The impact of skin tone on performance of pulse oximeters used by NHS England COVID Oximetry @home scheme: measurement and diagnostic accuracy study

Abstract Objectives To assess the impact of skin tone on the measurement and diagnostic accuracy of five fingertip pulse oximeters used by patients in the NHS (National Health Service) England COVID Oximetry @home scheme. Design Measurement and diagnostic accuracy study (exploring pulse oximeter accuracy across skin tones—EXAKT). Setting Twenty four intensive care units in England between June 2022 and August 2024. Participants 903 critically ill adults admitted to intensive care units screened for or enrolled into a trial evaluating different approaches to oxygen therapy. Interventions Pulse oximetry derived peripheral oxygen saturation (SpO 2 ) measurements were compared with paired arterial oxygen saturation (SaO 2 ) measurements from arterial blood analysed by co-oximetry (gold standard). Skin tone (individual typology angle) was objectively measured using a handheld spectrophotometer. Main outcome measures Pulse oximeter measurement accuracy was assessed for bias, precision, and overall accuracy. Diagnostic accuracy for identifying SaO 2 ≤92% was assessed by false negative and false positive rates for SpO 2 using thresholds of ≤92% and ≤94%, and the area under the receiver operating characteristic curve, and by the presence of occult hypoxaemia (SaO 2 92%). Results 11 018 paired SpO 2 -SaO 2 measurements were analysed. All tested pulse oximeters overestimated at lower values and underestimated at higher values of SaO 2 . On average, SpO 2 readings were 0.6-1.5 percentage points higher for patients with darker skin tone (individual typology angle −44°) than for those with lighter skin tone (46°). At both SpO 2 thresholds assessed, false negative rates increased with darker skin tones; the proportion of SpO 2 measurements >94% despite a paired SaO 2 ≤92% ranged from 5.3 to 35.3 percentage points higher for patients with darker skin tones than for those with lighter skin tones (7.6-62.2% v 1.2-26.9%, rate ratio 2.3-7.1). By contrast, false positive rates decreased with darker skin tones. Conclusions Five pulse oximeters provided by the NHS England COVID Oximetry @home scheme yielded higher SpO 2 measurements for patients with darker skin tones compared with those with lighter skin tones, which could translate into potentially clinically important differences in false negative and false positive rates for detecting hypoxaemia. Trial registration ClinicalTrials.gov NCT05481515