DRIFTS-based multivariate calibration and prediction of low-concentration polymorphic impurities in multiple lots of an active pharmaceutical ingredient, and outlier criteria.

•DRIFTS-based multivariate calibrations of two polymorphic impurities in ∼0 to 2% wt/wt range produce very different cross validations errors.•The predictions of both polymorphs via spectra of numerous independent API lots are biased and with numerous outliers, per Q-residuals.•Updating models with the spectra from some of the independent API batches increases calibration errors but reduces bias and eliminates outliers.