Background: The 7-valent pneumococcal conjugate vaccine (PCV7) was introduced routinely in the UK in September 2006 and replaced by PCV13 from April 2010. Aims: To evaluate the impact of PCV7 on the incidence of all-cause community-acquired pneumonia (CAP) in children. Also to investigate the aetiology of CAP before and after the introduction of PCV as well as serotype the pneumococcal infections. Methods: Enrolled children were from North East England (excluding Cumbria) who were aged 0–16 years and presented with clinical and radiological features suggestive of pneumonia. Epidemiology survey was prospectively undertaken in 2008–2009 at 11 hospitals in North East England. Data were compared to those from a similar survey undertaken in the same hospitals in 2001–2002. Aetiology studies were prospectively conducted in 2001–2002 (pre-vaccine) and 2009–2011 (post-vaccine) in Newcastle and Middlesbrough. Investigations included culture, serology, immunofluorescence antibody, urinary pneumococcal antigen and PCR assays. Epidemiology Results: A total of 542 children were enrolled, of which 74% were aged <5 years. PCV7 uptake was 90.7%. The annual incidence of pneumonia was 11.8/10 000 (95% CI 10.9–12.9), and the hospitalisation rate was 9.9/10 000 (95% CI 9.0–10.9). Compared to 2001, there was a 19% (95% CI 8–29) reduction in the annual rate of CAP in those aged <5 years, and in those <2 years a 33.1% (95% CI 20–45) reduction in the annual incidence of CAP and 38.1% (95% CI 24–50) reduction in hospitalisation rates. However, for those unvaccinated aged ≥5 years, there was no difference in the annual incidence of CAP and hospitalisation rate between both surveys. Since 2001, the overall reduction in annual incidence was 17.7% (95% CI 8–26) and for hospitalisation 18.5% (95% CI 8–28). Aetiology Results A total of 401 children were enrolled; 241 and 160 respectively in the pre- and post- vaccine studies (73% aged <5 years), for whom at least one diagnostic investigation had been performed. Identification of a definite pathogen was higher post-vaccine (61%) than pre-vaccine (48.5%) [p=0.019]. Rates of bacterial infections were not different between post- and pre-vaccine (17.5% versus 24%, p=0.258). Viral (31%) and mixed infections (12.5%) found more often post-vaccine than pre-vaccine (19.5% [p=0.021] and 5% [p=0.015] respectively). Pneumococcal detection post-vaccine was substantially improved when PCR assays were used compared to culture (21.6% versus 6%, p=0.0004). A serotype was identified in 75% (18/24) post-vaccine including serotypes 1 (44.4%), 3 (27.8%), 19A (22.2%) and 7A/F (5.6%). Conclusions: PCV7 has reduced both the annual incidence and rate of hospitalisation of pneumonia in children, particularly those aged <2 years. Pneumococcal serotypes which are included in PCV13 but not PCV7 predominated. This suggests that the replacement with PCV13 likely to be associated with a reduction in the incidence of pneumococcal-related pneumonia. Continued surveillance is required to monitor for emerging serotypes.

Historically, empyema thoracis has been a major cause of morbidity and mortality in children. It became the focus of considerable attention following its resurgence globally in the 1990’s. The factors driving this change remain uncertain. In addition, there remains significant controversy over the best method of management of the condition. This thesis aimed to define the epidemiology of paediatric empyema thoracis, to understand the factors contributing to the rise in the incidence of the disease. Secondary aims included investigation of the impact of the pneumococcal conjugate vaccine on paediatric empyema and evaluation of the effectiveness of different treatment methodologies in the condition. A progressive framework of multivariate time series models and wavelet analysis was used to investigate relationships between empyema and pneumonia, activity of Streptococcus pneumoniae, Staphylococcus aureus, Streptococcus pyogenes and Mycoplasma pneumoniae and time. The spatial epidemiology of the both conditions in North East England was defined and the impact of the introduction of routine pneumococcal vaccination investigated using an interrupted time series analysis. Multivariate survival models were used to investigate outcomes following different treatment methods. Hospitalisations due to empyema increased significantly in England between 1997 and 2006, underpinned by an increase in bacterial pneumonia. Isolations of S. pneumoniae and S. pyogenes were positive predictors of empyema nationally. No spatial variation in the risk of empyema was detected. Introduction of pneumococcal vaccination did not decrease empyema hospitalisations. Children who underwent primary surgical treatment for their empyema had a 40% reduction in hospital stay and a lower risk of readmission or of any complication. The increase in the incidence of paediatric empyema in England was driven predominantly by an increase in pneumococcal and streptococcal pneumonia. Primary surgery in empyema allowed earlier discharge, but further research is needed to establish which outcomes are most acceptable to patients and their families.

