Predictors Of Hospital Admission Research Articles

Predicting hospital admissions for upper respiratory tract complaints: An artificial neural network approach integrating air pollution and meteorological factors.

This study uses artificial neural networks (ANNs) to examine the intricate relationship between air pollutants, meteorological factors, and respiratory disorders. The study investigates the correlation between hospital admissions for respiratory diseases and the levels of PM10 and SO2 pollutants, as well as local meteorological conditions, using data from 2017 to 2019. The objective of this study is to clarify the impact of air pollution on the well-being of the general population, specifically focusing on respiratory ailments. An ANN called a multilayer perceptron (MLP) was used. The network was trained using the Levenberg-Marquardt (LM) backpropagation algorithm. The data revealed a substantial increase in hospital admissions for upper respiratory tract diseases, amounting to a total of 11,746 cases. There were clear seasonal fluctuations, with fall having the highest number of cases of bronchitis (N = 181), sinusitis (N = 83), and upper respiratory infections (N = 194). The study also found demographic differences, with females and people aged 18 to 65years having greater admission rates. The performance of the ANN model, measured using R2 values, demonstrated a high level of predictive accuracy. Specifically, the R2 value was 0.91675 during training, 0.99182 during testing, and 0.95287 for validating the prediction of asthma. The comparative analysis revealed that the ANN-MLP model provided the most optimal result. The results emphasize the effectiveness of ANNs in representing the complex relationships between air quality, climatic conditions, and respiratory health. The results offer crucial insights for formulating focused healthcare policies and treatments to alleviate the detrimental impact of air pollution and meteorological factors.

The accuracy of predicting hospital admission by emergency medical service and emergency department personnel compared to the prehospital MEWS: a prospective multicenter study

IntroductionOvercrowding in the emergency department (ED) is a global problem. Early and accurate recognition of a patient’s disposition could limit time spend at the ED and thus improve throughput and quality of care provided. This study aims to compare the accuracy among healthcare providers and the prehospital Modified Early Warning Score (MEWS) in predicting the requirement for hospital admission.MethodsA prospective, observational, multi-centre study was performed including adult patients brought to the ED by ambulance. Involved Emergency Medical Service (EMS) personnel, ED nurses and physicians were asked to predict the need for hospital admission using a structured questionnaire. Primary endpoint was the comparison between the accuracy of healthcare providers and prehospital MEWS in predicting patients’ need for hospital admission.ResultsIn total 798 patients were included of whom 393 (49.2%) were admitted to the hospital. Sensitivity of predicting hospital admission varied from 80.0 to 91.9%, with physicians predicting hospital admission significantly more accurately than EMS and ED nurses (p < 0.001). Specificity ranged from 56.4 to 67.0%. All healthcare providers outperformed MEWS ≥ 3 score on predicting hospital admission (sensitivity 80.0-91.9% versus 44.0%; all p < 0.001). Predictions for ward admissions specifically were significantly more accurate than MEWS (specificity 94.7–95.9% versus 60.6%, all p < 0.001).ConclusionsHealthcare providers can accurately predict the need for hospital admission, and all providers outperformed the MEWS score.

Development and validation of a risk prediction model for hospital admission in COVID-19 patients presenting to primary care

Background There is a paucity of prognostic models for COVID-19 that are usable for in-office patient assessment in general practice (GP). Objectives To develop and validate a risk prediction model for hospital admission with readily available predictors. Methods A retrospective cohort study linking GP records from 8 COVID-19 centres and 55 general practices in the Netherlands to hospital admission records. The development cohort spanned March to June 2020, the validation cohort March to June 2021. The primary outcome was hospital admission within 14 days. We used geographic leave-region-out cross-validation in the development cohort and temporal validation in the validation cohort. Results In the development cohort, 4,806 adult patients with COVID-19 consulted their GP (median age 56, 56% female); in the validation cohort 830 patients did (median age 56, 52% female). In the development and validation cohort respectively, 292 (6.1%) and 126 (15.2%) were admitted to the hospital within 14 days, respectively. A logistic regression model based on sex, smoking, symptoms, vital signs and comorbidities predicted hospital admission with a c-index of 0.84 (95% CI 0.83 to 0.86) at geographic cross-validation and 0.79 (95% CI 0.74 to 0.83) at temporal validation, and was reasonably well calibrated (intercept −0.08, 95% CI −0.98 to 0.52, slope 0.89, 95% CI 0.71 to 1.07 at geographic cross-validation and intercept 0.02, 95% CI −0.21 to 0.24, slope 0.82, 95% CI 0.64 to 1.00 at temporal validation). Conclusion We derived a risk model using readily available variables at GP assessment to predict hospital admission for COVID-19. It performed accurately across regions and waves. Further validation on cohorts with acquired immunity and newer SARS-CoV-2 variants is recommended.

