Methods:

A cross-sectional database study was conducted, including entries from 11 March to 1 May 2020. COVID-19 case definition was compared to RT-PCR results using 2× 2 tables to estimate sensitivity, specificity, PPV, NPV, accuracy and likelihood ratios (LR). Categorical variables were analyzed using χ 2 test. Stratification by gender, age group, history of chronic disease and severity of presentation was performed to control for confounding.

Results:

From a total of 10216 entries, 897 patients with results from RT-CPR for SARS-CoV-2 were included. There was no significant association between COVID-19 case definition and RT-PCR result (χ 2= 0.750, df= 1, p= 0.386). A sensitivity of 87.45%, specificity of 10.61%, PPV of 61.69%, and NPV of 33.96% were calculated, with an overall accuracy of 58.42%, LR+ was 0.98 and LR-was 1.18. Stratifying by age, homogeneity of OR= 0.024 showed an association between case definition and test result particularly among those aged 16-39 years (p= 0.014). Only 4 individual symptoms had a statistical association with RT-PCR result. These were dyspnea (p< 0.001), odynophagia (p< 0.001), rhinorrhea (p< 0.001) and fever (p= 0.008).

Conclusions:

Satisfying criteria for COVID-19 case definition neither showed a significant association nor is it a powerful predictor for SARS-CoV-2 infection among the study population. However, this COVID-19 case definition better identifies those cases in those aged 16-39 years, suggesting a possible relation between the type of the clinical presentation and age.