Sleep patterns are linked to cardiovascular outcomes and psychological well-being, but gaps in knowledge remain. This thesis tested four aspects of the relationships between sleep, cardiovascular risk and well-being using different methods of investigation; an analysis of a large population dataset (the English Longitudinal Study of Aging (Study 1)), an investigation of affective and biological responses in everyday life of working women (Studies 2, 3), and a short-term well-being intervention (Study 4). Studies 1 and 2 tested whether direct biological dysregulation may be in part responsible for higher risk of cardiovascular outcomes in poor sleepers. Study 1 found that in older adults longer sleep was correlated with elevated inflammation, while short sleep was associated with low haemoglobin. Disturbed sleep was more prevalent among those with higher inflammation, lower dehydroepiandrosterone sulfate and haemoglobin as well as anaemia. These relationships were found mostly in men, but nonetheless they emphasise that self-reported sleep has important biological correlates in older adults. Study 2 extended data from experimental studies to real life settings, and found that disturbed sleep is related to lower heart rate variability (HRV). This suggests that lower HRV, a marker of dysfunctional autonomic activity, may be another pathway contributing towards higher risk of cardiovascular outcomes in poor sleepers. Study 3 compared objective and subjective measures of sleep efficiency and discovered that psychosocial characteristics including work stress and social support are related to underestimations of sleep efficiency, in comparison with objective measures. Thus associations between self-reported sleep and health-related factors may be overestimated in studies based on self-report. Study 4 aimed to induce positive well-being in a randomised controlled trial, to test whether this would lead to improvements in sleep. Well-being was increased post-intervention, but improvements in sleep were marginal. Importantly, changes in well-being were correlated with beneficial alternations in subjective sleep, tentatively suggesting that positive well-being may exert protective effects on (self-reported) sleep. In combination, these studies contribute to the research literature relating sleep problems with cardiovascular risk and poor psychological well-being.

The key to effective scaling-up of coverage with insecticide treated nets is multiple effective delivery systems that are complementary, each one adding incrementally to the overall coverage. Generally, individual systems have been studied. A methodology is needed for studying the effectiveness of individual delivery systems, mixes of delivery systems and their relative contribution to coverage within a defined delivery system context. The insecticide treated net voucher scheme in two regions of Ghana provided an opportunity to develop a method of delivery systems evaluation. The thesis consists of eight chapters. Chapter one is the introduction, and chapter two a review of the literature on the delivery of ITNs. Chapter three presents the study justification, aim, objectives, conceptual framework, a description of the study setting and the methods used in the study. The thesis has four results chapters. In the first of these the quantitative coverage outcome evaluation of the voucher scheme is presented. Delivery attribution is used to evaluate the success of the voucher scheme. In the second results chapter, the intermediate processes in the delivery system are defined and the effectiveness of each one is assessed overall and by geographic area and socio-economic groups. In the third results chapter, qualitative methods are used to interpret the quantitative findings and to describe and explain the impact of the delivery systems context on the effectiveness of the intermediate processes of the voucher scheme. In the fourth of the results chapters' recommendations on a methodology of delivery system evaluation for ITNs and other public health interventions are made. The final chapter is a discussion of the findings of the study in the two regions of Ghana and their implications for the evaluation of delivery systems for ITNs and other public health interventions particularly in relation to malaria control.

No description available.

Diseases have the capacity to not only influence the dynamics of their hosts, but also interacting species like predators, prey and competitors. Likewise, interacting species can influence disease dynamics by altering the host's dynamics. The combination of these two effects is often called eco-epidemiology, the interaction of ecology and epidemiology. In this thesis, we explore this interplay of infectious diseases and predator--prey interactions, where the predator is a specialist. We start with an introductory chapter on modelling eco-epidemiology, with a particular focus on the myriad of different possible assumptions mathematical models in eco-epidemiology can have. In Chapter 2, we consider the effect predator--prey oscillations have on the endemic criteria for an infectious disease. In Chapter 3, we find a great variety of complex dynamics like tristability between endemic and disease-free states, quasi-periodic dynamics and chaos in a predator--prey model with an infectious disease in the predator. In Chapter 4, we consider the impact an infectious disease has on a group defending prey. Here, we find that the disease not only can coexist with a predator, it can actually help the predator survive where it could not in the absence of the disease, in stark contradiction to the principle of competitive exclusion which states that two exploiters should not coexist on a single resource. Lastly, in Chapter 5, we consider a spatial predator--prey model with a disease in the prey and focus on how preytaxis (the movement of predators along prey gradients) can alter various invasion scenarios. Through all these chapters, there is a common focus on the impact (endogenous) oscillations have in eco-epidemiology.