Using Machine Learning to Predict Unplanned Hospital Utilization and Chemotherapy Management From Patient-Reported Outcome Measures.

Adverse effects of chemotherapy often require hospital admissions or treatment management. Identifying factors contributing to unplanned hospital utilization may improve health care quality and patients' well-being. This study aimed to assess if patient-reported outcome measures (PROMs) improve performance of machine learning (ML) models predicting hospital admissions, triage events (contacting helpline or attending hospital), and changes to chemotherapy. Clinical trial data were used and contained responses to three PROMs (European Organisation for Research and Treatment of Cancer Core Quality of Life Questionnaire [QLQ-C30], EuroQol Five-Dimensional Visual Analogue Scale [EQ-5D], and Functional Assessment of Cancer Therapy-General [FACT-G]) and clinical information on 508 participants undergoing chemotherapy. Six feature sets (with following variables: [1] all available; [2] clinical; [3] PROMs; [4] clinical and QLQ-C30; [5] clinical and EQ-5D; [6] clinical and FACT-G) were applied in six ML models (logistic regression [LR], decision tree, adaptive boosting, random forest [RF], support vector machines [SVMs], and neural network) to predict admissions, triage events, and chemotherapy changes. The comprehensive analysis of predictive performances of the six ML models for each feature set in three different methods for handling class imbalance indicated that PROMs improved predictions of all outcomes. RF and SVMs had the highest performance for predicting admissions and changes to chemotherapy in balanced data sets, and LR in imbalanced data set. Balancing data led to the best performance compared with imbalanced data set or data set with balanced train set only. These results endorsed the view that ML can be applied on PROM data to predict hospital utilization and chemotherapy management. If further explored, this study may contribute to health care planning and treatment personalization. Rigorous comparison of model performance affected by different imbalanced data handling methods shows best practice in ML research.

Radiologists predict differential resource utilization but not clinical outcome in emergency department patients imaged with ultrasound for right upper quadrant pain.

Radiologists with diverse training, specialization, and habits interpret imaging in the Emergency Department. It is necessary to understand if their variation predicts differential value. The purpose of this study was to determine whether attending radiologist variation predicts major clinical outcomes in adult Emergency Department patients imaged with ultrasound for right upper quadrant pain. Consecutive ED patients imaged with ultrasound for RUQ pain from 10/8/2016 to 8/10/2022 were included (N = 7097). The primary outcome was prediction of hospital admission by signing attending radiologist. Secondary outcomes included: ED and hospital length of stay (LOS), 30-day mortality, 30-day re-presentation rate, subspecialty consultation, advanced imaging follow up (HIDA, MRI, CT), and intervention (ERCP, drainage or surgery). Sample size was determined a priori (detectable effect size: w = 0.06). Data were adjusted for demographic data, Elixhauser comorbidities, number of ED visits in prior year, clinical data, and system factors (38 covariates). P-values were corrected for multiple comparisons (false discovery rate-adjusted p-values). The included ultrasounds were read by 35 radiologists (median exams/radiologist: 145 [74.5-241.5]). Signing radiologist did not predict hospitalization (p = 0.85), abdominopelvic surgery or intervention within 30 days, re-presentation to the Emergency Department within 30 days, or subspecialty consultation. Radiologist did predict difference in Emergency Department length of stay (p < 0.001) although this difference was small and imprecise. HIDA was mentioned variably by radiologists (range 0-19%, p < 0.001), and mention of HIDA in the ultrasound report increased 10-fold the odds of HIDA being performed in the next 72h (odds ratio 10.4 [8.0-13.4], p < 0.001). Radiologist variability did not predict meaningful outcome differences for patients with right upper quadrant pain undergoing ultrasound in the Emergency Department, but when radiologists mention HIDA in their reports, it predicts a 10-fold increase in the odds a HIDA is performed. Radiologists are relied on for interpretation that shapes subsequent patient care, and it is important to consider how radiologist variability can influence both outcome and resource utilization.