Autistic children often lack social behaviours which are normally present by 8-12 months (Klin et al, 1992; Mundy et al, 1986), although current 'top down' theories about autism hypothesise later-developing conceptual difficulties in social cognition. Research indicates that there are abnormal 'bottom up' perceptual processes in autism (Moore, Hobson &amp; Lee, 1995). Processing of unexpected dynamic visual information may occur to a diminished extent (Courchesne, 1987), whilst in normal infancy, causal perception of the movements of animate. and inanimate objects is likely to be important for social cognition and affective relationships (Shultz, 1989; Premack &amp; Dasser, 1990). It has been suggested that autistic children have difficulties with the unpredictable nature of perceived social information (Moody &amp; Sigman, 1989b; Dawson &amp; Lewy, 1989). On the basis of such previous research, it was proposed that the early social abnormalities of autistic children are a result -of them tending not to notice, or process further, brief dynamic visual information about events unless these follow simple predictable patterns. To test this hypothesis, seven young autistic and seven developmentally delayed children, matched pairwise for verbal comprehension, were initially habituated to two computer-generated displays, of a cartoon-like 'boy' (Runner) running up to a 'wall' and back, and of a 'ball' moving towards the wall and apparently re-bounding back. Visual habituation was reliable and due to information encoding. The autistic children looked relatively less at the Runner habituation display, perhaps due to relatively diminished processing of animate dynamic information. After habituation criterion was reached, a time delay was introduced prolonging contact with the wall, so that the Ball display became 'impossible' whereas the Runner remained 'possible'. As predicted, recovery of visual fixation indicated that the delay was discriminated, and the autistic children recovered relatively less to the novel Ball display compared with the developmentally delayed children. The autistic children may not have perceived the anomaly in the novel 'impossible' Ball event. They either may have had a general expectation for inanimate objects to move independently, or a tendency not to notice altered significance in a repetitive visual event. However, the group differences could also have been due to the autistic children tending not to have real life expectations of cartoon images. Consistent with the hypothesis, the scores of all 14 children both for pre-11 month social behaviours and for joint attention, were associated with relative recovery for the novel Ball display. As predicted from previous research, the autistic children engaged infrequently in these social behaviours. It is proposed that how children perceive the dynamic animate and inanimate world affects how they develop socially during their first year of life. Relevant areas for future research and implications for early interventions are discussed.

Within the context of increasing prevalence of childhood obesity over a number of decades, alongside a trend towards increasingly 'obesogenic' environments, this thesis explored cross-sectional and prospective relationships between maternal feeding behaviour and child adiposity at age one year and at age five years. The research aims were addressed over two empirical studies. Study 1 collected data on breastfeeding through structured interviews with mothers (N=290) of one-year-old infants, observed mother-infant feeding interactions, and measured infant weight and length. Study 2 followed up the same sample (N=197) through a survey at age five, to examine cross-sectional relationships between maternal feeding behaviours (both practices and styles), child eating behaviours, and child adiposity (Study 2 Part A) as well as longitudinal associations with breastfeeding, maternal feeding behaviours and adiposity in infancy (Study 2 Part B). Age five height and weight were provided by a child health administrative database. Several limitations in the current knowledge base were also addressed, including the over-reliance on maternal self-report of feeding behaviour, the relative paucity of research examining the role of breastfeeding, and the lack of prospective studies beginning in infancy. Overall, findings suggest that feeding practices may be a response to child attributes (in terms of adiposity and eating behaviour), rather than a cause. Further, feeding practices are distinguishable from 'feeding styles', which are established early and may not function as a response to child attributes. Findings suggest that breastfeeding may promote a more responsive feeding style, as mothers are accustomed to sharing control over food intake with their child. This effect may extend into later infancy, during the transition to family meals and self-feeding, and may, in turn, facilitate appetite regulation over the short- and longer-term, which protects against excess weight gain.

