Understanding the ‘Traditional’ Mediterranean Cuisine: Relationship to Cognitive Health and Advances in Measurement of Adherence Sue Radd-Vagenas A thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy Physical Activity, Lifestyle, Ageing and Wellbeing Research Group Faculty of Health Sciences The University of Sydney 2019 “The doctor of the future will no longer treat the human frame with drugs but rather will cure and prevent disease with nutrition.” – Thomas Edison (1847-1931) ii CANDIDATE’S CERTIFICATE I, Sue Radd-Vagenas, hereby declare that the work contained in this thesis is my own and does not contain any material submitted for the award of another degree at any institution or university. I, Sue Radd-Vagenas, further confirm I was the principal researcher of all the work contained in this thesis, including work published with co-authors acknowledged within the text. I, Sue Radd-Vagenas, understand that if I am awarded a higher degree for my thesis entitled “Understanding the ‘Traditional’ Mediterranean Cuisine: Relationship to Cognitive Health and Advances in Measurement of Adherence” being submitted herewith for examination, the thesis will be lodged in the University Library and will be available for immediate use. I agree that the University Librarian (or in the case of the department, the Head of Department) may supply a photocopy or microform of the thesis to an individual for research or study or to a library. Sue Radd-Vagenas 25th February 2019 iii SUPERVISOR’S CERTIFICATE As supervisors for Sue Radd-Vagenas’ doctoral work, we certify that the thesis entitled “Understanding the ‘Traditional’ Mediterranean Cuisine: Relationship to Cognitive Health and Advances in Measurement of Adherence” is in a form ready for examination. Professor Victoria M Flood Faculty of Health Sciences The University of Sydney 25th February 2019 Professor Maria A Fiatarone Singh Faculty of Health Sciences The University of Sydney 25th February 2019 iv ACKNOWLEDGEMENTS This thesis would not have been possible without the unwavering support and encouragement of my husband, Nicholas Constantine Vagenas. Thanks ‘agapi mou’ for believing in me and the annual trips to Greece so I could immerse myself in Mediterranean culture. Also, for sharing your ‘traditional’ Greek recipe ideas, passed down from your mother. I found it delicious and enlightening to also be able to take the science into the kitchen. Secondly, I’d like to acknowledge my Croatian family. My father, for instilling a strong work ethic in me and always pushing me to reach greater heights. My mother, for her soothing advice and practical support when I was stressed and needed it most. I am indebted to Professors Vicki Flood and Maria Fiatarone Singh, my distinguished supervisors. You brought diverse and complementary insights and feedback during my candidature as well as challenging and inspiring me to complete a PhD – the first in my family! To librarians Elaine Tam and Kanchana Ekanayake: thank you for sharing your tips for database searches and to master EndNote. You helped boost my confidence. I also thank the co-authors mentioned in this thesis. Your collaboration and comments that emanated from various disciplines enabled me to write well thought out papers, which were published in leading nutrition journals during my candidature. This was a dream come true. It takes special people to volunteer for research projects. I am very grateful to all the participants from The Study of Mental and Resistance Training (SMART) and the v Maintain Your Brain (MYB) trial for completing dietary surveys and keeping food records. Without you, none of this would have been possible. Thank you also to my wonderful staff at the Nutrition and Wellbeing Clinic for keeping my private practice afloat while I focussed on my doctorate. Your ability to mostly work independently has been another blessing. Finally, to friends with advanced technical skills and an eye for detail, who freely gave of their time to assist me, I give thanks. Robert Clark for help in Adobe with figures for my systematic review and Rebecca Lister for checking on my formatting and layout within Word and teaching me tricks to improve my thesis presentation. You have contributed to making this thesis more beautiful and enjoyable to read. vi DEDICATION In loving memory of my dearest grandmother, Nevenka Lazic (1923-2016), who inspired me in all things related to good food and nutrition, ever since I could remember holding on to her apron strings in her Croatian Mediterranean kitchen. vii ABSTRACT The Mediterranean diet is a popular construct and already promoted by dietary guidelines as a healthy way to eat. It is also the most widely studied dietary pattern within nutrition science, having been