Dataset Information

Association between waiting times and short term mortality and hospital admission after departure from emergency department: population based cohort study from Ontario, Canada.

ABSTRACT:

Objective

To determine whether patients who are not admitted to hospital after attending an emergency department during shifts with long waiting times are at risk for adverse events.

Design

Population based retrospective cohort study using health administrative databases. Setting High volume emergency departments in Ontario, Canada, fiscal years 2003-7.

Participants

All emergency department patients who were not admitted (seen and discharged; left without being seen).

Outcome measures

Risk of adverse events (admission to hospital or death within seven days) adjusted for important characteristics of patients, shift, and hospital.

Results

13,934,542 patients were seen and discharged and 617,011 left without being seen. The risk of adverse events increased with the mean length of stay of similar patients in the same shift in the emergency department. For mean length of stay ≥ 6 v <1 hour the adjusted odds ratio (95% confidence interval) was 1.79 (1.24 to 2.59) for death and 1.95 (1.79 to 2.13) for admission in high acuity patients and 1.71 (1.25 to 2.35) for death and 1.66 (1.56 to 1.76) for admission in low acuity patients). Leaving without being seen was not associated with an increase in adverse events at the level of the patient or by annual rates of the hospital.

Conclusions

Presenting to an emergency department during shifts with longer waiting times, reflected in longer mean length of stay, is associated with a greater risk in the short term of death and admission to hospital in patients who are well enough to leave the department. Patients who leave without being seen are not at higher risk of short term adverse events.

SUBMITTER: Guttmann A

PROVIDER: S-EPMC3106148 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

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.

Project description:ImportanceEmergency department (ED) use in pregnancy is common and occurs for a variety of reasons, including obstetrical complications, exacerbated underlying conditions, and inadequate outpatient health care access. People with disabilities have elevated rates of certain medical, psychiatric, and obstetrical conditions as well as inadequate access to prenatal care; their risk of ED use in pregnancy is not known, however.ObjectiveTo compare the risk of ED use in pregnancy among people with physical, sensory, and intellectual or developmental disabilities with those without disabilities.Design, setting, and participantsPopulation-based cohort study leveraging linked administrative health data sets in Ontario, Canada, April 2003 to March 2019. Analysis included all recognized pregnancies to people with a preexisting physical, sensory, intellectual or developmental, or 2 or more (multiple) disabilities, and those without a disability. Data were analyzed from May 2022 to January 2023.ExposureDisability was ascertained using algorithms applied to 2 or more outpatient physician visits or 1 or more ED visits or hospitalizations before conception.Main outcomes and measuresModified Poisson regression-generated adjusted relative risks (aRR) and 95% CIs for any ED visit in pregnancy, from the estimated conception date up to the end of the pregnancy, adjusted for age, parity, income quintile, rurality, immigrant status, and preexisting chronic conditions, mental illness, and substance use disorders.ResultsThe cohort included 2 659 895 pregnant people with physical (221 739 participants; mean [SD] age, 29.8 [6.1] years), sensory (71 891 participants; mean [SD] age, 29.1 [6.4] years), intellectual or developmental (3877 participants; mean [SD] age, 26.1 [6.7] years), and multiple disabilities (14 359 participants; mean [SD] age, 29.5 [6.5] years), and pregnant people without a disability (2 348 023 participants; mean [SD] age, 29.4 [5.9] years). The rate of ED visits in pregnancy was 25.4% in people without a disability (596 771 visits). Relative to these individuals, the aRR for ED use was elevated in people with physical (aRR, 1.26; 95% CI, 1.25-1.27), sensory (aRR, 1.15; 95% CI, 1.14-1.17), intellectual or developmental (aRR, 1.33; 95% CI, 1.28-1.38), and multiple disabilities (aRR, 1.43; 95% CI, 1.40-1.46).Conclusions and relevanceIn this population-based study, people with disabilities were at elevated risk of ED use in pregnancy. This finding underscores the need for research on the benefits of proactive strategies to manage preexisting conditions in these individuals, improve their access to outpatient obstetrical and medical care, and prepare them for when ED visits occur.