Search

Background: Pain is a highly complex sensory and emotional experience; the biological, psychological and social aspects must each be considered. The intersection between the phenomenon of pain, the unpredictable pre-hospital environment and children is highly convoluted. Studies have shown that pre-hospital pain management in children is poor, despite access to pain management being considered a fundamental human right. Without effective pain treatment, children may suffer long-term psychological changes (e.g. altered pain perception) and are at risk of developing post-traumatic stress disorder. The aim of this thesis was to identify predictors, barriers and facilitators associated with effective pre-hospital pain management in children suffering acute pain and to identify ways to improve the quality of care. Methods: A postpositivist paradigm was adopted for the study, with a critical realist ontology and a modified objectivist epistemology. A mixed methods sequential explanatory design was adopted, informed by a systematic mixed studies review. The initial quantitative study employed a multivariable logistic regression analysis using routinely collected clinical data to identify predictors of effective pain management. The final qualitative study used face-to-face semi-structured interviews with ambulance clinicians to help explain the identified predictors, identify barriers and facilitators and explore ways to improve the quality of care. Interviews were audio recorded and transcribed verbatim with thematic analysis used to analyse the data. Results: The systematic mixed studies review included 13 studies (8 quantitative and 5 qualitative) and highlighted the importance of analgesic administration. The initial quantitative study included 2312 clinical records; only 39% of children suffering acute pain achieved effective pain management. Predictors of effective pain management included children who were younger, administered analgesics, attended by a paramedic or living in an area of low or medium deprivation. The final qualitative study included 12 ambulance clinicians (9 paramedics and 3 emergency medical technicians) who provided possible explanations for these disparities. Novel barriers and facilitators were also identified along with ways to improve pain management. Meta-inferences were developed which provided a more comprehensive understanding of this complex phenomenon. To improve pre-hospital pain management in children, the following recommendations were made; 1) explore methods to increase rates of analgesic administration, perhaps by utilising the intranasal and inhaled route; 2) reduce fear and anxiety in children, perhaps by using child friendly uniform, non-pharmacological techniques and more public interaction and 3) reduce fear and anxiety in clinicians, by enhancing training, optimising crew mix and developing a more pragmatic pain assessment tool. A theoretical model of pre-hospital pain management in children was developed as part of this thesis. Conclusion: Pre-hospital pain management in children may be improved by increasing rates of analgesic administration and reducing the fear and anxiety experienced by children and clinicians. Future research should explore the experience of the child and determine the most important outcome measures. Robust clinical trials are needed to determine the efficacy and safety of intranasal (fentanyl/ketamine) and inhaled (methoxyflurane) analgesics in the pre-hospital setting. Investment in future research and intervention development is imperative; we need to make children's pain in the pre-hospital setting matter.

Emergency readmission is defined within the NHS as an emergency admission to hospital within 30 days of discharge. Excess readmissions are undesirable in terms of care quality and efficiency; yet, despite financial incentives for improvement, reports of increasing readmission rates continue. There is evidence that pharmacist intervention can prevent medication errors, discrepancies and adverse drug events; which can each contribute to readmission. The purpose of the work in this thesis was to develop a model based on routinely collected prescription information to enable the pharmacy team to estimate readmission risk in the clinical setting, thereby facilitating appropriate prioritisation of potentially preventative intervention. A multiple logistic regression model for estimating readmission risk using routinely recorded prescription information among patients discharged home from the medical short stay units of one NHS Trust was developed, and survival analysis was undertaken to characterise readmission behaviour in relation to the predictors. The readmission rate was 18% (220/1240). Readmission risk increased with increasing age and polypharmacy: each additional medicine prescribed increased the odds of readmission within 30 days by eight per cent and each additional year of age increased the odds of readmission within 30 days by two per cent. Each additional medicine prescribed decreased the time to readmission by seven per cent and each additional year of age decreased the time to readmission by one per cent. Over one-third of readmissions occurred within one week (73/200) and more than half (114/200) occurred within two weeks, supporting that identification of those at risk and intervention to prevent readmission should be provided promptly. The predictive model developed is suitable for application on admission and could therefore enable clinicians to identify the patients most likely to require intervention to prevent readmission before they are discharged home from hospital, thereby maximising the time available to organise and/or provide the necessary support. Although the logistic regression model improved accuracy by 36% compared to indiscriminate intervention whilst identifying 70% of patients who would be readmitted, it had relatively weak discriminative capability (c-statistic 0.637). It may be the case that clinical intuition is as effective for predicting readmission and further research should be undertaken to confirm whether this is the case.

This thesis examines how clinical handovers between ambulance and emergency services were conducted. Clinical handovers are an exchange of patient information and responsibility from one healthcare team to another. They have been a key area of concern due to potential patient safety issues arising through non-technical human factor skills of communication and teamworking. However, there remains a lack of research that has examined clinical in the context of multidisciplinary settings. To redress this gap in knowledge this thesis adopted a video analysis methodology of pre-existing handover videos. Data was derived from television programs and provided naturally occurring instances of handovers being conducted. The video analysis approach was underpinned by conversation analysis, which allowed an examination of interactional features used by team members to structure handover activity. The three analytical chapters that make up this thesis how the handover activity was carried out: 1) the clinical handover structure, 2) epistemic knowledge claims, and 3) embodied actions. Overall the findings illustrated the different conversational tools used by interdisciplinary team members which shaped how information was communicated. It showed how speakers would use resources to develop engagement during discussions that aided the handover process. This showed that the handover went beyond a structured institutionally derived activity, but one shaped by the interactants. This thesis has added to the knowledge of how the interactional order of clinical handovers predominate irrespective of the continuous attempts to standardize the activity. These findings showed how handovers are carried out through an analytical lens of “work as done” rather than “work as imagined” providing understandings that can inform practice and shape future research directions.

The Emergency Department (ED) is a challenging environment to deliver high quality and safe care. Clinicians are continuously challenged by uncontrolled volume, uncertain acuity, limited information compounded by constant time pressure. Within this environment experts successfully navigate these challenges and resolve complexity. These highly calibrated clinicians have acquired procedural, technical and cognitive skills to predict, mitigate and resolve clinical problems for individual patients to multi-trauma resuscitations. Expert Decision Making in this domain remains opaque but exploring and understanding these tacit skills has the opportunity to bridge the novice to expert divide with appropriate educational interventions and improve safety and performance in the ED. This thesis presents a series of three interlinked studies designed to explore expert decision making in Emergency Medicine (EM). This is supported by a narrative review of the literature relating to EM as a foundation, on which I then set out my epistemological stance and justify my selection of methods. The first study examines the perspective of Foundation Doctors and the challenges they face working in the ED. Using an electronic survey their experience of clinical decision making, the opportunities to optimise their decision making by training and enhanced understanding of expert (Consultant) Decision making are explored. The second study explores how expert decision makers based in three health boards describe their decision making using semi-structured interviews. Interpretative phenomenological analysis enables the cognitive constructs deployed by fifteen Consultants in EM to be explored. The final study uses Video Reflexive Phenomenology as method to triangulate and refine the data developed during study 2. In this study ten consultants from the ED in the Royal Infirmary in Edinburgh select and then review a challenging clinical event. I will then propose a model of Expert Decision Making that has been developed and refined during the course of the studies. The strengths and limitations, implications and finally recommendations for future study will then be shared.

