Text this: Preventing false discovery of heterogeneous treatment effect subgroups in randomized trials