No description available.

This thesis investigates the application of classical and contemporary statistical methods in medical research attempting to bridge the gap between statistics and clinical medicine. The importance of using simple and advanced statistical methods in constructing and assessing risk models in medicine will be demonstrated by empirical studies related to vascular complications: namely abdominal aortic aneurysm and diabetic retinopathy. First, data preprocessing and preliminary statistical analysis are examined and their application is investigated using data on abdominal aortic aneurysm. We illustrate that when dealing with missing data, the co-operation between statisticians and clinicians is necessary. Also, we show advantages and disadvantages of exploratory analysis. Second, we describe and compare classification models for AAA selective screening. Tow logistic regression models are proposed. We also show that it is important to assess the performance of classifiers by cross-validation and bootstrapping. We also examine models that include other definitions of abnormality, weighted classification and multiple class models. Third, we consider the application of graphical models. We look at different types of graphical models that can be used for classification and for identifying the underlying data structure. The use of Naïve Bayes classifier (NBC) is shown and subsequently we illustrate the Occam’s window model selection in a statistical package for Mixed Interactions Modelling (MIM). The EM-algorithm and multiple imputation method are used to deal with inconsistent entries in the dataset. Finally, modelling mixture of Normal components is investigated by graphical modelling and compared with an alternative minimisation procedure. Finally, we examine risk factors of diabetic sight threating retinopathy (STR). We show the complexity of data preparation and preliminary analysis as well as the importance of using the clinicians’ opinion on selecting appropriate variables. Blood pressure measurements have been examined as predictors of STR. The fundamental role of imputation and its influence on the conclusions of the study are demonstrated. From this study, we conclude that the application of statistics in medicine is an optimisation procedure where both the statistical and the clinical validity need to be taken into account. Also, the combination of simple and advanced methods should be used as it provides additional information. Data, software and time limitations should be considered before and during statistical analysis and appropriate modifications might be implemented to avoid compromising the quality of the study. Finally, medical research should be regarded for statisticians and clinicians as part of a learning process.

Introduction: People with established renal failure have difficult treatment decisions to make. Pre-dialysis education that leads to treatment decision-making has traditionally been ad-hoc, with programmes lacking a theoretical or evidence basis. The two key aims of this study were to design and deliver a theory-based predialysis educational intervention and explore individuals' pre-dialysis experience. Understanding the interplay, between the two, informed recommendations about how best to intervene to help people make good treatment decisions. The MRC Complex Intervention Framework guided study development but it has only been possible to address the first two phases within the scope of this thesis. Subsequent phases will be the focus of post-doctoral research. In the Theory Phase renal and long-term condition self-management literature was reviewed and self-efficacy theory was predominant in the literature. To address a significant gap identified in the literature the 'Retrospective Patient Views of Pre-Dialysis Education Study' sought 29 participants perspective of pre-dialysis educational components, using semi-structured interviews. Clinicians' advice and written information were valued and unplanned vicarious learning began to emerge. The Modelling Phase was founded on the theory phase findings. Firstly, a multidisciplinary and service user Participant Action Research study developed and delivered a self-efficacy theory-based pre-dialysis education intervention. The 'Service Users Pre-Dialysis and Treatment Decision–Making Experience Study' explored 20 participants pre-dialysis year, at 3 time points, resulting in 54 interviews. Findings were analysed using a grounded theory approach revealing that vicarious learning, the clinical setting and written information influenced treatment decisionmaking. Conclusion: Decision-making is dynamic and requires dynamic educational inputs tailored to individual needs. The 'Conceptual Model of Influences Impacting the Pre-Dialysis Journey' brings together the themes emerging from the grounded theory analysis. From this the original contribution to extending self-efficacy theory was developed, namely that vicarious learning and not mastery is key at this stage of individuals' care journey and this has not previously been recognised. The practical output from the research is the 'Pre-Dialysis Education Curriculum'. Recommendations for future research include an exploratory trial of self-efficacy theory-based pre-dialysis education.

