Despite structured enforcement of food safety requirements known to prevent foodborne disease outbreaks, catering businesses continue to be the most common setting for outbreaks in England and Wales. Limited published evidence suggests that the way catering businesses are managed and operated may contribute to food safety control failures which in turn can result in an outbreak. The purpose of this study, funded by the Food Standards Agency1,2 was to identify the underlying management factors that may contribute to or prevent outbreaks in the catering industry. A matched case control study compared the management and operational practices of 148 catering businesses associated with foodborne outbreaks with 148 control catering businesses. High response rates were achieved: case businesses 90%, and control businesses 93%. To minimise false inferences from chance associations analysis followed a predefined hypothesised causal pathway. Hazard analysis critical control point systems and formal food hygiene training were found not to be protective and food hygiene inspection scores were not useful in predicting which businesses were likely to be associated with outbreaks. Larger small medium sized (SME) businesses were more likely to be associated with outbreaks compared to micro SME businesses. Operational and management practices did not differ significantly between case and control businesses when adjusted for SME size. SME size was not explained by other staff employment and management variables. However, businesses associated with Salmonella outbreaks were significantly more likely to use regional egg suppliers, the only significantly independent operational practice associated with Salmonella outbreaks. Regional egg suppliers were also more likely to supply businesses associated with outbreaks that were attributed to food vehicles containing eggs. Businesses associated with egg outbreaks were less likely to use eggs produced under an approved quality assurance scheme suggesting that the underlying risk associated with using regional suppliers may relate to the use of contaminated eggs.

Certain mould genera were found to predominate and built form was observed to be associated with mould concentration. PIV reduced the indoor humidity and it reduced the likelihood of mould reoccurring in the intervention homes.

Background Public health interventions need to both improve health and reduce health inequalities, whilst using limited health care resources efficiently. Well-established ethnic differences in cardiovascular disease (CVD) raise the possibility that CVD prevention policies may not work equally well across ethnic groups. The aim of this thesis was to explore whether there are ethnic differences in the potential impact of two CVD prevention policy choices – the choice between mass and targeted screening for high cardiovascular risk, including the use of area deprivation measures to target screening, and the choice between population and high-risk approaches. Methods Cross-sectional data from the Health Survey for England 2003 and 2004 were used. Three sets of analyses were carried out – first, calculation of ethnic differences in the utility of area deprivation measures to identify individual socioeconomic deprivation; second, investigation of ethnic differences in the cost-effectiveness of mass and targeted screening for high cardiovascular risk; third, analysis of ethnic differences in the potential impact of population and high-risk approaches to CVD prevention. Results Area deprivation measures worked relatively effectively and efficiently at identifying individual socioeconomic deprivation in ethnic minority groups compared to the white group. In ethnic groups at high risk of CVD, cardiovascular risk screening programmes were a relatively cost-effective option, screening programmes targeted at deprived areas were particularly cost-effective, and population approaches were found to be an effective and equitable way of preventing CVD despite potential underestimation of their impact. Discussion This thesis found that ethnic minority groups in the UK are unlikely to be systematically disadvantaged by a range of CVD prevention policies that have been proposed, or implemented, for the general population. Additional CVD prevention policies, in particular those based on the population approach, should be implemented.

