BackgroundAbout 90 % of patients with intra-thoracic malignancy experience breathlessness. Breathing training is helpful, but it is unknown whether repeated sessions are needed. The present study aims to test whether three sessions are better than one for breathlessness in this population.MethodsThis is a multi-centre randomised controlled non-blinded parallel arm trial. Participants were allocated to three sessions or single (1:2 ratio) using central computer-generated block randomisation by an independent Trials Unit and stratified for centre. The setting was respiratory, oncology or palliative care clinics at eight UK centres. Inclusion criteria were people with intrathoracic cancer and refractory breathlessness, expected prognosis ≥3 months, and no prior experience of breathing training. The trial intervention was a complex breathlessness intervention (breathing training, anxiety management, relaxation, pacing, and prioritisation) delivered over three hour-long sessions at weekly intervals, or during a single hour-long session. The main primary outcome was worst breathlessness over the previous 24 hours (‘worst’), by numerical rating scale (0 = none; 10 = worst imaginable). Our primary analysis was area under the curve (AUC) ‘worst’ from baseline to 4 weeks. All analyses were by intention to treat.ResultsBetween April 2011 and October 2013, 156 consenting participants were randomised (52 three; 104 single). Overall, the ‘worst’ score reduced from 6.81 (SD, 1.89) to 5.84 (2.39). Primary analysis [n = 124 (79 %)], showed no between-arm difference in the AUC: three sessions 22.86 (7.12) vs single session 22.58 (7.10); P value = 0.83); mean difference 0.2, 95 % CIs (–2.31 to 2.97). Complete case analysis showed a non-significant reduction in QALYs with three sessions (mean difference –0.006, 95 % CIs –0.018 to 0.006). Sensitivity analyses found similar results. The probability of the single session being cost-effective (threshold value of £20,000 per QALY) was over 80 %.ConclusionsThere was no evidence that three sessions conferred additional benefits, including cost-effectiveness, over one. A single session of breathing training seems appropriate and minimises patient burden.Trial registrationRegistry: ISRCTN; Trial registration number: ISRCTN49387307; http://www.isrctn.com/ISRCTN49387307; registration date: 25/01/2011Electronic supplementary materialThe online version of this article (doi:10.1186/s12916-015-0453-x) contains supplementary material, which is available to authorized users.