Objective: Association of SARS-CoV2 burden in the aerodigestive tract with the disease is sparsely understood. We propose to elucidate the implications of SARS-CoV2 copies in concurrent nasopharyngeal swab (NPS), whole mouth fluid (WMF) and respiratory droplet (RD) samples on disease pathogenesis/transmission. Methods: SARS-CoV2 copies quantified by RT-PCR in concurrent NPS, WMF and RD samples from 80 suspected COVID-19 patients were analysed with demographics, immune response and disease severity. Results: Among the 55/80 (69 %) NPS-positive patients, SARS-CoV2 was detected in 44/55 (80 %) WMF (concordance with NPS-84 %; p = 0.02) and 17/55 (31 %) RD samples. SARS-CoV2 copies were similar in NPS (median:8.74 × 10^5) and WMF (median:3.07 × 10^4), but lower in RD (median:3.60 × 10^2). The 25–75 % interquartile range of SARS-CoV2 copies in the NPS was significantly higher in patients who shed the virus in WMF (p = 0.0001) and RD (p = 0.01). Multivariate analyses showed that hospitalized patients shed significantly higher virus copies in the WMF (p = 0.01). Hospitalized patients with more severe disease (p = 0.03) and higher IL-6 values (p = 0.001) shed more SARS-CoV2 virus in the RD. Conclusions: WMF may be used reliably as a surrogate for diagnosis. High copy numbers in the NPS probably imply early disease onset, while in the WMF and RD may imply more severe disease and increased inflammation.