This study explored gender differences in quality of life and depressive symptoms over time among people with coronary heart disease (CHD) and compared them with a healthy population. Using three waves of the English Longitudinal Study of Ageing (2002-03, 2004-05 and 2006-07) methodological problems such as missing data and sources of error and uncertainty which may arise from reliance on a self-reported measure of CHD were addressed. A simulation study was set up to compare three techniques for dealing with missing data: full information maximum likelihood, multivariate normal imputation and a two-fold fully conditional specification. Results supported the use of the latter technique which outperformed the other two techniques in recovering the targeted parameters, especially with a binary outcome. Results based on imputed data showed that compared to people from the healthy population, men and women with CHD had on average lower levels of quality of life. Men with CHD were also at higher risk of having depressive symptoms than men from the healthy population. Women with CHD were as likely as women from the healthy population to have depressive symptoms. Trajectories over time of quality of life had a different shape from trajectories of depressive symptoms after the onset of CHD. Men’s quality of life declined over time and no changes in depressive symptoms were found. Women’s quality of life declined only between baseline and four year follow-up, while in the same period their risk of having depressive symptoms reduced. A sensitivity analysis based on an external validation study and a deterministic sensitivity analysis helped understand the impact that misclassification of the self-reported CHD measure could have on the results of this thesis. It was found that the reliability of the results presented could be affected by false positive and false negative cases of CHD.

Background: Traditional measures for planning in dentistry estimate dental workforce requirements based solely on normative approach. In contrast, the sociodental approach combines normative and subjective needs assessments and also incorporates behavioural propensity. The sociodental model has been recommended as a more rational approach to assessing dental needs. Much lower and more realistic levels of dental treatment needs have been reported using the sociodental approach compared to normative measures. This study compares the two approaches and applied the sociodental approach to different skill mix scenarios. Objectives: 1) To estimate and compare dental treatment needs and dental workforce requirements for a sample of Malaysian adults using the traditional normative and the sociodental approaches. 2) To estimate workforce requirements using different skill mix scenarios. Methods: This cross-sectional study was carried out on a selected sample of 732 adults aged 30-54 years who were employees of a public university in Kuala Lumpur, Malaysia. The participants’ sociodental need was assessed at three different levels; i) Normative Need, where their treatment needs were assessed based on professional judgements; this is equivalent to the traditional normative need estimates, ii) Impact-Related Need, where those who had normative need were assessed on their level of oral impacts using the Oral Impacts on Daily Performances (OIDP) index, and iii) Propensity-Related Need, where those who had both normative need and oral impacts associated with their treatment needs were assessed on their level of behavioural propensity to determine the type dental interventions most appropriate for them. The estimates of sociodental approach based on the integration of Normative, Impact-related and Propensity-related needs were compared to normative approach. Then, the requirements for dental workforce per 100,000 adults were assessed and compared between the different methods of assessing needs. Next, five different skill mix scenarios were developed where different ranges of dental tasks were delegated from dentists to professionals complementary to dentistry (PCDs). Results: The sociodental approach which comprises the assessments of Normative, Impact-Related and Propensity-Related needs resulted in significantly lower estimates than the conventional approach which uses normative assessment alone. The percentage differences in needs estimates between the sociodental and normative approaches were 91% for periodontal treatment and 89%-91% for prosthodontic treatment. Consequently, there were also differences in the number of dentists needed to treat 100,000 people. For restorative treatment, the number of dentists needed were 12.54 (normative approach) and 12.12 (sociodental approach), for periodontal treatment the respective figures were 14.43 (normative approach) and 2.32 (sociodental approach) and for prosthodontic treatment the need for dentists was 10.26 (normative approach) and 0.98 (sociodental approach). There was considerable potential for delegation of care to the PCDs, whereby the required number of dentists decreased and the required number of PCDs increased for varying levels of delegation. Conclusions: The sociodental approach to assessing dental treatment needs resulted in much lower estimates of oral health and workforce requirements than the normative needs approach. The numbers of dentists needed to deal with the dental needs of Malaysian adults can be markedly reduced by using PCDs.

No description available.

No description available.

