The authors aim to highlight the pitfalls of different validated methods used for the assessment of drugs' effect on QT duration. Digital 12-lead Holter electrocardiograms were recorded at baseline and after a single dose of sotalol in 39 healthy subjects (age = 27.4 +/- 8.0 years). Using both time- and rate-based approaches, the authors obtained averaged QRS-T complexes every minute ("time bins") and at different RR intervals ("rate bins"). Time bins were corrected for heart rate using a subject-specific approach. The individual alpha coefficients increased from placebo (0.309 +/- 0.052) to sotalol (0.454 +/- 0.136), P < .0001. When the placebo individual alpha coefficients were applied to correct the QT interval on sotalol, the changes were >5 ms smaller than those obtained using the ON drug alpha coefficients. The "rate"-averaging process leads to a complete loss of the time course of drug effect. In conclusion, the individual correction formula calculated from the placebo condition cannot always be used for QT correction on the drug.