This is an investigation into lipodystrophy syndrome (LS), defined as specific body fat alterations (BFA) and/or metabolic abnormality (MA), in HIV-infected children. Antiretroviral therapy (ART) has resulted in improved disease-free survival in HIV-infected individuals, but is an established risk factor for LS. The metabolic profile seen in LS patients is similar to that seen in the early stages of cardiovascular disease. As ART coverage improves, and because treatment is life-long, the importance of LS will become more relevant, especially as treated children survive into adulthood, and accumulate longer durations of exposure to multiple drugs. Data from a prospective European multi-centre cohort of 426 HIV-infected subjects aged 2-18, has been analysed. The manifestation of LS using clearly-defined phenotypes is described, with estimates of prevalence and incidence. Risk factors for LS were identified in logistic, and proportional hazards regression models. The temporal relationship between BFA and MA was explored, and the impact of LS on serum lipids modelled using a multi-level approach. Finally interventions used to manage symptoms of LS were described. The prevalence of LS at recruitment (median age 12.2 years) was 56.5% (95% CI: 51.7, 61.3): half had BFA alone, one-quarter had MA alone, and one-quarter had both. Over follow-up (median: 4.2 years) the incidence of BFA was 8.0 per 100 person-years (95% CI: 6.0, 10.7) and that of MA was 4.1 per 100 person-years (95% CI: 2.8, 5.9). Significant, independent risk factors associated with increased risk of LS outcomes included protease inhibitor (PI) or nonnucleoside reverse transcriptase inhibitor (NNRTI) use, age, clinical status, and White ethnicity. Decreased risk was associated with immunosuppression and detectable viral load. Several factors were associated with increased pro-atherosclerotic lipid concentrations over follow-up: however, NNRTI use and female sex were significantly associated with increased antiatherosclerotic HDL-cholesterol. This thesis used data from a unique large prospective cohort to underline the increasing significance of LS in HIV-infected children and adolescents.

Background: Despite concerns about widening social inequalities during the past 20 years of economic stagnation in Japan, evidence on health inequalities is sparse. Whether health inequalities are widening or narrowing, and what factors contribute to inequalities, remains unclear. Aim: To describe temporal trends in health inequalities between 1986 and 2007 and to investigate the contribution of material, behavioural, psychosocial and social relational factors to health inequalities in Japan. Methods: A series of eight triennial nationally representative sample surveys was analysed (n=398,303). Household income and a novel theory-driven social classification were used to calculate trends in relative and slope indices of inequality [RII and SII, respectively] in self-rated fair or poor [suboptimal] health. The contribution of mediating factors to the social gradient in suboptimal health was investigated in the 2001 sample. Results: In men, temporal trends in income RII narrowed over the period (RII declined 1.2% per year, p=0.008). Stable inequalities were observed in women’s income SII. Men’s income SII and women’s income RII showed marginally significant narrowing time trends. Inequalities by social class were constant in both genders. After imputation for missing household income, narrowing trends in income RII and SII were evident (annual declines: men 1.2%, women 1.1% for RII; both genders 0.1% for SII; all p<0.05, n=490,632). Overall, there were V-shaped time trends in age-standardised self-rated suboptimal health in both genders (quadratic term: men p<0.001, women p=0.005), with the lowest prevalence in early/mid 1990s. Mediating factors analysed altogether accounted for 20% in men’s and 44% in women’s income inequalities in self-rated suboptimal health in 2001. Conclusions: Health inequalities according to household income showed narrowing trends, but persisted over the study period. The prevalence of suboptimal health increased since the early/mid 1990s. Changes in the distribution of mediating factors over the period might have influenced the time trends observed.