Older people are often portrayed as a disadvantaged and silent group in society, whose views have been largely ignored. Demographic studies suggest the number of people over 75 years of age, as a percentageo f the population in coming years is likely to substantially increase, which will place greater demands on healthcare services. In the last two decades, health policy has focused on delivering high quality services based on individuals' needs, with a greater emphasis placed on individuals being involved in decisions about their care. This policy direction has facilitated a change in power relationships between patients and professionals and will require providers of healthcare to focus on delivering 'patient-centred' care at times and places that meet individual's needs and expectations. The aim of this researchw as to evaluatet he current medical model provision of NHS podiatry with the biopsychosocial model which claims to provide 'holistic', patient-centredc are. An important aim of this research was to provide a greater and more informed understandingo f what older people communicate about their 'lived' experiences, the significance of those experiences on care-seeking and their expectations of appropriate podiatry care. The research was undertaken with older people living in east Gloucestershire, who were 75 years old or over, and had requested NHS podiatry. The study was underpinned by a qualitative methodology, strengthened by a desire to change current clinical practice and inform health policy. The research methodology included involvement of participants in an innovative reminiscence technique, and as a consequence the 'podiatry patient career' was constructed. The texts generated from the participants were examined using an interpretative phenomenological analysis to ensure a 'person-centred' focus because it was imperative to hear the voices of the 'Participants' and not just the medical model 'patients' narrative. A portrait was revealed of older people who were conscious of their position in the life course and their own mortality, together with the effect this had on how they conducted their lives. The participants' raised consciousness of their 'self' affected their expectations, feelings, and interaction with others. For many of the participants there appeared a vicious circle of impending frailty that led to a diminishing circle of contacts which had an effect on their wider social activities and relationships. At this stage, participants perceived a resolution of their foot-care needs to be of great value and importance in maintaining their well-being which, assisted by the podiatrist, resulted in a handing over of the responsibility for their care. The conclusion is that neither model delivers 'person-centred' care to meet participants' expectations and foot-care needs. A new model is presented where differing and changing priorities, at different times of the participant's lived world will be relevant to meet their expectations and needs. The research concluded that the requirement for podiatry care can be taken as an early indicator of failing independence. The importance of the participant podiatrist relationship was also identified as cental to the delivery of 'person-centred' podiatry. The research findings depict older people who want to be involved in their care rather thm being 'a burden to the state. Recognition is also given to the changing nature of caring relationships in the next decade, and how NHS podiatry services will have to profoundly transform if they are to deliver a holistic, person-centred service in the future.

Progression of chronic kidney disease (CKD) in diabetes can be slowed by strict blood pressure and blood sugar control, prescription of medicines that modify the renin-angiotensin system and lifestyle changes, such as smoking cessation. Because of the large numbers of people with diabetes whose condition progresses (and eventually require dialysis or transplantation), it is possible that the management of their diabetes remains sub-optimal. The overall purpose of this thesis is to develop, test and evaluate an educational package to help people self-manage their risk of CKD progression. This thesis contains a case study, a critical review of literature, the main research study and an artefact (the self-management package). The case study developed from a three-month observation period in six general practitioner (GP) practices. The literature review evaluates the effect of patient education and selfmanagement on diabetes control and outcomes. The research project develops and evaluates the self-management package. Development of the package was informed by the findings of the case study and literature review, and also through interviews with 15 people at high risk of CKD progression. The resulting self-management package comprises written information; a 20-minute DVD filmed with patients; a fridge magnet (with key messages); a monitoring diary; and a blood pressure machine if required. Testing of the package was undertaken in the same six practices mentioned above, with one additional control practice. Patients with Type 1 or Type 2 diabetes at risk of kidney disease were included. Data on renal function (serum creatinine, eGFR and proteinuria), systolic and diastolic blood pressure (BP), glycated haemoglobin (HbA1c), body mass index (BMI) and smoking status were collected at six time points, before, during and after the intervention. Outcomes in patients in the participating surgeries who did receive a pack (n=116) were compared with patients in the control group (n=61). At time point 4 mean systolic BP in the intervention group was 129.2 ± 19.2 mmHg vs. 134.6 ± 15.0 mmHg in the control group (p=0.057). At time point 5 there was mild significance (p=0.053) in mean diastolic BP. At the end of the study (time point 6) the intervention group had a mean systolic BP of 132.1 ± 14.2 mmHg vs. 136.2 ± 16.4 mmHg and mean diastolic BP of 74.9 ± 8.5 mmHg vs. 77.6 ± 9.1mmHg in the control group (p=ns). There were no significant differences in HbA1c and BMI at any time period. The results of the research project have shown the importance of self-management techniques to control blood pressure, which in turn can slow the rate of CKD progression and reduce cardio-vascular risk. Following evaluation by patients, the self-management package has been amended and strategies for local and national dissemination of the package have been put in place.

