Background The Self-Efficacy Theory has been widely applied to research exploring chronic stress and in recent decades has proved promising in explaining the experiences of family carers of individuals with dementia. Aim To investigate the role of self-efficacy in the health-related quality of life of family carers of individuals with dementia, with particular consideration of the positive and negative impact related to caring, coping strategies and social support. Methods The sample comprised 289 family carers and individuals with dementia. The data collected were the baseline data for the SHIELD (Support at Home: Interventions to Enhance Life in Dementia) Carer Supporter Programme and included caring domain-specific self-efficacy beliefs, coping strategies, social support, positive and negative impact related to caring, and carer health-related quality of life. Analytic methods included correlations, t-tests, ANOVA, multiple regression and path analysis. Results Carers reported experiencing positive and negative emotions related to the caring experience and self-efficacy beliefs were found to have a “dual action” in both increasing positive impact related to caring, whilst buffering against negative impact. Carer characteristics and psychosocial resources were found to be determinants of self-efficacy beliefs, although caring stressors were not. In addition, it was found that self-efficacy, particularly for obtaining respite and controlling upsetting thoughts exerts a direct effect on carer quality of life, however only self-efficacy for responding to disruptive behaviours moderates the association between behavioural and psychological symptoms of dementia and carer quality of life. Finally, mediating effects of psychosocial resources including self-efficacy, coping strategies and social support were found in the association between stressors and carer quality of life. Conclusions Theoretically, findings have implications for informing conceptual models of carer coping and support the application of Self-Efficacy Theory to the caring experience. Clinically, findings assist in the design of effective carer interventions, specifically those directed towards enhancing self-efficacy.

Adolescence is a period in life that is characterised by substantial changes in the social environment. Compared to childhood, relationships with peers gain more importance and adolescents are particularly sensitive to peer influence. Adolescents, but not adults, show increased levels of risk-taking when in the presence of peers relative to when alone. Experimental studies suggest that heightened levels of risk-taking during adolescence might be specific to affective contexts (e.g. the presence of peers), while risk-taking in non-affective contexts remains stable or decreases. In Chapter 2 of this thesis the development of the impact of two decision variables (risk and valence) on decision-making in a non-affective context during adolescence is investigated in a behavioural risk-taking task. Chapter 3 employs a functional magnetic resonance imaging (fMRI) approach to examine the development of the neural correlates of social influence during adolescence. Previous studies have focussed on peer influence on risk-taking and little is known about the neural mechanisms of peer influence. This thesis examines whether heightened levels of sensitivity to peer influence during adolescence extend to the influence of a peer audience on tasks with either high-level (reasoning) or low-level (perceptual) cognitive components (Chapters 4 and 5). Chapter 4 investigates the effect of the presence of a peer audience on performance in a cognitively challenging task (relational reasoning), the development of this effect during adolescence and whether this effect is dependent on the identity of the audience (peer or non-peer). Chapter 5 examines the effect of the presence of a peer audience on performance in a low-level perceptual task to test whether peer audience effects would also extend to a low-level cognitive task. Chapter 6 investigates the modulation of brain activity during a high-level cognitive task (relational reasoning) by an evaluative peer audience in adolescents and adults.

Background: The varied effectiveness of Cognitive-Behavioural Therapy (CBT) and other psychological therapies in the treatment of persistent pain has led researchers to examine a range of factors that may predict treatment response. Studies have identified a diverse set of personal (i.e., age, gender, and marital status), specific (i.e., level of depression, treatment expectancy, and catastrophising), and non-specific (i.e., adherence to treatment) variables that appear to be associated with treatment outcome, although few with any consistency. This review summarises the available literature on predictors of outcome for persistent pain and discusses limitations of such studies.;Data Sources and Review Method: The persistent pain literature was examined to identify proposed variables that predict response to multicomponent treatment for persistent pain. PsycINFO and Medline databases were searched spanning articles published from 1990 to 2006 of treatment studies and review articles.;Results: Very few studies that focused on predictors of outcome following CBT for persistent pain in the adult population were found through the database searches. Therefore, in addition, a hand-search of key journals, of reference lists from searched articles, and of relevant books was completed and articles prior to 1990 were consequently included in the review at the author's discretion. All types of persistent pain were included except headache.;Conclusions: Main results: The review showed that of the many personal variables investigated to influence outcome, only socioeconomic status (SES), social support, and applying for compensation or disability pension consistently predicted outcome. Of the specific variables, changes in catastrophising, self-efficacy, pain-related fear, pain tolerance, pain helplessness and perceived disability have each been shown to consistently predict outcome. Few studies were found that had investigated the role of adherence during or after treatment on treatment outcome. Finally, suggestions for future research are discussed.;Keywords: Predictor(s), factor(s), cognitive-behavio(u)ral therapy, chronic pain, persistent pain, treatment outcome.