Exercise Referral Schemes (ERSs) are a widespread multi-agency intervention in which patients are referred to a programme of supervised sessions of subsidised exercise at a local leisure centre. National guidance states that schemes should employ strategies to engage people from disadvantaged groups. While people from such groups are known to attend primary care more frequently than those from more advantaged socioeconomic groups, research suggests that they are less likely to use preventive and specialist health services. This thesis aims to evaluate whether access to and use of ERSs is equitable through an examination of socioeconomic differences in referral, uptake and completion of the service. Firstly, the thesis presents a case study of key research, policy and practice events concerning the development of ERSs. Secondly, the thesis details findings of a scoping review undertaken across all ERSs in Greater London to identify schemes with suitable routine data collection to participate in the equity analysis. Thirdly, the thesis presents a cross-sectional analysis of 7985 patients referred by general practices to ERSs operating in six PCTs between April 2004 - March 2006. The main outcome measures were i) risk ratios for referral by general practice deprivation quintile ii) odds ratios for uptake of ERSs and iii) odds ratios for completion of ERSs by patient deprivation quintile. Fourthly, an exploration of the added value of using a geodemographic segmentation tool to enhance understanding of socioeconomic inequalities in service utilisation at small-area level is described. This research found that general practices within deprived areas were more likely to refer patients to ERSs than their counterparts in more advantaged areas. There was no evidence of an association between socioeconomic circumstance and likelihood of either taking up or completing the scheme. The implications of this research for policy, practice and future research are discussed.

Cognitive development in childhood is a key factor affecting adult life chances, including educational and occupational success. Intergenerational continuity in cognitive ability is often observed. Thus the persistence of poor cognitive outcomes across generations may lead to a ‘cycle of disadvantage’ that is difficult to break. In this thesis, intergenerational associations in cognitive ability between parents and first-born offspring were examined longitudinally. 1,690 members of the British 1946 birth cohort with at least one offspring constituted the study sample. Cognitive ability was measured at age eight years in parents and offspring. Social mobility and parenting practices were examined for their affects on the transmission of cognitive ability across generations. Offspring of parents who improved upon the occupational social class of their own fathers by the time they were aged 26, as well as offspring of parents who remained in a non-manual class, had higher cognitive scores than those whose parents remained in a manual social class, or who showed negative intergenerational mobility. Upwardly mobile and stable non-manual parents were also more likely to use positive parenting practices. Four measures of parenting were shown to mediate part of the intergenerational relationship in cognitive ability. The intellectual home environment, parental aspirations and cognitive stimulation were positively related with cognitive outcomes in the second generation, while coercive discipline was negatively associated with offspring ability. Path analyses revealed that maternal education, but not occupation, was an important predictor of offspring cognition. The educational attainment of fathers indirectly influenced the cognitive development of the next generation through its effect on occupational social class. For those parents with the lowest and highest ability scores, the quality of the intellectual environment enabled their offspring to ‘escape’ or replicate parental cognitive ability respectively. Cognitive stimulation and paternal aspirations helped offspring to avoid repeating the poor cognitive outcomes of their parents. These data add to the relatively few studies that examine intergenerational continuity and discontinuity in cognitive ability. The results provide a basis for understanding some of the processes by which parenting practices may influence intergenerational relationships.

Cognitive function has been associated with many physical and mental health conditions, as well as mortality. Cognitive decline is one aspect of ageing that causes anxiety among the general population. Understanding the risk factors which affect cognitive function over the life course is therefore important. One potential risk factor is socioeconomic position (SEP). This thesis investigates the impact of SEP across the life course on crystallized cognitive function and memory decline. The 1946 British birth cohort study and Whitehall II study of British civil servants were used for these analyses. Missing data is a potential source of bias in longitudinal studies, with both SEP and cognitive function predictive of dropout. This thesis therefore considers the impact of methods for dealing with missing data on the findings. A complete case analysis is compared with multiple imputation and Heckman selection models. To compare the suitability of these methods a simulation study was carried out. The Heckman selection method did not perform well in the simulation study. Multiple imputation was the best method of the three considered for data missing not at random. The impact of SEP on cognitive function varied by cohort, as well as SEP and cognitive measures, with father‟s occupational SEP, but not childhood household amenities, associated with crystallized cognitive function in the NSHD after adjustment for later life SEP. Accumulation models were usually supported when considering the life course hypotheses. In some analyses the conclusions varied depending on the missing data methodology utilized. Overall, there was no consistent conclusion as to whether childhood SEP remained a significant predictor of cognitive function in adulthood, but it was not a significant predictor of cognitive decline in Whitehall II after adjustment for later life SEP. Multiple imputation was found to be an appropriate method of dealing with missing data in most situations.

