Related Experiment Video
Updated: Jan 15, 2026
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Hospital stay duration prediction score system for covid-19 patients: Evaluation of proposed practical models
Artur V Kurakh1, Mykhailo M Hechko1, Khrystyna A Hechko1
1UZHHOROD NATIONAL UNIVERSITY, UZHHOROD, UKRAINE.
Insights
New prognostic models for COVID-19 patients show promise in predicting hospital stay duration. Models II and III demonstrated significant prognostic value, aiding in primary care and triage scenarios.
Area of Science:
- Medical Prognostics
- Infectious Disease Epidemiology
- Health Services Research
Background:
- Accurate prediction of hospital stay duration for COVID-19 patients is crucial for resource allocation and patient management.
- Existing prognostic tools may not fully capture the complexities of disease progression and recovery.
- The need for validated scales to guide clinical decision-making in primary care and triage settings is evident.
Purpose of the Study:
- To evaluate the prognostic capability of newly developed scales in predicting hospital stay duration for hospitalized COVID-19 patients.
- To establish baseline values for these scales to facilitate their integration into primary care.
- To identify the most effective models for practical clinical application.
Main Methods:
- Analysis of data from 140 hospitalized COVID-19 patients treated between the start of the pandemic and January 2022.
- Development and evaluation of four distinct prognostic models based on a point system.
- Linear regression analysis to correlate accumulated points with actual hospital stay duration.
Main Results:
- Models I and IV exhibited low practical utility due to low R-squared values.
- Models II and III demonstrated significant prognostic value, with R-squared values of 0.0703 and 0.111, respectively.
- Statistically significant increases in hospital stay were observed for scores of 2 and 5 in Models II and III compared to a score of 0.
Conclusions:
- Models II and III show the highest practical utility and prognostic value for COVID-19 scenarios.
- Middle and maximum scores on these models significantly correlate with hospital stay duration.
- These validated models can be beneficial for practical use in triage and primary care settings, even with limited data on specific comorbidities like asthma and COPD.
Objective:
Aim: Evaluate the prognostic value of the proposed scale on determining hospital stay duration of patients with COVID-19. Establish baseline values of the scale for further implementation into primary care.
Patients And Methods:
Materials and Methods: We gathered and analyzed data of 140 hospitalized patients who underwent treatment in Uzhhorod CMCH and had a positive COVID-19 test from the start of the pandemic and until January of 2022. 4 models were proposed, each patient was analyzed based on a point system determined by the model, to compare the number of accumulated points with hospital stay duration.
Results:
Results: Linear regression results showed that models I and IV provide a low practical usage as per R2. Models II and III are most promising showing a high prognostic value (R2 are 0.0703 and 0.111 respectively). Models II and III showed a statistically significant increase in hospital stay for scores 2 and 5 compared with 0. The accumulated data shows that scores 2 and 5 show a significant increase in hospital stays, while 1, 3 and 4 did not show any significant increase for practical usage.
Conclusion:
Conclusions: Overall, this analysis showed that middle scores and maximum scores showed significance compared to hospital stays, while Models II and III showed the highest practical usage coefficients in COVID-19 scenarios. Despite the fact that the number of asthma and COPD patients was low, these models may prove useful for practical usage in triage scenarios and primary care.