Two landmark trials conducted more than 35 years ago provided scientific evidence that, under very specific circumstances, long-term oxygen therapy (LTOT) may prolong life. These two trials enrolled 290 patients with chronic obstructive pulmonary disease and severe daytime hypoxemia documented by direct arterial blood gas measurement. From that time, LTOT became a standard of care, and the indications for oxygen therapy expanded to include nocturnal oxygen therapy for isolated nocturnal oxygen desaturation, ambulatory oxygen to correct exercise-induced desaturation, and short-burst oxygen to relieve dyspnea. In most cases, the rationale for broadening the indications for oxygen therapy is that, if hypoxemia exists, correcting it by increasing the FiO2 should help. However, with the exception of LTOT in severely hypoxemic patients with chronic obstructive pulmonary disease, randomized controlled trials of oxygen therapy have failed to demonstrate clinically significant benefits. Also, adherence to LTOT is usually suboptimal. Important areas for future research include improving understanding of the mechanisms of action of supplemental oxygen, the clinical and biochemical predictors of responsiveness to LTOT, the methods for measuring and enhancing adherence to LTOT, and the cost-effectiveness of oxygen therapy. A standardization of terminology to describe the use of supplemental oxygen at home is provided.