Acute undifferentiated febrile illnesses in the tropics and sub-tropics are caused by a wide range of infectious diseases that often have indistinguishable clinical features. In developing countries there may also be insufficient microbiology facilities to identify these infections leading to missed diagnoses, inefficient use of healthcare resources, over-use of empirical treatments, a lack of information on antimicrobial resistance and inaccurate epidemiological data for guiding prevention strategies. These problems occur in Sri Lanka, but a prospective, systematic, representative and comprehensive study of febrile illnesses has never been performed. The Ragama Fever Study was performed at a major hospital in western Sri Lanka that served both urban and rural areas. Its aims were to identify the causes of febrile illnesses in a large sample of patients admitted to the hospital over a 1-year period, develop clinical prediction rules that could distinguish between the most common infectious diseases and assist in the evaluation of rapid (point-of-care) diagnostic tests that were appropriate to this setting. 617 (86.7%) of 711 febrile patients admitted to a quarter of the hospital medical wards were recruited. 56.4% had confirmed infections with organisms identified including dengue (22.2%), chikungunya (16.7%), leptospirosis (5.2%), various bacteraemias (4.2%), Q fever (2.9%), rickettsial infections (2.3%), tuberculosis (1.1%) and urinary tract infections (0.8%). 7.6% had confirmed infections with no organisms identified including cellulitis (2.4%), respiratory tract infections with radiographic changes (2.1%) and pulmonary tuberculosis with radiographic changes (1.6%). 4.1% had confirmed non-infectious diseases and 37.2% had unconfirmed diseases including “viral fever” (13.3%), undifferentiated fever (7.8%), respiratory tract infections (6.8%), urinary tract infections (3.4%), leptospirosis (2.8%) and gastroenteritis (1.0%). Clinical prediction rules for identifying dengue fever and chikungunya were developed using imputation, multiple logistic regression, scoring algorithms and receiver operating characteristic (ROC) curve analysis. The dengue fever rule had sensitivity = 49.6%, specificity = 93.9%, positive predictive value (PPV) = 70.8% and negative predictive value (NPV) = 86.1%. The chikungunya rule had sensitivity = 35.0%, specificity = 95.0%, PPV = 60.0% and NPV = 87.1%. ROC curve analysis could not identify any probability cut-offs that would produce clinical prediction rules with acceptable combinations of both sensitivity and specificity. A commercial (Panbio) rapid serology test for dengue fever showed sensitivity = 43.4%, specificity = 88.8%, PPV = 54.6% and NPV = 83.5% on samples from admission and significantly better diagnostic performance on follow-up. When repeated in conjunction with a PanBio rapid NS1 antigen detection test, the diagnostic performance improved with sensitivity = 89.9%, specificity = 75.0%, PPV = 69.0% and NPV = 92.3% on admission. This study confirmed the wide range of infections that present as febrile illnesses in Sri Lanka and showed the limitations of clinical prediction rules and rapid diagnostic tests in identifying these on admission. I hope that it will prove a foundation for further work on these important topics.

