Representative population health surveys : improving public health through rigour, diversity of methods and collaboration

Prevention and slowing the progression, of chronic diseases ( such as cancer, heart disease, arthritis, diabetes, asthma, osteoporosis, dementia and incontinence ), and influencing risk factors and health behaviours of a population, relies on the best available data - driven evidence. The quality of measurement techniques to collect representative population health survey and surveillance data is, as a consequence, brought under scrutiny. The presentation of this thesis is the culmination of 17 years work that has been focused on contributing to improving public health in South Australia. It is premised on the understanding that continual epidemiological assessment using representative population health surveys can deliver evidence - based information needed by health policy makers, health planners and health promoters to make appropriate, timely and efficient evidence - based decisions. The objective of the portfolio of published papers was to demonstrate the contribution to producing quality data - driven evidence using population surveys through rigour in collecting self - reported data, diversifying surveillance data collection methods and facilitating collaboration. This portfolio presents papers that have addressed a range of methodological and chronic disease and risk factor epidemiological issues. In terms of demonstrating rigour the publications have addressed the bias associated with non - response, the methodological rigour inherent in face - to - face surveys, the differences in estimates that can occur based on mode of administration, the science of telephone surveying and the importance of good questionnaire design to produce valid and meaningful data. The literature presented has also demonstrated the first South Australian population - wide prevalence survey dealing with the consequences of domestic violence and associated issues ( for males and females ) in the community, and in doing so, demonstrated the use of the telephone to collect large - scale data in Australia on domestic violence and associated factors in the population. In addition, the first time the importance of undertaking an array of methodological precautions during the data collection phase associated with collecting data on sensitive health issues on the telephone was demonstrated in Australia as was the assessment of the bias obtained in health estimates dependent upon which telephone - based sample was used. In demonstrating the need for diversity in data collection the research submitted within this thesis has demonstrated the range of telephone surveying development issues and challenges in Australia and the benefits and the value of both face - to - face and telephone as survey data collection tools in Australia. The publications also made a significant contribution to the literature in the survey methodology area, in particular, within the systematic error in questionnaire design, the measurement error in BMI self - reported measurements, validity of self - reported height and weight, and the overall CATI methodology area. Epidemiological collaborative research in particular in the areas of social capital, HRT, mental health, suicide ideation, osteoporosis, interpersonal violence, chronic disease epidemiology and risk factor epidemiology was demonstrated. As a consequence of my research, surveying populations about their health is now entrenched into public health and health service sectors in SA. Rigour in collecting self - reported data, diversifying survey and surveillance data collection methods and facilitating collaboration, has produced quality date - driven evidence for South Australia.Thesis (Ph.D.)--University of Adelaide, School of Medicine, Discipline of Medicine, 2006