The complex: and important task of clinical assessment includes the completion of observations to inform consequent judgments and management decisions. The current study investigated the impact of priming and professional variables on psychologists' judgments and decision making about a child they observed. A novel experimental method was employed where participants were assigned to a primed or unprimed condition (through simulated referral letters), and then viewed a video of a simulated "assessment" session where the child's behaviour was constructed to be deliberately ambiguous. Participants provided hypotheses and ratings of the child's behaviour, predicted their planned actions as professionals, and judged their confidence in their assessment. Results suggested that priming did not have a significant influence on participant ratings of the child's behavior. Although, priming significantly increased participant confidence in their assessment, it did not significantly affect proposed actions. Participant variables (e.g., age, gender, years of experience) did not affect behavior ratings, but professional training did significantly influence hypotheses generation, whereby qualified psychologists' hypotheses had more non-pathologising content and increased focus on difficulties within the parent-child relationship, than did their pre-qualified counterparts. These results are discussed within their theoretical and empirical context, as are the study's limitations and potential to inform clinical assessment and professional training for psychologists.

Exploration and discovery are core components of play. For children with autism, whose needs are complex and diverse, the potential that technology affords for developmentally appropriate play is under researched. Many software programmes focus on operational routines and fail to maximise on the capricious, idiosyncratic and emergent fun that can evolve when children are relaxed and able to freely discover their interests. There are very few studies that reveal how children with autism benefit, in terms of their confidence and self esteem, when technology is utilised simply as a trigger for play and how this particularly influences emotional regulation, social communication and learning. The aim of this PhD is two-fold: to contribute to knowledge on the challenges of designing for play for children with autism and to make a contribution to research in the autism field by demonstrating the expressive communication abilities revealed by children through playful interaction with a software interface. The PhD overview critiques eight published works, which analyse and discuss the design methods, the nature of discovery-led interaction and the impact on children's play and communication. The early publications are cohered around participatory design, and how the relationships with end user populations influenced early prototypes and shaped initial evaluation. These studies paved the way for deeper investigations into the multi sensory capacities of technology in relation to the characteristics of autism. These studies furthered the contextualisation of the work by locating design in and with affordable everyday environments. The emphasis on situated-ness is understood from a philosophical perspective in the most recent and concluding studies, which draw on phenomenology for the meta-analysis of action and reaction. The final paper provides a confident argument for play as a means of exploiting the physical directness of tactile and auditory input. The studies reveal how even the most hard to engage child demonstrated creative and communicative ability beyond the expectation of his carers. The papers contribute towards an understanding of playfulness as an emergent, unpredictable and ultimately rewarding experience for children severely affected by autism. In the design field, the work represents a contribution to new knowledge in participatory and collaborative methods for developing responsive sensory interfaces for marginalised groups. Key words: Autism, interaction, tangible interaction, play, playfulness, sensory, prototypes, generative, therapeutic, repetition, inspiration, participatory design, collaboration, embodiment, colour.

Autism is a spectrum of developmental disorders, with onset in early childhood affecting social, communicative and imaginative development. Numerous theories have been proposed regarding the aetiology of autistic spectrum disorders (ASD), yet the condition remains poorly understood. Children with ASD tend to suffer from severe dietary and/or gastrointestinal (GI) problems (including abdominal pain, constipation, diarrhoea and bloating). These symptoms may be associated with disruption of the indigenous gut flora promoting overgrowth of potentially pathogenic (toxin producing) microorganisms. The primary aim of this research was to provide a detail overview of the gut microbiota of ASD children and study the effect of selected probiotics on the gut microbiota, gut function and associated symptoms (i. e. behaviour) of ASD children. The gut microbiota of ASD children was compared to that of non-ASD control groups (siblings and unrelated children). Changes in bacterial gut populations were assessed using fluorescence in situ hybridisation (FISH) targeting relevant gut bacterial groups in a culture independent manner. The gut microbiota of ASD children contained higher levels of Clostridium histolyticum subgroup (Clostridium clusters I and II) bacteria than that of non-ASD children. Screening probiotic bacteria for their ability to inhibit the growth of Clostridium perfringens was assessed in vitro. Six probiotics (namely, L. plantarum NCIMB 41114, L. acidophilus JVT5, L. pentosus JCM 1558, L. rhamnosus GG 1003, L. plantarum 903 and L. plantarum WCFS1) showed statistically significant inhibition of C. perfringens (although many others were also tested). These probiotics were further tested in batch culture fermentation systems to investigate whether they could beneficially modulate the faecal bacterial populations of children with ASD. Probiotic supplementations increased the levels of lactobacilli/enterococci populations and demonstrated a degree of anti-microbial action against C. histolyticum subgroup. L. plantarum WCFS 1 was then used in a double-blind, placebo-controlled, randomized cross-over feeding study in ASD. Administration of L. plantarum WCFSI significantly increased numbers of lactobacilli/enterococci in the faecal flora of ASD children compared to a placebo. No modulation of the C. histolyticum subgroup was seen during the study. A significant improvement in bowel function was observed with probiotic administration. Moreover, the total behaviour problem score was significantly lower after L. plantarum WCFS 1 administration compared to the placebo. This study supports the hypothesis that probiotic therapy may modulate the, GI microbiota composition in ASD individuals and alleviate GI symptoms of ASD individuals.