This paper reviews research studies examining befriending as an intervention for adults with mental health problems. It first considers contextual issues and psychological theories relevant to befriending. Because of the small body of published papers on befriending, the review used deliberately broad inclusion criteria which were based on: (a) the characteristics of the befriending intervention, (b) the target problem, and (c) the research design. Fourteen studies met the inclusion criteria. The studies fell into three types according to their methodological design: (1) group comparison designs, (2) process evaluation and user satisfaction survey studies, and (3) befriending as a control comparison. Overall, the evidence from these studies suggested that receiving befriending support has potential benefits for adults with mental health problems, for example, increased self esteem and social functioning, and remission of symptoms. However, little is known about the processes occurring within the befriending relationships that may lead to the benefits suggested by the current literature and the literature gives limited indications about for whom and under what circumstances befriending might be most beneficial. Implications for further research are discussed.

Cognitive impairment is a core feature of schizophrenia, and has been found to be associated with an individual's ability to function independently in the community. This thesis will being by reviewing the existing literature in this area. This will cover general features of the cognitive impairment that is associated with schizophrenia, as well as the three-way relationship between cognitive impairment, ability to function in the community and psychotic symptomatology. Because of the role that cognitive deficits have been found to play in determining an individuals functional outcome, it will be argued that these deficits need to be routinely assessed for all patients with schizophrenia. The empirical paper describes a study that involved looking at the relationship between performance on a brief measure of cognitive functioning, the Brief Cognitive Assessment (BCA) (Velligan et al., 2004), and ability to function in the community in a sample of patients with schizophrenia. The use of the BCA as a measure of cognitive impairment in this population was explored, and contributions of both symptoms and cognitive deficits to functional outcome were investigated. In the final part of this thesis, the critical review, the importance of routine cognitive assessment for individuals with schizophrenia will be restated, based on the findings of the literature review and the empirical study. This section will include a personal reflection on the experience of both the research process and the use of the BCA with this population, and a summary of the strengths and weakness of this measure.

Persecutory delusions can be conceptualised as beliefs about the intentions of others. Contemporary models of persecutory delusions have different approaches to understanding interpersonal processes in paranoia. The current paper provides a critical review of the theoretical approaches and the empirical evidence on interpersonal processes across the continuum of paranoia. The three main models of persecutory delusions (Bentall and colleagues' Delusions-as-defence model, Freeman and Colleagues' Threat anticipation model and Trower and Chadwick Interpersonal theory of the self) propose that the social environment is involved in the development of persecutory delusions but they put forward different mechanisms to explain this. The evidence on interpersonal processes is organised in five main areas: interpersonal life events, attachment, schematic beliefs about relationships, social cognition and interpersonal behaviour. The review concludes with a summary of the evidence in relation to the three theoretical approaches, a discussion of methodological issues and the implications for future research.