Hospitalization forecast to inform COVID-19 pandemic planning and resource allocation using discrete event simulation

Purpose: This study aims to address the pressing need for accurate forecasting of healthcare resource demands during the COVID-19 pandemic. It presents a novel approach that combines a stochastic Markov model and a discrete event simulation model to dynamically predict hospital admissions and daily occupancy of hospital and ICU beds.Design/methodology/approach: The research builds upon existing work related to predicting COVID-19 spread and patient influx to hospital emergency departments. The proposed model was developed and validated at San Juan de Alicante University Hospital from July 10, 2020, to January 10, 2022, and externally validated at Hospital Vega Baja. The model involves an admissions generator based on a stochastic Markov model, feeding data into a discrete event simulation model in the R programming language. The probabilities of hospital admission were calculated based on age-stratified positive SARS-COV-2 results from the health department's catchment population. The discrete event simulation model simulates distinct patient pathways within the hospital to estimate bed occupancy for the upcoming week. The performance of the model was measured using the median absolute difference (MAD) between predicted and actual demand.Findings: When applied to data from San Juan hospital, the admissions generator demonstrated a MAD of 6 admissions/week (interquartile range [IQR] 2-11). The MAD between the model's predictions and actual bed occupancy was 20 beds/day (IQR 5-43), equivalent to 5% of total hospital beds. For ICU occupancy, the MAD was 4 beds/day (IQR 2-7), constituting 25% of ICU beds. Evaluation with data from Hospital Vega Baja showcased an admissions generator MAD of 2.42 admissions/week (IQR 1.02-7.41). The MAD between the model's predictions and actual bed occupancy was 18 beds/day (IQR 19.57-38.89), approximately 5.1% of hospital beds. The ICU occupancy MAD was 3 beds/day (IQR 1-5), making up 21.4% of ICU beds.Practical implications: The dynamic predictions of hospital admissions, ward beds, and ICU occupancy for COVID-19 patients proved highly valuable to hospital managers, facilitating early and informed planning of resource allocation. Additionally, this study underscores the importance of utilizing simulation techniques to predict and manage hospital occupancy levels, thereby enhancing decision-making in hospital bed management, not only during pandemics but also during regular periods.Originality/value: This study introduces a novel hybrid approach that combines stochastic modeling and discrete event simulation to forecast healthcare resource demands during the COVID-19 pandemic. The methodology's effectiveness in predicting admissions and bed occupancy contributes to improved resource planning and situational awareness.

A novel scale for triage assessment of frailty in the emergency department (ED-FraS): a prospective videotaped study

BackgroundRapid recognition of frailty in older patients in the ED is an important first step toward better geriatric care in the ED. We aimed to develop and validate a novel frailty assessment scale at ED triage, the Emergency Department Frailty Scale (ED-FraS).MethodsWe conducted a prospective cohort study enrolling adult patients aged 65 years or older who visited the ED at an academic medical center. The entire triage process was recorded, and triage data were collected, including the Taiwan Triage and Acuity Scale (TTAS). Five physician raters provided ED-FraS levels after reviewing videos. A modified TTAS (mTTAS) incorporating ED-FraS was also created. The primary outcome was hospital admission following the ED visit, and secondary outcomes included the ED length of stay (EDLOS) and total ED visit charges.ResultsA total of 256 patients were included. Twenty-seven percent of the patients were frail according to the ED-FraS. The majority of ED-FraS was level 2 (57%), while the majority of TTAS was level 3 (81%). There was a weak agreement between the ED-FraS and TTAS (kappa coefficient of 0.02). The hospital admission rate and charge were highest at ED-FraS level 5 (severely frail), whereas the EDLOS was longest at level 4 (moderately frail). The area under the Receiver Operating Characteristic curve (AUROC) in predicting hospital admission for the TTAS, ED-FraS, and mTTAS were 0.57, 0.62, and 0.63, respectively. The ED-FraS explained more variation in EDLOS (R2 = 0.096) compared with the other two methods.ConclusionsThe ED-Fras tool is a simple and valid screening tool for identifying frail older adults in the ED. It also can complement and enhance ED triage systems. Further research is needed to test its real-time use at ED triage internationally.

Is Insurance Payer Associated With Hospital Admission of Emergency Department Patients With Mandible Fractures?

