Methicillin-resistant Staphylococcus aureus (MRSA) is a global public health problem and is a major cause of morbidity and mortality worldwide, imposing serious economic costs on patients and hospitals. Prior to the mid-1990s, MRSA was largely a healthcare-associated pathogen, causing infection predominantly in people with frequent or recent contact with healthcare facilities (HA-MRSA). Since then, community-associated MRSA (CA-MRSA), which often causes infection among healthy children and young adults with no exposure to the healthcare setting, has become increasingly prevalent. Worryingly, there is evidence that CA-MRSA is penetrating the healthcare MRSA reservoir, and even replacing traditional HA-MRSA strains. This highlights the need to keep abreast of the changing epidemiology of MRSA in order to implement effective infection control strategies. To investigate the composition of the healthcare MRSA reservoir and ascertain the extent to which CAMRSA has penetrated this reservoir, a countywide, population-based cohort study of MRSA in hospital inpatients and nursing home residents was conducted in Orange County (OC), California, covering a total of 46 facilities. CA-MRSA was found to be fully mixed with HA-MRSA in the hospital setting. The predominant CA-MRSA clone in the US, USA300, was the most commonly isolated MRSA clone in OC hospitals. In OC nursing homes, HA-MRSA (specifically a variant of USA100 that is also very common in OC hospitals but has not been reported elsewhere) predominates, but USA300 made up just over a quarter of the isolates and was the second most frequently isolated clone. Both OC hospitals and nursing homes were dominated by the same three strains: USA300, USA100 and a variant of USA100. Not only are community-based infection control strategies needed to stem the influx of community associated strains, in particular USA300, into the hospital setting, but also strategies tailored to the complex problem of MRSA transmission and infection in nursing homes, to minimise the impact of the unique nursing home MRSA reservoir on overall regional MRSA burden. A key component of effective infection control strategies is prompt isolation of MRSA carriers, facilitated by rapid diagnostics. PCR-based methods of MRSA detection offer a much faster alternative to traditional culture techniques, but are expensive and often complex to operate. A novel nucleic acid amplification technique developed by my industrial sponsor, TwistDx Ltd, called recombinase polymerase amplification (RPA), has been incorporated into a probe based detection system called TwistAmp MRSA, and offers a simple and cheap alternative to current commercial PCR-based assays, amplifying MRSA to detectable levels within 20 minutes. I tested the assay with diverse collections of MRSA and discovered that 4% of isolates from a UK MRSA collection could not be detected by the assay. I subsequently developed RPA primers for their detection. Nonetheless, TwistAmp MRSA was able to detect most MRSA strains, and was comparable to current commercial assays in this respect. Despite a very high analytical sensitivity of approximately 20 CFU/swab, the clinical sensitivity of TwistAmp MRSA was lower than expected with respect to the current market leader, Xpert MRSA. I investigated lysis and filtration methods to improve the assay's clinical sensitivity, but found that such methods did not currently warrant inclusion in the TwistAmp MRSA protocol. While TwistAmp MRSA performance is in line with current assays, and is a faster, cheaper and simpler assay, a problem faced by all molecular methods of MRSA detection is the constant emergence of undetectable MRSA strains, necessitating continual assay evaluation and improvement where possible.