The three chapters of this thesis are united under a common theme of exploring mindfulness, specifically with reference to older people. Chapter one: The systematic literature review critically evaluates the evidence for mindfulness-based stress reduction (MBSR) for use with older adults (aged 60 years and older). 12 articles met pre-specified criteria for inclusion. In studies of relatively healthy, high functioning older adults, there were few benefits observed. For samples of older people with clinical levels of mood and physical health difficulty, benefits were observed in some studies across a range of psychosocial domains. The present review indicates that MBSR is both a viable and a feasible approach for use with older people. As methodological concerns were evident, further research could usefully build upon this initial evidence-base. Chapter two: The empirical paper was a cross-sectional survey of community dwelling, English-speaking younger adults aged 18-34 years (n = 162) and older adults aged 65-93 years (n = 134), using a group difference design. In accordance with initial hypotheses and theories of lifespan development, the present findings indicated higher levels of dispositional mindfulness, affect and well-being for older adults, compared to younger adults. Future research and clinical implications are discussed. This may inform the development of age-appropriate mindfulness interventions. Chapter three: In the reflective paper, I share my experiences of conducting research with people in later life and of researching the topic of mindfulness. I then provide a reflective and reflexive account of the stages of research and go on to discuss implications for my professional and personal development.

Early data from North America supports the use of educational cardiac arrest debriefing as a strategy to improve the quality of cardiopulmonary resuscitation (CPR) in the hospital setting. As some debriefing approaches are challenging to deliver in the NHS setting, there was a need to develop debriefing approaches that are both effective and suited to NHS working practices. This thesis is modelled on the Medical Research Council framework for the development and evaluation of complex interventions. Undertaken between October 2011 and January 2015, it describes the development and feasibility assessment of three cardiac arrest debriefing approaches, which were specifically designed to be deliverable in NHS hospitals. Development work comprised three work packages (systematic review, process evaluation, qualitative study). These studies provided evidence to support the use of cardiac arrest debriefing, but showed that weekly group debriefing is undeliverable in many NHS hospitals. Through qualitative work, I identified six distinct mechanisms by which debriefing may affect clinical practice. Synthesis of these data led to the development of three cardiac arrest debriefing approaches (monthly group debriefing, individual oral debriefing, written feedback). We tested the feasibility of delivering these interventions by implementing them in three NHS hospitals (one intervention per hospital). In a before/after study, it was demonstrated that, despite practical challenges, interventions were deliverable in NHS hospitals. However, they were found to have no effect on either CPR quality or patient outcome. This finding was attributed to high performance in study hospitals at baseline. This thesis demonstrates that the developed cardiac arrest debriefing interventions are deliverable in NHS hospitals. It has also generated important new theory about the mechanisms by which debriefing may affect clinical practice. This thesis lays the foundation for future work to evaluate the clinical and cost-effectiveness of these cardiac arrest debriefing interventions.

Previous research has identified guilt as a significant emotion for both people with dementia and their caregivers. As guilt has been associated with psychopathology and depression, it is important to explore the nature, prevalence and clinical implications of this self-conscious emotion within the context of the dementia caregiving dyad. Chapter 1 presents a critical review of the quantitative and qualitative literature exploring feelings of guilt and caregiver burden in informal caregivers of people with dementia. The paper particularly focuses on evidence regarding the relationship of guilt to the construct of caregiver burden, the conceptualisation and measurement of guilt and burden in dementia caregivers and the factors associated with caregiver guilt and burden. Methodological limitations are discussed in relation to the clarity of the results. Clinical implications and future research suggestions are identified. Chapter 2 presents a mixed methods research paper on the development and validation of a measure of guilt for people with dementia. The results reveal strong item-total correlation in the new scale. Good reliability and convergent validity of the measure are also demonstrated. Study limitations, clinical implications and future directions are discussed. Chapter 3 offers a reflective account of my experience of the research process as well as my reflective learning, personal and professional development during this process and clinical training.