BackgroundThere is a lack of consensus on the optimal triage pathway for emergency department (ED) patients with mandibular fractures. It remains unclear if patient insurance payer predicts hospital admission given potentially competing logistical and health systems incentives. PurposeTo generate nationally representative estimates of the frequency of hospital admission and its association with primary insurance payer for ED patients with mandible fractures. Study DesignThis retrospective cohort study used the 2018 Nationwide Emergency Department Sample, the largest all-payer database in the United States, to identify patients with mandible fractures. The database includes a stratified sample with discharge weights to generate nationally representative estimates. Patients with other facial fractures and/or concomitant injuries that independently warranted admission were excluded. PredictorThe primary predictor variable was primary payer (public, private, self-pay, other/no-charge). Outcome VariableThe primary outcome variable was hospital admission(yes/no). CovariatesCovariates included patient-, medical/injury-, and hospital-related variables. Analyses: Descriptive statistics, along with bivariate and multivariate logistic regression with Bonferroni correction, were used to produce national estimates and identify predictors of admission. P<0.01 was considered significant. ResultsThe cohort included 27,238 weighted encounters involving isolated mandible fractures, of which 5,345(20%) were admitted. The payer for admitted patients was 46% public, 25% private, 22% self-pay, and 7% no charge/other. In bivariate analyses, public insurance was associated with a higher likelihood of admission than private insurance (RR 1.24, 95% CI 1.06-1.45), though there was no association in the multivariate model (OR 1.03, 95% CI 0.83-1.28). In multivariate analysis, higher Charlson Comorbidity Index (OR 1.32, 95% CI 1.18-1.48), alcohol-related disorder (OR 3.47, 95% CI 2.74-4.39), substance-related disorder (OR 1.43, 95% CI 1.20-1.71), and more mandible fractures (OR 3.08, 95% CI 2.65-3.59) were associated with admission. Compared to body fractures, sub-condylar (OR 3.83, 95% CI 2.39-6.14), angle (OR 3.53, 95% CI 2.84-6.09), and symphysis (OR 4.14, 95% CI 2.84-6.09) fractures had higher odds of admission. Finally, level I (OR 4.11, 95%C I 2.41-6.98) and level II (OR 3.16, 95% CI 1.85-5.39) trauma centers had higher odds of admission. ConclusionsIn 2018, 20% of ED patients with isolated mandible fractures were admitted. Several patient and hospital characteristics were predictors of admission. Insurance status was not associated with admission.

A flexible approach to measure care coordination based on patient-sharing networks

BackgroundEffective care coordination may increase clinical efficiency, but its measurement remains difficult. The established metric “care density” (CD) measures care coordination based on patient-sharing among physicians, but it may be too rigid to generalize across disorders and countries. Therefore, we propose an extension called fragmented care density (FCD), which allows varying weights for connections between different types of providers. We compare both metrics in their ability to predict hospitalizations due to schizophrenia.MethodsWe conducted a longitudinal cohort study based on German claims data from 2014 through 2017 to predict quarterly hospital admissions. 21,016 patients with schizophrenia from the federal state Baden-Württemberg were included. CD and FCD were calculated based on patient-sharing networks. The weights of FCD were optimized to predict hospital admissions during the first year of a 24-month follow-up. Subsequently, we employed likelihood ratio tests to assess whether adding either CD or FCD improved a baseline model with control variables for the second follow-up year.ResultsThe inclusion of FCD significantly improved the baseline model, Χ2(1) = 53.30, p < 0.001. We found that patients with lower percentiles in FCD had an up to 21% lower hospitalization risk than those with median or higher values, whereas CD did not affect the risk.ConclusionsFCD is an adaptive metric that can weight provider relationships based on their relevance for predicting any outcome. We used it to better understand which medical specialties need to be involved to reduce hospitalization risk for patients with schizophrenia. As FCD can be modified for different health conditions and systems, it is broadly applicable and might help to identify barriers and promoting factors for effective collaboration.