Section one of this thesis is a meta-synthesis of qualitative literature exploring the emotional experiences of emergency ambulance personnel. Thorough searches of the literature were conducted in key databases, resulting in 16 studies being included in the synthesis. A meta-ethnography approach was followed, resulting in three overarching themes: moral distress, the emotional intensity of ambulance work and emotional regulation. The findings are considered in relation to related literature and theories. Several clinical implications and recommendations for future research are discussed. Section two is a qualitative study exploring the emotional experiences of emergency ambulance call handlers. Call handlers are the public’s initial point of contact to UK ambulance services and have critical information to obtain, in order to arrange an appropriate response. Twelve call handlers participated in semi-structured interviews, which were analysed using Interpretative Phenomenological Analysis. Three superordinate themes were identified: being more than a robot: “you’re not a robot and neither is the person on the end of the phone and neither is the patient”, belonging to a team: “we call each other the green family” and the unknown: “if you try and think about what’s going to happen next, you’d not be able to come to work”. The results are discussed in relation to theoretical perspectives, relevant literature, clinical implications and future research recommendations. Section three is a critical appraisal of the research process offering reflections on the ambiguity around the call handler role, the researcher’s previous experience, call handlers’ interest in the research and the potential relationship between emotional suppression and identifying or naming emotions. The complexities of managing the roles of a researcher and a practising clinician and the potential role for clinical psychologists in ambulance services are also discussed.

With an increasing use of health services in England there is a political desire to alter demand patterns and re-design services to reduce spend. With a yearly 4% increase in ambulance service use a variety of generic health education initiatives have been tried to reduce demand with seemingly limited effect. Calling an ambulance when you have a perceived need can be considered a health behaviour and to effectively intervene to alter this behaviour requires an understanding of variation in population use. This study analysed 769,376 emergency calls in the East of England including those not conveyed to hospital by three acuity levels. Population use of the ambulance service was considered in relation to sociodemographic factors derived from the census. A significant regression was found (F(27,3582)=44.81, p < 0.001), with an adjusted R² of 0.2469. Population factors related to higher utilisation included age groups 85-90 and 90 & over, mixed and black ethnicity, and those who had never worked or were long-term unemployed. The study examined, using moderated regression analysis, the influence of self efficacy, access to services, general health status and social networks on the relationship between socio-demographics and ambulance utilisation. The findings suggest that access to services and general health status of the population act as significant moderators, but self efficacy and social networks are not significant moderators. Policy interventions suggested involve targeting specific population types related to higher utilisation to manage demand. The original contribution of this study is the further development of understanding ambulance utilisation through the use of large datasets within the English NHS context.

Since the September 11, 2001 terrorist attacks in the USA, government agencies and professional societies have focused greater attention towards the hospital response, particularly Emergency Departments (EDs), for emergency preparedness and response. In the UK, the Civil Contingencies Act lays out responsibilities on NHS organisations to uphold civil protection. The ED is a complex, hectic, and high-pressured environment. Chemical, Biological, Radiological, and Nuclear (CBRN) events are multi-faceted emergencies and present numerous challenges to ED staff (first receivers). This thesis used Hierarchical Task Analysis (HTA) to explore whether there were differences between operational procedures and practice using the Human Factors/Ergonomics (HF/E) approach for Work as Imagined (WAI; procedures) and Work as Done (WAD; practice). The aims were to 1) Explore whether HTA could be used to map complex socio-technical systems and 2) Make recommendations to improve the ED response to CBRN events. Study 1 (WAI) took a top down approach for document analysis of standard operating procedures (emergency plans) at two Type 1 NHS Trust EDs. The plans were analysed thematically to create colour coded HTA representations of the CBRN plans. Differences in General Organisational Responsibilities (GORs) were found between the two Trusts; Trust A had 13 GOR themes and Trust B had 20 GOR themes, of which 13 overlapped. The emergency plans include role-specific action cards used to support individual tasks; each card (n=30) was analysed and represented as HTAs. An Ergonomics review of the HTAs as part of the coding process produced higher level (standardised) themes at both Trusts as: 1. Prepare to respond to CBRN incident (Trust A and B). 2. Respond to CBRN incident (Trust A and B). 3. Initiate recovery from CBRN incident (Trust A). 4. Document CBRN incident (Trust A). Study 2 (WAD) collected data with semi-structured interviews from 57 first receivers. Scenario cards were used as prompts to explore clinical and operational practice. The field notes were converted to HTAs and thematically analysed. The synthesis of these themes provided a bottom up perspective to consider standardisation based on WAD (how first receivers respond to CBRN events). Similarities between the Trusts were evident in the importance placed on actions for isolate and contain, liaise and communicate, and escalate. Differences were found in the importance placed on Personal Protective Equipment (PPE), decontamination, treatment, and investigation of the presentation. The results for WAI and WAD were compared. WAI focused on actions such as documentation, checking, and timing; WAD prioritised patient needs through assessment, treatment, and diagnosis. Finally, the findings were synthesised into standardised recommendations as: • A CBRN planning checklist. • Action card template. • Example of a meso-system. • CBRN response framework. The WAI vs WAD framework provided a robust theoretical framework to identify gaps and improvement opportunities in the ED CBRN response. This research shows that HTA can be used to unpack a multifaceted event in a complex environment.

