A transformation of oxygen saturation (the saturation virtual shunt) to improve clinical prediction model calibration and interpretation

ABSTRACT Background The relationship between peripheral oxygen saturation (SpO 2 ) and the inspired oxygen concentration is non-linear. SpO 2 is frequently used as a dichotomized predictor, to manage this non-linearity. We propose the saturation virtual shunt (VS) as a transformation of SpO 2 to a continuous linear variable to improve interpretation of disease severity within clinical prediction models. Method We calculate the saturation VS based on an empirically derived approximation formula between physiological VS and SpO 2 . We evaluated the utility of the saturation VS in a clinical study predicting the need for facility admission in children in a low resource health-care setting. Results The transformation was saturation VS = 68.864*log 10 (103.711 − SpO 2 ) −52.110 . The ability to predict hospital admission based on a dichotomized SpO 2 produced an area under the receiver operating characteristic curve of 0.57, compared to 0.71 based on the untransformed SpO 2 and saturation VS . However, the untransformed SpO 2 demonstrated a lack of fit compared to the saturation VS (goodness-of-fit test p-value <0.0001 versus 0.098). The observed admission rates varied non-linearly with the untransformed SpO 2 but varied linearly with the saturation VS. Conclusion The saturation VS estimates a continuous linearly interpretable disease severity based on SpO 2 and improves clinical prediction.