Table MTE1tg. Two group t-tests (TOST) of equivalence in means (equal n’s)

This sample size table is for a two one-sided t-test (TOST) procedure used in a two-group design for a test of the null hypothesis that the means are not equivalent (i.e. that the test drug mean differs from the standard drug mean by an amount ΔL or more in one direction or ΔU or more in the other direction). The alternative hypothesis is that the means are equivalent, defined as differing by the amount μT – μS (usually 0). Computed using bivariate non-central t distribution; see Manual Appendix and Help for definitions of the two non-centrality parameters. See nQuery Advisor® manual Appendix for details of computing methods. Results for this table will agree exactly with those for Table MTE0 when the opposite equivalence limit is set to an extreme value in MTE1tg. Also the sample size obtained in MTE1tg will agree with that obtained for MTE0 when μT – μS =0 and the power used to calculate n in MTE1tg is 100-2(100- P) where P is the power entered for MTE0. Figure MTE1tg shows computations for five examples. Additional extreme values tested: significance level .001, .20 one- sided by power of 50% and 99% and small and large equivalence limits to n, values of n equal to 3 and 2000 and small and large effect sizes back to power. Multiple examples checked against MTE0, PASS, spreadsheet calculations with the normal approximation, and tables in Phillips, K.E. “Power of the two one-sided tests procedure in bioequivalence” Journal of Pharmacokinetics and Biopharmaceutics 18(1990) pp. 137-143 with CV in Phillips divided by the squareroot of 2 (or SD diff in MTE1co divided by 2).

Figure MTE1tg. In columns 1, 2, 3 and 5, nQuery Advisor® 7.0 has solved for n, in column 4 for Power.