This research examines public behaviours, preferences and perceptions in relation to three key issues: 1) preparing for and 2) responding to a major incident; and 3) how emergency services interact with the public during their response. These areas are of great importance for reducing risk and improving public safety but are challenging to study due to difficulties in gaining access to casualties and the unpredictable nature of major incidents. Chapter one of this thesis provides a narrative review of literature in the field, comprised of 45 research articles that discuss how casualties behave and how they are managed during major incidents, along with how prepared people are for these extreme events. Chapter two is a methodological chapter, which explains the benefits and limitations to collecting data in live exercises. This chapter details the three unique, large-scale live exercises that were used to collect data for analysis in chapters three, four and five. Data collected from these three live exercises was made into one dataset that analyses all three data chapters and comprises of 265 interviews from people playing the role of casualties and 499 surveys from people playing casualties in the live exercises. Chapter three explores the experiences of people and the types of behaviours exhibited during these exercises and why this may be. A thematic analysis of the dataset for this chapter revealed 3 superordinate themes. One superordinate theme was 'expected and experienced emotions' as people expected to feel or observe certain emotions, such as panic, but that was not observed or felt by anybody in the exercises. People also expected to observe different experiences such as chaos, but instead the environment seemed calmer meaning people at times had to alter their expectations of a major incident. It was also found that people felt the exercise was realistic and did feel negative emotions such as uncertainty but also felt more positive and appreciative towards the emergency services following their participation in the exercises. 'Expected communication' was the second superordinate theme as the type of information people expected to be told was frequently discussed. Overall more information was requested but it was felt that basic information, reassurance and clearer instructions would especially enable people to cope better and feel less unsure and uncertain. Lastly, the third superordinate theme of 'experiences and expectations from the environment' detailed how people were feeling unsettled and unfamiliar with the major incident environment especially whilst awaiting rescue. This chapter also supports the idea of casualties helping other people whilst awaiting for the first responders. Chapter four looks at casualty behaviour in relation to interacting with each other and focuses on the benefits of helping one another in major incidents. A thematic analysis of this chapter revealed 5 main themes (i) physical benefits to casualties helping each other, for example somebody could limit blood loss before the emergency services arrive (ii) psychological benefits to casualties helping each other, such as it can avoid feeling isolated (iii) benefits of helping for the recovery phase of the incident (iv) limitations reported by casualties when helping others, difficult personalities for example and (v) how to enable people to help each other, examples include providing people with adequate instructions or equipment. Chapter five uses a mixed approach, looking at both the perception (how people felt about the emergency services) and preference (how people would want the response of the emergency services to be) aspect of the research question by drawing on preand post-exercise surveys and interview data to examine people's experiences of emergency service interactions. Wilcoxon signed rank tests reveal that for 2 out of 3 exercises the level of trust and confidence people had for the Police, Ambulance and Fire and Rescue Service decreased post event, although it still remained high. A thematic analysis of this chapter supports the quantitative analysis and previous research by finding that a lack of emergency communication impacts the public perception of confidence in emergency services, and willingness to comply. The majority of themes indicate that there were a number of preferences people had that would improve the perception they hold of the emergency services; these themes were to (i) be friendly and reassuring (ii) be informative and iii) to give clear instructions. The remaining two themes of 'improving perceptions' and 'current perception', show how people believe that more live exercises could help improve the relationship between the public and the emergency services, although people still have a high amount of respect, trust and confidence for the services. Chapter six uses data collected from members of the public taking part in a survey or focus group and aims to examine what factors affect motivation to engage in preparedness action and people's preferences for receiving information on how to prepare in advance of a major incident. The 121 surveys collected and 9 focus groups (consisting of a total number of 36 participants) were analysed using a content analysis. Findings show that people have a low awareness of risk information and a mixed perception of receiving information due to beliefs that some of the information kept from them. Preferences for receiving information included receiving actionable (what can I do) information through a range of different formats to reach different people. Furthermore, from the qualitative analysis it was found that assessment of threat is a motivator that's affects people's decision to become prepared and to take preparedness action as people can be complacent if they do not believe the threat is credible. Other motivators to becoming prepared that were found include sense of community, personal responsibility and self-efficacy. Overall, whilst most of the findings are from members of the public role-playing casualties in simulated rather than real major incidents, findings extend current theories of casualty behaviour. In particular, findings support the Elaborated Social Identity Model of Collective Resilience (Drury & Reicher, 1999; Drury & Reicher, 2000)., and disprove myths that the common public response to emergencies is to panic. Findings regarding public preferences and perceptions pose implications for emergency services, providing concrete, evidence driven recommendations for how casualty communication could be improved in order to increase compliance, cooperation and public perceptions of emergency services.

Background: The nature of responding to or experiencing situations where first aid is indicated during off-duty time has been the subject of anecdotal debate in the healthcare professions. Despite this, very limited primary evidence exists in nursing, midwifery and the wider professions. Aim: To explore the nature of UK nurses’ and midwives’ experiences, beliefs and perceptions about responding at off-duty situations where first aid may be required with a view to developing an in-depth understanding of the area. Methodology: A constructivist grounded theory approach was employed to provide an in-depth exploration of 16 nurses’ and midwives’ views about responding to off-duty situations where first aid may be required. Within this broad context the research focus was one of open inquiry due to the paucity of primary evidence. The main sample was selected via a participant referral process that took place over a 2-year period. Loosely structured interviews enabled the discovery of rich data resulting in theme construction that led to the development of a substantive grounded theory. Findings: A core enduring in vivo theme, 'The Right Thing to Do', emerged as a central and consistent conceptual reality constructed via three key in vivo themes; 'Something I've Heard', 'Am I Covered?', 'Just Who I Am', each with a number of sub-themes. A pervading anxiety about responding at off-duty situations requiring first aid was persistently evident across these themes. Discussion: The study illuminates an area that has previously been the subject of largely anecdotal debate. The substantive theory of 'doing "The Right Thing" in a climate of anxiety' explores and illustrates the issues and tensions that exist surrounding the off-duty response. Implications and recommendations for practice and education curricula focus on the fostering of knowledge and understanding of professional identity, position in law and scope of practice, together with potential future research directions. This study provides the first qualitative primary evidence, and the second overall research study in the area, iv contributing a significant new perspective to a key area of practice, both nationally and globally. Limitations are acknowledged and outlined. Conclusion: This study found a strong sense of moral agency among nurses and midwives, despite a powerful underlying feeling of anxiety surrounding broader issues of urban myth, protection, and personal and professional identity. The substantive theory emerged as 'doing "The Right Thing" in a climate of anxiety'.

Introduction: Each year, 35% of people over the age of 65 experience falls and approximately 45% of those are aged over 85 and live in the community. Local unpublished data from an English ambulance service found that following ambulance call-outs, 24-32% of the patients were not conveyed to hospital and of those, only 32-52% were referred to the local Falls Prevention Service for a specialist multifactorial assessment (NICE, 2013). Background: The literature review undertaken supports the need for a falls pathway (guidance for practitioners) to be in place which is fully utilised, by all practitioners. However, it did not uncover the reasons for low referral rates of non-conveyed patients to Falls Prevention Services. This thesis will outline the design of a study which aims to, gain an in-depth understanding of the falls patient journey from patient and ambulance crew perspectives and generate a clear understanding of the ambulance service customs and practices which could inform improvement of the existing falls pathway. Methods: A critical ethnographic approach enabled participants’ values, behaviours and beliefs to be explored. The methods were participatory observation, semistructured interviews of patients/carers and ambulance crew and in-depth field notes. The sample was people over 50 years of age who had fallen or their carers (n= 10) who had been seen by the ambulance service. Ambulance crew were also recruited (n= 10). Results: The study gained an in-depth understanding of the experiences of crew and patients/carers. For both ambulance crew and patients/carers, there are two shared themes ‘falls journey’ & ‘falls not being a problem for patients’. For crew ‘training’ was a theme and for patients/carers ‘language’ and ‘patient transport’ are themes. The study’s unique contribution is that it has gained an in-depth understanding of the patient journey from patient and ambulance crews’ perspectives and a better understanding of the falls pathway. In summary, the falls pathway in the local area was clearly understood and followed by all crew observed. The study shared the challenges of recruiting in a hard-to-reach group.

