D Some theory on equivalence trials

D.1 Type I error in equivalence trials

Assume we have a continuous outcome with variance σ2 and two treatment groups of size nA and nB, respectively. Now consider the difference ˉd of the mean outcomes as treatment effect, so ˉdN(Δ,σ2λ2) where the true treatment difference is Δ and λ=1/nA+1/nB.

Equivalence is established if the γ100 % confidence interval for the treatment difference is within the pre-specified interval of equivalence (δ,δ), i.e. ˉd±z(1+γ)/2σλ(δ,δ). With z=z(1+γ)/2 we obtain the equivalent requirement ˉd(ξ,ξ) where ξ=δz(1+γ)/2σλ

to establish equivalence. This happens with probability

Pr(ˉd(ξ,ξ))=Φ(ξΔλσ)Φ(ξΔλσ)=Φ(δΔλσz(1+γ)/2)Φ(δΔλσ+z(1+γ)/2).

Now evaluate (D.1) at Δ=δ to obtain the Type I error rate (more precisely an upper bound on the Type I error rate)

α=Φ(z(1+γ)/2)Φ(z(1+γ)/22δλσ).

Likewise, the power 1β is derived from (D.1) with Δ=0:

1β=2Φ(δλσz(1+γ)/2)1 so 1β/2=Φ(δλσz(1+γ)/2).

Note that 1β/2=Φ(z1β/2), and therefore

δλσ=z(1+γ)/2+z1β/2.

With (D.2) we obtain

αΦ(z(1+γ)/2)=1(1+γ)/2=(1γ)/2

for relatively small β (say β30 %), since the second term in (D.2)

Φ(z(1+γ)/22δλσ)=Φ(z(1+γ)/22(z(1+γ)/2+z1β/2))=Φ(z(1+γ)/22z1β/2)

is then very close to zero. Note that (D.3) differs from superiority trials, where α=1γ.

D.2 Sample size calulations in equivalence trials

Assume equal group sizes, i.e. n=nA=nB, then λ=2/n.

With (D.3) we then obtain

n=2σ2(z(1+γ)/2+z1β/2)2δ2=2σ2(z1α+z1β/2)2δ2

as the required sample size in each group for Type I error rate α and power 1β.