Aquatic therapy for boys with Duchenne muscular dystrophy (DMD): an external pilot randomised controlled trial

Table 10 Results are based on 10,000 simulations

True treatment difference	Probability future RCT declares AT plus optimised land-based exercises superior to optimised land-based exercises alone
True treatment difference	N = 20	N = 30	N = 40	N = 60
0	0.10	0.10	0.11	0.10
δ/2	0.27	0.33	0.38	0.39
δ	0.52	0.64	0.73	0.75
3δ/2	0.75	0.88	0.94	0.93

N is the total sample size divided equally between interventions. Primary endpoint of a future Bayesian trial would be change from baseline at 6 months in linearised NSAA score, and the ‘true treatment effect’ refers to the underlying difference between average outcomes on each intervention. Future trial data are simulated according to the model \( \widehat{\theta} \) ~ N(θ, 4σ²/N) and s² ~ (σ²/N) χ ²_N _− 2 setting σ = 15 and δ = 9, where \( \widehat{\theta} \) is the difference between sample mean outcomes on each treatment arm (and θ is the true treatment difference) and s ² is the pooled sample response variance

ISSN: 2055-5784