Aim. Limited evidence exists on the effects of life stress on behaviour change, especially amongst middle aged people. This study aims to assess the impact of life stress from stressful life events and being diagnosed with chronic diseases on change in smoking, alcohol use, diet, and physical activity in a national sample of middle aged people. Method. This study used data from 3 waves of the MRC National Survey of Health and Development when the sample was aged 36, 43 and 53 years. A change was defined as the change in the status of particular health behaviour between two consecutive waves (age 36 and 43 years, and age 43 and 53 years). The changes in health behaviours (outcomes) examined were stopping smoking, smoking relapse, increased risk of having an alcohol drinking problem, increased alcohol consumption, change in dietary behaviour (index score), increased physical activity and decreased physical activity. Stressful life events and health related life events were assessed using scores derived from a stressful life events inventory. Diagnosis of a medical condition was the self reported diagnosis of 5 chronic conditions: hypertension, angina, heart attack, stroke, and diabetes. Associations were assessed at age 43 years, 53 years, and the aggregate of both observations at ages 43 and 53 years. The influence of three social support factors: perceived support, social network, and social participation, and demographic factors: sex, social class, and education level on the association were also assessed. Results. The analysis found that stressful life events was associated with greater odds of smoking relapse in the ex-smokers and lower odds of increased physical activity in the cohort members who were not physically active at baseline. Health related life events were found to be associated with lower odds of increased physical activity. Being diagnosed with at least one medical condition was associated with greater odds of stopping smoking in current smokers, lower odds of smoking relapse in ex-smokers, and increased physical activity. It was also found to be associated with improved diet behaviour in the men. The social support factors were found to influence the effects of the life stressors in some of the associations. Conclusion. Life stress from life events experience does influence health behaviours change amongst a middle aged national sample. However, particular life stress from health related life events, specifically, from being diagnosed with a chronic medical condition can motivate health protective behaviour change. Social support factors buffer the effect of life stress to some extent.

This thesis investigated the influences of early life socioeconomic circumstances (SEC) on cardiovascular disease (CVD) risk in Russia, Poland and the Czech Republic. Early life SEC are inversely associated with CVD risk factors and outcomes in western countries, but the same relationships have not been evaluated in Central and Eastern Europe. CVD is a major cause of morbidity and mortality in this region. I used data from the first round of the HAPIEE (Health, Alcohol and Psychosocial factors In Eastern Europe) study, which randomly selected men and women aged 45 to 69 years from population registers in Novosibirsk (Russia), Krakow (Poland) and six towns in the Czech Republic. Nearly 29,000 people were recruited, with an overall response rate of 61%. Blood pressure (systolic and diastolic blood pressure, hypertension), lipids (total and HDL cholesterol), adiposity (BMI, waist circumference, waist to hip ratio), smoking habits (starting and quitting) and CVD risk, as measured by SCORE, were assessed in relation to retrospectively collected direct and proxy measures of early life SEC. The reliability of adult anthropometric measures (height, leg length and trunk length) as proxy markers of early life SEC was confirmed by investigating the relationships with three direct measures of early life SEC (maternal and paternal education and household ownership of six assets at age ten years). Higher childhood SEC were linked to an increased likelihood of women starting smoking, a decreased likelihood of men quitting smoking, and reduced adiposity in both genders. The remaining CVD risk factors and overall CVD risk did not appear to be consistently influenced by childhood SEC. The results of the thesis suggest that early life socioeconomic influences on classical CVD risk may not be consistent across cultures, and that they may vary by the stage of the epidemiological transition and by local context.