This thesis is an examination of the association between alcohol use and mental health. Typically, research on alcohol use and mental health has concentrated on clinical disorders such as alcohol dependency and major depression. Less attention has been paid to drinking patterns as risk factors for, or consequences of, poor mental health. This thesis aimed to explore the relationship between hazardous patterns of alcohol consumption (such as ‘binge’ drinking) and mental health longitudinally - taking into account both directions of causality whilst also modelling a reciprocal relationship over time (i.e. alcohol use and mental health influencing changes in each other). The main purpose of this work was to compare competing models of causation to conclude which “best fits” – i.e. whether hazardous alcohol use leads to poor mental health, whether poor mental health leads to engaging in hazardous alcohol consumption, or a combination of both. Data from three existing sources was drawn upon to achieve these goals: the Whitehall II study of British Civil Servants, the Birmingham Untreated Heavy Drinkers (BUHD) project and the Health, Alcohol and Psychosocial factors In Eastern Europe (HAPIEE) two-wave survey of participants from the Czech Republic, Russia and Poland. All datasets have repeated measures of a number of alcohol related variables and standardised mental health scales. Questions pertaining to basic demographic and other lifestyle variables are also available. Using dynamic modelling techniques, no significant effects were observed for any measure of alcohol consumption (weekly units and frequency of heavy drinking days) on changes in mental health, however, mental health was found to impact upon changes in drinking behaviours suggesting that mental health is the dominant underlying process in the relationship between alcohol use and mental health (no evidence was found for a reciprocal relationship/feedback loop over time).

The equitable provision of care is a core principle of the NHS. Previous research suggests that older cancer patients may be less likely to use specialist palliative care than younger patients, but studies have failed to fully define and measure clinical need. The aim of this study was to examine use of specialist palliative care in relation to age, after controlling for need. I used a mixed methods approach, grounded in a pragmatic philosophy and drawing upon a health capability account of equitable healthcare. I undertook a focused ethnography of three specialist palliative care services, using documentary evidence, observation of meetings, and interviews to investigate conceptualisations of need for care. I derived two models of need. The first ‘aspirational’ model encompassed physical, psychological, social and spiritual care for patients and carers. However, with limited resources, a predominantly physical model of need was applied. Additionally, observations suggested that care may vary in relation to patient characteristics including age. To locate a suitable measure of need, I conducted a systematic literature review and critical and content appraisal of health-related quality of life instruments. I chose the EORTC QLQ-C30 instrument as the indicator of need in a cross-sectional survey of patients and carers, conducted to measure use of specialist palliative care in relation to age. 252 patients and 137 carers attending four outpatient lung cancer clinics participated. 39% received specialist palliative care. Age was not associated with use of specialist palliative care; metastatic disease, global quality of life (‘need’) and the clinic where treatment was provided were. These findings suggest equitable use of specialist palliative care. However, a comprehensive account of equity must consider both use and quality of care. There were some suggestions that, within a resource-limited context, the quality of care may vary. Future equity research should prospectively consider variations in use and quality of specialist palliative care for different patient groups across all care settings, and from diagnosis to death.

The key aims of this thesis were to investigate the role of social support and marriage in adjustment and recovery in coronary heart disease (CHD). Declining death rates in CHD due to medical and surgical advances combined with increasing prevalence rates have contributed to a large and steadily growing population of chronic CHD patients, many of whom have suffered an acute cardiac event. In the context of this population, there is considerable need to determine factors that improve both adjustment and prognosis. Aspects of social support and marriage have been robustly associated with morbidity and mortality in CHD. Exploration of the potential psychological and biological pathways that link these factors forms the core of this thesis. Data from two separate studies are presented with the majority of analyses originating from data gathered in the Tracking Recovery after Acute Coronary Events (TRACE) study, a longitudinal study exploring diverse correlates of adjustment and recovery in 298 ACS patients. Associations between social support, marital satisfaction, distress, quality of life and HRV among ACS patients followed up from hospital admission to 12 months following discharge are presented. Data were also derived from a second study which explored psychobiological factors in a sample of 88 suspected coronary artery disease (CAD) patients and the analysis focused on marital influence on HRV. The overall thesis objective was to identify significant relationships between social and marital support, and various psychobiological factors that may contribute to adjustment and, ultimately, influence CHD prognosis.

