The International Global H-mode Threshold Database has been revised andcleaned from inadequate data for some tokamaks. This significantly reducesthe scatter for the corresponding datasets. The resulting new analysesincluding all ten tokamaks yields results comparable toprevious work. In this case, the uncertainties are essentiallydetermined by the inability of the linear models to take the differentdependencies into account with sufficient accuracy.However, restricting the database to a more homogeneous subsetof medium and large size tokamaks with single-null elongatedplasma shapes yields expressions with much lower scatter.In this case the contribution from the experimental scatterin the individual machines plays a role.These regressions are stable against excluding one tokamak in turn.