QT interval correction methods.

All QT intervals were corrected for heart rate before further analyses were conducted. Four different correction methods were used to calculate QTc intervals. The most commonly used formulas are the Fridericia correction (equation 4) and Bazett correction (equation 5). The QTc_F and QTc_B intervals were used as a reference in this study.

where QT represents the measured QT interval, in milliseconds, and RR represents the RR interval, in seconds. Study-specific correction factors were investigated using the observed data generated in this study. Pretreatment QT and RR measurements were transformed into their natural logarithms and used to estimate a study-specific correction factor by applying a simple linear regression analysis (equation 6).

where β represents the intercept and α represents the slope of the regression. The estimated slope of the regression model was implemented as the correction factor and applied to all QT interval measurements in the study (equation 7):

Furthermore, QT interval measurements were also stratified on the day of measurement (i.e., day 1, day 3, and day 7), and a specific correction factor was estimated for each day and applied as a correction for data collected on that day (QTc_DAYS). QT intervals, corrected by the four methods (QTc_F, QTc_B, QTc_SSB, and QTc_DAYS), were used to investigate the effect of piperaquine on QTc prolongation. The slopes of the linear regression between QTc and RR intervals were used to evaluate the performance of each correction method (21). A slope close to zero represents a complete correction of the QTc calculations across the heart rate range and a consistent and appropriate performance of the method.