PULSE OXIMETRY MEASUREMENT ERROR, OCCULT HYPOXEMIA, AND ORGAN DYSFUNCTION ACROSS RACIAL AND ETHNIC GROUPS IN CRITICALLY ILL ADULTS IN UNITED STATES
Author(s)
Minseol Jang, PharmD1, Hae Sun Suh, MA, MS, PhD2.
1MS Student, Kyung Hee University, Seoul, Korea, Republic of, 2Kyung Hee University, Seoul, Korea, Republic of.
1MS Student, Kyung Hee University, Seoul, Korea, Republic of, 2Kyung Hee University, Seoul, Korea, Republic of.
OBJECTIVES: Pulse oximetry may overestimate arterial oxygen saturation in some racial and ethnic groups, potentially delaying recognition of hypoxemia. We evaluated associations between race/ethnicity, pulse oximetry measurement error, and subsequent organ dysfunction.
METHODS: We conducted a retrospective cohort analysis of the Blood-gas and Oximetry Linked Dataset (BOLD). Eligible intensive care unit (ICU) records had paired pulse oximetry oxygen saturation (SpO2) and arterial oxygen saturation (SaO2) measurements and race/ethnicity recorded as White, Black, Hispanic or Latino, Asian, or American Indian/Alaska Native. Pulse oximetry measurement error was defined as SpO2 minus SaO2. Occult hypoxemia was defined as SpO2 of at least 92% with SaO2 below 88%. Multivariable linear and logistic regression adjusted for age, sex, body mass index, data source, arterial oxygen tension, pH, arterial carbon dioxide tension, hemoglobin, and prior 24-hour Sequential Organ Failure Assessment (SOFA) score. Clinical analyses evaluated change in SOFA score from the 24 hours before to the 24 hours after the paired measurement.
RESULTS: Among 45,519 ICU admissions, mean (SD) measurement error was 1.13 (3.81) percentage points in White patients and 1.76 (4.43) in Black patients. Occult hypoxemia occurred in 2.4% and 3.6% of White and Black patients, respectively. After adjustment, Black patients had greater measurement error than White patients (adjusted difference, 0.64 percentage points; 95% CI, 0.50-0.77) and higher risk of occult hypoxemia (risk ratio, 1.25; 95% CI, 1.03-1.52). The American Indian/Alaska Native subgroup showed both higher adjusted mean SOFA change (β = 1.06, 95% CI 0.51-1.61) and higher adjusted risk of at organ dysfunction (RR 1.32, 95% CI 1.14-1.52) compared with White patients.
CONCLUSIONS: In this study, pulse oximetry measurement error and adverse clinical outcomes differed across racial and ethnic groups. These findings underscore ongoing disparities in oxygenation assessment and critical care and support efforts to improve equitable monitoring practices.
METHODS: We conducted a retrospective cohort analysis of the Blood-gas and Oximetry Linked Dataset (BOLD). Eligible intensive care unit (ICU) records had paired pulse oximetry oxygen saturation (SpO2) and arterial oxygen saturation (SaO2) measurements and race/ethnicity recorded as White, Black, Hispanic or Latino, Asian, or American Indian/Alaska Native. Pulse oximetry measurement error was defined as SpO2 minus SaO2. Occult hypoxemia was defined as SpO2 of at least 92% with SaO2 below 88%. Multivariable linear and logistic regression adjusted for age, sex, body mass index, data source, arterial oxygen tension, pH, arterial carbon dioxide tension, hemoglobin, and prior 24-hour Sequential Organ Failure Assessment (SOFA) score. Clinical analyses evaluated change in SOFA score from the 24 hours before to the 24 hours after the paired measurement.
RESULTS: Among 45,519 ICU admissions, mean (SD) measurement error was 1.13 (3.81) percentage points in White patients and 1.76 (4.43) in Black patients. Occult hypoxemia occurred in 2.4% and 3.6% of White and Black patients, respectively. After adjustment, Black patients had greater measurement error than White patients (adjusted difference, 0.64 percentage points; 95% CI, 0.50-0.77) and higher risk of occult hypoxemia (risk ratio, 1.25; 95% CI, 1.03-1.52). The American Indian/Alaska Native subgroup showed both higher adjusted mean SOFA change (β = 1.06, 95% CI 0.51-1.61) and higher adjusted risk of at organ dysfunction (RR 1.32, 95% CI 1.14-1.52) compared with White patients.
CONCLUSIONS: In this study, pulse oximetry measurement error and adverse clinical outcomes differed across racial and ethnic groups. These findings underscore ongoing disparities in oxygenation assessment and critical care and support efforts to improve equitable monitoring practices.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
CO57
Topic
Clinical Outcomes, Real World Data & Information Systems
Topic Subcategory
Clinical Outcomes Assessment
Disease
Injury & Trauma