associated with multiple health benefits for chronic disease. One of the least well researched areas, however, relates to brain health. Finding ways to prevent, slow or reverse cognitive decline has been identified as a priority for society and governments due to increasing rates of dementia within an ageing global population. Therefore, determining the effect of the Mediterranean diet on cognition and brain morphology and function, particularly for at-risk groups, is considered of paramount importance. However, there are inconsistencies in the definition of the Mediterranean diet, and these have impeded comparisons across trials. The terminology for the Mediterranean diet came into existence after the Seven Countries Study conducted in the late 1950s and early 1960s. It therefore referred to the ‘traditional’ Mediterranean cuisine during that time period. Yet over the last 50 years the term ‘Mediterranean diet’ has been loosely used, often without specificity to the archetypal diet. Hence, Mediterranean diet interventions and adherence tools have variably represented the ‘traditional’ dietary pattern. In addition, no adherence tool exists that has been specifically developed and validated for online and in-person use to assess adherence to the ‘traditional’ Mediterranean dietary pattern and aspects of cuisine within at-risk Western cohorts, including individuals diagnosed with mild cognitive impairment (MCI). To address these gaps, this thesis had four specific aims. First, to empirically describe the ‘traditional’ Mediterranean cuisine, improving its understanding within the literature. Second, to systematically investigate and synthesise findings from existing clinical trials using a Mediterranean diet intervention for cognition and brain morphology or function outcomes, and inform future research requirements. A particular focus of this review was viii on the quality of interventions used and how they related to the ‘traditional’ dietary pattern, in terms of types and quantities of prescribed foods. Third, to develop a new diet index tool to assess, more broadly, adherence to the ‘traditional’ Mediterranean dietary pattern and aspects of cuisine for use within Western cohorts. Fourth, to test the reliability and validity of this tool for online administration to healthy Australians [based on n=84 from the Maintain Your Brain (MYB) validity study cohort] and in-person use among individuals from the same country diagnosed with MCI [based on n=68 from the Study of Mental and Resistance Training (SMART) cohort]. The goal of the discussion chapter was to sum up the thesis findings and discuss their implications, and delineate remaining research gaps to progress the field. For the first aim, a narrative review was conducted investigating the evolution of the Mediterranean diet and cuisine from ancient to modern time periods (see Chapter Two). This review identified some additional concepts from the ‘traditional’ cuisine, which have previously been poorly captured, overlooked or not fully examined, but which may be potentially important and influence health outcomes. It also found that popular Mediterranean diet index tools rarely assess such elements. To address the second aim, a pre-defined protocol was developed and the first systematic review exclusively of randomised controlled trials was conducted on the effect of a Mediterranean diet on cognition and brain morphology and function (see Chapter Three). This review established that previous interventions varied considerably and did not necessarily represent the ‘traditional’ diet and cuisine. Overall, the results showed no benefit for a Mediterranean diet intervention, with mostly non-significant and small effect sizes for cognition, and no evidence for brain morphology or connectivity outcomes across the limited experimental evidence. However, in the most robustly-designed trial, which also included a more ‘traditional’ intervention, there was evidence for attenuation of cognitive decline over four years of ageing compared to a lower fat diet, with significant effect sizes for three domain composite scores representing Memory, Frontal and Global function. This finding persisted with the recent corrections within the original paper for a sub- cohort of this study. Hence, promising clinical evidence exists to support that a more ‘traditional’ Mediterranean diet may slow progression of cognitive decline in older adults with cardiovascular risk factors. To achieve the third aim, a new Mediterranean diet index ix tool was developed (see Chapter Four). This was informed by earlier findings described in Chapter Two and the reported limitations of existing tools, which may be reducing their ability to consistently predict health outcomes. The tool, named