Abstract Design This randomised crossover trial compared nocturnal auto-adjusting continuous positive airway pressure (APAP) and nocturnal oxygen therapy (NOT) in adults and children with sickle cell anaemia, with patient acceptability as the primary outcome. Secondary outcomes included pulmonary physiology (adults), safety, and daily pain during interventions and washout documented using tablet technology. Methods Inclusion criteria were age > 8 years and the ability to use an iPad to collect daily pain data. Trial participation was 4 weeks; week 1 involved baseline data collection and week 3 was a washout between interventions, which were administered for 7 days each during weeks 2 and 4 in a randomised order. Qualitative interviews were transcribed verbatim and analysed for content using a funnelling technique, starting generally and then gaining more detailed information on the experience of both interventions. Safety data included routine haematology and median pain days between each period. Missing pain day values were replaced using multiple imputation. Results Ten adults (three female, median age 30.2 years, range 18–51.5 years) and eleven children (five female, median age 12 years, range 8.7–16.9 years) enrolled. Nine adults and seven children completed interviews. Qualitative data revealed that the APAP machine was smaller, easier to handle, and less noisy. Of 16 participants, 10 preferred APAP (62.5%, 95% confidence interval (CI) 38.6–81.5%). Haemoglobin decreased from baseline on APAP and NOT (mean difference −3.2 g/L (95% CI −6.0 to −0.2 g/L) and −2.5 g/L (95% CI −4.6 to 0.3 g/L), respectively), but there was no significant difference between interventions (NOT versus APAP, 1.1 (−1.2 to 3.6)). Pulmonary function changed little. Compared with baseline, there were significant decreases in the median number of pain days (1.58 for APAP and 1.71 for NOT) but no significant difference comparing washout with baseline. After adjustment for carry-over and period effects, there was a non-significant median difference of 0.143 (95% CI −0.116 to 0.401) days additional pain with APAP compared with NOT. Conclusion In view of the point estimate of patient preference for APAP, and no difference in haematology or pulmonary function or evidence that pain was worse during or in washout after APAP, it was decided to proceed with a Phase II trial of 6 months APAP versus standard care with further safety monitoring for bone marrow suppression and pain. Trial registration ISRCTN46078697. Registered on 18 July 2014