There is an increased incidence of young people with stroke (age ≤45years) in Human immunodeficiency virus (HIV) endemic countries; this has been largely attributed to hypertension. However, hospital based surveys in countries like Malawi and South Africa have shown that the prevalence of hypertension in these young people is lower than expected, but HIV infection is substantially higher, implicating HIV as a risk factor. For many years a link between HIV and stroke has been postulated, but the relationship is uncertain. Whilst HIV may be a risk factor for stroke directly through mechanisms linked with HIV-associated vasculopathy, or indirectly through opportunistic infections, the drugs that treat HIV infection may also increase the risk of stroke because of their metabolic effects. Many studies, almost all retrospective, have failed to separate the direct effect of HIV infection from the indirect effects, including combined antiretroviral therapy, on cerebrovascular risk. HIV infection increases the risk of stroke mimics such as intracranial toxoplasma infection. The Recognition of Stroke in the Emergency Room (ROSIER) score is commonly used to screen for a stroke and triage patients for computer tomography (CT) of the brain. However, the accuracy of the ROSIER score and CT brain to reliably differentiate a stroke diagnosis from those with a stroke mimic in people with HIV infection is uncertain. I found that the ROSIER score and CT brain imaging had poor diagnostic accuracy in an HIV positive population. Therefore, in my thesis, every patient with an acute neurological symptom was fully assessed for a stroke as part of the screening process and confirmation was by magnetic resonance brain imaging. I subsequently investigated the risk factors and aetiology of stroke through a prospective case-control study in an HIV endemic country. Through this work, I showed that HIV infection is associated with cerebrovascular disease. Although hypertension was the leading risk factor in the population overall, HIV infection and its treatment was the second most important, and the most important in younger patients. Unexpectedly, I found that starting combined antiretroviral therapy in a subgroup of people living with HIV infection independently increased the risk of stroke. In this cohort, ischaemic stroke was the predominant stroke type and opportunistic infections only accounted for less than a third of these cases. The heterogeneity of HIV stroke with respect to risk factors for stroke, the degree of immunosuppression and HIV activity, and prior or current opportunistic infection has made it difficult to generalise epidemiological findings in some studies to populations at large. My study, to some extent unravels some of this ambiguity. I speculate that HIV related strokes evolves through the introduction of cART and then transitions into an aging population, accelerating atherosclerotic stroke and potentially contributing to an anticipated stroke epidemic in countries like Malawi.

Contemporary health policy and funding bodies are placing increasing emphasis on patient and public involvement (PPI) in healthcare and health research, advocating PPI in all stages of the research process. Currently, however, there is limited empirical evidence critiquing different approaches to PPI or exploring its associated benefits and challenges. Without this information researchers and patient/public representatives cannot make informed decisions about best practice. The principal aim of this thesis was to generate a detailed understanding of the implementation of PPI in health research. To accomplish this broad aim, I focused on a specific health condition, epilepsy, and the research structures underlying health research in the UK, namely, research networks. I achieved this using a multi-sited, ethnographic approach, incorporating multiple qualitative data collection methods, including 47 interviews, 35 observations, fieldnotes and document analysis. My in-depth thematic analysis of the data found that PPI is conceptualised in terms of ‘meaningful’ and ‘tokenistic’ involvement by those engaged in the process, rather than how it is depicted in the current models of involvement. Having first explored these terms I identified five components that can help to ensure that PPI is meaningful and not tokenistic. Having compared and contrasted multiple approaches to PPI I conclude that there is not one single ‘best approach’ for implementing PPI. Rather, to achieve high ‘quality’ PPI there is a need to incorporate seven methodological factors that overarch approaches and ensure that there is an alignment of approach and purpose. Both the professionals and the patient/ public representatives within my research appeared to be highly aware of the moral and political motivations of PPI, but were primarily motivated by pragmatic or consequentialist reasons. Professionals were motivated almost exclusively by the goal of improving the applicability or relevance of the research. This goal was important for representatives too but they were also motivated by a range of personal reasons, including the wish to feel they were making a difference; the opportunity to learn about epilepsy and epilepsy research; and the opportunity to interact with others. The perceived benefits of PPI were also identified and discussed in depth, and appeared to be largely congruent with those reported in the literature. However, my work has identified some challenges and barriers around PPI that have not previously been explored including: adverse emotional effects; organisational practicalities; concerns about ‘representativeness’ and ‘tokenism’; the ‘blurring’ of roles and the erosion of patient-clinician boundaries. I conclude by recommending that there should be an increased focus on appropriate, ‘meaningful’, involvement rather than endeavouring to implement PPI in all stages of the research process, as currently advocated in policy documents. The insights into the challenges of PPI that my work has provided will allow them to be addressed from the outset, improving the PPI experience and consequently the likelihood of PPI being successfully implemented.

