Background: Approximately 14% of UK hospital in-patients receive supplemental oxygen therapy, only 57% have valid prescriptions. Oxygen must be optimally prescribed to ensure maximal therapeutic response whilst minimizing adverse outcomes (including fatality). This study investigates prescription compliance. Methods: All adults admitted to medical wards (18 February to 3 March 2020) were included. Analyses present proportions, descriptive statistics, and hypothesis testing. Ethical approval was unnecessary for this audit. Results: Of the 636 patients admitted, 66 (10%) were receiving oxygen therapy. Ages ranged from 34 to 100 years with 36 (54.5%) males and 30 (45.5%) females. The prescription was not documented in the oxygen section of the drug chart (n = 37, 56.1%, p = 0.389), nor did it have the physicians signature (n = 40, 60.6%, p = 0.110) nor date (n = 46, 69.7%, p = 0.002). Thirteen chronic obstructive pulmonary disease (COPD) patients (19.7%) were at risk of hypercapnic failure (p = 1.582x10 −6). Target oxygen saturation (SpO2) range had been documented for 30 (45.5%) patients. A target SpO2 range of 88-92% was documented for 9 patients (13.6%), a 94-98% range documented for 11 patients (16.7%). All patients had an invalid prescription. Conclusion: We present real-world practice in naturalistic settings, immediately before pandemiclockdown. Enhanced compliance is advocated to reduce risks of harm and mortality.