In late 2019, SARS-CoV-2 emerged in the Wuhan province of China. Rapid global spread led to the Covid-19 pandemic. Rapid and accurate detection of SARS-CoV-2 has become a vitally important tool in controlling the spread of the virus. Lateral flow devices (LFDs) offer the potential advantage of speed and on-site testing. The sensitivity of these devices compared to the gold standard RT-PCR has been questioned. We compared the performance of the Innova lateral flow kit, recommended by the UK government, with our rapid in-house RT-PCR protocol using stored positive patient samples. The LFD device was found to be 6,000-10,000 times less sensitive than RT-PCR for the detection of SARS-CoV-2. Overall, the LFD detected 46.2% of the positives detected by RT-PCR. 50% of the LFD results were observed to be weak positives, only visible after careful examination by experienced laboratory staff. At lower viral loads, such as 10,000-100,000 RNA copies/ml, the LFD detected 22.2% of positives. In addition, two strong positives (3 and 1.5 million RNA copies/ml) were not detected by the LFD. The argument for use of LFD kits, despite their lack of sensitivity, is that they detect infectious virus and hence contagious individuals. At present, there is a lack of scientific evidence supporting this claim. The LFD used in the UK fails to identify individuals with considerable viral loads and has been subject to a class I recall by the US FDA but is still approved and recommended for use by the UK government. We believe that using LFD testing for assessing SARS-CoV-2 infection risk is a strategy which has risks that outweigh any benefits.