FATORES PREDITORES DE HOSPITALIZAÇÃO E ÓBITO EM IDOSOS COM COMPROMETIMENTO COGNITIVO: UM ESTUDO LONGITUDINAL

ABSTRACT Objective: to identify sociodemographic factors, clinical conditions and sarcopenia parameters that predict hospital admission and death in older adults with cognitive impairment. Method: this is a longitudinal observational study carried out with 170 older adults with cognitive impairment assessed between 2019 and 2021. Predictor variables were sociodemographic characteristics, clinical conditions and sarcopenia parameters. Sarcopenia was operationalized through handgrip muscle strength (dynamometry), muscle mass (calf circumference) and the Timed Up and Go test. Occurrence of hospital admission and death within one year after assessment of older adults were the predicted variables. Analyzes were carried out using descriptive statistics, independent Student’ t-test, Mann-Whitney U test, chi-square test and univariate logistic regression. Results: most participants were female (±77.57 years old), with low education, sedentary, 15.9% with sarcopenia and 13% with a history of hospital admission. It was identified that education level had an effect on occurrence of hospital admission (U=1,423.5, p=0.027) and death (U=647.0, p=0.025) within one-year follow-up. Furthermore, there is an association between history of hospital admission in the last 6 months and occurrence of hospital admission [χ2(1)=4.729; p=0.030] and death [χ2(1)=3.919; p=0.048] within one year follow-up. It was identified that history of hospital admission in the last 6 months was associated with occurrence of hospital admission within one-year follow-up (OR=2.963; 95%CI 1.076-8.165, p=0.036). Conclusion: history of hospital admission in the last six months is associated with occurrence of hospital admission over a year in older adults with cognitive impairment.

Predicting early hospital admissions for emergency department patients at the time of triage

Objective: This study aimed to determine factors predicting hospital admissions at King Abdulaziz Medical City in Emergency Department (ED), Riyadh, Saudi Arabia. Methods: A retrospective study was conducted at the Emergency Department of King Abdulaziz Medical City among adult patients aged &amp;ge;14 years triaged in ED and arriving by either paramedic transport or self-referral and presented to ED from 01/01/2021 to 01/01/2022. The data were collected from the patient&amp;apos;s medical records and a pre-designed checklist was prepared to collect data. Results: The majority of patients (88.7%) arrived at the ED walk-ins, as for the patient acuity, a substantial segment of the population (86.9%) fell into category 3. Exactly half of the patients reported no known comorbidities. The prior visit to the emergency department within the past 72 hours was documented in 12.3% of the cases. As for the patients&amp;apos; outcomes post-triage, a large proportion (78.4%) were discharged after their ED visit. The majority of emergency visits were associated with no complications (92.5%), while morbidity was reported among 4.2% and mortality among 3.4% of the visits. The rate of hospital admission was significantly higher among patients aged 25-34 years and 55-64 years, who arrived walk-in, who had patient acuity category &amp;quot;3&amp;quot; and had no previous ED Visit &amp;lt; 72 hours (p=˂0.05). Conclusion: It was found that being bedridden, having dementia, having no prior ED visits within 72 hours, and having end-stage renal disease on dialysis were risk factors for hospital admission.

Diagnostic and prognostic value of an ejection fraction corrected for myocardial remodelling

Abstract Introduction Despite clinical guideline recommendations to use left ventricular ejection fraction (EF) as a marker of left ventricular function in heart failure (HF), it is prone to measurement error, is load dependent and is skewed during geometric cardiac remodelling. EF also cannot distinguish between patients with HF with preserved EF (HFpEF) and those with a normal heart. By evaluating EF at the ventricular mid-wall layer (the corrected EF (EFc)) instead of the endocardial surface, the geometric dependency shortcomings of EF can be overcome. Purpose We evaluated whether EFc can provide additional diagnostic and prognostic information compared with traditional EF. Methods We conducted a retrospective cohort study using a dataset of all consecutive patients who were admitted to hospital following the request of a troponin across three centres between 2010 and 2017. Patients who underwent an echocardiogram either within 3 months before or 5 months after their hospital admission were included. Patients were classified according to whether their index hospital admission was due to HF, and further classified according to EF ((i) EF&amp;lt;40, HF with reduced EF (HFrEF); (ii) 40≤EF&amp;lt;50, HF with mid-range EF (HFmrEF); (iii) EF≥50, HFpEF). All-cause mortality was assessed as a competing risk to HF admission. Cox-proportional regression models were constructed to compare the EFc to EF in evaluating risk of HF admission. Results The study population consisted of 2752 individuals (56.5% male, age 69.3±16.4 years). Comparing patients with HFpEF to those with an index admission without HF, there was no difference in EF between the groups (62.3% versus 64.2%, p = 0.79), however a significant difference was observed when comparing EFc (56.6% versus 61.8%, p = 0.0007) (Figure 1). Both low EF (&amp;lt;50%) and low EFc (&amp;lt;50%) were associated with a high HF readmission risk, which were comparable. However, in the cohort with a normal EF (EF≥50%), predictive models using EFc were significantly more accurate in predicting HF readmissions within 3 years, where the leave one out cross validation ROC analysis showed a 18.6% reduction in errors when EFc replaces EF in the risk prediction models. Table 1 shows the specific IDI and NRI scores to quantify the improvements to predictions when EF or EFc are added to the baseline model (age, gender and creatinine), for assessing readmissions within 3 years in the EF≥50% cohort. The higher IDI and NRI scores for EFc compared to EF shows that adding EFc to "baseline" improves predictions more than adding EF to "baseline". The results also show that using EFc instead of EF will lead to 12.2% more patients in the event group (HF readmission) being correctly identified as higher risk, and 16.6% more patients in the non-event group (no HF readmission) being correctly identified as lower risk. Conclusions In comparison to EF, the EFc improves the prediction of hospital admissions due to HF for patients where the EF≥50%.Figure 1Table 1

