The 'ephemeral' intensive care units that saved the French healthcare system, a new concept to be preserved

Predictive Modeling of COVID-19 Intensive Care Unit Patient Flows and Nursing Complexity: A Monte Carlo Simulation Study

This study aimed to develop a Monte Carlo simulation model to forecast the number of ICU beds needed for COVID-19 patients and the subsequent nursing complexity in a French teaching hospital during the first and second pandemic outbreaks. The model used patient data from March 2020 to September 2021, including age, sex, ICU length of stay, and number of patients on mechanical ventilation or extracorporeal membrane oxygenation. Nursing complexity was assessed using a simple scale with three levels based on patient status. The simulation was performed 1000 times to generate a scenario, and the mean outcome was compared with the observed outcome. The model also allowed for a 7-day forecast of ICU occupancy. The simulation output had a good fit with the actual data, with an R2 of 0.998 and a root mean square error of 0.22. The study demonstrated the usefulness of the Monte Carlo simulation model for predicting the demand for ICU beds and could help optimize resource allocation during a pandemic. The model's extrinsic validity was confirmed using open data from the French Public Health Authority. This study provides a valuable tool for healthcare systems to anticipate and manage surges in ICU demand during pandemics.

Triage and monitoring of COVID-19 patients in intensive care using unsupervised machine learning

BACKGROUND

We designed an algorithm to assess COVID-19 patients severity and dynamic intubation needs and predict their length of stay using the breathing frequency (BF) and oxygen saturation (SpO₂) signals.

METHODS

We recorded the BF and SpO₂ signals for confirmed COVID-19 patients admitted to the ICU of a teaching hospital during both the first and subsequent outbreaks of the pandemic in France. An unsupervised machine-learning algorithm (the Gaussian mixture model) was applied to the patients' data for clustering. The algorithm's robustness was ensured by comparing its results against actual intubation rates. We predicted intubation rates using the algorithm every hour, thus conducting a severity evaluation. We designed a S₂₄ severity score that represented the patient's severity over the previous 24 h; the validity of MS₂₄, the maximum S₂₄ score, was checked against rates of intubation risk and prolonged ICU stay.

RESULTS

Our sample included 279 patients. . The unsupervised clustering had an accuracy rate of 87.8% for intubation recognition (AUC = 0.94, True Positive Rate 86.5%, true Negative Rate 90.9%). The S₂₄ score of intubated patients was significantly higher than that of non-intubated patients at 48 h before intubation. The MS₂₄ score allowed for the distinguishing between three severity levels with an increased risk of intubation: green (3.4%), orange (37%), and red (77%). A MS₂₄ score over 40 was highly predictive of an ICU stay greater than 5 days at an accuracy rate of 81.0% (AUC = 0.87).

CONCLUSIONS

Our algorithm uses simple signals and seems to efficiently visualize the patients' respiratory situations, meaning that it has the potential to assist staffs' in decision-making. Additionally, real-time computation is easy to implement.

Impact of ICU transfers on the mortality rate of patients with COVID-19: insights from comprehensive national database in France

Background

The first wave of the COVID-19 pandemic confronted healthcare systems around the world with unprecedented organizational challenges, particularly regarding the availability of intensive care unit (ICU) beds. One strategy implemented in France to alleviate healthcare pressure during the first COVID-19 wave was inter-hospital transfers of selected ICU patients from overwhelmed areas towards less saturated ones. At the time, the impact of this transfer strategy on patient mortality was unknown. We aimed to compare in-hospital mortality rates among ICU patients with COVID-19 who were transferred to another healthcare facility and those who remained in the hospital where they were initially admitted to.

Method

A prospective observational study was performed from 1 March to 21 June 2020. Data regarding hospitalized patients with COVID-19 were collected from the Ministry of Health-affiliated national SI-VIC registry. The primary endpoint was in-hospital mortality.

Results

In total, 93,351 hospital admissions of COVID-19 patients were registered, of which 18,348 (19.6%) were ICU admissions. Transferred patients (n = 2228) had a lower mortality rate than their non-transferred counterparts (n = 15,303), and the risk decreased with increasing transfer distance (odds ratio (OR) 0.7, 95% CI: 0.6–0.9, p = 0.001 for transfers between 10 and 50 km, and OR 0.3, 95% CI: 0.2–0.4, p < 0.0001 for transfer distance > 200 km). Mortality decreased overall over the 3-month study period.

Conclusions

Our study shows that the mortality rates were lower for patients with severe COVID-19 who were transferred between ICUs across regions, or internationally, during the first pandemic wave in France. However, the global mortality rate declined overall during the study. Transferring selected patients with COVID-19 from overwhelmed regions to areas with greater capacity may have improved patient access to ICU care, without compounding the short-term mortality risk of transferred patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13613-021-00933-2.