This thesis comprises a literature review, an empirical paper and a reflective paper. The systematic literature review involved critical assessment of eleven randomised controlled trials in which cognitive-behavioural therapy (CBT) for anxiety or depression was modified for older adults. The aim was to identify older adult-specific modifications as well as effectiveness in symptom reduction. Most commonly, modifications aimed to support understanding and homework completion, but few were age-specific. Studies comparing modified and standard CBT are lacking. Modified CBT may be more effective than wait-list, usual treatment and minimal contact. Efficacy is less clear compared with medication or an active control, and unknown compared with an alternative therapy. Future research should consider qualitative and quantitative evaluations of treatment effectiveness, and directly compare standard and modified CBT alongside an alternative therapy. The empirical paper describes an exploration of the lived experiences of volunteer befrienders to people with dementia. Individual semi-structured interviews were conducted with nine befrienders aged between 25 and 66 years. Transcripts were analysed using Interpretative Phenomenological Analysis. The notion of the person underlies the three main themes: ‘The person as the centre’, ‘A paradoxical relationship’ and ‘Core existential experiences’. Befrienders valued their befriendees’ happiness and individuality, and experienced some emotional challenges. The befriending relationship was experienced as special, paradoxical and complex, and enabled befriendees to reconnect with their individuality, community and humanity. Future research should explore issues including effectiveness, attrition and support. The reflective account summarises the helpful and challenging boundaries I experienced in conducting my doctoral research. The challenge for me was in knowing when to adapt something that was already there and when to be adaptable myself. Clinical training has developed a wide range of skills and competencies, which will be valuable in making adaptations and being adaptable as a qualified Clinical Psychologist.

Background: Saudi Arabia is ranked sixth for the highest prevalence of T2DM worldwide. There is very little information on dietary factors associated with the prevalence of T2DM in Saudi Arabia. There is an urgent need for the identification of culturally specific T2DM risk factors to then develop culturally tailored public awareness programs and interventions, to reduce the prevalence of T2DM. Objectives: To conduct a systematic review of the association between dietary factors and T2DM in the Middle Eastern region. To determine the association between dietary factors, including anthropometric measures, selected food items and beverages, selected dietary biomarkers and T2DM in Saudi adults. To identify culturally specific barriers to healthy eating in Saudi adults, with and without T2DM. Methods: The systematic review was conducted by searching several electronic databases and contacting authors, libraries, and research centres in the Middle East. Included studies assessed potential dietary factors for T2DM in Middle Eastern adults. An existing cross-sectional survey (n = 2631), which is part of a larger Biomarkers Screening Survey conducted in the urban area of Riyadh, Saudi Arabia (2009), of Saudi adults aged ≥ 18 years, was used to examine the association between dietary factors and T2DM. Anthropometric measures (n = 2355) included body mass index (BMI), waist circumference (WC), waist to hip ratio and sagittal abdominal diameter. Selected food/beverage included 17 items collected from a food freque ncy questionnaire (FFQ) (n = 1867). The FFQ was validated in this study against two 24 hour dietary recalls (n = 98). Dietary biomarkers included vitamin D and selenium (n = 567). Barriers to healthy eating were assessed by face-to-face interviews (n =108) carried out using modified predeveloped Saudi questionnaires and an additional questionnaire developed and piloted for the purpose of this study. Results: The systematic review highlighted the large gap in evidence of associations between dietary factors and T2DM in the Middle Eastern region in general and Saudi Arabia specifically. For the cross-sectional analyses, the overall sample was 2631 adults (females: 1280, males: 1351) and the prevalence of T2DM was 29.3% (females: 25.4%, males: 32.9%). WC was associated with T2DM independently of BMI, specifically in males. The intake of dates showed an inverse association with T2DM in males. Vitamin D levels were significantly higher in females with diabetes in comparison to nondiabetics. Serum selenium was associated positively with lipid parameters in females and fasting insulin in males. However, selenium was not associated with diabetes. Lack of dietary knowledge and culturally specific barriers (lack of social support, lack of will power and reliance) were barriers to healthy eating in participants with and without T2DM. Conclusion: The findings of this thesis highlight culturally specific factors associated with T2DM in Saudi adults. Further dietary studies in relation to T2DM are required in Saudi Arabia. Cultural issues should be incorporated when designing health awareness campaigns to address Saudis specific needs.

