The aim of the study was to investigate compliance with long-term oxygen therapy (LTOT), the consumption of mobile oxygen and factors that might influence on these measures. The study included 182 patients with LTOT and 125 patients answered a questionnaire on daily activities. Information on arterial gas tension, lung function, and diagnosis was obtained from the medical file. Data on oxygen systems and the actual usage of oxygen were obtained from the oxygen suppliers. Seventy-six per cent of the patients had a concentrator and 35% had both stationary and mobile oxygen. Good compliance (use of oxygen > 15 h day-1) was seen in 65%. Of the tested factors, only mobile oxygen had an influence on the effective usage. Outdoor activity was reported in 65% of the patients. Only 48% of these patients had mobile units, while 21% of the home-bound patients had mobile units. Of the patients with mobile oxygen, only 39% used their mobile unit > 2 h week-1. Poorer usage of stationary oxygen was observed in patients with outdoor activity (2.4 hours day-1 less). Although only 16 (13%) patients reported a usage less than 15 h day-1, the actual consumption was less than 15 h day-1 in 29%. In conclusion, patients on LTOT tended to overestimate their oxygen usage. Acceptable compliance was observed in 65% of the patients. Only mobile oxygen had a significant impact on the overall compliance. As more patients had outdoor activities than expected, and as these activities had a negative impact on the usage of stationary oxygen, more attention must be paid on detecting outdoor activities. However, as fewer than half of the patients with outdoor activity and mobile oxygen used their mobile systems, there is a need for measures to improve compliance with mobile oxygen. Using lists referring to the actual usage and discussing the usage with the patients may improve the compliance and cost-benefit.