The aim of this thesis was to identify and prioritise evidence-based interventions that reduce emergency hospital admissions and explore their implementation in clinical practice. To achieve its aim, this thesis first identified evidence-based interventions that reduce emergency admissions using a systematic review of systematic reviews. Second, it determined which of the evidence-based interventions had been recommended in national clinical guidelines, developed into quality of care indicators, and featured in national audits in England. Thirdly, it explored the extent to which evidence- and guideline-based interventions were being optimally utilised in primary care in the United Kingdom. Fourthly, it evaluated an innovative access to care initiative implemented in primary care to reduce hospital admissions. Fifthly, it explored the use of evidence by decision makers when designing and implementing local interventions as part of national campaigns to reduce admissions. Overall, the findings showed that there are a variety of evidence-based interventions that reduce emergency hospital admission rates. However, the thesis showed gaps in the promotion and use of these interventions in clinical practice in the United Kingdom. It also demonstrated that increasing access to primary care using innovative interventions may not be an effective or evidence-based approach to reducing hospitalisations. Stakeholders seeking to reduce admission rates may consider optimising use of the evidence- and guideline-based interventions prioritised in this thesis. This may be done using existing quality improvement infrastructure or by developing or expanding health care services that feature the interventions. Furthermore, efforts should be made to robustly evaluate novel interventions implemented in clinical practice to reduce hospital admissions. The results of such evaluations should be systematically disseminated to facilitate peer-peer learning, avoid duplicated efforts, and promote use of evidence-based interventions.

The aim of this thesis is to advance the understanding of how the relationship between health and socioeconomic status changes over the life-course and over time. Three different studies which address several important aspects of the socioeconomic inequalities in health are presented. First, the existence of a socioeconomic gradient in functional health at the end of life in England is investigated. The objective is to contribute to the empirical evidence concerning the existence of a disappearing association between socioeconomic status and health in old age, by analysing whether socioeconomic status is associated with disability trajectories in the final years of life. The second study examines how the changing demographic composition of different educational groups over time can help to explain the widening educational gap in self-reported health in the United States, connecting family demography with health inequalities. The third study explores whether, and to what extent, education has a causal effect on cancer outcomes in England and Wales. This thesis contributes to the existing literature in at least three ways. First, it conceptualises the last years of life as a life stage where socioeconomic inequalities in health might emerge. Differences in the disability-related pathways preceding death were found only wealth. Second, it highlights the increasingly important role of demographic factors and marital status in particular, when studying health inequalities over time. Indeed, the findings suggested an increasingly relevant contribution of marital status over time for educational inequalities in health in the United States, despite marriage representing only a small part of the explanation. Furthermore, it investigated for the first time the causal effect of education on survival after cancer diagnosis, providing evidence for the absence of a causal impact of an additional year of compulsory schooling on cancer incidence and survival after two educational reforms.

Actions taken immediately following a life-threatening incident are critical for the survival of the patient. In particular, the timely arrival of ambulance crew often makes the difference between life and death. As a consequence, ambulance services are under persistent pressure to achieve rapid emergency response. Meeting stringent performance requirements poses special challenges in metropolitan areas where the higher population density results in high rates of life-threatening incident occurrence, compounded by lower response speeds due to traffic congestion. A key ingredient of data-driven approaches to address these challenges is the effective modelling of ambulance movement thus enabling the accurate prediction of the expected arrival time of a crew at the site of an incident. Ambulance mobility patterns however are distinct and in particular differ from civilian traffic: crews travelling with ashing blue lights and sirens are by law exempt from certain traffic regulations; and moreover, ambulance journeys are triggered by emergency incidents that are generated following distinct spatial and temporal patterns. We use a large historical dataset of incidents and ambulance location traces to model route selection and arrival times. Working on a road routing network modified to reflect the differences between emergency and regular vehicle traffic, we develop a methodology for matching ambulances Global Positioning System (GPS) coordinates to road segments, allowing the reconstruction of ambulance routes with precise speed data. We demonstrate how a road speed model that exploits this information achieves best predictive performance by implicitly capturing route-specific patterns in changing traffic conditions. We then present a hybrid model that achieves a high route similarity score while minimising journey duration error. This hybrid model outperforms alternative mobility models. To the best of our knowledge, this study represents the first attempt to apply data-driven methodologies to route selection and estimation of arrival times of ambulances travelling with blue lights and sirens.

The aim of this thesis was to evaluate the centralisation of emergency medical services (EMS) in different local hospitals into a single specialised emergency care hospital in terms of costs and quality of care. It also aimed to investigate preferences and trade-offs that individuals' were willing to make to receive treatment at the centralised specialised emergency hospital. The economic evaluation method builds upon a systematic review of economic evaluation methods and types of economic evaluation that have been used to assess the performance of centralised healthcare specialities. A discrete choice experiment (DCE) was used to investigate preferences for centralised EMS and the trade-offs individuals were willing to make to receive treatment at the centralised hospital. The DCE identified preferences for: shorter travel times to the hospital; shorter waiting times; fewer days in hospital; low risk of death; low risk of readmission; and outpatient follow-up care in local hospitals. However, people were willing to trade-off increased travel time and waiting time for higher quality of specialised emergency medical care in the centralised hospital. A Markov model was developed to evaluate the costs and effectiveness of centralisation of EMS compared with non-centralised care. Multiple sensitivity analyses were carried out to assess whether centralisation had an impact on cost, quality and cost-effectiveness over the short and longer term. The incremental cost per QALY at one year (deterministic estimate £1,004 per QALY) and 10 years (deterministic estimate £636 per QALY) were both well below the threshold used by the National Institute for Health and Care Health Excellence (£20,000-£30,000 per QALY). There were fewer deaths in the centralised EMS compared with noncentralised services (deterministic estimate: 31.47 fewer deaths at 1 year, 31.57 fewer deaths at 10 years). Discounting the costs and outcomes at different rates did not alter conclusions. The economic evaluation suggested that centralisation of EMS into fewer more specialised units could be cost-effective, although cost-effectiveness may vary in specific population sub-groups. Sub-group analyses suggest that centralised EMS would be more cost-effective for elderly patients, the most economically deprived patients and those presenting with diseases of the circulatory system. These findings ii support the recommendations to centralise urgent and EMS in England. However, a cost benefit analysis that incorporated the results of the DCE into the economic evaluation suggested that centralised EMS could have negative societal value when compared with services provided in local hospitals. The implications of these findings, potential limitations of the methods used in this thesis and recommendations for future research are discussed.

