The absence of standardized assessment protocols with well- defined measurement properties limits comparison of outcomes among those receiving long-term oxygen therapy (LTOT). We describe simple protocols for a hospital test, a simulated home test, and an actual home test, their reliability and relationship to each other. Stable patients with exercise hypoxemia participated. In 74 patients who completed four exercise tests, correlations between tests ranged from 0.85 to 0.78. Of these 27.0% had the same prescription from all four tests. In 46% prescriptions were within 1 L/ min and in 27% within 2 L/min. During exercise the hospital tests suggested slightly higher oxygen prescriptions than did the simulated home tests (2.5 L/min versus 2.0 L/min, p < 0.001). In 23 patients who participated in actual home assessments, the correlations between the home test, the hospital, and the simulated home tests were 0.22 (95% CI -0.24 to 0.67) and 0.27 (95% CI -0.18 to 0.72). In conclusion, standardizing tests for the assessment of LTOT is important. We describe simple hospital and simulated home tests that are reproducible, easy to carry out, and correlate well with each other.