Potential source of bias in AI models: lactate measurement in the ICU in sepsis patients as a template
ObjectiveHealth inequities may be driven by demographics such as sex, language proficiency, and race-ethnicity. These disparities may manifest through likelihood of testing, which in turn can bias artificial intelligence models. We aimed to evaluate variation in serum lactate measurements in the int...
Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-07-01
|
| Series: | Frontiers in Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmed.2025.1606254/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849320518398771200 |
|---|---|
| author | Pratiksha Pradhan Pratiksha Pradhan Fredrik Willumsen Haug Fredrik Willumsen Haug Nebal S. Abu Hussein Nebal S. Abu Hussein Dana Moukheiber Lama Moukheiber Mira Moukheiber Sulaiman Moukheiber Luca Leon Weishaupt Jacob G. Ellen Helen D'Couto Ishan C. Williams Leo Anthony Celi Leo Anthony Celi Leo Anthony Celi Joao Matos Joao Matos Joao Matos Tristan Struja Tristan Struja Tristan Struja |
| author_facet | Pratiksha Pradhan Pratiksha Pradhan Fredrik Willumsen Haug Fredrik Willumsen Haug Nebal S. Abu Hussein Nebal S. Abu Hussein Dana Moukheiber Lama Moukheiber Mira Moukheiber Sulaiman Moukheiber Luca Leon Weishaupt Jacob G. Ellen Helen D'Couto Ishan C. Williams Leo Anthony Celi Leo Anthony Celi Leo Anthony Celi Joao Matos Joao Matos Joao Matos Tristan Struja Tristan Struja Tristan Struja |
| author_sort | Pratiksha Pradhan |
| collection | DOAJ |
| description | ObjectiveHealth inequities may be driven by demographics such as sex, language proficiency, and race-ethnicity. These disparities may manifest through likelihood of testing, which in turn can bias artificial intelligence models. We aimed to evaluate variation in serum lactate measurements in the intensive care unit (ICU) in sepsis.MethodsUtilizing MIMIC-IV (2008–2019), we identified adults fulfilling sepsis-3 criteria. Exclusion criteria were ICU stay < 1-day, unknown race-ethnicity, < 18 years of age, and recurrent ICU-stays. Employing targeted maximum likelihood estimation analysis, we assessed the likelihood of a lactate measurement on day 1. For patients with a measurement on day 1, we evaluated the predictors of subsequent readings.ResultsWe studied 15,601 patients (19.5% racial-ethnic minority, 42.4% female, and 10.0% limited English proficiency). After adjusting for confounders, Black patients had a slightly higher likelihood of receiving a lactate measurement on day 1 [odds ratio 1.19, 95% confidence interval (CI) 1.06–1.34], but not the other minority groups. Subsequent frequency was similar across race-ethnicities, but women had a lower incidence rate ratio (IRR) 0.94 (95% CI 0.90–0.98). Patients with elective admission and private insurance also had a higher frequency of repeated serum lactate measurements (IRR 1.70, 95% CI 1.61–1.81 and 1.07, 95% CI, 1.02–1.12, respectively).ConclusionWe found no disparities in the likelihood of a lactate measurement among patients with sepsis across demographics, except for a small increase for Black patients, and a reduced frequency for women. Subsequent analyses should account for the variation in biomarker monitoring being present in MIMIC-IV. |
| format | Article |
| id | doaj-art-5d6c99a56a50418c9feda94cfd366186 |
| institution | Kabale University |
| issn | 2296-858X |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Medicine |
| spelling | doaj-art-5d6c99a56a50418c9feda94cfd3661862025-08-20T03:50:05ZengFrontiers Media S.A.Frontiers in Medicine2296-858X2025-07-011210.3389/fmed.2025.16062541606254Potential source of bias in AI models: lactate measurement in the ICU in sepsis patients as a templatePratiksha Pradhan0Pratiksha Pradhan1Fredrik Willumsen Haug2Fredrik Willumsen Haug3Nebal S. Abu Hussein4Nebal S. Abu Hussein5Dana Moukheiber6Lama Moukheiber7Mira Moukheiber8Sulaiman Moukheiber9Luca Leon Weishaupt10Jacob G. Ellen11Helen D'Couto12Ishan C. Williams13Leo Anthony Celi14Leo Anthony Celi15Leo Anthony Celi16Joao Matos17Joao Matos18Joao Matos19Tristan Struja20Tristan Struja21Tristan Struja22Laboratory for Computational Physiology, Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, United StatesCollege of Engineering, Northeastern University, Boston, MA, United StatesLaboratory for Computational Physiology, Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, United StatesHarvard John A. Paulson School of Engineering and Applied Sciences, Boston, MA, United StatesLaboratory for Computational Physiology, Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, United StatesPulmonary Critical Care Sleep Medicine Division, Yale School of Medicine, New Haven, CT, United StatesLaboratory for Computational Physiology, Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, United StatesLaboratory for Computational Physiology, Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, United StatesPicower Institute for Learning and Memory, Massachusetts Institute of Technology, Cambridge, MA, United StatesDepartment of Computer Science, Worcester Polytechnic Institute Computer Science, Worcester, MA, United StatesLaboratory for Computational Physiology, Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, United StatesHarvard Medical School, Boston, MA, United StatesDivision of Pulmonary, Critical Care, and Sleep Medicine, Georgetown University Hospital, Washington, DC, United StatesSchool of Nursing, University of Virginia, Charlottesville, VA, United StatesLaboratory for Computational Physiology, Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, United States0Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA, United