Abstract 11578: Predicting 30- and 90-day Hospital Admission Rates Using 3-Dimensional Echocardiography Volume-Curve Parameters

Introduction: This study compares impact of 3D echocardiography-derived novel volume-time parameters to established diastolic dysfunction grades with the aim to improve diagnostic and prognostic prediction of hospital admissions. Methods: Subjects (100 heart-failure (HF) and 11 controls) completed 3D-echocardiograms. Under 2016 ASE Guidelines ‘normal’ and ‘grade 1’ were classified as ‘lower-grade’ (n=64), while ‘grade 2’ and ‘grade 3’ as ‘higher-grade’ (n=38). Early passive filling rate to late passive filling rate ratio was calculated, with early-filling and atrial contraction volumes normalized to total filling volume and body surface area. VC based analysis was classified into ‘normal’ (n=37; atrial contraction enhanced, normal, early enhanced) and ‘abnormal’ (n=74; mid-filling enhanced, delayed filling and restrictive-constrictive). Both methods were compared for 30- and 90-day hospital admissions. Results: The mean age is 67.4 years (SD: 11.1) with 51.4% male. Overall, 9.0% (n=10) and 16.2% (n=18) of the subjects were admitted within 30-days and 90- days, respectively. For 30-day admission, 60% and 40% were lower-grade and higher-grade groups by ASE, respectively, (hazard ratio [HR], 0.91; 95% confidence interval [CI], 0.26 to 3.23, p-value = 0.89); For admission within 90-days, 50% were in lower-grade and 50% in higher-grade groups of ASE, respectively, (HR 1.37; 95% CI, 0.54 to 3.45, p-value = 0.51). Kaplan-Meier curves and log rank test showed significant differences between VC groups with (p-value =0.021 and p-value&lt;0.001) in at risk of admission within 30- and 90-days, respectively. Conclusions: Volume curve analysis provides a better diagnostic and prognostic tool for predicting near term hospital admissions in HF patients. Kaplan-Meier Curves depicting admission within 30- and 90- days by VC and ASE Group

Wastewater surveillance provides 10-days forecasting of COVID-19 hospitalizations superior to cases and test positivity: A prediction study

BackgroundThe public health response to COVID-19 has shifted to reducing deaths and hospitalizations to prevent overwhelming health systems. The amount of SARS-CoV-2 RNA fragments in wastewater are known to correlate with clinical data including cases and hospital admissions for COVID-19. We developed and tested a predictive model for incident COVID-19 hospital admissions in New York State using wastewater data. MethodsUsing county-level COVID-19 hospital admissions and wastewater surveillance covering 13.8 million people across 56 counties, we fit a generalized linear mixed model predicting new hospital admissions from wastewater concentrations of SARS-CoV-2 RNA from April 29, 2020 to June 30, 2022. We included covariates such as COVID-19 vaccine coverage in the county, comorbidities, demographic variables, and holiday gatherings. FindingsWastewater concentrations of SARS-CoV-2 RNA correlated with new hospital admissions per 100,000 up to ten days prior to admission. Models that included wastewater had higher predictive power than models that included clinical cases only, increasing the accuracy of the model by 15%. Predicted hospital admissions correlated highly with observed admissions (r = 0.77) with an average difference of 0.013 hospitalizations per 100,000 (95% CI = [0.002, 0.025]) InterpretationUsing wastewater to predict future hospital admissions from COVID-19 is accurate and effective with superior results to using case data alone. The lead time of ten days could alert the public to take precautions and improve resource allocation for seasonal surges.