Background: Obesity is a major threat to public health given its strong links with cardiometabolic morbidity and premature mortality. One-third of obese adults are metabolically healthy, but little is known about modifiable determinants of this state or its progression over time. Aims: To determine whether physical activity and sedentary behaviour distinguish healthy from unhealthy obesity, and whether healthy obese adults have increased risk for developing metabolic ill-health and type 2 diabetes. Methods: Data were drawn from up to 5427 men and women participating in the Whitehall II cohort study. Normal-weight, overweight, and obese adults were considered healthy if they had < 2 of 5 metabolic risk factors (hypertension, low HDL-cholesterol, high triglycerides, high blood glucose, and insulin resistance). Associations of self-reported moderate-to-vigorous physical activity and leisure sitting time with prevalence and 15-year incidence of metabolic risk factor clustering were examined among healthy obese adults. Differences in accelerometer-assessed total physical activity were also examined between healthy and unhealthy obese groups. Metabolic risk factor incidence among initially healthy obese adults was described, and published risk estimates of incident type 2 diabetes were systematically searched and meta-analysed. Results: Neither high self-reported moderate-to-vigorous physical activity nor low self-reported leisure sitting was associated with health among obese adults. Higher total physical activity among healthy versus unhealthy obese adults was evident through accelerometer assessment only (p=0.002). After 20 years, 52% of initially healthy obese adults were unhealthy obese, with insulin resistance being most commonly incident. Meta-analyses of 8 studies indicated that healthy obese adults have 4.03 (95% CI=2.66-6.09) times greater risk of incident type 2 diabetes than healthy normal-weight adults. Conclusions: Higher physical activity rather than lower sedentary behaviour distinguishes healthy from unhealthy obesity. Healthy obesity is strongly linked with future insulin resistance and type 2 diabetes, suggesting that it is not a harmless condition.

This thesis reports two strands of work leading to the creation of a smartphone application harnessing identity change to promote smoking cessation during pregnancy. The first strand studied the role of smoker identity in smoking cessation. Using data from a national prospective study of smokers in England, Study 1 found that most young women do not have a positive smoker identity but that those who do are less likely to try to quit smoking, independent of other smoking-related attitudes. Study 2 used prospective national survey data and showed that adopting a non-smoker identity after a quit attempt predicts medium-term abstinence. Study 3 was a meta-ethnography of smoker identity in young adults and found that multiple smoker identities evolve depending on the context and that these do appear to play a role in cessation. The second strand related to the development of the application itself. Using interviews with pregnant smokers, Study 4 found that, from their perspective, stopping smoking would require strong enough motivation, improved ability and a supportive social and physical environment. Study 5 involved focus groups of health professionals working with pregnant smokers and solicited recommendations regarding the design and delivery of digital cessation aids in pregnancy. Study 6 used data from a pilot randomised controlled trial of the smoking cessation website, MumsQuit, to assess associations between exposure to particular components and four-week continuous abstinence. Six behaviour change techniques were associated with abstinence: promoting problem solving, providing feedback on behaviour, promoting self-monitoring of outcomes of the behaviour, providing feedback on outcomes of the behaviour, providing social reward, and promoting self-reward. The thesis concludes with a description of the development of a smartphone application, SmokeFree Baby, based on preceding work.

Incidence of cervical cancer has fallen dramatically since the introduction of the NHS Cervical Cancer Screening Programme, yet this largely preventable disease disproportionately affects women of lower socioeconomic status (SES). This thesis set out to explore the evidence for, and possible mechanisms of, socioeconomic inequalities in cervical screening uptake in England. Studies 1-3 were cross-sectional observational studies exploring the relationship between routinely collected Primary Care Trust (PCT)-level cervical and breast screening coverage and area-level deprivation, and testing for associations between population- and programme-delivery characteristics and cervical screening coverage. Study 4 used semistructured qualitative interviews to explore the views of health professionals on the factors that support or hinder cervical screening uptake. Study 5 used cross-sectional observational survey data to assess whether perceived benefits of cervical screening explained the association between SES and screening attendance. Study 6 was a crosssectional, observational analysis of the relationship between colposcopy attendance and area deprivation. Analysis of variance showed no significant reduction in cervical screening inequalities from 2007-12 but an improvement in breast screening coverage among lower SES women. Regression analyses revealed that population factors explained more of the variation in PCT-level cervical screening coverage than did programme-delivery factors. Health professionals considered programme-specific factors to support, and population factors to hinder, cervical screening participation. Women from lower SES backgrounds were more sceptical about the benefits of cervical screening but these beliefs explained little of the variance in screening attendance. Logistic regression analysis demonstrated that although colposcopy attendance was high, it was lower in income-deprived areas. This thesis demonstrated persistent SES inequalities in cervical screening attendance despite efforts to address the problem, and delayed uptake of colposcopy among women living in deprived areas. The work points strongly to the influence of population factors in explaining variation in cervical screening in women of all ages. Programme-delivery factors were also important for screening uptake in younger women. Some PCTs (now CCGs), were identified as exemplars of good practice and others as requiring further support. Women of lower SES delay attendance at colposcopy appointments, but earlier attendance may be achieved with support.

