Vital Signs as Predictors of Oxygen Requirements in COVID-19 Outpatients: The Development of a Streamlined Risk Score

TL;DR

The study developed a simple risk score using vital signs to predict which outpatients with COVID-19 will need oxygen therapy, helping clinics triage patients efficiently.

Contribution

A streamlined clinical risk score based solely on vital signs to predict oxygen requirements in COVID-19 outpatients was developed and validated.

Findings

Body temperature, respiratory rate, SpO2, and pneumonia on CT scans were significant predictors of oxygen need.

A risk score based on vital signs alone achieved high accuracy (AUC 0.937) in predicting oxygen requirements.

The score does not require detailed medical history or CT scans, making it practical for resource-limited settings.

Abstract

Background: With COVID-19 becoming a common disease, primary care facilities such as clinics are required to efficiently triage patients at high risk of severe illness within the constraints of limited medical resources. However, existing COVID-19 severity risk scores require detailed medical history assessments, such as evaluating the severity of pneumonia via chest CT and accounting for past and comorbid conditions. Therefore, they may not be suitable for practical use in clinical settings with limited medical resources, including personnel and equipment.

Purpose: The goal is to identify key variables that predict the need for oxygen therapy in COVID-19 patients and develop a simplified clinical risk score based solely on vital signs to predict oxygen requirements.

Patients and methods: A retrospective observational study of 584 outpatients with COVID-19 confirmed by polymerase chain reaction test visited Sasebo Chuo Hospital between April 28, 2022, and August 18, 2022. Analyses were conducted after adjustment for background factors of age and sex with propensity score matching. We used the Fisher test for nominal variables and the Kruskal-Wallis test for continuous variables.

Results: After adjusting for age and sex, several factors significantly correlated with the need for oxygen within seven days including body temperature (p < 0.001), respiratory rate (p = 0.007), SpO2 (p < 0.001), and the detection of pneumonia on CT scans (p = 0.032). The area under the receiver-operating characteristic curve for the risk score based on these vital signs and CT was 0.947 (95% confidence interval: 0.911-0.982). The risk score based solely on vital signs was 0.937 (0.900-0.974), demonstrating the ability to predict oxygen administration with no significant differences.

Conclusions: Body temperature, advanced age, increased respiratory rate, decreased SpO2, and the presence of pneumonia on CT scans were significant predictors of oxygen need within seven days among the study participants. The risk score, based solely on vital signs, effectively and simply assesses the likelihood of requiring oxygen therapy within seven days with high accuracy. The risk score, which utilizes only age and vital signs and does not require a detailed patient history or CT scans, could streamline hospital referral processes for admissions.

Figures