This thesis investigates an aspect of Southampton's history not previously explored, the effects of smallpox on the town and its environs during the eighteenth century. The work provides a new viewpoint on the town's efforts to establish and maintain itself as a sea bathing and spa health resort, while at the same time supporting sick and wounded military personnel, prisoners of war and billeted troops. The study undertakes a detailed analysis of the town's inoculation records, held within the `Inoculation Book' and from this produces new information on the prevailing attitudes towards the poor, smallpox and inoculation in the town. Brief comparisons with Salisbury and Winchester demonstrate two alternative attitudes towards outbreaks of the disease and the use of inoculation, within these communities. The thesis attempts to assess the efficacy of each approach. This would merit further detailed investigation in the future. Throughout the eighteenth century there were reports of inhabitants fleeing from the towns to rural areas during smallpox outbreaks. The thesis investigates the plausibility of this premise for the Southampton area, and drawing on modem scientific research together with established ecological observation, places these combined findings within the historical context. This has resulted in an entirely new and important evaluation of the role of the rural ecological environment in the survival of earlier generations and would benefit from further investigation in other areas of the country.

Background: Previous research has demonstrated a strong association between labour market status and minor psychiatric morbidity (MPM). This PhD thesis aims to uncover the role of mediating factors, and the extent to which the relationship varies over space and time. In addition, this research seeks to establish the direction of causality and to differentiate between secure and insecure employment, and between various forms of joblessness. Methods: MPM was measured using the General Health Questionnaire (GHQ-12). Analyses were undertaken using British Household Panel Survey (BHPS) data from 1992-2008. Firstly, unstratified and gender-stratified series of nested linear and logit autoregressive random effects models were run to assess the role of confounding and mediating factors in the relationship between labour market status and MPM. Secondly, three complementary multilevel modelling approaches were used to assess the extent to which independent variation in GHQ-12 scores existed at the Local Authority District (LAD) level, and whether area-level unemployment rate was independently predictive of MPM. Thirdly, unstratified and age-group stratified fixed effects models were run in order to assess the effects of labour market transitions on MPM and therefore to investigate causality and age effects. Results: Across both genders it was shown that after adjustment for a range of confounding factors: insecure employment, unemployment, permanent sickness and other inactivity were significantly predictive of MPM compared to secure employment. Transition analyses suggest that this relationship is causal. Virtually no independent variation in GHQ-12 scores was found at the LAD level, but unemployment was comparatively less distressing for those living in high unemployment areas. Age was found to moderate the relationship between labour market status and MPM to some degree. Conclusions: This research deepens our understanding of the causal processes underlying the relationship between labour market status and psychological wellbeing, whilst considering the roles of spatial, temporal and macroeconomic context.

The association between parental separation occurring during childhood and adult psychological distress is well established, however the potential mechanisms involved in translating the experience of parental separation into the increased risk of reporting psychological distress is unclear and little investigated. Previous literature indicates that material and relational factors may be involved and that these two pathways are likely to be linked across the life course. The identification of the mechanisms involved in the association between parental separation and psychological distress will offer suggestions as to how families and children who undergo separation can best be supported in order to prevent long-term adverse consequences for psychological health. The diversification of family forms since the mid-20th century and in particular the increased chances that a child experiences parental separation in more recent years stimulated the investigation of whether the association between parental separation and adult psychological distress, and the mediating material and relational pathways, has changed over time. It is thought that as separation becomes more common it will have less of an effect upon the children involved. There are few studies which have investigated this with respect to psychological distress and this thesis extends those which do exist methodologically. Analysis of data from three British birth cohorts finds that parental separation is associated with increased chances of reporting psychological distress and this does not differ by gender, age of child or cohort. Examination of mediating pathways shows that both material and relational factors are involved, although material factors particularly so, and that these differ for men and women, and also by cohort. The inter-linkage of material and relational factors across the life course was found to be complex. These findings suggest a need to support separating families, particularly through their educational careers, in order to minimise the long-term consequences for children.

