Demographic Factors Associated with Triage Acuity, Admission and Length of Stay During Adult Emergency Department Visits

Helena Coggan, Pradip Chaudhari, Yuval Barak-Corren, Andrew M. Fine, Ben Y. Reis, Jaya Aysola, William G. La Cava · 2025

Read Demographic Factors Associated with Triage Acuity, Admission and Length of Stay During Adult Emergency Department Visits by Helena Coggan, Pradip Chaudhari, Yuval Barak-Corren, Andrew M. Fine, Ben Y. Reis, Jaya Aysola, William G. La Cava (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-22781.txt.

Introduction

Helena Coggan and co-authors published this 2025 study on how patient traits relate to emergency department care. They used records from one US medical center covering 2011 to 2019. The team asked whether gender, race, and age track how urgent a case is judged, whether the patient is admitted, and how long the visit lasts.

Matching methods balanced the groups before comparison, and models adjusted for vital signs and diagnoses. Black patients were more likely than White patients to wait long before admission, though not before discharge. Men were more often judged urgent or admitted, yet stayed shorter times. Patients over thirty were also judged more urgent and admitted more, but stayed longer. The paper documents uneven patterns worth probing in triage practice.

Catalog record

CREATORHelena Coggan, Pradip Chaudhari, Yuval Barak-Corren, Andrew M. Fine, Ben Y. Reis, Jaya Aysola, William G. La Cava
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-22781

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Demographic Factors Associated with Triage Acuity, Admission and Length of Stay During Adult Emergency Department Visits Helena Coggan, Pradip Chaudhari, Yuval Barak-Corren, Andrew M. Fine, Ben Y. Reis, Jaya Aysola, William G. La Cava Abstract Objective: To describe the association of demographic factors with triage acuity, hospital admission rates, and length of stay (LOS) for adult patients in the emergency department (ED). Methods: We performed a retrospective cross-sectional analysis using publicly available electronic health records describing visits to the ED of a single US medical center during 2011-2019. The primary exposures of interest were self-reported gender, race/ethnicity, and age. The outcomes studied were triage acuity, admission to hospital, and LOS in the ED. Odds ratios were calculated using propensity-score matching. Analyses were adjusted for confounding variables, including vital signs and diagnoses. Key Results: Black patients were more likely than White patients to experience long stays before admission but not before discharge. Men were more likely than women to be triaged as urgent or admitted, and had shorter stays. Patients over 30 were likelier to be triaged as urgent or admitted, but had longer stays. keywords emergency, fairness, demographics 1 Introduction Persistent disparities have been observed in the quality of healthcare received by patients from different demographic groups 1 , including in the emergency department (ED) setting 2 , 3 , 4 . These inequities are particularly stark in patients from marginalised racial groups: for example, Black patients are more likely than White patients to experience extended stays before admission to hospital, in both adult 5 , 6 , 7 and pediatric contexts 8 , 9 . Research has also consistently shown that Black patients are assigned lower triage scores than White patients, receive less thorough medical investigations, and wait longer to see a physician 10 , 11 , 12 . Black and Hispanic race, Medicaid insurance, and a preference for Spanish as a primary language have also been established as risk factors for deprioritisation in the ED 13 . Although the factors driving these care disparities are not fully understood, they may be partly driven by implicit provider bias, as shown by recent mixed-methods work finding that physicians are less likely to communicate well when their race differs from that of the patient 5 . Women are also more likely than men to report difficulties in obtaining timely care in the ED 14 . Outcome inequalities on the basis of protected characteristics such as gender, race and age are especially dangerous in the context of emergency medicine, where they can have immediate and life-changing consequences. Efforts to address these disparities first require them to be carefully quantified. We investigated differences in ED outcomes between demographic groups using a large, publicly available single-site dataset. In doing so, we sought to address a number of key gaps in the literature. Previous studies of racial and ethnic disparities in the US healthcare system have generally focused on comparing White, Black and Hispanic patients, lacking the sample size to investigate disparities in other ethnic groups. Our dataset contained more than 17,000 visits by Asian patients, allowing us to investigate whether differences in outcome observed amongst other patients of color extended to this group. We also sought to examine whether the effects of gender, race and age on rates of admission and LOS persisted when we accounted for changes in a patient’s condition after triage, a factor not included (to our knowledge) in previous studies. When measuring differences in admission rates and LOS, we also accounted both for a patient’s chief complaint (i.e. the nurse’s initial assessment of the reason for visit) and for the actual diagnosis eventually attached to their ED stay, to account for differences in a patient’s ability to communicate their condition; previous work has generally included one or the other, which may not give an accurate picture of what is known at various points about a patient’s condition. We investigated whether patients from marginalised groups waited longer to be discharged, as opposed to merely waiting longer to be admitted, as longer wait times to discharge may be associated with a loss of trust in healthcare systems and dissuade patients from seeking prompt care when necessary. To our knowledge this is the first study of its kind to consider discharge times, and the first to consider outcomes of acuity, admission and LOS in parallel. Finally, as previous studies have investigated patterns of disparity in non-public datasets 15 , 5 , 2 , 13 , 12 , 10 , 11 or in a single dataset without information on vital signs 6 , 15 , 9 , we examined a newly-available, public single-site dataset, MIMIC-IV-ED 16 . Healthcare disparities observed in such datasets have the potential to be reproduced by machine learning and artificial intelligence (AI) models trained on these data for clinical decision support 17 . Several data-driven AI tools for the ED already exist, for example to predict admissions 18 , 19 and clinical orders 20 , which in turn may improve patient flow and case management. Given the high potential for current and future models to leverage readily available ED data, we sought to address our research questions using MIMIC-IV-ED 16 , 21 . This is a large, publicly available database of visits by adult patients to the ED of Beth Israel Deaconess Medical Center, a tertiary academic medical center in Boston, Massachusetts. It is vital that researchers designing these tools are aware of possible sources of bias within the data itself, to ensure that they do not reinforce existing inequalities. This is especially true in the light of the final rule on Section 1557 of the Affordable Care Act, that holds health providers responsible for discrimination arising from their use of AI-based decision support interventions 22 . 2 Methods and Materials 2.1 Data Sources and Study Design We conducted a single-center retrospective cross-sectional study of ED visits by adults to BIDMC. Visits were excluded if they did not have a valid recorded ‘in-time’ (time of arrival to the ED) or ‘out-time’ (time at which they left the ED); to be valid, an in-time had to precede the recorded out-time. We obtained vital signs taken at triage, and excluded visits for which such data was not available, or where vitals fell outside a reasonable range (e.g. a temperature of 200 degrees Farenheit). We obtained vital signs recorded during a patient’s stay, but included stays without such information, as many patients may be admitted too quickly to allow repeated vital assessments. Age was obtained to the nearest year by linkage to MIMIC-IV, assuming a birthday of January 1st for each patient. We converted ICD codes relating to Charlson comorbidity scores 23 (a measure of severity) using the R package ‘comorbidity’ 24 . Visits were excluded if no ICD code was available; linked ages were observed for all patients. In total, 35,252 visits, or 8.3% of those in the database, were excluded due to missing or incomplete information. Patient-reported pain, generally reported as a number between 0 and 10, was converted to a score within this range (i.e. interpreting 90 as out of 100 and thus converting it to 9, or ‘flipping a coin’ to convert 6.5 to either 6 or 7). Pain

Read the complete text