How and why depression confers greater risk of impaired adaptation following coronary artery bypass graft (CABG) surgery is not well understood. This PhD aimed to address these issues by: developing and piloting a longitudinal study (the ARCS study) in order to track the recovery of patients undergoing CABG surgery; and conducting an extended version of the ARCS study on which to test the association between pre-operative depression and post-operative recovery and the underlying social-behavioural, cognitive and biological pathways. Outcomes were studied in the early and short term, namely three to five days (216 participants) and two months (154 participants) following surgery. The results indicated that greater pre-operative depression symptoms were predictive of poorer recovery, including longer in-hospital stays, greater emotional distress, physical symptoms and pain in the early term, and greater emotional distress, physical symptoms and pain and impaired health status in the short term, independent of demographic and disease severity factors (all p <0.05). Some associations were limited to subtypes of depression symptoms. Specifically, somatic/affective, but not total or cognitive/affective, depression symptoms were predictive of short-term outcomes. Mediators of the depression-recovery relationship were tested. Social support and behavioural factors were not shown to be mediators, but instead physical activity, body mass index and smoking status all had independent effects on recovery. Cognitive mediation was shown, with greater pre-operative negative illness perceptions mediating the relationship between pre-operative depression symptoms and post-operative anxiety and physical symptoms in the early term, and affective pain and physical symptoms in the short term. Biological mediation was not shown: although depression symptoms were related to neuroendocrine and inflammatory patterns suggestive of poorer physical functioning, these patterns did not consistently relate to recovery. Further work is needed to translate these findings into new ways to approach the measurement, diagnosis and treatment of depressed cardiac surgery patients.

No description available.

Plasmodium ookinetes are an invasive stage of the malaria life cycle that occurs in the mosquito vector. Ookinetes traverse the mosquito midgut epithelium where they come to rest under the basal lamina and transform into oocysts. The close association of the developing parasites with the basal lamina has led to proposals that the basal lamina may provide signals to induce transformation of ookinetes into oocysts. This hypothesis was tested using an in vitro culture system for the murine malaria parasite, Plasmodium berghei, and transformationw as found to occur independentlyo f the presenceo f laminin. The putative role of laminin in invasion was also investigated in vitro; however, the insect cell lines used in co-cultures, that had been previously described to be invaded by ookinetes, were found instead to phagocytose ookinetes. Thus, phagocytosis of ookinetes by four insect cell lines is reported. Finally, laminin has been shown to bind to Plasmodium ookinete proteins and oocysts have been reported to incorporate this molecule into the oocyst capsule. It has been hypothesised that Plasmodium incorporation of laminin, and other basal lamina components, into the oocyst capsule would enable the parasite to mask itself from the innate immune system of the vector. Anopheles stephensi mosquitoes were infected with P. berghei and incorporation of laminin into the oocyst capsule was investigated using immunogold labeling and electron microscopy techniques. Laminin was observed to coat oocysts and was found to be incorporated in the oocyst capsule. The close proximity of the vector stages of malaria parasites to the insect basal lamina have led to several suggestions that important interactions may occur between the parasite and the basal lamina that are essential for parasite development. The results obtained demonstrate that laminin is not important in the early stages of sporogonic development.

