Healthcare associated infection surveillance: Examining influences on reliable and valid data collection and analysis

Healthcare settings are dangerous places. Individuals who receive healthcare may be subject to unintended harm as a consequence. One potential adverse event is a ‘healthcare associated infection’. This contemporary term refers to any infection which is acquired in healthcare facilities or any infection that occurs as a result of healthcare interventions. This thesis is concerned with the topic of healthcare associated infections. The effects of healthcare associated infections are felt not only by individual patients through increased morbidity and mortality but also by health services faced with higher costs associated with infections. The prevention of infection requires a multifaceted approach which is underpinned by healthcare-associated infection surveillance. Surveillance is used to influence practice and policy as well as to evaluate the effectiveness of strategies to reduce healthcare associated infections. Surveillance of healthcare associated infections is a critical element of any infection control program and it is crucial that healthcare-associated infection surveillance data are reliable and valid. In this thesis, three individual studies are presented. The three studies focus on two specific healthcare-associated infections: Staphylococcus aureus bacteraemia and Clostridium difficile infection. The aim of this thesis is to explore the epidemiology of these two infections and, in doing so, to examine methodological influences on reliable and valid healthcare associated infection data collection and analysis. The first study – an examination of the epidemiology of Staphylococcus aureus bacteraemia in Tasmania, Australia – used a descriptive, observational, population-based study design. This is the first known Australian study to capture and analyse data from all cases of SAB at a population-based level and represent this as an incidence. Four key findings can be identified from this study. First, the incidence of Staphylococcus aureus bacteraemia at a population level was accurately determined for the first time in Australia and was found to be 21.26 per 100,000 population, with 42% of Staphylococcus aureus bacteraemia being healthcare associated. Second, 55% of healthcare associated Staphylococcus aureus bacteraemia was associated with intravascular device management. Third, case definitions for healthcare associated Staphylococcus aureus bacteraemia have an influence on detection. Sixty-eight per cent of healthcare associated Staphylococcus aureus bacteraemia occurred in persons hospitalised less than 48 hours but had other criteria which resulted in them being defined as healthcare associated. Therefore, in cases where no criteria other than timeframe are used to define cases of SAB, approximately 30% of cases of SAB would be incorrectly identified as community associated. Fourth, 11% of Staphylococcus aureus bacteraemia were identified in private hospitals which fall outside the scope of almost all Staphylococcus aureus bacteraemia surveillance programs in Australia....