Nutritional factors affecting bone health throughout the lifecvcle has received considerable attention in the literature but little is known about the influence of nutrition on stress fracture (SF) risk. Royal Marine (RM) recruits undergo a 32-week arduous military training programme, where the prevalence of SF has been ~5% over recent years. The Institute of Naval Medicine was tasked by Surgeon General to investigate risk factors associated with stress fracture during RM training at the Commando Training Centre Royal Marines. The present thesis focuses on the nutritional aspects of this work, with the primary aim to identify the influence of past and present dietary intake and nutritional status on SF risk and bone health in RM recruits. A bespoke dietary assessment tool was developed to investigate nutritional intake relative to SF risk and training success in a cohort of RM recruits (n=545). No aspects of diet during training were associated with SF risk. However, recruits with a higher energy intake during training were more likely to successfully complete training (P<0.05). Poor aerobic fitness, low body mass (less than 65 kg) and small thigh girth were independent risk factors for SF (n=27; P<0.05). In a larger cohort of RM recruits (n=1090), childhood (0-12 y), adolescent (12-18 y) and pre-RM training diets were assessed with a food frequency questionnaire. Vitamin 0 and micronutrient status were assessed in serum blood samples. Childhood milk intake and adolescent fruit and vegetable intakes were associated with bone quality (assessed by Broadband Ultrasound Attenuation) (P<0.05). Low milk intake during childhood, and high intake of fizzy drinks during adolescence, were associated with increased SF risk (n=75; P<0.05). Importantly, serum 25(OH)D (as a marker of vitamin 0 status) <60 nrnol.L" was associated with increased SF risk (P<0.05). Stress fractured recruits (n=65) and matched controls (n=65) underwent Dual Energy X-ray Absorptiometry and peripheral Quantitative Computed Tomography scanning. Stress fractured recruits had lower bone density of the lumbar spine and hip, and narrower tibiae than their matched controls. These novel data suggest an important role for lifestyle factors in the pathogenesis of SF in a military population, with concomitant bone density/structure differences at key fracture sites. Further research (including a possible vitamin D supplementation intervention) is warranted.

The health-related benefits of a high consumption of fmit & vegetables on a variety of diseases, including skeletal health have been advocated by health organisations. The effect of high acidity, weight-reducing diets on bone health also remains undefined This project involved two key aspects: a population-based study and an intervention itudy. The principal aims of the population based study were to: (i) estimate the ;id/alkaline potential of the diet in a representative group of British elderly from )nal Diet and Nutrition Survey (NDNS) datasets using several validated algorithms for calculating the potential renal acid load (PRAL) and estimated net endogenous acid production [NEAP], (ii) compare and characterise tiie aforementioned algorithms by specific nutrients/food groups likely to influence dietary acid load in tiie abovementioned population group and a representative sample of tish adults (18-64 years) using North South Ireland Food Consumption Survey (NSEFCS) datasets.

The aim of this study was to examine the food access, selection and acquisition issues affecting consumers aged over sixty years living in Northern Ireland. People of pensionable age make up 16.3% of the total Northern Ireland population and by 2040 the proportion of people aged 60+ will have doubled (Age Concern, 2007). Therefore this is a significant consumer segment and one worthy of study and scrutiny. Older people in Northern Ireland are considered disadvantaged in terms of food related provision and their behaviour and experiences warrant particular attention. Diversity within the older population must be taken into account; there is a lack of informative data on the over 60 age group and a tendency to treat all persons who have reached this age milestone as one homogeneous group (Wait and Harding, 2006). Due to the changing nature of consumer behaviour, current research results are necessary as one cannot rely on past experiences of older consumers to determine the behaviour of present and future generations (Carrigan, 1998). "Older consumers of tomorrow will be different from their counterparts today" (BCSC, 2006, p7). To date, the food related behaviour of older consumers in Northern Ireland has not received significant attention and their behaviour has not been extensively documented. This is a timely study with relevance to an increasing consumer segment that represents a major customer base for food retailers. Capitalising on this market through recognising and meeting the needs of older customers will lead to large rewards for food retailers and result in a more satisfying shopping experience for consumers themselves. It will also serve to inform local food and retail policy which must take into account the needs, wants and abilities of all consumer groups. A pragmatic research approach was taken to data collection with a combination of both qualitative and quantitative techniques employed. Qualitative focus group discussions were carried out with older consumers (n=4) and semi-structured interviews undertaken with food retail managers (n=6). These were then followed by a large-scale quantitative consumer questionnaire (n=791). The results reveal that older people face a range of issues when accessing, selecting and acquiring food and their behaviour and satisfaction is influenced by individual consumer characteristics. Greatest difficulty is experienced once in-store where the physical retail environment was found not to accommodate older consumers. From the retailer interviews it was found that store management recognise the importance of older shoppers and are aware of the issues they face. However, there is a mismatch in perceptions between the retailer viewpoint on current provision and that of the consumer, and considerable scope for retailers to improve their retail offering and better serve the older population. It is also evident that older consumers are a heterogeneous population group with differences in behaviour and experiences shaped by a range of lifestyle characteristics. A number of recommendations have been proposed which would overcome the difficulties currently experienced by aged consumers and strengthen retail provision.