Maternal psychosocial well-being (MPW) is a comprehensive concept that covers the psychological (e.g., depression, distress, anxiety, coping, mental health,) and social (e.g., family and community support, empowerment, relationships, culture) aspects of motherhood. High rates of poor maternal mental health, with maternal depression the most prevalent condition, have been reported in the low and middle income countries, including Pakistan. Though evidence based interventions exist to address maternal depression, these have not been translated into policy because of various implementation barriers. Integration of these interventions into existing maternal and child health (MCH) programmes has been suggested as a strategy to provide accessible care to mothers. In the current study we developed and integrated a cognitive behavioural therapy–based MPW intervention (the 5 pillars approach) into a child nutrition and development program. Following qualitative research with community health workers (CHWs) and families, CHWs were trained in (1) empathic listening, (2) family engagement, (3) guided discovery using pictures, (4) behavioural activation, and (5) problem solving. A qualitative feasibility study in one area demonstrated that CHWs were able to apply these skills effectively to their work, and the approach was found to be useful by CHWs and mothers. This work provides vital information on the lessons learnt in the implementation of a maternal psychosocial wellbeing intervention for universal use. The facilitating factors included mothers being the central focus of the intervention, utilizing existing local CHWs whom the mothers trust, simple training and regular supervision, and an approach that facilitates, and does not add, to the CHWs’ work.

Background: Saudi Arabia is one of the Gulf Cooperative Council (GCC) countries which have common characteristics such as high-income governments, dominant expatriate populations, and under-developed healthcare systems, including healthcare financing. The dominance of the expatriate working population raises the question of how to find a mechanism that ensures expatriates have appropriate access to medical care whilst the employers bear the responsibility of healthcare expenses. Saudi Arabia is one of the few GCC countries to have reformed its private healthcare system through a Compulsory Employment-Based Health Insurance (CEBHI). The CEBHI was designed to mitigate some of the disadvantages of the Employment Sponsored Insurance scheme previously implemented in the United States; and this is the first study to investigate the impact of this form of private health insurance on access to medical care, in a country such as Saudi Arabia. The main aim of the study was to explore the influence of health insurance on access to medical care, in order to assist the Saudi Government in their deliberations about making CEBHI compulsory for all people (citizens and expatriates) within Saudi Arabia. This aim was investigated through the following objectives: 1) to review health financing in Saudi Arabia and compare it with other GCC countries and elsewhere in the world; 2) to compare the access to medical care of insured and uninsured expatriates in Saudi Arabia; 3) to develop a framework for understanding the complex relationship of health insurance and access to healthcare, 4) to make policy-relevant recommendations regarding the key question as to whether compulsory health insurance in Saudi Arabia should be expanded. Methods: Two methods were used to tackle the study objectives. Firstly, a framework for country-level analysis of healthcare financing arrangements was used to compare and analyse the national expenditure on healthcare within the GCC and other developing/developed countries. Secondly, a logistic regression analysis of data from a cross-sectional survey was undertaken to investigate the impact of health insurance on access to medical care, considering the main workplace and personal characteristics of the expatriates. Three access measures, access to usual medical care (Access 1), inability to access medical care (Access 2), and utilization of medical care (Access 3), were used to evaluate access to medical care for the expatriate population. Prior to the implementation of CEBHI the expatriate population accessed medical care through a variety of different avenues. These modes of access were used as classification of the expatriate population into four groups. Two of these groups were insured but had a different Previous Method of Paying for Healthcare (PMPHC) (Group B=insured, not paid, and Group D=insured and paid) and two groups were not insured but also had different PMPHC (Group A=not insured, not paid and Group C=not insured, but paid). A multistage stratified cluster sampling was used, and a sample selected from each sector and company size proportionately. The total sample size was 3,278. A simple conceptual framework for studying access to medical care was developed to guide the multi-variate regression techniques, and greatly assisted interpretation of the results. Results: The GCC characteristics impact on the healthcare financing strategies of GCC countries in three ways. First, GCC governments provide the majority share of the health budget, similar to high-income countries. Second, GCC countries use different strategies to control expatriates costs, but some of these strategies lead to increased out-of-pocket expenses, which is a characteristic of low-income countries. Third, health care financing systems in GCC countries are still being developed as they finance most of their public services, including health care services, with revenue from natural resources (i.e. oil or gas). Additionally, some of their health care indicators are identifiable with those from below upper-middle income countries. In addition, after CEBHI, private expenditure did not change but remained around 22.4%, which does not reflect the huge number of people having access to medical care though private sector only. However, there was a shift in the means of private sector expenditure from Out Of Pocket payments to private insurance expenditure. OOP expenditure decreased from 32.3% in 2006 to 28.4% in 2008, and private insurance expenditure increased as a percentage of private sector expenditure from 26.2% in 2006 to 36.7% in 2008. Analysis of the data from the survey demonstrates that health insurance is strongly associated with access to medical care, as measured by the three different access measures). Compared to uninsured workers, being enrolled in CEBHI increased the possibility of an expatriate’s access to usual medical care and utilisation of medical care by more than 10 (8.709-12.299, 95%), and 2.3 (1.946-2.750, 95%) respectively. However, the influence of PMPHC is greater than the influence of insurance alone on reducing the inability to access medical care (health insurance reduced the inability to access medical services by 42% (0.515-0.995, 95%), whereas PMPHC reduced the inability to access medical services by more than 65.% (0.273-0.436, 95%)).Therefore, the impact of health insurance on access to medical care is much greater for those expatriates previously having had healthcare costs met by their employer, than for those who had not. These impacts remained, when the odds ratios were adjusted for both workplace and personal characteristics. Conclusion: CEBHI has a clear positive impact on reducing out of pocket payments and increasing private insurance expenditure. However, overall, private healthcare expenditure has increased insignificantly. This indicates that the main impact of CEBHI on private expenditure, is the change in the mode of payment from out of pocket payments to private insurance expenditure. However, the actual impact on private sector expenditure is still minor. Access to medical care is influenced by health insurance. In addition, it is also influenced by PMPHC as a contributory role to play in the influence of health insurance on access to medical care. Workplace and personal characteristics play a small part in mediating the influence of health insurance on access to medical care. A framework was developed for understanding the complex relationship of health insurance and access to healthcare, which will be useful for further investigations regarding the influence of health insurance on access to medical care. Both long and short-term recommendations are proposed for increasing the expatriate population’s access to medical care, whilst reducing the burden on healthcare financing.