The Incidence and Factors Affecting the 28-day Hospital Admission Among Adult Ambulatory COVID-19 Patients

Objective: This study aimed to determine the 28-day hospital admission incidence and risk factors related to the hospitalization of COVID-19 patients who were followed as outpatients in a university hospital. Methods: A retrospective cohort study was designed in which the sociodemographic characteristics, symptoms on the first visit, presence of comorbidities, and viral load predictor cycle threshold (Ct) value were defined as independent variables and hospital admission in the first 28 days after the first visit was defined as dependent variable. Factors related to hospital admission were evaluated with univariate and multivariate analyses. Cox regression analysis was used for multivariate analyses, and the effect size was determined with a hazard ratio (HR). Statistical significance was defined as p&lt;0.05. Results: 368 patients were included in this study. The median (25-75th percentile) age was 36 (28-45) years, and 46.1% of the patients were female. Sixty-five patients (17.7%) were hospitalized in the first 28 days. When age ≤29 referenced, age≥50 (HR=4.1, 95% confidence interval [CI]=1.7-9.6), 40-49 (HR=3.0, 95% CI=1.3-6.6) and 30-39 (HR=1.6, 95% CI=0.6-3.6), fever or chills, (HR=2.3, 95% CI=1.3-4.1), dyspnea (HR=2.0, 95% CI=1.1-3.4), fatigue (HR=1.9, 95% CI=1.0-3.5), vomiting (HR=3.0, 95% CI=1.5-5.8), and sore throat (HR=0.4, 95% CI=0.2-0.8) were defined as independent risk factors according to multivariable analysis. Hypertension was the only comorbidity independently predicting hospital admission (HR=2.2, 95% CI=1.0-4.4). Conclusion: Advanced age, systemic and lower respiratory tract infection signs, and hypertension independently increased the 28-day hospital admission. The presence of a sore throat was not a factor for hospital admission. A lower risk of hospital admission in patients with sore throats may indicate that the patient will have a mild upper respiratory tract infection and will not progress to severe disease. Due to missing data, this study could not fully evaluate the hospital admission risk factors. Studies that include clinical, laboratory, and radiological findings are needed to generate a more accurate model predicting hospital admission risk.

153 Social Domains of Needs Score on Predicting Hospital Admissions for Patients in a Community Paramedicine Program

153 Social Domains of Needs Score on Predicting Hospital Admissions for Patients in a Community Paramedicine Program

Noninvasive Assessment of Impaired Gas Exchange with the Alveolar Gas Monitor Predicts Clinical Deterioration in COVID-19 Patients.

The COVID-19 pandemic magnified the importance of gas exchange abnormalities in early respiratory failure. Pulse oximetry (SpO2) has not been universally effective for clinical decision-making, possibly because of limitations. The alveolar gas monitor (AGM100) adds exhaled gas tensions to SpO2 to calculate the oxygen deficit (OD). The OD parallels the alveolar-to-arterial oxygen difference (AaDO2) in outpatients with cardiopulmonary disease. We hypothesized that the OD would discriminate between COVID-19 patients who require hospital admission and those who are discharged home, as well as predict need for supplemental oxygen during the index hospitalization. Patients presenting with dyspnea and COVID-19 were enrolled with informed consent and had OD measured using the AGM100. The OD was then compared between admitted and discharged patients and between patients who required supplemental oxygen and those who did not. The OD was also compared to SpO2 for each of these outcomes using receiver operating characteristic (ROC) curves. Thirty patients were COVID-19 positive and had complete AGM100 data. The mean OD was significantly (p = 0.025) higher among those admitted 50.0 ± 20.6 (mean ± SD) vs. discharged 27.0 ± 14.3 (mean ± SD). The OD was also significantly (p < 0.0001) higher among those requiring supplemental oxygen 60.1 ± 12.9 (mean ± SD) vs. those remaining on room air 25.2 ± 11.9 (mean ± SD). ROC curves for the OD demonstrated very good and excellent sensitivity for predicting hospital admission and supplemental oxygen administration, respectively. The OD performed better than an SpO2 threshold of <94%. The AGM100 is a novel, noninvasive way of measuring impaired gas exchange for clinically important endpoints in COVID-19.

Clinical Outcomes of Children With COVID-19 by SARS-CoV-2 Strain: A Cohort Study.

