Dataset Information

Examining the association between paramedic transport to the emergency department and hospital admission: a population-based cohort study.

ABSTRACT:

Background

Increasing hospitalization rates present unique challenges to manage limited inpatient bed capacity and services. Transport by paramedics to the emergency department (ED) may influence hospital admission decisions independent of patient need/acuity, though this relationship has not been established. We examined whether mode of transportation to the ED was independently associated with hospital admission.

Methods

We conducted a retrospective cohort study using the National Ambulatory Care Reporting System (NACRS) from April 1, 2015 to March 31, 2020 in Ontario, Canada. We included all adult patients (≥18 years) who received a triage score in the ED and presented via paramedic transport or self-referral (walk-in). Multivariable binary logistic regression was used to determine the association of mode of transportation between hospital admission, after adjusting for important patient and visit characteristics.

Results

During the study period, 21,764,640 ED visits were eligible for study inclusion. Approximately one-fifth (18.5%) of all ED visits were transported by paramedics. All-cause hospital admission incidence was greater when transported by paramedics (35.0% vs. 7.5%) and with each decreasing Canadian Triage and Acuity Scale level. Paramedic transport was independently associated with hospital admission (OR = 3.76; 95%CI = 3.74-3.77), in addition to higher medical acuity, older age, male sex, greater than two comorbidities, treatment in an urban setting and discharge diagnoses specific to the circulatory or digestive systems.

Conclusions

Transport by paramedics to an ED was independently associated with hospital admission as the disposition outcome, when compared against self-referred visits. Our findings highlight patient and visit characteristics associated with hospital admission, and can be used to inform proactive healthcare strategizing for in-patient bed management.

SUBMITTER: Strum RP

PROVIDER: S-EPMC8506085 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Abstract Objective To evaluate the association between initiation of fluoroquinolones and hospital admission or emergency department visit for suicidality. Design Population based cohort study. Setting IBM MarketScan database, USA. Participants 2 756 268 adults (≥18 years) who initiated an oral fluoroquinolone (ciprofloxacin, levofloxacin, moxifloxacin, gemifloxacin, ofloxacin, gatifloxacin, norfloxacin, lomefloxacin, besifloxacin) or comparator antibiotic (January 2003 to September 2015) and had at least six months of continuous health plan enrollment and a diagnosis of pneumonia or urinary tract infection (UTI) three days or less before the drug initiation date. Comparator antibiotics were azithromycin in the pneumonia cohort and trimethoprim-sulfamethoxazole in the UTI cohort. Participants were matched 1:1 within each cohort on a propensity score, calculated from a multivariable logistic regression model that included 57 baseline covariates. Main outcomes measure Primary outcome was hospital admission or emergency department visit for suicidal ideation or self-harm within 60 days after treatment initiation. Cox proportional hazard models were used to estimate hazard ratios and 95% confidence intervals. Results The pneumonia cohort included 551 042 individuals, and the UTI cohort included 2 205 526 individuals. During the 60 day follow-up, 181 events were observed in the pneumonia cohort and 966 in the UTI cohort. The adjusted hazard ratios for fluoroquinolones were 1.01 (95% confidence interval 0.76 to 1.36) versus azithromycin in the pneumonia cohort and 1.03 (0.91 to 1.17) versus trimethoprim-sulfamethoxazole in the UTI cohort. Results were consistent across sensitivity analyses and subgroups of sex, age, or history of mental illnesses. Conclusion Initiation of fluoroquinolones was not associated with a substantially increased risk of admission to hospital or emergency department visits for suicidality compared with azithromycin or trimethoprim-sulfamethoxazole.

Project description:ObjectiveTo predict hospital admission at the time of ED triage using patient history in addition to information collected at triage.MethodsThis retrospective study included all adult ED visits between March 2014 and July 2017 from one academic and two community emergency rooms that resulted in either admission or discharge. A total of 972 variables were extracted per patient visit. Samples were randomly partitioned into training (80%), validation (10%), and test (10%) sets. We trained a series of nine binary classifiers using logistic regression (LR), gradient boosting (XGBoost), and deep neural networks (DNN) on three dataset types: one using only triage information, one using only patient history, and one using the full set of variables. Next, we tested the potential benefit of additional training samples by training models on increasing fractions of our data. Lastly, variables of importance were identified using information gain as a metric to create a low-dimensional model.ResultsA total of 560,486 patient visits were included in the study, with an overall admission risk of 29.7%. Models trained on triage information yielded a test AUC of 0.87 for LR (95% CI 0.86-0.87), 0.87 for XGBoost (95% CI 0.87-0.88) and 0.87 for DNN (95% CI 0.87-0.88). Models trained on patient history yielded an AUC of 0.86 for LR (95% CI 0.86-0.87), 0.87 for XGBoost (95% CI 0.87-0.87) and 0.87 for DNN (95% CI 0.87-0.88). Models trained on the full set of variables yielded an AUC of 0.91 for LR (95% CI 0.91-0.91), 0.92 for XGBoost (95% CI 0.92-0.93) and 0.92 for DNN (95% CI 0.92-0.92). All algorithms reached maximum performance at 50% of the training set or less. A low-dimensional XGBoost model built on ESI level, outpatient medication counts, demographics, and hospital usage statistics yielded an AUC of 0.91 (95% CI 0.91-0.91).ConclusionMachine learning can robustly predict hospital admission using triage information and patient history. The addition of historical information improves predictive performance significantly compared to using triage information alone, highlighting the need to incorporate these variables into prediction models.