This inter-disciplinary project investigates the relationship between family structure and early child health. The two main aims are: (1) to determine whether family structure and changes in family structure are associated with children's physical health in the Millennium Cohort Study; (2) to explore potential pathways through which these associations operate. In spite of much public debate around families, marriage, and child outcomes, UK literature on this topic remains incomplete. This thesis aims to fill two gaps: first, testing whether there is a link with children's physical health, rather than more commonly reported outcomes such as cognitive function or education achievements. Physical health outcomes included are respiratory health, childhood growth, and unintentional injuries. Second, few studies use prospective, longitudinal data and methods. Cross sectional studies cannot examine the direction of the relationship, nor capture the dynamics of changes in family structure. Here, longitudinal techniques test a complex model made up of variables ordered a priori. In unadjusted analyses, family structure presented a consistent gradient in child health: cross-sectionally, children living with married parents had better health than those living with cohabiting parents, while those living with lone parents had the worst health. Longitudinally, those who experienced changes in family structure fared worse than those living with continuously married parents, with some important exceptions, such as those living with cohabiting parents who subsequently married. Socio-economic factors were important predictors of family structure and child health. Proximal pathways through which socio-economic characteristics and family structure affected child health varied according to health outcome. Maternal mental health appeared to be important across outcomes. Concluding, this work shows the importance of using nuanced definitions of family, particularly when it comes to capturing its fluidity over time. Children who experienced changes in family structure were a heterogeneous group with diverse backgrounds and outcomes. Socio-economic factors emerged as important antecedents to both family structure and child health.

The field of epidemiology is now experiencing a data deluge, demanding appropriate methods to efficiently analyse large amounts of data. In this thesis we present advances towards data-intensive epidemiology, introducing novel methods and applications of data mining in this field. We focus on two distinct applications. Our first application is the task of risk of bias assessments of systematic reviews. At present these are a highly manual process, where reviewers identify relevant parts of research articles for a set of methodological elements that affect the risk of bias, in order to make a risk of bias judgement for each of these elements. We use text mining to identify relevant sentences within the text of included articles, to rank articles by risk of bias, and to reduce the number of risk of bias assessments the reviewers need to perform by hand. The application of text mining to risk of bias assessments also led to the following methodological contributions. We introduce the concept of a rate-constrained ranking task, of which ranking articles for rapid reviews is an example. We derive a novel metric, the rate-weighted area under the ROC curve (rAVC) , to evaluate ranking models for rate-constrained ranking tasks. Furthermore, we derive a method to generate confidence bounds around ROC curves, that is particularly appropriate for these types of tasks. Our second application is the task of choosing hypotheses to test in epidemiological analyses. Currently researchers use prior knowledge about the composition of causal pathways, and their own research interests and preconceptions, to decide which hypotheses to test. Where no strong priors exist it may be preferable to use a systematic approach to identify those to follow up. We present a novel screening step that uses Mendelian randomisation to systematically search a large number of hypotheses for potentially causal relationships that should be investigated further. As an exemplar we search for the causal effects of body mass index (BMI) and find many associations with outcomes that are supported in the literature.

This thesis explores the ethical issues raised by linkage to routine administrative data specifically in relation to longitudinal birth-cohort studies. An established birth-cohort based at the University of Bristol, the Avon Longitudinal Study of Parents And Children (ALSPAC) forms the case study for this investigation. In undertaking this research I used an integrated empirical ethics approach, conducting a qualitative investigation of stakeholder perspectives in the Project to Enhance ALSPAC through Record Linkage (PEARL), and utilising the data generated to reflect upon relevant issues raised in the research ethics literature. Semi-structured interviews were conducted with key stakeholders in PEARL: researchers (n=6), individuals involved in the ALSPAC Ethics and Law Committee (n=6), and young people who are members of the cohort (n=24). The aims were to explore the experiences and perspectives of these participants to gain a contextualised and nuanced understanding of the ethical issues raised by the use of administrative data in longitudinal birth-cohort research. The data were thematically analysed and core themes of understanding, special trusting relationships and intimacy were identified that explained the ALSPAC research environment. The significant ethical issues pertaining to data linkage were identified in the main themes of benefits, burdens and informed consent. It was found that the specific research environment of ALSPAC was important in shaping the nature of ethical issues in relation to data linkage, and influential in terms of how participants understood and acted in relation to these issues in practice. By reflecting on the findings of this study and the research ethics literature, recommendations are made for the practice of data linkage. This study adds both to empirical and ethical knowledge about the ethical issues raised by data linkage, by specifically examining it in relation to longitudinal birth-cohort studies and stakeholders; which are currently under-researched areas in this field.