We sought to investigate the disease outcomes and predictors of severe outcomes among children infected with the Delta variant of SARS-CoV-2 compared with pre-Delta strains. Single-center retrospective cohort study in an emergency department located within an urban academic children's hospital. Patients included children (0-18 years) who tested positive for SARS-CoV-2. Main outcomes measured include need for hospital admission or COVID-directed therapies. There was a trend toward decreased hospital admission and no significant difference in the severity of outcomes in the Delta cohort relative to the pre-Delta cohort. The Delta cohort had lower odds of hospital admission (odds ratio [OR], 0.79; 95% confidence interval [CI], 0.51-1.23), but the result was not statistically significant. Logistic regression analyses showed that overall, age 1 to 4 years (OR, 2.35; 95% CI, 1.23-4.57) and public insurance (OR, 1.80, 95% CI, 1.08-3.01) were predictors of hospital admission. Within the Delta cohort, the presence of any comorbidity increased the odds of admission (OR, 2.52; 95% CI, 1.09-6.04). Black children had lower odds of admission than white children (overall OR, 0.53; 95% CI, 0.31-0.90; pre-Delta OR, 0.50; 95% CI, 0.26-0.95). The severity of measured disease outcomes was similar in pediatric patients when comparing children infected with the pre-Delta and Delta variants of SARS-CoV-2, even among children with comorbidities once adjusting for acuity.Ongoing research is essential to determine disease severity and risk for children with comorbidities because SARS-CoV-2 continues to mutate, including with Omicron subvariants.

Prehospital prediction of hospital admission for emergent acuity patients transported by paramedics: A population-based cohort study using machine learning.

The closest emergency department (ED) may not always be the optimal hospital for certain stable high acuity patients if further distanced ED's can provide specialized care or are less overcrowded. Machine learning (ML) predictions may support paramedic decision-making to transport a subgroup of emergent patients to a more suitable, albeit more distanced, ED if hospital admission is unlikely. We examined whether characteristics known to paramedics in the prehospital setting were predictive of hospital admission in emergent acuity patients. We conducted a population-level cohort study using four ML algorithms to analyze ED visits of the National Ambulatory Care Reporting System from January 1, 2018 to December 31, 2019 in Ontario, Canada. We included all adult patients (≥18 years) transported to the ED by paramedics with an emergent Canadian Triage Acuity Scale score. We included eight characteristic classes as model predictors that are recorded at ED triage. All ML algorithms were trained and assessed using 10-fold cross-validation to predict hospital admission from the ED. Predictive model performance was determined using the area under curve (AUC) with 95% confidence intervals and probabilistic accuracy using the Brier Scaled score. Variable importance scores were computed to determine the top 10 predictors of hospital admission. All machine learning algorithms demonstrated acceptable accuracy in predicting hospital admission (AUC 0.77-0.78, Brier Scaled 0.22-0.24). The characteristics most predictive of admission were age between 65 to 105 years, referral source from a residential care facility, presenting with a respiratory complaint, and receiving home care. Hospital admission was accurately predicted based on patient characteristics known prehospital to paramedics prior to arrival. Our results support consideration of policy modification to permit certain emergent acuity patients to be transported to a further distanced ED. Additionally, this study demonstrates the utility of ML in paramedic and prehospital research.

Predictors of hospital admission when presenting with acute-on-chronic breathlessness: Binary logistic regression.

Breathlessness due to medical conditions commonly causes emergency department presentations and unplanned admissions. Acute-on-chronic breathlessness is a reason for 20% of emergency presentations by ambulance with 69% of these being admitted. The emergency department may be inappropriate for many presenting with acute-on-chronic breathlessness. To examine predictors of emergency department departure status in people with acute-on-chronic breathlessness. Secondary analysis of patient-report survey and clinical record data from consecutive eligible attendees by ambulance. Variables associated with emergency department departure status (unifactorial analyses; p<0.05) were included in a binary logistic regression model. The study was conducted in a single tertiary hospital. Consecutive survey participants presenting in May 2015 with capacity were eligible. 1,212/1,345 surveys were completed. 245/1,212 presented with acute-on-chronic breathlessness, 171 of whom consented to clinical record review and were included in this analysis. In the final model, the odds of admission were increased with every extra year of age [OR 1.041 (95% CI: 1.016 to 1.066)], having talked to a specialist doctor about breathlessness [9.262 (1.066 to 80.491)] and having a known history of a heart condition [4.177 (1.680 to 10.386)]. Odds of admission were decreased with every percentage increase in oxygen saturation [0.826 (0.701 to 0.974)]. Older age, lower oxygen saturation, having talked to a specialist, and having history of a cardiac condition predict hospital admission in people presenting to the emergency department with acute-on-chronic breathlessness. These clinical factors could be assessed in the community and may inform the decision regarding conveyance.