Eight schools, located in Scotland were involved in this study. Four different types of data were collected in the following order: first, 183 semi-structured interviews with a range of staff and pupils across the schools, the interviews covering questions relating to health education, promotion and ethos including quality of relationships; second, a school audit of health education and health promotion in all schools; third, Researcher observations for all schools; and, fourth, questionnaire data collected from 446 pupils across the schools. The Health Promoting School (HPS) concept is based on the belief that schools have the potential to influence their students' health and health behaviour through the school's social organisation, culture and physical environment, as well as through the formal curriculum. To date, there is little empirical evidence to test the effectiveness of the HPS, at least evidence that adjusts for known predictors of the behaviours, a standard set by the more advanced area of 'school effects' research on educational outcomes. This thesis will add to that evidence base. The aims of this study have three main components: first, to quantify 'school effects' on a range of pupils' health behaviours comprising current smoking, weekly alcohol drinking, ever tried drugs and physical activity: second, to assess the extent to which the health behaviour profile of schools are related to health promotion activity as evidenced by an audit: third, to select and analyse qualitative data from three case study schools. The purpose of the second and third aims is to investigate the extent to which school processes are associated with 'school effects' on pupils' health behaviours, triangulating data from different methodologies. The questionnaire data indicated that a strong school effect existed for smoking and drinking to a lesser degree, but not for drugs or physical activity. This addressed the first aim of this study and, in addition, provided the means by which three case study schools were selected. These were the two schools with the lowest (added value) and highest odds (lost value) for smoking after adjustment for known predictors of the health behaviours. Plus, a third school which was significantly different from the school with lowest smoking and located in the same town, as this eased interpretation of the results. Relating to the second aim, the pattern of the 'school effects' on smoking were triangulated with data from three different data sources. First, in the audit, higher levels of action on health education and health promotion were associated with lower (adjusted) rates of smoking. Second, the three case study schools were used to explore the Researcher's observations; the school with added value for smoking was rated more highly than the two with lost value. Regarding the third aim, based on qualitative data from a range of staff and pupils, the analysis showed that the school with added value had progressed furthest towards functioning as a whole school, performing best across all the areas explored. These findings theoretically triangulated with the schools low smoking rates according to the HPS concept. These results confirm the importance of school processes on students' health behaviour, particularly smoking, and support a school-wide or "Health Promoting School" approach to improving health behaviours.

Malaria remains the most important parasite-related public health problem globally, with the majority of burden occurring in sub Saharan Africa. Increased political and financial support has resulted in rapid scale up of malaria prevention measures, so that disease burden has been substantially reduced in many African countries. However, behavioural change by malaria vector populations, so that a greater proportion of human exposure to bites occurs outdoors, threatens to undermine the impact of malaria control with existing front line interventions such as insecticide treated nets (ITNs) and indoors residual spraying (IRS) because both act indoors. Also, progress towards lower transmission levels poses substantive entomological monitoring challenges because most standard methods fail to detect low levels of vector density and malaria transmission. The overall goal of this study was to enhance understanding of the potential and limitations of ITNs for reducing malaria transmission by outdoor biting mosquitoes, and to develop a safe, sensitive, practical and effective malaria vector surveillance tool that enables sustained entomologic monitoring of intervention impact. An existing mathematical model was adapted to examine the possibility that ITNs can achieve community suppression of malaria transmission exposure, even when mosquitoes avoid them by feeding on people while they are outdoors. Simulations indicated that ITNs may provide useful levels of community suppression of malaria transmission, even when outdoor biting rates exceed indoor biting rates and slightly more than half of bites occurred at times and places when using ITNs is not feasible. This suggests that ITNs should not be deprioritized as a malaria control tool simply because local vector species prefer to feed outdoors. Nevertheless, complementary interventions that target outdoor- and early-biting mosquitoes should be prioritized, especially for going beyond malaria control to achieve elimination. Cross over and Latin Squares experimental designs were used to compare the sensitivity of multiple trapping techniques for catching malaria vectors, under conditions of both high and low mosquito density, in rural Kilombero and urban Dar es Salaam, respectively. A new tent style trapping device called the Ifakara Tent Trap was successfully developed and proved to be safe and more efficacious than any other commonly used alternative to human landing catch for catching Anopheles gambiae s.l. in the low transmission setting of urban Dar es Salaam. Its sampling efficiency appeared to be independent of vector density in a rural setting with high mosquito abundance but increased as mosquito densities decreased in an urban area of low mosquito density where it exceeded that of HLC at lowest densities. This density- dependence of the trap implies that this tool may have particular potential for monitoring malaria in low transmission settings. It was also demonstrated to be effective when used by unsupervised community members under programmatic conditions and it is currently the only technique used for routine adult mosquito surveillance by the Urban Malaria Control Programme of Dar es Salaam. However, it cannot be used to determine how bites upon humans are distributed between indoor and outdoor exposure components.