Fruit may not be perceived as a viable snack option due to a number of barriers (Jack et al, 1997; 1998). The aim of this thesis was to determine the reasons that fruit is not chosen as a snack (while other snack foods are), to examine these in depth and to make recommendations for further intervention. It was discovered via an exploratory diet diary study that the four most frequent reasons given for choosing a non fruit item as a snack (pleasure, availability, hunger/satiety, convenience) were also the four most frequent reasons for not consuming fruit as a snack. Each was examined in turn. Flavour nutrient conditioning was used to attempt to increase pleasure ratings in participants who were hungry (in the context of a starter) and less hungry (snacking context), with only main effects of time giving an increase in pleasure ratings. A location manipulation of availability in a profit driven environment resulted in an increase in fruit purchase frequency in a snacking context when fruit was placed at the point of purchase. In an investigation of hunger/satiety, a preload study revealed that high (dried) and low (fresh) energy density fruit provided satiety equivocal to a biscuit, while the fresh fruit was perceived as more filling. In a second study, where fresh fruit and dried fruit were provided as snacks, dried fruit was chosen more frequently when hungry in free living individuals. Finally, fruit in a more convenient form (pre prepared and packaged apple) had no long term effect on fruit consumption. These findings suggest that manipulations of availability and hunger/satiety are more likely to be beneficial compared to manipulations of pleasure or convenience. Future interventions should focus on hunger/satiety and availability as these may have the greatest potential to increase fruit consumption as a snack.

Non-communicable diseases are a huge burden worldwide, with increased fruit and vegetable (FV) intake being associated with reduced risk. Although antioxidants are thought to be the beneficial components in FV, substitutions of less healthy foods may also occur when increased FV are consumed. A meta-analysis pooling data from published randomised intervention studies (RIS) where FV were increased showed micronutrients, fibre and carbohydrate intakes increased; whilst fat intakes potentially decreased. No significant difference was found in energy. An RIS of increased FV in a diabetic population confirmed these findings. However, 4 pooled FV studies revealed modest increases in energy when FV were increased, suggesting inconsistencies in energy; however this could be due to heterogeneity between studies. Furthermore when FV were increased, variety was also likely to increase. Whilst much interest has focused on general FV consumption, the health benefits of single fruit andlor vegetables may differ, and it is therefore of interest to examine the potential of individual foods and compounds contained within this food group. Apples are rich in polyphenols, particularly chlorogenic acid and epicatechin however a number of factors are likely to affect this polyphenol content including SUMMARY 2 - PAGE 3 storage or cooking. Analysis showed no effect of storage, cooking, or processing on polyphenol content but between-apple variability showed significant differences in golden and red delicious apples. Furthermore, epicatechin-rich apples in various forms were tested in participants at increased risk of CVD, to assess benefit on CVD risk factors such as oxid ised LDL and blood pressure. This pilot intervention showed no effect of epicatechin at a dose of 80mg/day on CVD risk. In conclusion , increased FV intake showed improvement in diet profile. Differences in storage, cooking and processing showed no effect on polyp he no I content in apples, however between-apple variability was observed. Finally epicatechin-rich apples and apple products showed no effect on risk factors for CVD.

This thesis employed a number of methods to examine the relationship between fruit and vegetable (FV) intake, dietary patterns and vascular health. A major aim was to examine the secondary end-point measurements within Fruit, vegetable and Insulin Resistance STudy (FIRST). Overweight participants at high cardiovascular disease (CVD) risk were randomised to 2,4 or 7 portions FV for 12 weeks and CVD risk factors were measured pre- and post- intervention. Eighty nine participants completed the study and despite evidence of good compliance, increased FV intake did not have a significant effect on blood pressure, lipid levels or C-reactive protein. Quantification of the retinal vessel caliber (RVC) is an emerging method to examine the microvasculature in vivo. Increasing FV intake had no significant effect on measured RVC in a subsample of FIRST participants. A second aim was to explore associations between dietary patterns and RVC in a cross-sectional study involving an elderly population (n= 288). Three major dietary patterns were identified using principle component analysis (PCA), however no significant associations were observed between dietary patterns and measured RVC. A third aim was to conduct a review of observational studies investigating the association between a posteriori-derived dietary patterns and risk of type 2 diabetes (T2DM). A meta- analysis was performed of prospective studies that investigated the association between PCA-derived dietary patterns and T2DM risk in approximately 271,702 participants and 16,020 incident cases of T2DM. The results demonstrated a significant reduction in T2DM risk for individuals in the highest category of healthy dietary pattern compared with those in the lowest category (OR 0.86; 95% CI:0.80, 0.91; P < 0.00001). Furthermore, a significant increase in risk ofT2DM was observed for those in the highest unhealthy dietary pattern category when compared with those in the lowest category (OR 1.42; 95% Cl: 1.32, 1.53, P < 0.0001).