Background: A growing body of evidence shows that social capital may affect the health and well-being of older adults. A number of studies also suggest that social capital is a determinant of oral health. However, the evidence for these claims is weak in terms of causal inference criteria such as temporality because previous studies are mostly based on cross-sectional analysis. Aims: The aims of this thesis were to examine whether social capital is a determinant of oral health among older adults, and whether this association is explained by socio-demographic, socio-economic, health and behavioural factors. Methods: Secondary data from waves 3 (2006-07) and 5 (2010-11) of the English Longitudinal Study of Ageing (ELSA) were analysed with 6,977 adults aged 50 and over in the cross-sectional analysis, and 5,385 and 5,114 older adults in the longitudinal and modelling of change analyses respectively. Four measures of social capital were derived from the ELSA study, reflecting structural (membership in organisations and volunteering) and functional (number of close ties and social support) dimensions of the concept. Oral health outcomes were assessed using measures of self-rated oral health, oral health-related quality of life and edentulousness. Binary and multinomial logistic regression models were used to estimate the odds of poor oral health for different levels of social capital, sequentially adjusted for socio-demographic, socio-economic, health and behavioural factors. Results: There was some evidence that lower levels of social capital were associated with poorer oral health among older adults in the cross-sectional and longitudinal analyses. The size of the statistically significant associations ranged from odds ratios of 1.21 (95%CI:1.01-1.46) to 2.14 (95%CI:1.62-2.84) independent of other dimensions of social capital and several measures of oral health. Poor oral health at baseline (2006-07) also predicted lower levels of social capital 4 years later. There was weaker evidence that positive/negative changes in social capital were associated with improving/worsening oral health. The only consistent finding was the association between low social support and poor self-rated oral health, with odds ratios of 1.36 (95%CI:1.11-1.66) in the cross-sectional analysis, 1.27 (95%CI:1.01-1.60) in the longitudinal analysis, and 1.46 (95%CI:1.13-1.90) in the modelling of change analysis. Conclusions: Overall, the results of the thesis found weak evidence that low social capital is a determinant of poor oral health among older adults. There was some evidence of longitudinal associations between functional dimensions of social capital and subjective oral health, but little evidence for other measures of social capital and oral health. One key area of further research is the mechanisms and interventions by which older adults are able to generate and maintain social support and close ties in later life.

Childhood obesity adversely affects children’s physical and psychosocial health. Studies suggest that obesity treatment can improve children’s health. This thesis describes the process of developing and testing an intervention for overweight and obese 8-12 year old children. A systematic review, with strict parameters on the timing of outcomes and nature of comparison groups, found moderate evidence that treatment is effective compared to control and evidence was strongest for treatments based on behavioural therapy. The most widely cited programme at the time this study was developed was a behavioural treatment developed in the United States. Whilst this treatment had produced encouraging outcomes, the literature was severely limited by an absence of controlled trials and limited ethnic and social diversity in participants. The process for adapting the treatment for UK families was then described. The treatment was piloted with 33 obese children and was well received. Children attending treatment significantly reduced adiposity and experienced improved mood and self-esteem. The randomisation procedure was not tested, but the recruitment and assessment protocols appeared feasible. A randomised controlled trial was then conducted; the primary outcome was difference in BMI and BMI SDS at 6 months. 72 children were randomised (37 and 35 in the treatment and control groups respectively). Intention to treat analyses (including all children with baseline data) using baseline values where 6 month data were missing resulted in data being available for 58 children. Between-group differences in adiposity were not significant at 6 months, and there were no differences for secondary outcomes (body composition and blood pressure) and only marginal differences in exploratory analyses of psychological outcomes. The lack of a significant effect appeared due to the control group experiencing equivalent change to the treatment group. The implications and potential applications of the findings from the trial and systematic review are discussed.

While the rate of AIDS-related death has declined, as a consequence of the effectiveness of antiviral treatment for HIV, HIV/HCV coinfection and in particular liver-related death (LRD) has assumed increasing importance. This thesis aims to analyse important epidemiological areas of HIV/HCV coinfection to improve the knowledge base of the subject and provide guidance to clinicians in a fast moving area of research. Data for this thesis are from the EuroSIDA study, which is a large multi-centre pan-European prospective observational cohort study with over 18,000 HIV-positive individuals including approaching 5,000 HIV/HCV coinfected individuals. The study was initiated in 1994 and continues to expand and diversify to meet current research needs. Results from the studies included in this thesis have shown that treatment for HIV in coinfected individuals can also have a beneficial effect on the natural course of HCV, with HCV viral load remaining stable over time in those treated for HIV compared with increasing HCV viral load in those not yet treated. The incidence of treatment for HCV has steadily increased in Europe to 4.7 per 100 PYFU in 2010, but remains low with just 25% of eligible patients receiving treatment. LRD accounts for more than a fifth of deaths in this population, with significant liver fibrosis and those triple infected with HBV at increased risk. The 5-year probability of LRD is low for those with F0/F1 fibrosis (2.2%), but increases substantially for those with F2/F3 (10.3%) and F4 (14.0%) fibrosis. With potent new treatments for HCV coming to market, it is clear that while they remain prohibitively expensive they should be targeted at those at the greatest risk of LRD. The prognostic LRD score derived here will help clinicians to make difficult decisions on who should be prioritised for HCV treatment.