The study aimed at optimally assessing the impact of indoor residual spraying (IRS) and insecticide treated nets (ITNs) on vector species abundance, their infectivity and resistance status, and Plasmodium falciparum prevalence, malaria deaths and case fatality rates in the human population. Malaria prevalence surveys were conducted and routine surveillance data was retrospectively analyzed. The average P. falciparum prevalence in children between the ages of 1 and 14 years was below 10% across the study period. The intervention effect was more pronounced in IRS areas than in ITNs localities but with an incremental protective effect of their combined use. Age-specific comparison showed better intervention effect on children below 5 years than older children 5 to 14 years old. While the average number of deaths and case fatality rates in children under the age of five plunged precipitately, the reductions were more significant in IRS districts than in ITNs districts. Results indicate the need for supplementing parasite prevalence survey data with routine surveillance data in low transmission intensity areas and demonstrate the significance of evidence-based age-specific deployment of interventions. To monitor vector species abundance and infectivity, mosquitoes were collected daily using exit window traps. The three major vectors; An. gambiae s.s, An. arabiensis and An. funestus s.s, and three potential vectors of malaria, An. nili, An. rivulorum and An. funestus-like species were identified. Overall, the biggest impact of IRS and ITNs was on An. gambiae s.s, and An. funestus abundance. No An. gambiae s.s was collected in IRS localities, thus validating the fact that An. gambiae s.s and An. funestus are characteristically more amenable to control by IRS and ITNs than An. arabiensis. The transmission potential for all malaria vectors, as expressed by the calculated transmission index, was zero as none of the trapped mosquitoes tested positive for P. falciparum sporozoites. The identification of An. nili, An. rivulorum and An. funestus-like necessitate further research to determine their role in malaria transmission in the country. The low numbers of mosquitoes collected also indicate a compromise in the efficiency of exit window traps in low transmission settings, suggesting the need for their replacement with a more robust collection tool like the CDC light trap. While the persistence of An. arabiensis suggests the presence of resistance segregating in this population or, that this outdoor species is not in contact with IRS or ITNs, it could as well imply that it’s the one species perpetuating malaria transmission in these meso-to hypo- endemic areas. To determine the impact of interventions on insecticide resistance status of malaria vectors, susceptibility assays using the WHO standard protocol were conducted in 17 localities. High levels of resistance were detected in both An, gambiae s.l and An, funestus s.l to pyrethroids and DDT but with 100% susceptibility to malathion and bendiocarb. The level of resistance was significantly higher in IRS areas than in ITN areas. These findings indicate that resistance has been selected for following extensive vector control. Resistance to both DDT and deltamethrin in IRS localities and ITN areas with intense cotton growing was detected suggesting selection due to either historical use of DDT, gene flow or cross-resistance. All An. gambiae s.s were molecular s-forms and only the west (leu-phe) kdr was detected. Complete susceptibility to the organophosphates and carbamates provides a possibility to switch to these alternative insecticide classes for IRS. The detected increases in the malaria prevalence in localities with high insecticide resistance levels indicate vector control failure. These findings point to the need for information on underlying biochemical and molecular resistance mechanisms to make possible the design of an effective resistance management strategy, and for the assessment of the impact of resistance on interventions. The results indicate that the impact of malaria control can be optimally assessed by using a combination of epidemiological (routine surveillance and prevalence data) and entomological indicators, in the context of a malaria decision support system, to enhance policy formulation for objective implementation of malaria control interventions and rational use of available resources.

Introduction: Individuals practicing unprotected receptive anal intercourse are at particularly high risk of HIV infection. Men who have sex with men in the developed and developing world continue to have disproportionate and increasing levels of HIV infection. The last few years have seen important progress in demonstrating the efficacy of oral pre-exposure prophylaxis, vaginal microbicides, and treatment as prevention but there has also been significant progress in the development of rectal microbicides. The purpose of this thesis is to summarise the status of rectal microbicide research, to identify opportunities, challenges, and future directions in this important field of HIV prevention, and to describe the results of a recently completed Phase 1 rectal microbicide study (MTN-007). Methods: MTN-007, a Phase 1, randomised, partially blinded, rectal safety study was undertaken to determine whether a reduced glycerin formulation of tenofovir 1% gel was safe and acceptable to men and women with a history of practicing receptive anal intercourse. The study was conducted at three clinical trial sites in the United States (Pittsburgh, Pennsylvania; Boston, Massachusetts; and Birmingham, Alabama). Study participants were randomized to one of three gel arms (tenofovir gel, a hydroxyethyl cellulose placebo gel, or a 2% Nonoxynol gel) or a no treatment arm and received a total of eight rectal daily doses of the study product. In addition to collecting conventional clinical safety and acceptability data, the study also included extensive mucosal safety assays to determine whether product administration was associated with changes in mucosal biology that might predispose to increased risk of HIV acquisition associated with unprotected receptive anal intercourse. Results: Sixty-five participants (45 men and 20 women) were recruited into the study. There were no significant differences between the numbers of ≥ Grade 2 adverse events across the arms of the study. Likelihood of future product use (acceptability) was 87% (reduced glycerin formulation of tenofovir 1% gel), 93% (hydroxyethyl cellulose placebo gel), and 63% (Nonoxynol-9 gel). Fecal calprotectin, rectal microflora, and epithelial sloughing, did not differ by treatment arms during the study. Suggestive evidence of differences was seen in histology, mucosal gene expression, protein expression, and T cell phenotype. These changes were mostly confined to comparisons between the Nonoxynol-9 gel and other study arms. Microarray analysis of the mucosal transcriptome provided preliminary evidence that topical application of tenofovir 1% gel was associated with decreased mitochondrial function within the rectal mucosa. Conclusions: The MTN-007 study demonstrated that, using conventional criteria, tenofovir gel is safe and acceptable and should be advanced to Phase 2 development as a potential rectal microbicide. However, microarray analysis of mucosal tissue suggested that use of tenofovir gel may modulate mucosal mitochondrial function. This observation will require further evaluation in future studies.

