are more short of breath than usual. Compared to a normal day, can you still speak in full sentences? Are you using your oxygen right now, and what is it set to? Do you have any chest pain, blue coloring in your lips or fingers, or a fever?”

Those are targeted questions drawn from the patient’s action plan, not a generic script. Based on the answers, the AI either walks the patient through the documented rescue steps or escalates immediately.

The categories AI handles well: equipment troubleshooting confirmed against device guides, medication and inhaler questions answered from the care plan, and administrative requests that never needed a physician.

The categories AI does not decide: anything suggesting respiratory decompensation. Severe breathlessness, chest pain, cyanosis, confusion, falling oxygen saturation, inability to speak in sentences. These route to on-call coverage immediately with a structured summary prepared.

The escalation protocol

A structured AI triage for pulmonology after-hours works like this.

The patient calls. The AI identifies them and pulls their chart – diagnosis, oxygen requirement, rescue action plan, recent exacerbation history, and the practice’s escalation thresholds.

Structured intake begins. What is the symptom? How does it compare to a normal day? Can you speak in full sentences? Any chest pain, cyanosis, or fever? The AI works through the red-flag list from the patient’s action plan, with thresholds set conservatively because respiratory distress moves fast.

If the responses indicate a stable chronic pattern, the AI walks the patient through their documented action-plan steps, offers a morning callback, and logs the interaction to the chart in real time.

If any red-flag appears, the AI immediately connects the on-call physician with a structured summary: patient name, diagnosis, oxygen requirement, reported symptom, how it compares to baseline, and the full conversation. The physician picks up already briefed and ready to decide whether this patient needs the ED tonight.

The athenahealth integration advantage

For pulmonology practices on athenahealth, native EHR integration is what makes accurate breathing triage possible.

Without integration, an AI agent works from whatever is passed at call setup – not enough to judge a respiratory complaint. With athenahealth integration, the AI has the diagnosis, oxygen orders, medication list, action plan, and exacerbation history before the first word. That is the difference between a generic breathlessness script and one that knows this patient’s baseline and escalates the moment they fall below it.

Integration also closes a documentation gap. Every after-hours interaction, AI-handled or escalated, is logged back to the chart. When the patient comes in for their next visit, the pulmonologist can see they called about worsening breathlessness, what they were guided to do, and whether it resolved. Continuity between after-hours contact and clinic follow-up is a known weak point in chronic respiratory care. Automatic charting closes it.

What implementation requires

Deploying AI for pulmonology after-hours coverage requires several things done right.

Conservative, action-plan-based thresholds. Breathing complaints are not the place for aggressive automation. Escalation rules should be built around each patient’s documented rescue plan, with red-flag thresholds set by the pulmonologists who take call. When in doubt, escalate.

Physician buy-in before go-live. The on-call pulmonologist needs to trust that the AI catches decompensation reliably. The setup phase should include physicians reviewing and approving escalation rules before any patient is routed through the system.

Transparency with patients. Patients should know they are speaking with an AI that asks structured questions and connects them to a clinician for anything urgent. Respiratory patients are anxious for good reason; clarity builds cooperation.

Morning review as a standard step. Every after-hours call should queue for care-team review the next morning. This creates accountability, catches edge cases, and improves the protocol over time.

Why this matters beyond call volume

The on-call burden in pulmonology is real, and the mix of routine equipment calls and genuine emergencies makes it draining. A system that handles CPAP and inhaler questions without a page keeps the on-call physician sharp for the exacerbation that calls at 3 a.m. That is a safety argument as much as a quality-of-life one.

The documentation benefit compounds. When a patient calls about a CPAP leak and gets walked through a fix, that interaction is on the chart before the next visit. The pulmonologist can address recurring equipment problems proactively instead of hearing about them for the first time months later. Better continuity, better adherence, better outcomes for a patient population whose treatment depends on consistent device use.

Answering services deliver none of that. The physician still gets paged, and the interaction disappears.

Key takeaways

Pulmonology after-hours calls range from routine CPAP and inhaler questions to genuine respiratory distress, and breathing is the symptom where under-triage is most dangerous
Answering services cannot triage respiratory complaints safely because they lack the patient’s baseline, oxygen requirement, and action plan
AI integrated with athenahealth knows the patient’s respiratory context before the call and applies conservative, action-plan-based escalation
Genuine red flags route to on-call physicians immediately with a full structured summary prepared
Every interaction is charted in real time, closing the continuity gap in chronic respiratory care
Conservative thresholds and pulmonologist sign-off on escalation rules are non-negotiable in this setting

Pulmonology after-hours call volume is not going away as long as patients are managing chronic respiratory disease at home. The question is who handles the routine equipment and medication majority and how reliably the decompensating patient gets through. AI triage built on chart context and conservative escalation can take the CPAP and inhaler calls, protect the on-call pulmonologist, and make sure the COPD patient sliding into an exacerbation reaches a clinician fast.

Sources

Management of COPD exacerbations. Global Initiative for Chronic Obstructive Lung Disease overview. https://pubmed.ncbi.nlm.nih.gov/30592252/
CPAP adherence in obstructive sleep apnea. Documents adherence challenges and follow-up patterns. https://pubmed.ncbi.nlm.nih.gov/27070243/

Internal links