The aim of this thesis was to investigate the retrodictive mentalising abilities (a kind of backwards inference from a mental state to its causal antecedent in order to make sense of others’ behaviours) of people with and without Autism Spectrum Disorder (ASD). A new experimental paradigm was developed in order to examine people’s ability to make sense of others’ behaviours in a way that closely resembles the intricacies of real-world settings. The stimuli utilised in this thesis portrayed people’s spontaneous and genuine responses during four specific events (scenarios). People were told a joke in the Joke scenario whereas in the Story scenario the researcher related a series of unfortunate mishaps that she experienced earlier in the day. In the Compliments scenario, people were told a series of compliments while in the Waiting scenario the researcher performed irrelevant tasks during an experiment whilst the person was kept waiting. Participants viewed brief videoclips of these behavioural responses and were asked to determine which event had previously occurred to the people in the videoclips. Participants eye movements were recorded to ascertain the visual strategies used. Typically developing individuals successfully inferred the events that occurred by viewing brief samples of behavioural reactions of typically developing individuals (Experiment 1). It was found that scenario experienced did not impact how targets self-reported their level of empathic ability (Experiment 2). While people with ASD were able to infer people’s behavioural responses, their performance on the task was inferior as compared to typically developing individuals (Experiment 3). Participants varied their gaze strategies depending on the event experienced by the people in the videoclips and they had a tendency to focus more on the mouth compared to the eye region of the face (Experiment 1 and 3). When participants viewed videoclips of behavioural responses of people with and without ASD to the same events, they were more successful at inferring the reactions to the events which occurred when viewing videoclips of neurotypical individuals as opposed to individuals with ASD (Experiment 4). Furthermore, participants were unable to identify the reactions to two of the four events when viewing videoclips of people with ASD.