Norovirus is the commonest cause of outbreaks of gastrointestinal disease in the U.K. Most reported outbreaks occur in health care settings, such as hospitals and nursing homes, and can cause severe disruption through ward closures, cancelled operations and staff sickness. Previous studies estimated these outbreaks cost the NHS around £115 million a year. Despite previous studies some questions remain. What is the burden of norovirus in hospitals - how many outbreaks occur and how many people are hospitalised each year as a result of norovirus infection? Do published reports of outbreaks provide evidence of what works in infection control? Can the factors facilitating norovirus transmission during outbreaks in hospitals be identified? These questions were answered through a series of inter-linked studies that explored mortality, morbidity, transmission pathways and aspects of infection control. The introduction of a new surveillance system provided greater insights into the heavy burden that norovirus imposes on English hospitals. In the years 2009-2011, 3,980 reports of outbreaks of suspected and confirmed norovirus were received. There was little difference in the epidemiology of outbreaks from one season to the next. On average outbreaks were associated with 13,000 patients and 3,400 staff becoming ill, 8,900 days of ward closure and the loss of over 15,500 bed-days annually. Analysis of mortality data demonstrated a clear association between norovirus infection and mortality in the elderly (65 years and over) with an estimated 80 deaths per year in this age group. The number of deaths increased in years where norovirus activity was higher but this was not associated with increased pathogenicity of the virus. Norovirus was the only pathogen that had a significant association with mortality in the regression models. Modeling of routine hospital admission data demonstrates that norovirus accounted for around 3,000 norovirus admissions a year to English hospitals, two thirds of which were in the elderly. A review of published papers did not provide clear evidence for the effectiveness of infection control measures. However, this was largely because the reporting of outbreaks was poor and that the introduction of more rigorous reporting protocols would improve this. Analysis of 3,500 outbreaks of norovirus demonstrated that closing a ward or bay promptly (within three days of the first person becoming ill) is beneficial. The duration of outbreak, the total duration of disruption were shorter, and fewer patients overall were affected, if closure occurred promptly. When closure occurred 7 or more days after the first onset date outbreaks were twice as long as those where closure was prompt. The duration of outbreak was also increased by ward size and in outbreaks occurring in winter time. Outbreaks were longer if they occurred on care of the elderly wards. A strategy of prompt closure is beneficial, particularly in larger wards and during winter time. The time between the first two cases of each outbreak was used to estimate the serial interval for norovirus in a hospital setting and was estimated to be 1.86 days. This distribution and dates of illness onset were used to calculate epidemic trees for each outbreak. A permutation test found strong evidence that proximity was a significant driver of outbreaks (p < 0.001). Patients occupying the same bay as patients with symptomatic norovirus infection are at increased risk of becoming infected by these patients compared with patients elsewhere in the same ward. In summary, there is a demonstrable association with mortality in older people, and around 3,000 admissions to hospital each year. Over 3,900 outbreaks were reported in three years (2009-2011). On average 13,000 patients were affected each year leading to 8,900 days of ward closures. Vomiting appears to be an important driver of outbreaks. Acting quickly by closing affected areas appears to be beneficial in controlling outbreaks caused by norovirus. This is especially the case in larger wards during the winter.