Research Aim: To understand crises in the homes of people with dementia living with their family carers and crisis interventions which support families in crisis. Methods: A range of research methodologies were undertaken with a variety of stakeholders to understand crises in the homes of people with dementia. These included three systematic reviews, focus groups, narrative inquiry analysis, national stakeholder questionnaire and two discrete choice experiments (DCEs). Results: The systematic review looking at risk factors associated to hospital admissions for people with dementia found that physical health problems, such as falls/ fractures and infections were highly associated with admissions for people with dementia. Family carers often accepted BPSD symptoms as part of their relative’s dementia diagnosis and it was the physical health problems and vulnerability issues, such as risks/ hazards in the home which concerned people with dementia and their families. Health/ social care professionals’ opinions were similar to the family carers but the DCE revealed that behavioural problems such as aggressive behaviour were significant risk factors of crisis according to staff. The systematic reviews suggested that crisis interventions help to reduce institutionalisation and number of admissions to hospital. The other studies revealed that highly costly medical interventions did not significantly improve matters for family carers and their relatives and there was a strong preference towards interventions which enabled families to maintain their independence for longer at home as well as empower them to cope more effectively in a crisis through education & training, practical home adaptations and assistive technology. Conclusion: It is hoped that the model of crisis process, risks and interventions which has been proposed in this PhD may become a useful tool for researchers, policy makers and clinicians to help understand the complex process of crises involving people with dementia and how to intervene with a crisis.

Rationale: Whilst relapse to drug taking, even after a substantial period of abstinence, is common, laboratory-based research documenting precursors to relapse in humans is limited.;Objectives: The purpose of this literature review is to evaluate evidence from both animal and human studies on the effect of acute stressors on drug craving and relapse.;Method: Attempts at conceptualizing the definition of stress have been unsuccessful (Chrousos & Gold, 1992). For the purposes of this review, stress was broadly defined in line with Piazza & Le Moal (1998) as coerced exposure to environmental conditions or events that would normally be considered aversive enough to motivate avoidance. This review considered both physical and psychological stressors and is limited to opiate (specifically heroin and morphine) and psychostimulant drugs (e.g. cocaine, amphetamine) in animals, and alcohol, opiates, nicotine, and cocaine in humans.;Conclusions: Results from animal studies of acute stress and relapse are inconsistent. Human studies, despite providing more consistent findings, have a range of methodological limitations. A greater understanding of how stress precipitates relapse is likely to have a significant impact on the way clinical interventions are offered to substance misusing individuals.

Introduction: Provision of support for people bereaved by suicide has become a key priority for suicide prevention strategies in many developed countries. Few studies have measured whether suicide bereavement increases risk of suicidal behaviour compared with bereaved controls. Methods: I sampled 659,572 staff and students at 37 UK higher education institutions in 2010. Via mass email, I invited adults who had experienced a sudden bereavement to complete an online survey measuring post-bereavement suicidal ideation and attempts, and other psychosocial outcomes. Inclusion criteria were: current age 18-40 years, and sudden bereavement of a close contact since the age of 10 years. Multivariable regression was used to compare those bereaved by suicide to two reference categories: those bereaved by natural causes and those bereaved by unnatural causes. Results: Of 3,432 eligible respondents, 614 adults were bereaved by suicide, 712 by sudden unnatural causes of death, and 2,106 by sudden natural causes. Compared with adults bereaved by natural causes, adults bereaved by suicide had a similar risk of suicidal thoughts, poor social functioning, non-suicidal self-harm and incicdent depression, but a significantly increased risk of suicide attempts (AOR=1.65; 95% CI=1.12-2.42; p=0.01), drop-out from work or education, and subjective stigma. Compared with adults bereaved by unnatural causes, adults bereaved by suicide had a similar risk of all the above outcomes, except for an increased risk of poor social functioning and subjective stigma. For all the associations identified, risks were elevated in both relatives and non-relatives. Discussion: My findings suggest that there are some risk similarities in suicidality between young adults exposed to suicide bereavement and those exposed to other violent bereavements. Implications: The needs of young adults in the UK bereaved by unnatural deaths may also need consideration in suicide prevention policy.

Conceptualisations of recovery from psychosis have evolved over time, from medically defined models of symptom reduction, to more recent definitions of subjective recovery. First-episode psychosis (FEP) individuals present as a group who are in the early stages of adjustment to the experience of a psychotic illness. Due to the typically early age of onset, they are also in the process of adjusting to major life and role changes. This paper addresses how recovery from FEP has been conceptualised and measured in the literature, and focuses on the validity of applying such measures and concepts specifically to an FEP group.;Key terms: First episode psychosis, subjective recovery, recovery and early intervention in psychosis.