Automated analysis of the electrocardiogram (ECG) has been of increasing importance over the past five decades. Establishment of the normal limits of the ECG in any population is the first step before computer application in ECG interpretation in that population. Normal limits of the ECG have been established for different racial groups in the Western world using automated methods and subsequently followed up with the design of appropriate computerbased diagnostic software for ECG analysis. Unfortunately, this has not been the case for the indigenous black African nations even though this group represents a quarter of the world population. Earlier studies, using manual ECG analysis in Africans and indeed, more recent comparative studies between White and African Americans on the one hand and British and Chinese populations on the other have convincingly demonstrated racial variation in the ECG. The present study is therefore aimed at deriving normal limits of the ECG from a large population of healthy individuals living in Nigeria. This ultimately will serve as a template for the evolution of appropriate computer software applicable to ECG interpretations in black Africans. In the same vein, by using automated methodsmethods, normal limits of the ECG have been shown to be age and sex-dependent in different racial groups. All previous work on the ECG recorded in Nigeria and indeed, the West African sub region, has mostly quoted a single normal range of ECG variables for all ages. This is probably because robust computer application was never used to analyse the ECGs. For the first time in the indigenous black Africans, the present study aimed to establish age and sex-specific normal limits of the ECG in this population. The thesis initially reviews the number of ECGs that are recorded worldwide annually as a way of underscoring the clinical significance of setting out to establish the age and sex-dependent normal limits for Africans. Thereafter, a chronological review of the history and evolution of the ECG is presented, bringing into perspective, how, at each stage of evolution, the need and genuine desire to make things better had propelled creativity and invention. The history starts with the first recording of the human ECG in 1887, through the early 1900s when buckets of electrolytes served as ECG electrodes and ends at the modern day recording of the 12-Lead resting ECG using portable and sophisticated machines equipped with microprocessor chips for ECG interpretation. Subsequently, other modern applications of the ECG such as the signal averaged ECG, Holter ECG, exercise ECG and vectorcardiography are briefly reviewed before a discussion on the current status of the ECG and its challenges in today’s world which is propelled largely by technology. The historical evolution of computer applications in the field of ECG interpretation is also thoroughly reviewed since the present work is the first comprehensive application of automated ECG analysis undertaken in an apparently healthy Nigerian population and indeed, in any indigenous black African population. Different automated ECG analysis programs from different centres and applied to different racial populations are also presented and in the process, highlight the fact that there has been none adapted for black Africans, in spite of the well established racial dissimilarity in the ECG. The Common Standards for Quantitative Electrocardiography (CSE) project is also discussed to underscore how computer application has come a long way in the analysis and diagnostic interpretation of the ECG. The University of Glasgow ECG Analysis program (Uni-G) is also extensively discussed as this is the program which has been deployed in the analysis of the ECGs in this study. Since the program has already been designed to receive race as an input, the opportunity exists, after this study, to extend the program for the African population, using the new set of normal limits established in this study. The normal electrocardiogram is also reviewed from basic physiology to computer analysis in order to set the appropriate frame-work for the scope of the work done in the study. The normal ECG in different racial groups is also discussed before focusing on the history of ECG recording among Nigerians dating back to the early 1960s. iv iv In the present study, 12 lead ECGs were recorded using a Cardiac Science Atria 6100 electrocardiograph in and around Ilorin, Nigeria. A detailed description of the study site is provided in the Thesis. Apparently healthy volunteers were recruited from the University of Ilorin and from surrounding villages. Each was medically examined by the author and a detailed medical history obtained. A total of 1500 participants were initially recruited into the study of whom 239 were excluded from the final analysis due to technical inadequacies in ECG recording or the presence of an unexpected ECG abnormality such as a conduction defect. ECGs recorded in Nigeria were transferred locally to a PC and sent on a CD to the Central Core ECG Laboratory in Glasgow Royal Infirmary for further analysis, using the Uni-G ECG Analysis program. The ECG variable measurements output from the program were subjected to statistical analysis using SAS v9.1 in the Robertson Centre for Biostatistics. Plots and summary statistics were used to assess the relationship with age and sex. Regression techniques were used to assess formal relationships. Normal ranges were established by splitting the data into age and sex subgroups and by calculating the 96th percentile range within each subgroup. This was done for each and every one of the ECG variables which totalled well over 600. The normal limits of the electrocardiogram in Nigerians are presented and the results compared with existing databases for British and Chinese populations derived in the same laboratory where the analysis was done using the same ECG computer analysis program. The results presented confirm, beyond reasonable doubt, that there indeed are striking racial differences in the normal ECG. The study included 782 males and 479 females, all apparently healthy, with a relatively even spread of ages between 20 and 87 years. The normal limits of the R amplitude in lead V5 ranged between 0.64 to 3.45mV and 0.83 to 3.80 mV in males and females respectively. For S amplitude in lead V2, the normal limits ranged between -3.91 to -0.33mV and -3.1 to -3.8mV in males and females respectively. Voltages were generally higher in males than females except in lead I. The younger age group tended to exhibit higher amplitudes for the R wave but not for the S wave which tended to increase with advancing age, particularly in females. The Cornell product {(RaVL+SV3)xQRSd} was higher in all male age groups compared to corresponding female groups. It decreased with increasing age in males but the reverse was true in females. This is at variance with the trend in the British and Chinese populations with which this data was compared. The Nigerian population, in this study, generated higher voltages than either the British or Chinese populations. The 96th percentile range for the heart rate was 55 to 112 beats per minute in females and 53 to 105 beats per minute in males. Ninety percent of the entire study population had a heart rate between 60 and 100 beats per minute. Heart rate was higher in females than males in all age groups and increased with age. The mean heart rate was 81+14 beats per minute in females and 75+13 beats per minute in males (p<0.0001). A similar trend was observed in the British and Chinese populations. Nigerians, however, had the highest mean heart rates for the different age groups among the three populations. Mean QRS duration in males was 87.8+9.4ms and 83.4+7.7ms in females (p<0.0001). Variation of QRS duration across age groups was not statistically significant.