This thesis documents the development and evaluation of an mHealth intervention for sexual and reproductive health (SRH) referral from drugstores to health facilities in Mwanza Tanzania. SRH is an important factor for human development. Over the last 2 decades, provision and accessibility to SRH services has benefited from international and national health promotion interventions. In developing countries, use of close to community providers (CTC providers), such as village health workers, has been a key component of health promotion. This has been especially true in Tanzania where up to 70% of formal primary health care facilities lack health staff. CTC providers have been promoted through international initiatives such as the World Health Organization’s (WHO) task-shifting initiative. WHO’s recognition of CTC providers ranges from cadres at the grassroots level such as village health workers and drugstores to formal auxiliary providers based in health facilities, such as medical aides and nurses. In Tanzania, drugstores provide a range of SRH services ranging from simple advice on how to use a condom to complex prescriptions of antibiotics for STI treatment. Evidence has shown that drugstores – though more likely to have health-related training than any other informal CTC providers – lack skills necessary for provision of SRH services. This may contribute to poor SRH outcomes, such as increase in prevalence of sexually transmitted infections (STIs)/human immunodeficiency virus (HIV) and antibiotic resistance. Accessing SRH services at the formal health facility level is key to improving these outcomes. To create SRH service linkages and integration between drugstores and health facilities in Mwanza, Tanzania, an intervention that pioneered an mHealth SRH referral from drugstores to health facilities was designed and implemented for 18 months from September 2012 to February 2014. Referral between these two SRH providers using mHealth tools had not been attempted before. The intervention provided an electronic platform accessible to 52 drugstores and 18 health facilities in two districts of Mwanza region. Through a toll-free number and password, drugstores referred patients with SRH conditions to health facilities using the text-messaging feature on their mobile phones. From the platform’s in-built data collection tool, SRH uptake data demonstrated that 38% of patients referred from drugstores accessed HIV, STIs, family planning and maternal health services at the health facility level. A follow-up randomised household survey found that 72% of the participants would accept such type of referral in future, and among those who had ever visited drugstores for SRH services, 15% had heard about the intervention. At the end of the intervention, drugstores and health facilities confirmed that it was beneficial to their SRH service provision and that they would like to continue implementing it. In conclusion, this text messaging intervention pioneered community referral from drugstores to health facilities for SRH treatment by using mobile phones which appeared to be acceptable and effective in Mwanza Tanzania. With the growing use of mobile phones in Africa and the need to provide SRH services beyond the Millennium Development Goals (MDGs) era, mobile phone-based community referral through CTC providers, such as drugstores, could make an important contribution to achieving Universal Health Coverage targets.