Humans come into contact and interact with potential infective agents. The innate immune system is the first line of response to ward off infection. Innate immunity is, in part, under genetic control. This genetic control may help us understand the differences between individuals in preventing infection or limiting infectious and inflammatory illness. Systemic inflammation is a complex disorder that is difficult to define. Current definitions are derived from consensus meetings. A need has been expressed for a more useful definition of systemic inflammation. The work presented here identifies some of the underlying hereditability in limiting or being more vulnerable to severe infectious and injurious insults. Individual differences in complement activation potential and endotoxin recognition underlie part of the observed differences in a systemic inflammatory response to severe infection and injury. An exploratory study using heart rate variability as a non-invasive method to distinguish infectious systemic inflammation from sterile systemic inflammation was inconclusive. Chapter 1 gives the background to this study and an introduction to the approaches taken in this thesis. Chapter 2 describes in detail the methods used in the genetic association study and physiological systems analysis. Chapter 3 goes into some detail about the potential pitfalls in genotyping association studies and how these were addressed in the current study. The areas of genotypi ng quality, linkage disequilibrium, ethnicity, sample size and validation of previously done work are discussed using MBL-2 and ACE as examples. Chapter 4 is a description of the work done on genetic variability in the endotoxin receptor complex and how in may result in the host response to severe infection and physical insults. TLR4 polymorphisms were associated with lower platelet counts in severe inflammation. The reasons for this are unclear but may point to a direct effect of the TLR4 pathways on platelets or indicate that platelet counts are a more sensitive marker of systemic inflammation than SIRS criteria. These data support the view that variation in TLR4 function influences the early inflammatory response. This phenomenon may be one aspect of reduced fitness in the capacity to respond appropriately to an insult. Chapter 5 reports the central role of complement in the acute phase response. Polymorphisms in two out of the three complement activation pathways were shown to have potential modifying effects in paediatric systemic inflammation. This chapter reports that polymorphisms in the CFH gene may modulate the acute inflammatory response and corroborates the previously reported finding that MBL-2 variant genotypes are a risk factor for the early occurrence of SIRS/sepsis in a large cohort of paediatric critical care patients, independent of other potentially important functional polymorphisms in the complement and innate immunity system. A better understanding of how these polymorphisms operate at the pathophysi ol ogi cal level is needed before these findings can be translated to clinically useful therapeutic modalities. This study demonstrates that genetic polymorphisms associated with reduced complement activation may be associated with early SIRS/sepsis. This is consistent with a view that appropriate complement activation occurring early following an infectious or inflammatory insult protects children from early SIRS/sepsis. Chapter 6 assesses the usefulness of full MBL-2 genotyping and compares the MBL-2 genotype and M BL serum levels between a cohort of healthy children and a cohort of critically ill paediatric patients. MBL2 genotyping did not render more information with regards to M BL serum level when all promoter and structural polymorphisms were identified over and above structural polymorphisms and the XY promoter polymorphism. The children admitted with infection did not have a surplus of M BL deficient genotypes as compared with healthy children. This suggests that M BL deficient genotypes do not predispose to severe infection. M BL serum levels in SIRS or sepsis were lower compared with critically ill children without systemic inflammation. M BL levels were most reduced in the acute phase response in those genotypes with intermediate serum levels, which may reflect a consumption of M BL in critical illness and an inability to maintain pre-insult M BL serum levels. Chapter 7 explores a novel way to discriminate SIRS from sepsis by means of heart rate variability analysis. In this small paired sample study no differences were seen in LF metrics to differentiate sterile SIRS from sepsis. Neither was there a difference in LF metrics between those children who went on to develop a nosocomial infection and those who did not. Normalised HF was significantly higher in sterile SIRS vs. sepsis. These preliminary finding require further validation and a longitudinal approach in a larger cohort. Finally, Chapter 8 discusses the findings of this thesis in the context of interpretation and of the findings and potential future approaches. This thesis supports the view that better metrics are required to discriminate systemic inflammation as well as the concept that in children control of an inflammatory threat is aided by a vigorous capacity to respond.