Background: Trauma is a significant public health problem both in the UK and beyond. It can have a devastating impact on individuals, their family and society. The care of injured patients has long been thought to be sub-standard in the UK and patient outcomes were noted to be worse than other developed countries. Between 2010-12 regional trauma systems were introduced in England, with the aim of improving survival and long-term outcomes of injured patients. The aim of this study was to critically assess the implementation of these trauma systems on processes and outcomes of care in two regions of England. Methods: A systematic review was undertaken to identify studies evaluating the association between an inclusive trauma system and patient outcome. A mixed-methods approach was used for the study. Data on trauma deaths were obtained from the UK Office for National Statistics (ONS). The Trauma Audit and Research Network (TARN) database was interrogated to provide data on all patients who presented to hospital within two years before and two years after trauma system implementation. A time-series analysis and a before and after study, using a comparator region to control for temporal trends, was undertaken for each region. Twenty semi-structured interviews with Emergency Department (ED) staff were conducted to gain a broader understanding of the effect of this change. Data were then merged and areas of convergence and discrepancy highlighted. Results: The systematic review identified eight observational studies that all demonstrated a significant fall in the odds of death when patients presenting with traumatic injuries were treated within such a system. However, they were deemed to represent a very low-quality body of evidence. ONS data demonstrated that whilst trauma mortality rates were stable, between 30- 50% die outside of hospital. Analysis of TARN data demonstrated that, following system implementation, a greater proportion of injured patients were seen at Major Trauma Centres (MTCs), quality of care indices such as time to CT scan improved and mortality fell. Analysis of the interviews revealed seven main themes and whilst all staff welcomed the commitment to improve trauma care, some, especially outside of the MTCs, expressed concerns about disengagement and being unable to provide the level of care expected. Conclusions: This study adds to the body of evidence supporting the role of inclusive trauma systems in improving quality of care indices and patient outcomes. Contrary to some other studies, this study has shown improvements within two years, particularly at MTCs. Whilst most ED staff interviewed corroborated this view, some barriers to delivering high quality trauma care were felt to remain. Whilst trauma was once seen as a disease of young men and motor vehicle collisions, it is now dominated by falls in the elderly population and trauma systems must be able to meet their needs. Further research is warranted to learn more about the large population of trauma patients that do not survive to reach hospital. Perhaps some of the greatest future improvements of trauma systems are to be found here.

Introduction: The most appropriate airway management technique for use by paramedics in out-of-hospital cardiac arrest is yet to be determined and evidence relating to the influence of airway management strategy on outcome remains equivocal. In cases where return of spontaneous circulation (ROSC) occurs following out-of-hospital cardiac arrest, patients may undergo direct transfer to a specialist heart attack centre (HAC) where the post resuscitation 12 lead ECG demonstrates evidence of ST elevation myocardial infarction. To date, no studies have investigated the role of airway management strategy on outcomes in this sub-set of patients. The AMICABLE (Airway Management In Cardiac Arrest, Basic, Laryngeal mask airway, Endotracheal intubation) study therefore sought to investigate the influence of prehospital airway management strategy on outcomes in patients transferred by the ambulance service directly to a HAC post ROSC. Methods: Adults with ROSC post out-of-hospital cardiac arrest who met local criteria for transfer to a HAC were identified prospectively. Ambulance records were reviewed to determine prehospital airway management approach and collect physiological and demographic data. HAC notes were obtained to determine in-hospital course and quantify neurological outcome via the Cerebral Performance Category (CPC) scale. Neurologically intact survivors were contacted post discharge to assess quality of life via the SF-36 health survey. Statistical analyses were performed via Chi-square, Mann Whitney U test, odds ratios, and binomial logistic regression. Results: A total of 220 patients were recruited between August 2013 and August 2014, with complete outcome data available for 209. The age of patients ranged from 22-96 years and 71.3% were male (n=149). Airway management was undertaken using a supraglottic airway (SGA) in 72.7% of cases (n=152) with the remainder undergoing endotracheal intubation (ETI). There was no significant difference in the proportion of patients with good neurological outcome (CPC 1&2) between the SGA and ETI groups (p=.286). Similarly, binomial logistic regression incorporating factors known to influence outcome demonstrated no significant difference between the SGA and ETI groups (Adjusted OR 0.725, 95% CI 0.337-1.561). Clinical and demographic variables associated with good neurological outcome included the presence of a shockable rhythm (p < .001), exposure to angiography (p < .001), younger age (p < .001) and shorter time to ROSC (p < .001). Due to an inadequate response rate (25.4%, n=15) analysis of SF36 data was limited to descriptive statistics. Limitations: The study only included patients who achieved ROSC and met the criteria for direct transfer to a HAC. Results are therefore not generalisable to more heterogenous resuscitation populations. Accuracy of clinical decision making and ECG interpretation were not assessed and therefore some patients included in the study may have been inappropriately transferred to a HAC. The low SF-36 survey response rate limited the level of neurological outcome analysis that could be undertaken. Conclusion: In this study, there was no significant difference in the proportion of good neurological outcomes in patients managed with SGA versus ETI during cardiac arrest. Further research incorporating randomised controlled trials is required to provide more definitive evidence in relation to the optimal airway management strategy in out-of-hospital cardiac arrest.

This research develops a comprehensive model for evaluating the efficiency and productivity of the sector of head trauma injury (HTI) care in England and Wales, in order to reduce the costs associated with trauma care. After assessing the advantages and disadvantages of various efficiency measurement approaches, the data envelopment analysis (DEA) methodology is chosen for this research, including both the DEA-based Malmquist index model and the bootstrapping DEA model. Since the variables selected for these models include some missing data, the approach known as multiple imputation by chained equations (MICE) is proposed to deal with such missing data situations, in order to ensure the accuracy of the inferential and predictive results that our analyses generate. In addition, an experimental study is provided to simulate this approach, in order to investigate its validity as a methodology for replacing such missing values within DEA applications. This experimental study is based on a real data set of 66 hospitals provided by the Trauma Audit and Research Network (TARN), within Salford Royal NHS Foundation Trust. The results of this experimental study show that MICE works well and gives an acceptable estimate of true efficiency. Furthermore, this research introduces a framework that combines DEA with structural equation modelling (SEM) in order to investigate the effects of uncontrollable variables on efficiencies. While the use of DEA provides valuable results, our SEM analysis reveals additional findings that were not identified in previous studies. For example, unlike previous second stage analysis studies in DEA that focused on only the direct effects of environmental factors on the efficiency scores, this study uses SEM to investigate further any indirect effects and the total effects of these uncontrollable factors on the efficiencies. This additional information is shown to be more useful and more informative than the results generated by the previous studies. The methodologies proposed and developed in this thesis are then applied to the full set of available TARN data in order to measure the efficiency and productivity of HTI care, demonstrating real possibilities for reducing the costs of head trauma care.

