Quantifying the transmission dynamics of a pandemic pathogen is paramount for informing an effective and proportionate public health response. However, traditional pandemic surveillance methodology can be prone to a multitude of biases, which can limit the validity of any inference based on such sources of data. For example, large-scale community testing can be heavily influenced by testing rates, test-seeking behaviours, and symptom-based testing criteria, all of which can change over time. In contrast random sampling, the process of testing a randomly selected subset of the population, can be used to accurately assess the level of community infections with far fewer biases. The REal-time Assessment of Community Transmission -1 (REACT-1) study is a community-based study that estimated the prevalence of SARS-CoV-2 swab-positivity in England over nearly two years of the pandemic (1 May 2020 – 1 April 2022), based on RT-PCR test results among random samples of the population. In this thesis I use mathematical models and data from the REACT-1 study to quantify the transmission dynamics of SARS-CoV-2 across the different phases of the pandemic in England. The periods presented cover the widespread circulation of the original SARS-CoV-2 strain, the emergence of the Alpha, Delta and Omicron variants and the rollout of England’s mass vaccination campaign. I estimate the temporal trends in infection prevalence, infection incidence, growth rate, time-varying reproduction number, infection fatality ratio and infection hospitalisation ratio. I then further explore the dynamics of variant competition using REACT-1 data to quantify (1) the establishment of the Alpha variant as it outcompeted the Beta variants, (2) the competition between sub-lineages of the Delta variant, and (3) the establishment of the Omicron variant and the resulting sub-lineage competition between BA.1 and BA.2. To highlight some of the potential future challenges in SARS-CoV-2 surveillance I additionally explore the dynamics of influenza variant competition from 2014 to 2020. High levels of immunity to SARS-CoV-2 have already been reached in many populations due to high rates of vaccination and natural infection. However, waning of immunity and the continued emergence of new variants will lead to recurrent epidemics far into the future. With pandemic surveillance capacity having already been scaled back, community surveillance based on a random sampling framework, could offer a mechanism to continue quantifying transmission dynamics even with greatly reduced testing resources.