RNA interference (RNAi) is a highly conserved post-transcriptional gene silencing process triggered by double-stranded RNA (dsRNA) in eukaryotic cells. Elucidation of the RNAi regulatory pathway and its components has led to the identification of endogenous dsRNA molecules, termed microRNAs (miRNAs), which are transcribed as a single hairpin molecule prior to their maturation into a cytoplasmic dsRNA. The efficient gene silencing achieved by these short hairpin RNA (shRNA) molecules and the cumulative understanding of the RNAi pathway has prompted the development of hairpin expression vectors capable of mediating stable gene silencing in vitro and in vivo. The aim of this thesis is to evaluate the efficacy of viral vector-mediated RNAi in the retina using recombinant adeno-associated viruses (AAV) and lentiviruses that contain silencing hairpin cassettes to target four genes in murine photoreceptors and the retinal pigment epithelium (RPE). A detailed assessment of the utility and extend of RNAi in the retina using different viral vectors and hairpin designs is presented in this thesis. Lentiviral and AAV vectors were firstly used to silence GFP in vitro and in vivo as a proof of concept for vector mediated RNAi in the retina. Subsequently, we used lentivirally-mediated RNAi to study disease processes in the retina concentrating on tight junction (TJ) modulators ZO-1 and ZONAB and their role in RPE homeostasis, cell-cycle progression and epithelial-mesenchymal transition (EMT). Here we demonstrated how TJ misregulation can lead to RPE loss, proliferation or dedifferentiation; processes involved in pathological conditions such as atrophic age-related macular degeneration (AMD) and proliferative vitroretinopathy (PVR). Whilst lentivirally-mediated RNAi was used to elucidate aspects of retinal function and disease, AAV-mediated RNAi was used to probe the therapeutic potential of shRNAs by silencing Peripherin-2 (Prph2), the second most abundant retinal protein, using a miRNA-based hairpin. AAV2/8 particles were used to target endogenous Prph2 and evasion of silencing was demonstrated using an engineered Prph2 cDNA that could be used in a suppression and replacement approach for the treatment of dominant retinal disorders.

Steroidogenic Factor-1 (SF-1, NR5A1) is a nuclear receptor transcription factor that plays a central role in adrenal and reproductive biology. In humans, SF-1 regulates adrenal development and disruption of SF-1 or its known targets is associated with impaired adrenal function. Therefore, the identification of novel SF-1 targets could reveal important new mechanisms in adrenal development and disease. This thesis describes three approaches to identifying SF-1 targets in NCI-H295R human adrenocortical cells. SF-1-dependent regulation of CITED2 and PBX1 was investigated since these factors regulate adrenal development in mice through pathways shared with Sf-1. Expression of CITED2 and PBX1 was confirmed in the developing human adrenal, and SF-1 was found to bind to and activate the CITED2 promoter and to cooperate with DAX1 to activate the PBX1 promoter. SF-1 binding was investigated using chromatin immunoprecipitation microarrays (ChIP-on-chip). These studies revealed that SF-1 binds to the extended promoter of 445 genes, including factors involved in angiogenesis. Angiopoietin 2 (ANGPT2) emerged as a key novel SF-1 target, confirmed by transactivational studies, suggesting that regulation of angiogenesis might be an important additional action of SF-1 during adrenal development and tumorigenesis. Global gene expression analysis following SF-1 overexpression revealed differential expression of 1058 genes, many of which are involved in steroidogenesis, lipid metabolism and cell proliferation. Bidirectional manipulation of SF-1 revealed a subset of positively regulated genes, including the known targets STAR and CYP11A1 and novel target SOAT1, a regulator of cholesterol esterification. Considering that defects in several SF-1 targets have been associated with adrenal disorders, mutational analysis of SOAT1 was performed in forty-three subjects with unexplained adrenal insufficiency but failed to reveal potentially disease-causing variants. Taken together, manipulation of SF-1 in human adrenal cells has expanded our knowledge of the many potential actions of SF-1 in the human adrenal gland.