The aim of this study was to investigate the discourses women use to describe their eating patterns in relation to comfort eating. The idea originated from the work of Susie Orbach (1978) which focussed upon women's eating patterns and 'fat' being rooted in powerlessness and self-denial. After undertaking a thorough search of the literature, the gap around emotional eating became apparent. This thesis examined the concept of women's relationship to food, and explored if, and if so how, women acknowledge emotional regulation through food and comfort eating. A qualitative research strategy was adopted. The process of data analysis consisted of 2 phases; phase 1 was a preliminary internet based collection of data and analysed using Thematic Analysis, and phase 2 was in the form of semi structured interviews and analysed using Foucauldian Discourse Analysis from an ontological framework of Social Constructionism. Overall, the multiple discourses in this study found that participants acknowledged the use of food for 'comfort', and that emotional distress was a mediator when describing their relationship to food. Participants constructed positions associated with; repressed anger, feelings of shame and failure to achieve an ideal, and lack of self-efficacy. This study also highlighted discourses around participants experiencing a non- tangible overwhelming 'urge' to eat that felt uncontrollable. This was in line with a sense of depleted energy and an inability to achieve the goal of weight loss. This study suggests that medium to long term psychotherapy could be a way forward to address the internal conflicts and inability to change highlighted in the participants' discourses. Inclusion of Psycho-education is also recommended for weight loss management groups in order to address the problematic area of obesity.

The aim of this thesis was to develop a conceptual understanding of consumer acceptance of novel foods, grounded in the attitudes, perceptions and behaviours of individuals to foods that were "novel" by virtue of an innovative method of production or newness within the diet. Novel food acceptance is dominated by psycho-sensory research and thus this thesis' behavioural approach provides an alternative perspective that contributes to the food choice and acceptance literature. The grounded theory method (GTM) was selected as the guiding methodology because of its theory building mandate. The GTM procedures were applied to a secondary data set of 45 transcripts drawn from 6 separate research projects. These projects were linked by the common use of the focus group technique to gene rate data together with a complementary subject matter, namely data relating to the attitudes perceptions and experiences of i11dividuals to cOnve11lional foods, novel foods and novel food technologies. The analysis showed novel food acceptance is a cyclical process of acceptance or rejection containing five 'acceptance states', rather than a single decision point. The five acceptance states derived from the analysis 'and related to the consideration, trial and sustained incorporation of a novel food into dietary practices that arc shaped by personal, product and situational factors are: i) conceptual acceptance, or the degree of engagement with and approval of a novel food; ii) connective acceptance, ,or the personal connections that an individual makes with a novel food by visualising its potential benefits and its role in satisfying a perceived personal need; iii) evaluative acceptance, or the trade-offs made when evaluating novel foods with available alternatives; iv) tri al acceptance, or the practices associated with the preparation of the novel food, the taste: of the novel food and its perceived impact on well-being; and v) dietary acceptance, or the complementarity of a novel food with the structure of existing meal patterns and the substitutability of the novel food with existing foods within the diet. An outcome of this behavioural model of consumer acceptance is the incorporation of the novel food into established dietary practices, which is referred to as the "acceptability of dietary change". These findings provide a framework for novel food acceptance research and have implications for food consumer research.

Many people fail to perform healthy behaviours, despite holding positive attitudes towards acting healthily. Examining moderators (e.g., individual difference variables and mood) and different types of attitudes (implicit/explicit attitudes) might explain variance in behaviour unaccounted for explicit attitudes alone. Using the Implicit Association Test, the five studies within this thesis tested these issues in relation to eating behaviour.

The consumption of flavonols, major sources of dietary flavonoids, have been linked to the reduction of risk factors for various chronic diseases. However, the bioavailability of flavonols would appear to be limited and not all compounds within the same subclass are absorbed with equal efficacy, thus compromising their nutritional relevance. The aim of this current study was to examine the absorption and metabolism of four major flavonols specifically quercetin, myricetin, kaempferol and galangin using both in vitro methods and healthy human volunteers to investigate how bioavailability was affected by the addition of other ingredients, specifically fat. Firstly, the enzymatic synthesis of flavonol metabolites was conducted using liver S9 homogenates, which demonstrated that all major flavonol aglycone generated several isomers of glucuronide or sulphate forms after the incubation period. The evidence for identification and characterisation of available authentic standards and synthesised flavonol metabolites using HPLC-DAD-MS-MS is presented. The success of this method was demonstrated through its application to assess the fate of flavonols in the cell culture transport experiments, which were examined using co-cultures of Caco-2 (enterocyte cell) and HT29-MTX (goblet cell) monolayers and the results partially substantiate that the absorption is most probably achieved by passive transport.