Predicting COVID-19 Cases Among Nurses Using Artificial Neural Network Approach

Area of Science:

• Epidemiology
• Public Health
• Health Informatics

Background:

• Frontline health workers, particularly nurses, face significant COVID-19 infection risks.
• Understanding risk factors is crucial for protecting healthcare personnel and maintaining healthcare system capacity.

Purpose of the Study:

• To develop a predictive model for identifying nurses at high risk of COVID-19 infection.
• To assess the impact of various factors on infection risk among nurses.

Main Methods:

• Multivariate regression analysis and classification algorithms were employed.
• Data from 220 nurses were analyzed, considering exposure, personal protective equipment (PPE) access and use, hand hygiene, stress, and training.
• An artificial neural network (ANN) was developed for classification.

Main Results:

• Access to PPE and training were associated with lower infection scores.
• Exposure to COVID-19 cases and stress were linked to higher infection probability.
• The ANN achieved 75.8% validation and 76.6% overall accuracy in classifying infected nurses.

Conclusions:

• Predictive modeling can identify nurses at higher risk of COVID-19 infection.
• Interventions focusing on PPE access, training, stress reduction, and exposure control are vital.
• ANNs offer a valuable tool for healthcare workforce risk management during pandemics.