Abstract Background A cost effective and efficient diagnostic tool for COVID-19 as near to the point of care (PoC) as possible would be a game changer in the current pandemic. We tested reverse transcription loop mediated isothermal amplification (RT-LAMP), a method which can produce results in under 30 min, alongside standard methods in a real-life clinical setting. Methods This prospective service improvement project piloted an RT-LAMP method on nasal and pharyngeal swabs on 21 residents of a high dependency care home, with two index COVID-19 cases, and compared it to multiplex tandem reverse transcription polymerase chain reaction (RT-PCR). We recorded vital signs of patients to correlate clinical and laboratory information and calculated the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of a single swab using RT-LAMP compared with the current standard, RT-PCR, as per Standards for Reporting Diagnostic Accuracy Studies (STARD) guidelines. Results The novel method accurately detected 8/10 RT-PCR positive cases and identified a further 3 positive cases. Eight further cases were negative using both methods. Using repeated RT-PCR as a “gold standard”, the sensitivity and specificity of a single novel test were 80 and 73% respectively. PPV was 73% and NPV was 83%. Incorporating retesting of low signal RT-LAMP positives improved the specificity to 100%. We also speculate that hypothermia may be a significant early clinical sign of COVID-19. Conclusions RT-LAMP testing for SARS-CoV-2 was found to be promising, fast and to work equivalently to RT-PCR methods. RT-LAMP has the potential to transform COVID-19 detection, bringing rapid and accurate testing to the PoC. RT-LAMP could be deployed in mobile community testing units, care homes and hospitals to detect disease early and prevent spread.