Background: Changing service demands require United Kingdom ambulance services to redefine their role and response strategies, in order to reduce unnecessary Emergency Department attendances. Treat and Refer guidelines have been developed with this aim in mind. However, these guidelines have been developed in the absence of reliable evidence or guiding mid-range theory. This has resulted in inconsistencies in clinical practice. One condition frequently included in Treat and Refer guidelines is hypoglycaemia. Therefore this thesis aimed to investigate prehospital hypoglycaemic emergencies in order to develop an evidence base for future interventions and guideline development. Research approach: A pragmatic and inductive applied health services research approach was employed. Multiple methods were used in a sequential explanatory design. Three linked studies were undertaken with the results of previous studies informing the development of the next. Study one: A scoping review of prehospital treatment of hypoglycaemic events. Aims: i) To describe the demographics of the patient population requiring ambulance service assistance for hypoglycaemic emergencies; ii) To determine the extent to which post-hypoglycaemic patients with diabetes, who are prescribed oral hypoglycaemic agents (OHA), experience repeat hypoglycaemic events (RHE) after being treated in the prehospital environment. Methods: A scoping literature review was conducted using an overlapping retrieval strategy that included both published and unpublished literature. Findings: Twenty-three papers and other relevant material were included. Hypoglycaemia related ambulance calls account for 1.3% to 5.2% of ambulance calls internationally. Transportation rates varied between studies (25%-73%). Repeat hypoglycaemic emergencies are experienced by 2-7% of patients within 48 hours. There was insufficient detail to determine any relationship between repeat events and OHA. The low quality of included papers means that the results should be cautiously interpreted. The safety of leaving patients on OHA at home post hypoglycaemic emergency is unknown. Consequently patients taking OHAs who experience a hypoglycaemic emergency should be transported to hospital for observation. There was a lack of knowledge about the Scottish demographics of the patient population. Study two: A retrospective cross-sectional observational study of diabetes related emergency calls. Aims: To investigate i) the patient demographics and characteristics of hypoglycaemia related emergency calls; ii) the incidence of repeat hypoglycaemic events; and iii) the factors associated with emergency calls that result in individuals being left at home. Methods: A retrospective observational cross-sectional study conducted using Medical Priority Dispatch System® call data from West of Scotland Ambulance Control Centre over a 12 month period. Data were extracted on age, gender, dispatch code, time of call, deprivation category, and immediate outcome (home or hospital). Multiple regression analysis was used to determine predictors of remaining at home. Findings: 1319 calls for hypoglycaemia were received. Patient demographics were similar to the scoping review findings. Most patients remained at home (N = 916 vs N = 380; p < .001). RHE’s were experience by 3.1% within 48 hours, and 10.6% within two weeks. The most significant independent predictor for patients remaining at home was a prior call to the ambulance service (OR of 2.4 [95%CI 1.5 to 3.7]). Patients’ reasons for remaining at home and the causes of subsequent severe events are unknown. It is likely that non-clinical factors may explain some of this behaviour. Study 3: Investigating patients’ experiences of prehospital hypoglycaemic care. Aim: To investigate the experiences of patients who are attended by ambulance clinicians for a hypoglycaemic emergency. Methods: In-depth interviews with adults with diabetes who had recently experienced a hypoglycaemic emergency treated by ambulance clinicians. Participants were recruited from Greater Glasgow and Clyde and Lanarkshire Health Board areas. Data were analysed using Framework Analysis. Findings: Twenty six patients were interviewed. Three key themes were developed. Firstly, an explanation for help seeking behaviour; patients’ impaired awareness of hypoglycaemia as well as the inability of friends and relatives to cope can contribute to an ambulance call-out. Secondly, the perceptions of ambulance service care; patients felt the service provided was good; however ambulance clinicians’ advice was inconsistent. Thirdly, the influences on uptake of follow-up care; patient preferences for follow-up care were influenced by previous experiences of home, hospital and primary care. Post-hoc analysis identified three psychological theories that may explain these findings and provide a useful basis for intervention development: Common Sense Model (Leventhal et al, 1998); Health Belief Model (Rosenstock, 1966); Ley’s cognitive hypothesis model of communication (Ley and Llewelyn, 1995; 1981). Conclusion: Most people treated for severe hypoglycaemia by ambulance clinicians remain at home and do not follow-up their care. A few experience repeat hypoglycaemic emergencies. Key causal, but modifiable factors, contributing to this include:- impaired awareness of hypoglycaemia; inconsistent delivery of ambulance clinician referral advice; and patients’ perceptions of the costs and benefits of follow-up care. Ambulance services cannot address all these factors in isolation. The studies in this thesis have generated an evidence base and identified plausible candidate theories. This will support the future development of novel interventions to improve severe hypoglycaemic emergency follow-up.

This work compares the procedures used in the Emergency Departments in the University of Benin Teaching Hospital (UBTH) in Nigeria and in Manchester Royal Infirmary (MRI) in the UK. It goes on to develop a discrete event model of the latter in Rockwell Arena®.Raw data from UBTH were obtained over a number of visits by interviewing senior administrators, clinicians and nursing staff and by tracking patients over a period of 2 months between 1 July and 29 August, 2011. Information from MRI was supplied through an approved ethical protocol to the National Research Ethics committee (REC Reference 13/NN/0175, IRAS ID 124168, dated March 4, 2013). This embraced patient journeys, locations, investigations and tests for the 98236 patients who attended the ED between April 2012 and March 2013. These (anonymised) data were obtained as spreadsheets from the original Symphony® records, which were then manipulated and analysed using the computer language, R. Anecdotal information on ED operations, patient flow and procedure duration times were also obtained from ED staff. All of this information identified similarities and differences between patient journeys in the two hospitals and were used to generate appropriate process maps. Proposals were made to improve the recoding and maintenance of patients’ records in UBTH. In the case of MRI, each patient’s journey was expressed as a journey-string, which was an ordered list of locations and milestones derived from the time-stamps recorded in the original spreadsheets. A large transition matrix (168 by 168) was generated from the set of journey strings and established the probability of a patient going from one location to any another. This reflects all the decisions which were made at each step of the patient’s journey. The number of destinations from a particular source reflects the options available at a particular instant in time, while the size of each probability reflects the preferred destination. The transition matrix together with the duration and resource requirement of the process associated with the destination is the key to the generation of a process map for each journey through the system. This methodology is original and can be applied generally. This was used as a basis for the journey-path model. In the final MRI model the 4h deadline was not included since the mechanism for its actual implementation was somewhat vague. Instead some isolated models based on patients’ priorities and resource re-allocation were described. From these it was inferred that changing the priority of a patient may not in itself be sufficient to alter the journey profile and in order to do so resources must be re-allocated. The only alternative would appear to be the fast-tracking of patients.

Rehabilitation outcomes are an important measurement of trauma system effectiveness. However, currently there is no clinically applicable trauma rehabilitation score or framework available to evaluate health and rehabilitation needs after trauma. The World report on Disability (2011) recommended the application of the World Health Organisation International Classification of Function, Disability and Health (ICF) as a framework for all aspects of rehabilitation. A standardised language, based on coded categories would aid in international efforts to evaluate health and disability globally. The ICF framework has not been applied in trauma rehabilitation or trauma systems to date. The objectives were to investigate rehabilitation needs of trauma patients and evaluate to what extent the ICF can be used as a framework to capture and assess health and rehabilitation outcome of patients following traumatic injuries. Two cohort studies with 103 and 308 patients respectively demonstrated the utility of the Rehabilitation Complexity Scale (RCS) in an acute trauma setting. The RCS outperformed other acute measures and rehabilitation complexity correlated with length of stay and discharge destination. A systematic review of 34 articles confirmed that outcome measures frequently used in trauma outcome studies represent only six percent of health concepts contained in the ICF. A quantitative international on-line questionnaire with expert clinicians working in trauma (n=217), identified 121 ICF categories pertinent to rehabilitation and health outcome of trauma patients. Qualitative patient interviews (n=32) identified nearly double the amount of ICF categories (n=234) compared to clinicians. Combined analysis of qualitative and quantitative data presents 109 ICF categories important for rehabilitation and health outcome assessment of trauma patients, using the ICF as a framework. This thesis describes the need for improved outcome evaluation of trauma patients. It demonstrates the acceptability of the ICF language and framework amongst clinicians and suggests the application of the ICF as a framework for trauma service delivery and outcome assessment.

