Predicting severe COVID-19 disease in adults: A single-centre cohort study during the first three pandemic waves in 2020-2021 in Vilnius, Lithuania.

BACKGROUND

Since its emergence, the COVID-19 infection has led to significant morbidity and mortality worldwide. Early identification of patients at risk for severe disease is essential for more effective triage, timely therapeutic intervention, and optimal resource allocation. Differences in population characteristics may contribute to variability in disease outcomes, which emphasizes the need for regional-level data, especially from underrepresented regions. The main aim of this study was to identify the demographic, clinical, and laboratory predictors of severe COVID-19, defined as the need for oxygen therapy, in Lithuania.

MATERIALS AND

METHODS

We conducted an ambispective observational cohort study at Vilnius University Hospital Santaros Klinikos in Vilnius, Lithuania, from March 2020 to December 2021. Adult patients with a confirmed diagnosis of COVID-19 and hospitalized longer than 24 hours were included in this study. Data were collected from the electronic medical records and patient interviews. To identify predictors of severe COVID-19 course, a multivariable binary logistic regression model was performed.

RESULTS

Among 495 patients, 52.9% were male, the median age was 55 years, and 61.2% had at least one underlying condition. The most common symptoms on admission were malaise (77.1%), subfebrile fever (65.9%), and cough (69.7%). CRP demonstrated the highest predictive value for severe COVID-19 (AUC = 0.84), followed by LDH (AUC = 0.80). Older age (OR 1.04 per year, 95% CI 1.00-1.08), obesity (OR 3.55, 95% CI 1.35-9.30), lymphopenia (OR 3.70, 95% CI 1.37-9.99), higher LDH (OR 1.008, 95% CI 1.00-1.01) and CRP (OR 1.021, 95% CI 1.01-1.04) levels were identified as the strongest predictors for severe COVID-19 disease course.

CONCLUSION

Older age, obesity, lymphopenia, and higher CRP and LDH were associated with developing severe COVID-19 disease, indicating that combining patient history and laboratory parameters can provide a practical risk stratification approach to help clinicians identify high-risk patients early upon hospitalisation.