Research into the health benefits of music has rapidly expanded over the last decade with recent studies showing early evidence of the ability of music to alter biomarkers of the neurochemical and neuroendocrine systems. However, it is not clear to what extent music can alter the response of the immune system. This thesis explores psychoneuroimmunological responses to music, in particular focusing on how participatory music interventions can modulate inflammatory responses. The biomarkers investigated include cortisol, a neuroendocrine stress marker, a range of pro- and anti-inflammatory cytokines within the immune system, the social bonding hormone oxytocin and the neuropeptide beta-endorphin. Study 1 involved a six-week drumming intervention for mental health service users, and showed that drumming was associated with short-term increases in positive affect and cytokine activity and reductions in cortisol, and longitudinal improvements in depression, mental and social wellbeing, and reduced pro-inflammatory response. Study 2 replicated study 1 with a control group, showing comparable results at 6 weeks but also showing that if the intervention is extended to 10 weeks, there are also reductions in anxiety and all results are then maintained for 3 months following the end of the intervention. Study 3 aimed to explore the mechanisms of these effects in more detail. A randomised control trial comparing group drumming to three different control conditions showed that drumming, unlike the control conditions, leads to changes in a range of moods and emotions and the accompanying biological responses show signs of being associated with these emotions rather than with the physical parameters of group drumming. Study 4 extended the work of the previous mental health studies to explore how participatory music interventions can interact with the psychobiology of both mental health and physical health by studying patients affected by a chronic disease: cancer. A single session of group singing was found to be associated with reduced levels of cortisol, increased cytokine activity and, surprisingly, reduced levels of both beta-endorphin and oxytocin, again with associations between biological responses and emotions. Given the prevalence of mental health conditions such as depression, either as a primary or secondary diagnosis, and evidence that such conditions are associated with heightened inflammation, participatory music interventions could offer novel opportunities for managing mental health and optimising immune function in patients.

Research has shown that married men and women have better physical and psychological health and greater longevity than their unmarried counterparts. However, the past 50 years have witnessed changes in the marriage and divorce rates, resulting in more people at older ages who are unmarried or with varied relationship histories. Given the strong association between marriage and health there could potentially be more people at older ages in poorer health, which may be particularly detrimental given the ageing population. Whilst there is much research looking at marriage and physical and psychological health there is little on marriage and physical capability. Physical capability is the capacity to perform the physical tasks of daily living and is predictive of mortality and future social care use. This PhD investigates the relationship between marriage and physical capability at mid to later life using two measures – grip strength and walking speed – from two nationally representative datasets of people aged 50 years and over in England and the USA. Cross-sectional associations between marriage and physical capability are investigated in a comparative analysis between England and the USA, and longitudinal associations through examining changes in walking speed over a ten year period in England. A descriptive analysis of early life circumstances and its association with entry into and exit out of marriage in England and the USA is also carried out. Findings show that married people had both higher levels of current physical capability and a slower decline in physical capability over time than their unmarried counterparts. Much of the “marriage advantage” is explained by their greater wealth, but there were some unexplained associations, particularly among widowed men. There were few gender and country differences in the association. The results of this thesis suggest that marriage is important for maintaining physical capability for people at mid to later life in England and the USA.

The conventional sources to assess cortisol levels are saliva, plasma or urine specimens, which are subject to a variety of factors and provide only momentary cortisol exposure. A rather new technique is the analysis of hair cortisol which might overcome some of the methodological issues associated with the other specimens. The aim of the PhD was two-fold. Firstly, I investigated the associations between hair cortisol concentration and different socioeconomic factors, work-related stress and psychosocial factors. Study I employed a design with two time points four years apart and therefore was able to evaluate psychosocial and socioeconomic factors cross-sectionally and as a dynamic entity. Lower income and worsening income change was associated with elevated hair cortisol. Moreover, an effect of status incongruity, a mismatch between education and income, on hair cortisol was found. Study III looked at the impact of examination stress on cortisol in hair and saliva in medical and law students. The student groups differed slightly in their hair cortisol levels at baseline and also in levels of anxiety, impeding proper conclusive findings. Perceived stress and anxiety were not related to hair cortisol but to salivary cortisol. The use of avoidant coping mechanisms was associated with elevated hair cortisol levels. My second aim was to evaluate the long-term consistency of cortisol in saliva and hair and also saliva-hair correlations over corresponding time-intervals. Using two distinct longitudinal studies (Study I and II), the findings indicate that a flatter rate of decline in salivary cortisol over the day was associated with elevated hair cortisol concentrations several years later, 4 (female sample) and 8 years ago (in men only), while no relationship could be found between hair cortisol in relation to the AUC and the CAR. Another dataset (Study III) assessing salivary and hair cortisol over corresponding intervals revealed positive associations between the AUC and hair cortisol.