This study was prompted by the reflective essays submitted by paramedics in which they recounted traumatic experiences from work which seemed unprocessed. I subsequently became interested in how paramedics maintained their resilience and ‘survived’ their work as measured by for example, job satisfaction, morale, attrition and mental and physical well-being. A qualitative research design was adopted which used a psycho-social interview technique called ‘free association narrative interviewing’ (FANI) as the means of data collection. Data collection included a one hour biographical interview initially and a second follow up 45 minute semi structured interview with seven participants recruited from a trust in England. In all instances data was collected at the university or participant’s home and interviews were digitally recorded. The interviews were transcribed and analysed thematically. The study was approved by UWE and NHS ethics committees. The analysis explored the key findings within the context of contemporary resilience theory drawing on psychological, sociological and psycho-social literature. However the main focus of the work was from a psycho-social perspective which is psycho-dynamically informed and uses psychoanalytic concepts and principles to explore core issues. The themes captured respondents’ motivations to become a paramedic, which embraced notions of gender differences and the ‘wounded healer’. The hidden toll of the work in relation to health, and the impact of recent performance targets and changes to skill mix which appeared to threaten traditional protective methods of support, emerged as other themes. Support from within the organisation was explored, focusing on the role of management, debriefing, peer support, humour, time out and the culture of shame and denial within the organisation. Outside support explored the role of friends, family and referral to outside agencies. In dealing with unprocessed encounters, paramedics appeared to use other strategies including emotional detachment, dissociation, displays of anger and inappropriate professional behaviour. Despite limitations to the study, the findings unveil a unique account of how paramedics survive and become resilient.

Aims: This study explored why patients seek unplanned follow up after treatment in an emergency department, if participants could identify the healthcare professional they were treated by, and whether the patient’s perception of, and confidence in, the healthcare professional had an impact on their subsequent decision to seek follow up. Methods: A mixed method pragmatic approach was used in order to quantify and explore the aims of the study. Participants were followed up two weeks later by telephone and finally invited to take part in a focus group to explore their experience and perceptions of their visit to the emergency department. Results: 18% of patients sought unplanned follow up in the 2 weeks following their initial visit, with no statistically significant difference between healthcare professional groups. 19% of patients incorrectly identified the healthcare professional treating them, with evidence of a gender bias. Discussion: The qualitative elements of the study explored the quantitative results. Participants were more likely to believe the healthcare professional was a doctor if they were male and had effective communication skills. A number of practical issues were identified in reducing unplanned follow up rates. The most common were issuing fitness to work certificates, explaining the trajectory of an illness or injury and addressing specific pain management issues. A change in policy would be required for non –medical health care professionals to be able to issue fitness to work certificates but in this study it was found to be the single most effective strategy to reduce reconsultantion rates. Conclusion: Patients seek unplanned follow up for a variety of reasons. This study shows that non-medical HCPs do not have a higher planned or unplanned follow-up rate, and they may have some advantages over junior medical staff in terms of effective consultation skills, high patient satisfaction and reduced reconsultation rates.

Overcrowding in Emergency Departments (ED) results in increased Lengths of Stay (LOS) and longer Waiting Times (WTs) for patients. These are ever-growing concerns in all hospitals around the world. Published literature shows many causes of this problem. This study, explores and identifies that one of the prominent causes can be attributed to human behaviour within EDs. It appears that patient behaviour has not been included in previous work as a cause of overcrowding. This thesis aims to present a method referred to as Discrete Event Simulation (DES) in an effort to explore and understand the effects of patient behaviour in the overcrowded ED of Tripoli Medical Centre (TMC). Firstly, a descriptive study was adopted, which collected data from different hospital sources, i.e. ED reports, ED's services time, staff and managers‘ opinions, and the observations of patients attending the ED of TMC. This identified the most significant behaviours impacting LOS and WT. Secondly, four different DES models were developed and analysed. Ultimately a hybrid model comprising of DES logic and Bayesian Network (BN) modelling was developed to capture and analyse a more accurate occurrence of human behaviour. Analysis of the hospital data reveals four behavioural factors are responsible for increasing WTs and LOS, leading to disruptions in patient flow. These factors include confrontation, challenges, passivity and illness belief. These behaviours are often appearing amongst the minor and non-urgent patients, who represent approximately 75% of all ED patients. The examination area within ED was found to be the place most commonly impacted by difficult behaviours, which is shown by the prolonged WTs in this area. A new strategy termed as Patient Behavioural Control (PBC) has been devised, developed and modelled. The PBC strategy aims to detect patient behavioural problems early and implements a revised patient flow procedure that results in overall reductions in LOS and WTs. This thesis contributes to the knowledge in this arena through consideration being directed towards patient behaviours from an operational perspective, utilising DES models in an effort to establish cause and effects, thus helping to devise a new approach in the healthcare sector. This new approach is not restricted in terms of application in only the healthcare arena, but can be adopted across other sectors for the management of human behaviours.

Emergency Medical Services (EMS) play a pivotal role in any healthcare organisation. Response and turnaround time targets are always of great concern for the Welsh Ambulance NHS Trust (WAST). In particular, the more rural areas in South East Wales consistently perform poorly with respect to Government set response standards, whilst delayed transfer of care to Emergency Departments (EDs) is a problem publicised extensively in recent years. Many Trusts, including WAST, are additionally moving towards clinical outcome based performance measures, allowing an alternative system-evaluation approach to the traditional response threshold led strategies, resulting in a more patient centred system. Three main investigative parts form this thesis, culminating in a suite of operational and strategic decision support tools to aid EMS managers. Firstly, four novel allocation model methods are developed to provide vehicle allocations to existing stations whilst maximising patient survival. A detailed simulation model then evaluates clinical outcomes given a survival based (compared to response target based) allocation, determining also the impact of the fleet, its location and a variety of system changes of interest to WAST (through ‘what-if?’ style experimentation) on entire system performance. Additionally, a developed travel time matrix generator tool, enabling the calculation and/or prediction of journey times between all pairs of locations from route distances is utilised within the aforementioned models. The conclusions of the experimentation and investigative processes suggest system improvements can in fact come from better allocating vehicles across the region, by reducing turnaround times at hospital facilities and, in application to South East Wales, through alternative operational policies without the need to increase resources. As an example, a comparable degree of improvement in patient survival is witnessed for a simulation scenario where the fleet capacity is increased by 10% in contrast to a scenario in which ideal turnaround times (within the target) occur.