More than 5.7 million people in India were estimated to have Human Inmmunodeficiency virus (HIV) infection in 2005. Large numbers of men, usually without their immediate families, migrate within the country in search of work. Data, mostly from sub-Saharan Africa, suggest migrant men form an important bridging group for HIV and sexually transmitted infection (STI) transmission from core groups into the general population. Migrant men have been targeted for HIV/STI prevention interventions although few data are available on the association of migration, sexual behaviour, HIV and STIs in India. This thesis describes a study on HIV, STIs and sexual risk behaviour among migrant men in India to help guide HIV/STI prevention interventions. Quantitative and qualitative methodologies were used. A cross-sectional survey with a stratified random sample of 846 participants was conducted between September 2005 and May 2006 amongst male migrant diamond and textile workers in Surat city, Gujarat. Data were collected on socio-demographic, lifestyle, sexual behaviour, knowledge, attitudes and perceptions, STI and HIV related signs and symptoms. Blood and urine samples were tested for HIV and STIs. Four focus groups with 37 migrant men were conducted between April and May 2007. To my knowledge, this study provides the first non-surveillance HIV and STI data in men in Gujarat. The study participants were found to form a group with an intermediate risk of HIV infection. Higher risk sexual behaviour, HIV and STI prevalence was low compared to other studies on migrant and mobile men. However the prevalence was generally higher than described in the Indian male general population. Condom use was low but more likely during riskier sexual behaviour. The quantitative results are interpreted within the context of the qualitative results and other literature. The implications for HIV/STI intervention work in Surat are discussed.

SCORE scale assesses the risk of fatal atherosclerotic cardiovascular disease (CVD), based on traditional risk factor levels. The high-risk SCORE version is recommended for Central & Eastern Europe/former Soviet Union (CEE/FSU). The aim of the thesis was to evaluate SCORE performance in CEE/FSU, using evidence from two large CEE/FSU studies. These studies – MONICA and HAPIEE – include cohorts from CEE/FSU countries which have relatively high but heterogeneous CVD mortality. MONICA subjects were followed for >=10 years from the mid-1980s. Ongoing HAPIEE follow-up (baseline 2002-2004) allowed preliminary assessment of SCORE performance in contemporary CEE/FSU settings. The present study included Czech, Polish-Warsaw, Polish-Tarnobrzeg, Lithuanian, and Russian MONICA samples (n=15,027), plus Czech, Polish, and Russian HAPIEE samples (n=20,517). Predicted 10-year CVD mortality was calculated with high-risk SCORE; observed mortality data came from local registers. While SCORE calibration was good in most MONICA samples (predicted to observed (P/O) mortality ratios approached 1.0), mortality risk was under-estimated in Russian men and women. In Cox regression analysis, SCORE >=5% significantly predicted 10-year CVD mortality: hazard ratios (HR) ranged from 1.7 to 6.3. The shorter HAPIEE follow-up meant that P/O ratios exceeded 1.0. These ratios were 2-3 times higher in Czech and Polish vs. Russian participants. Estimates of 10-year HAPIEE mortality confirmed this gap between Czech and Polish vs. Russian samples. SCORE significantly predicted CVD mortality in each HAPIEE sample (HR 2.6-10.5). Values of Harrell’s C-statistic, a summary discrimination measure, reached 0.6-0.7 in MONICA and HAPIEE. Adding socioeconomic parameters or alcohol consumption characteristics to the SCORE model failed to improve its predictive performance. High-risk SCORE discrimination was satisfactory in most MONICA and HAPIEE samples, despite risk under-estimation in Russian MONICA. HAPIEE data suggest that in contemporary Czech and Polish populations, high-risk SCORE might over-estimate CVD risk. SCORE extension by additional predictors did not improve its performance.