States1Division of Pulmonary, Critical Care and Sleep Medicine, Beth Israel Deaconess Medical Center, Boston, MA, United StatesLaboratory for Computational Physiology, Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, United States2Faculty of Engineering, University of Porto (FEUP), Porto, Portugal3Institute for Systems and Computer Engineering, Technology and Science (INESCTEC), Porto, PortugalLaboratory for Computational Physiology, Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, United States4Medical University Clinic, Kantonsspital Aarau, Aarau, Switzerland5Hospital Muri, Muri Aargau, SwitzerlandObjectiveHealth inequities may be driven by demographics such as sex, language proficiency, and race-ethnicity. These disparities may manifest through likelihood of testing, which in turn can bias artificial intelligence models. We aimed to evaluate variation in serum lactate measurements in the intensive care unit (ICU) in sepsis.MethodsUtilizing MIMIC-IV (2008–2019), we identified adults fulfilling sepsis-3 criteria. Exclusion criteria were ICU stay < 1-day, unknown race-ethnicity, < 18 years of age, and recurrent ICU-stays. Employing targeted maximum likelihood estimation analysis, we assessed the likelihood of a lactate measurement on day 1. For patients with a measurement on day 1, we evaluated the predictors of subsequent readings.ResultsWe studied 15,601 patients (19.5% racial-ethnic minority, 42.4% female, and 10.0% limited English proficiency). After adjusting for confounders, Black patients had a slightly higher likelihood of receiving a lactate measurement on day 1 [odds ratio 1.19, 95% confidence interval (CI) 1.06–1.34], but not the other minority groups. Subsequent frequency was similar across race-ethnicities, but women had a lower incidence rate ratio (IRR) 0.94 (95% CI 0.90–0.98). Patients with elective admission and private insurance also had a higher frequency of repeated serum lactate measurements (IRR 1.70, 95% CI 1.61–1.81 and 1.07, 95% CI, 1.02–1.12, respectively).ConclusionWe found no disparities in the likelihood of a lactate measurement among patients with sepsis across demographics, except for a small increase for Black patients, and a reduced frequency for women. Subsequent analyses should account for the variation in biomarker monitoring being present in MIMIC-IV.https://www.frontiersin.org/articles/10.3389/fmed.2025.1606254/fullsepsislactateMIMIC-IVcritical carehealth equity |
| spellingShingle | Pratiksha Pradhan Pratiksha Pradhan Fredrik Willumsen Haug Fredrik Willumsen Haug Nebal S. Abu Hussein Nebal S. Abu Hussein Dana Moukheiber Lama Moukheiber Mira Moukheiber Sulaiman Moukheiber Luca Leon Weishaupt Jacob G. Ellen Helen D'Couto Ishan C. Williams Leo Anthony Celi Leo Anthony Celi Leo Anthony Celi Joao Matos Joao Matos Joao Matos Tristan Struja Tristan Struja Tristan Struja Potential source of bias in AI models: lactate measurement in the ICU in sepsis patients as a template Frontiers in Medicine sepsis lactate MIMIC-IV critical care health equity |
| title | Potential source of bias in AI models: lactate measurement in the ICU in sepsis patients as a template |
| title_full | Potential source of bias in AI models: lactate measurement in the ICU in sepsis patients as a template |
| title_fullStr | Potential source of bias in AI models: lactate measurement in the ICU in sepsis patients as a template |
| title_full_unstemmed | Potential source of bias in AI models: lactate measurement in the ICU in sepsis patients as a template |
| title_short | Potential source of bias in AI models: lactate measurement in the ICU in sepsis patients as a template |
| title_sort | potential source of bias in ai models lactate measurement in the icu in sepsis patients as a template |
| topic | sepsis lactate MIMIC-IV critical care health equity |
| url | https://www.frontiersin.org/articles/10.3389/fmed.2025.1606254/full |
| work_keys_str_mv | AT pratikshapradhan potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT pratikshapradhan potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT fredrikwillumsenhaug potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT fredrikwillumsenhaug potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT nebalsabuhussein potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT nebalsabuhussein potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT danamoukheiber potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT lamamoukheiber potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT miramoukheiber potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT sulaimanmoukheiber potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT lucaleonweishaupt potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT jacobgellen potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT helendcouto potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT ishancwilliams potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT leoanthonyceli potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT leoanthonyceli potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT leoanthonyceli potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT joaomatos potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT joaomatos potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT joaomatos potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT tristanstruja potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT tristanstruja potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate AT tristanstruja potentialsourceofbiasinaimodelslactatemeasurementintheicuinsepsispatientsasatemplate |