This thesis examines issues related to cervical cancer epidemiology and prevention through screening, with the aim of informing policy regarding setting up an organised cervical screening programme in Hong Kong. There are five studies described here. The first, a case control study, indicated that screening is effective in preventing invasive cervical cancer among Chinese women. In addition, the main risk factors identified in other studies, were confirmed as risk factors in this population. Secondly, a cross-sectional study examined the pattern of cervical screening in Hong Kong. The screening system at that time achieved poor coverage, was inefficient, inequitable and potentially harmful. Thirdly, a cross-sectional study of practitioners showed the diversity in provision of services and the lack of consensus among practitioners in the management of abnormal smears. Fourthly, the use of an industrial quality management technique in monitoring quality, using inadequate smear rates as an indicator is assessed. It demonstrated that this is an efficient and useful method that can be applied to monitoring a screening programme. The last study was a randomised controlled trial showing that when women are given balanced information on cervical screening, with information on both the harms and benefits, relatively fewer chose to attend. The implications of these studies in relation to setting up a screening programme are discussed.

The increasing prevalence of obesity has been accompanied by an increase in the number of individuals at the extreme end of the obesity spectrum. The detrimental impacts of extreme obesity, defined as BMI ≥ 40.0kg/m2, on affected individuals' physical and psychological health have not been fully established. Furthermore, it remains unclear whether medical and behavioural interventions are effective at facilitating weight loss for individuals with extreme obesity. The efficacy of two treatment pathways within the Heart of England NHS Foundation Trust Specialist Weight Management Service were examined, with both demonstrated to facilitate clinically and statistically significant weight loss. A detailed profile of the characteristics of individuals entering the service highlighted the substantial physical and psychological co-morbidity associated with extreme obesity, revealing widespread impairment in quality of life and mental health. A systematic review of primary research examining the efficacy of medical and behavioural weight management interventions within lesser-researched extreme obese populations demonstrated the value of medically-supported programmes and also revealed the limited body of good quality research. This thesis has enhanced current understanding of extreme obesity, and recommendations generated from this work have been made in order to improve primary research examining weight management interventions and service provision for affected individuals.

Introduction: Smoking is a major risk factor for development of serious disease and smoking cessation greatly reduces this risk. The association between smoking, smoking cessation and mental health however, is less clear-cut, therefore this thesis aimed to further investigate this association. Methods: The first part of the thesis reports a systematic review and meta-analysis of longitudinal studies to determine the difference in change in mental health between quitters and continuing smokers. The second part of the thesis reports three prospective analyses of individual level-patient data from five trials for smoking reduction treatment. The first analysis examined the association between cessation and change in mental health using propensity score matching (PSM). The second analysis examined the association between cessation and risk of psychiatric disorder using PSM. The final analysis examined the association between change in mental health after quitting and odds of relapse. Results and interpretations: Cessation was associated with improvements in mental health compared with continuing smoking; there was no association between cessation and risk of psychiatric disorder, and no association between change in mental health after cessation and future relapse. Results support the misattribution hypothesis, and have implications for future research, smoking cessation treatment and public health policy.

Introduction: Obesity has been recognized as a major public health problem worldwide that requires preventive action. Prevention is best targeted at children, there is lack of quantitative and qualitative research on obesity prevention in children and most have been conducted in western countries. The aim of this study is to describe the prevalence of obesity among 6-7 years old school children, investigate contributing factors and identify potential components for an intervention programme to prevent obesity amongst children in the State of Qatar. Methods The study consisted of two distinct parts: cross sectional survey and focus groups with a range of stakeholders. Results: Whilst there was a high prevalence of overweight and obesity 16%, underweight was also prevalent (21.7%). The results of the focus group discussions indicate that causes of childhood obesity are multifactorial, and a multi-sector approach to prevention would be acceptable. Some of the important barriers that need to be considered in developing interventions were highlighted. Conclusion: In Qatar there is coexistence of underweight and obesity in primary school children. Qualitative results suggest that a multi-sector approach to prevention would be acceptable toward dietary and physical activity, and suggested potential components for an intervention programme in preventing obesity amongst children in the State of Qatar.