The value of genetic test feedback for common conditions is widely debated. This is principally because of the lack of impact on behaviour change of feedback for genes with small effect sizes, but also because of concern about the risks of fatalistic responses to positive test results or false reassurance from negative results. This thesis describes research using feedback for one gene, FTO, implicated in the development of obesity, as a model to investigate motivational and emotional reactions to testing for genetic susceptibility. It comprises a series of six studies examining the benefits and harms associated with genetic test feedback. They incorporated a mixture of qualitative and quantitative methodologies, used hypothetical and real genetic feedback, and tested predominantly normal-weight students and overweight/and obese individuals from a web panel. Fatalism or false reassurance in response to FTO genetic test feedback was not observed in any of the studies. Genetic test feedback was consistently perceived as motivating, and negative emotional effects of a higher-risk FTO gene result were minimal. Overweight and obese individuals found the test result helpful for alleviating guilt and stigma; although in response to an unexpected lower-risk genetic test result, some were disappointed. University is notoriously a life stage with risk of weight gain but not all students gain weight. One study examined associations between genetic risk status and weight gain, and found that students with at least one higher-risk allele were more likely to gain weight. The final study was a randomised controlled trial examining the effect of giving FTO feedback alongside simple weight control advice to first year students. Short-term (one month) results showed that weight control intentions were significantly higher in those randomised to receive FTO feedback and weight control advice than weight control advice alone, but there was no effect on weight or reported behaviour change. Although the studies in this thesis had many limitations, the findings indicate that people are unlikely to misinterpret or overstate the impact of genetic test results, at least in the context of a single gene implicated in a multifactorial condition. However, effects on behaviour remained elusive. This indicates the need for future research to learn how to harness the potential of genetic information to promote personalised prevention.

This thesis combines neuroimaging and epidemiological techniques to investigate the hypothesis that late-life depressive symptoms are partially caused by vascular risk factors. Magnetic resonance imaging (MRI) was used to study the structural brain changes associated with depressive symptoms, major depressive disorder and long-term exposure to vascular risk factors (hypertension, dyslipidaemia, diabetes, smoking and Framingham stroke risk). This was complemented by an epidemiological approach to investigate whether vascular risk factors are associated with depressive symptoms. A sample of participants from the Whitehall II study were invited to take part in the Whitehall Imaging sub-study at the University of Oxford. Participants recruited between April 2012 and June 2013 (n=229, mean age 69, age range 60-82 years, 83% male) underwent detailed cognitive testing, a clinical interview and a multi-modal 3 Tesla MRI brain scan. Depressive symptoms were measured at previous Whitehall II phases, and again in 2012-2013 using a structured assessment for DSM-IV mood disorder and a self-report questionnaire. Long-term exposure to vascular risk factors was measured at five collection phases between 1985 and 2009. Ten percent of participants (n=23) had current depressive symptoms and 13% (n=29) had late-onset depressive symptoms (depression onset after age 60). Current and late-onset depressive symptoms were associated with reduced white matter integrity in frontal-subcortical areas. Study of the MRI correlates of vascular risk factors also showed an association between long-term exposure to high fasting glucose (mean across five examinations between 1985 and 2009) and reduced white matter integrity in frontal-subcortical areas. However, long-term exposure to other vascular risk factors was not significantly associated with depressive symptoms. In conclusion, while vascular risk factors were not consistently related to late-life depressive symptoms, long-term exposure to high glucose levels and depressive symptoms were both associated with reduced white matter integrity in frontal-subcortical areas.