Large scale screening of health care workers and the general population for asymptomatic COVID-19 infection requires modalities that are amenable to testing at scale while retaining acceptable levels of sensitivity and specificity.This study evaluated a novel COVID-19 Direct-RT LAMP assay using saliva samples in asymptomatic individuals by comparison to RT-PCR. Additional studies were performed using VTM collected from routine diagnostic testing. Analytical sensitivity was determined for Direct RT-LAMP assay using the WHO International Standard. Finally, quantified results from RT-PCR testing of 9177 nose and throat swabs obtained from routine diagnostic testing were used to estimate the sensitivity of Direct RT-LAMP using the limit of detection curve obtained from the analytical sensitivity data.Results from saliva testing demonstrated a sensitivity of 40.91% and a specificity of 100% for Direct RT-LAMP. The sensitivity and specificity for nose and throat swabs were 44.85% and 100% respectively. The 95% limit of detection (LOD) for Direct RT-LAMP was log 7.13 IU/ml (95% 6.9–7.5). The estimated sensitivity for Direct-RT LAMP based on the results of 9117 nose and throat swabs was 34% and 45% for saliva and VTM respectively.The overall diagnostic sensitivity of Direct RT-LAMP was low compared to RT-PCR. Testing of nose and throat swabs and estimating the sensitivity based on a large cohort of clinical samples demonstrated similar results. This study highlights the importance of utilising the prospective collection of samples from the intended target population in the assessment of diagnostic sensitivity.