Featured Review · Pulmonary Medicine
Outpatient oxygen therapy should not begin with a tank. It should begin with one question: when does this patient actually become hypoxemic?
SMART Brevity: the signal first
Why it matters: Supplemental oxygen is common, expensive, and burdensome. The strongest evidence is for severe resting hypoxemia, while benefit is much less certain for moderate resting desaturation, isolated exertional desaturation, and nocturnal desaturation. At the same time, the delivery system itself can either preserve or reduce a patient’s independence.
The Pearl: Ask when the hypoxemia occurs—at rest, with exertion, or during sleep—and prescribe oxygen only when the evidence, the patient’s goals, and the delivery system all line up.
2½-minute capsule summary
Watch: Outpatient Long-Term Oxygen Therapy Made Simple
Prefer the visual version first? This short video walks through the same practical decision path: resting, exertional, and nocturnal hypoxemia.
What the evidence actually supports
The 2025 JAMA Internal Medicine review by Suen and colleagues makes one point especially clear: the evidence base for outpatient supplemental oxygen is uneven. Most high-quality evidence comes from COPD, and the mortality benefit is established primarily in patients with severe resting desaturation.
| Clinical situation | Practical interpretation |
|---|---|
| Severe resting desaturation SpO₂ ≤88% or PaO₂ ≤55 mm Hg | Strongest evidence for LTOT. In COPD, landmark trials demonstrated survival benefit. Typical recommendation is ≥15 h/day, titrated to keep SpO₂ above 89%. |
| Moderate resting or exertional desaturation SpO₂ 89%–93% | Routine LTOT is not supported in COPD. The LOTT trial did not improve mortality, hospitalization, quality of life, anxiety, depression, or functional status. |
| Severe exertional desaturation only | Evidence is mixed. Laboratory walk distance may improve modestly, but real-world symptom and functional benefit is inconsistent. Individualized testing and shared decision-making are appropriate. |
| Isolated nocturnal desaturation | Routine nocturnal oxygen is not supported for COPD by current evidence. Evaluate for obstructive sleep apnea and other causes of nighttime hypoxemia. |
A practical three-question framework
1. Is the patient hypoxemic at rest?
Start with room-air resting SpO₂. Severe resting desaturation—SpO₂ of 88% or less—meets the major evidence-based threshold for LTOT. In COPD, long-term oxygen should generally be used for at least 15 hours per day. The review also notes that patients with pulmonary hypertension or cor pulmonale may qualify at slightly higher thresholds, and oxygen targets may be higher in that setting.
By contrast, a resting SpO₂ of 89% to 93% is not, by itself, a reason to prescribe LTOT in COPD. This distinction matters because oxygen is not a benign convenience: it adds equipment, tubing, fall risk, cost, stigma, and logistical burden.
2. Does the patient desaturate with exertion?
If the patient is mobile and reports exertional breathlessness, cough, or fatigue, test during activity. The review favors an individualized approach because the benefit of oxygen for isolated severe exertional desaturation remains uncertain. Exercise oximetry is practical in clinic; a 6-minute walk test offers a more standardized assessment of desaturation, walk distance, and breathlessness.
For qualification testing, the authors emphasize documenting three values: resting room-air saturation, exertional room-air saturation, and exertional saturation while oxygen is titrated. If SpO₂ falls to 88% or less during walking, oxygen can be titrated to maintain the clinician-prescribed target.
3. Is the desaturation only nocturnal?
Nocturnal desaturation should not automatically trigger oxygen therapy. In COPD, the INOX trial did not show improvement in survival or progression to LTOT with nocturnal oxygen for isolated nighttime desaturation. The first question should instead be whether obstructive sleep apnea, obesity hypoventilation, or another cause of nocturnal hypoxemia is present.
Qualification is only half the prescription
The review’s most important contribution may be that it reframes oxygen as a person-centered intervention, not simply a saturation threshold. Approximately 1.5 million US adults use outpatient supplemental oxygen, yet many receive systems that do not adequately support mobility, work, caregiving, or independence.
A prescription therefore needs to answer more than “how many liters?” It should address:
- When oxygen is needed: rest, exertion, sleep, or continuous use.
- How much oxygen is needed: continuous-flow requirement versus a pulsed-flow setting.
- Whether the patient is mobile: work, appointments, exercise, stairs, errands, and caregiving responsibilities.
- Whether the device can actually deliver the prescribed oxygen: especially during exertion.
- Whether the patient can safely operate and carry the equipment: cognition, strength, gait stability, and fall risk all matter.
- Whether the patient and caregiver understand the equipment: troubleshooting, fire safety, tubing, batteries, and when to call for help.
Portable oxygen concentrators are not interchangeable with liters per minute
One practical source of confusion is the portable oxygen concentrator. Most portable concentrators use pulsed flow, and a device setting of “2” does not necessarily equal 2 L/min of continuous oxygen. Performance also varies between devices, particularly in patients who are tachypneic, weak, or require higher flow.
The review recommends confirming that the patient’s actual portable device maintains adequate oxygenation during exertion. For mobile patients who require more than 3 L/min of continuous flow, liquid oxygen may be preferable when available because it can provide higher continuous flows with greater portability.
The hidden harms of an oxygen prescription
Oxygen can relieve symptoms and provide reassurance, but patients also describe substantial burdens: heavy tanks, limited mobility, fear of running out, dependence on caregivers, social isolation, stigma, and tripping hazards from tubing. Some patients interpret starting oxygen as a sign that they are entering the final stage of illness.
That is why shared decision-making matters most when the evidence is uncertain. An active patient with severe exertional desaturation who wants to exercise may value ambulatory oxygen very differently from a frail patient with frequent falls and the same oximetry result.
Reassess—and sometimes stop oxygen
Oxygen started during an acute hospitalization should not automatically become permanent. The review highlights data showing that many patients are never reassessed after discharge, even though a substantial proportion no longer meet oxygen criteria when retested. Close follow-up should confirm ongoing need, adequacy of the device, symptoms, adherence, and whether oxygen is helping the patient achieve the goals that justified starting it.
The Pearl
Prescribe oxygen to a person, not to a pulse oximeter. First identify whether hypoxemia occurs at rest, with exertion, or during sleep. Then choose a delivery system that actually supports the patient’s physiology, mobility, goals, and daily life. Reassess the need over time.
Source review
Suen AO, Jacobs SS, Kitlowski MR, Branson RD, Iyer AS. A Person-Centered Approach to Supplemental Oxygen Therapy in the Outpatient Setting: A Review. JAMA Intern Med. 2025;185(6):720–733. doi:10.1001/jamainternmed.2025.0279.
Educational use: This review summary is intended for medical education and does not replace individualized assessment, payer-specific qualification requirements, or the complete source article and clinical guidelines.
