The effect of common infections on cognition and dementia in people with and without diabetes.

INTRODUCTION: Dementia is a major contributor to disability and dependence worldwide and is currently the leading cause of death in England. As the global burden of dementia continues to rise, identification of modifiable risk factors for dementia has become an urgent public health priority. Common infections may be associated with risk of dementia or cognitive impairment, but the nature, temporality or magnitude of any relationship is unclear. OBJECTIVES AND DATA SOURCES: The objectives of this thesis were to 1) systematically review evidence from longitudinal studies on the association between common clinically symptomatic bacterial infections and risk of incident dementia or cognitive decline, 2) examine the association of late-life infections with incident dementia using data from the Clinical Practice Research Datalink (CPRD) and Hospital Episode Statistics (HES), 3) investigate the association of midlife infections and incident dementia using data from the UK Biobank study, 4) investigate the association between midlife infections, cognitive decline and neuroimaging measures using data from the UK Biobank study. RESULTS: Evidence from 7 studies included in the systematic review (chapter 3) suggested that common clinically symptomatic bacterial infections were associated with an increased risk of dementia, with effect estimates ranging from HR 1.10 (95% CI; 1.02–1.19) to OR 2.60 (95% CI; 1.84–3.66) in a narrative synthesis of findings. However, studies were either from the United States or Taiwan, predominantly focused on hospitalised infections, mainly pneumonia or sepsis, and faced other methodological limitations including small sample sizes or inadequate confounder adjustment. Findings from my CPRD and HES study (chapter 5) showed that late life infections were associated with dementia risk in a population of 989,800 individuals aged 65 years and older. Dementia risk was higher for sepsis, pneumonia, infections resulting in hospital admission and among individuals with diabetes. In the third study (chapter 6) which compared the association of infections and dementia in a younger healthier population (the UK Biobank study), a lack of association was observed between the presence, site and setting of common midlife infections and dementia though consistent with the CPRD and HES study, an association was found for hospitalised infections. In the final study (chapter 7) which included 16,728 participants (median age 56.0 years [IQR 50.0-61.0]) midlife infections were not associated with cognitive decline on the mean correct response time, fluid intelligence and prospective memory tests. However, urinary tract infections and increasing numbers of infections were associated with slight declines in visual memory performance over time. No association was found between infections, hippocampal and white matter hyperintensity volume. CONCLUSION: Common late-life infections were associated with an increased risk of dementia, particularly for more severe infections (sepsis, pneumonia and infections resulting in hospital admission). However, mid-life infections were not associated with cognitive decline, hippocampal or white matter hyperintensity volume, with the exception of visual memory. Clinical trials are needed to investigate whether strategies to prevent infections lower subsequent dementia risk. In terms of cognitive decline, further studies with sufficient sample sizes for infection types and hospitalised infections, are needed to confirm the findings in this thesis.