Due to the fact that emergency situations can lead to substantial losses, both financial and in terms of human lives, it is essential that texts used in a crisis situation be clearly understandable. This thesis is concerned with the study of the complexity of the crisis management sub-language and with methods to produce new, clear texts and to rewrite pre-existing crisis management documents which are too complex to be understood. By doing this, this interdisciplinary study makes several contributions to the crisis management field. First, it contributes to the knowledge of the complexity of the texts used in the domain, by analysing the presence of a set of written language complexity issues derived from the psycholinguistic literature in a novel corpus of crisis management documents. Second, since the text complexity analysis shows that crisis management documents indeed exhibit high numbers of text complexity issues, the thesis adapts to the English language controlled language writing guidelines which, when applied to the crisis management language, reduce its complexity and ambiguity, leading to clear text documents. Third, since low quality of communication can have fatal consequences in emergency situations, the proposed controlled language guidelines and a set of texts which were re-written according to them are evaluated from multiple points of view. In order to achieve that, the thesis both applies existing evaluation approaches and develops new methods which are more appropriate for the task. These are used in two evaluation experiments – evaluation on extrinsic tasks and evaluation of users’ acceptability. The evaluations on extrinsic tasks (evaluating the impact of the controlled language on text complexity, reading comprehension under stress, manual translation, and machine translation tasks) Text Complexity and Text Simplification in the Crisis Management domain 4 show a positive impact of the controlled language on simplified documents and thus ensure the quality of the resource. The evaluation of users’ acceptability contributes additional findings about manual simplification and helps to determine directions for future implementation. The thesis also gives insight into reading comprehension, machine translation, and cross-language adaptability, and provides original contributions to machine translation, controlled languages, and natural language generation evaluation techniques, which make it valuable for several scientific fields, including Linguistics, Psycholinguistics, and a number of different sub-fields of NLP.

In recent years safety in medicine has been high on the agenda, both for government and for healthcare providers. This thesis starts by describing the evolution of patient safety and then goes on to explore error specifically in the Emergency Department (ED). Focus is drawn to the role of non-technical skills for improving safety. The initial broad aim of this research was to learn how the ED team could function better to improve patient care. An interview study is developed to investigate how ED staff change their behaviour during periods of high demand and to determine the direction of future research. This study highlights that staff would benefit from increased awareness of the nontechnical skills that contribute to effective teamwork and enhanced patient safety. The interviews also reveal the leadership role of the registrar is of particular importance. Therefore, a series of studies are developed to identify and describe the non-technical skills required by Emergency Medicine trainees, with a specific focus on leadership. The process of developing a provisional assessment tool for assessing non-technical skills in the ED is described. This draws on published literature and curricula as well as considering existing methods of assessment. The assessment tool is revised using re-analysis of staff interviews and a series of preliminary observations in the ED. Content validity of the tool is measured using a survey of expert opinion and this helps to further refine the tool components. An experimental study reveals that whilst adequate levels of inter-rater reliability are achievable, rater accuracy appears to be more problematic. Various sources or rater error are also explored and this leads onto a larger, multicentre observational study investigating use of the tool in the workplace. Further data for reliability is collected and field notes are analysed to provide a detailed description of the non-technical skills displayed by ED registrars. Findings of the studies are summarised and limitations, applications and further research are discussed.

Aim and objectives: To describe the ideal doctors' shift handover process in a systematic fashion, and to identify tasks that should be performed, but are not consistently done. To understand the types of communication problems that may occur during the handover process, their causes, their likelihood of occurrence and their effect on patient safety. Method: Three studies were conducted in two, Scottish Acute Medical Units. A Hierarchical Task Analysis was performed and data was collected by means of interviews and focus groups. Observations of doctors' actual shift handover process were compared against the description of doctors' ideal handover process. To examine potential failures modes, a Healthcare Failure Modes and Effects Analysis was performed using focus group interviews. Results: The handover process entailed the pre-handover, the handover, and the post- handover phases. Multiple critical steps in the process were omitted by outgoing shift doctors. The pre-handover was particularly vulnerable to information omission, with over 50% of its critical tasks not being performed across a total of 62 observations. Nonetheless, most of these omissions were typically caught during the handover meeting, especially if incoming doctors participated in pre-handover activities. Post-handover activities involved prioritizing and delegating clinical tasks. However these were observed not to happen consistently due to multiple interruptions. Thirty-four failure modes were identified, with eight of them posing a significant risk to patient safety. The studies found that interruptions, patient workload, and a lack of standardised procedures were the biggest causes for information loss during the handover process. Conclusions: There are key critical tasks necessary for an ideal doctors' shift handover process. A simple, handover process checklist may ensure critical handover tasks have been achieved prior to any shift change. Interruptions, patient workload, peer trust, and a lack of standard operating procedures are areas that future handover research should examine.

Aim: Using a mixed methods approach, this thesis seeks to explore the attitudes of emergency care staff towards young people (aged 12− 18 years) who self-harm and to gain an understanding of the basis of attitudes that exist. Background: This thesis has drawn on Strauss et al’s (1964), concept of the hospital as a negotiated order, a perspective that has latterly been applied to the organisation of hospital A&E services (Sbaih1997a&b 1998a&b, 2001, 2002). As the fundamental premise of emergency care work is the rapid assessment of patients’ needs, categorisation is an essential element of this work. This thesis therefore also draws on the sociological theories which have examined the categorisation of patients as ‘good’ or ‘bad’, as earlier sociological work has clearly demonstrated that practitioners working in emergency services judge patients based on their reasons for accessing the service (Roth 1972, Jeffery 1979, Dingwall & Murray 1983); patients who self-harm are amongst those adversely judged. However the extent to which these categorisations extend to young people was not wholly clear. Findings from earlier research that had considered this were inconclusive and inconsistent (Dingwall & Murray 1983, White 2002). Methods: A mixed methods approach, using a triangulation convergent design was employed. Staff employed in four emergency departments in South East London and five London Ambulance complexes that served these departments were surveyed; data from 143 questionnaires were analysed using SPSS. Qualitative data were obtained through semi-structured interviews with 12 practitioners, seven nurses and five paramedics, with thematic analysis undertaken. The two data sets were integrated and analysed to identify where the two data sets were consistent and whether/where discrepancies existed. Results: Findings from this study indicate that age, i.e. being a young person, does influence attitudes towards self-harm. Young people are less adversely judged as their self-harm is seen as symptom of distress, a coping mechanism or response to a stressor out with a young person’s control, thus as a consequence, attitudes towards young people who self-harm are benign. The findings lend support to previous research which has indicated that as an occupation, nurses have less positive attitudes than their peers working in emergency services. Although not statistically significant, the nurses surveyed in this study obtained lower scores on the scale used to measure attitudes than their medical and paramedical colleagues. The data from the interviews illustrated the difficulties and frustration the nurses faced in managing the care of young people who self-harm, which centred on the pressure to ‘move young people on’, pressures that were exacerbated by the need to do this within four hours. The paramedics interviewed did not face these challenges. Nurses faced considerable difficulty in securing admission to a children’s ward; the accounts of the nurse interviewees suggested that their ward colleagues expected and anticipated that young people who had self-harmed would be challenging in terms of their behaviours, whereas no such expectation existed with other adolescent patients. To this end the diagnostic label of self-harm had negative connotations Conclusions: The findings from this study have extended existing knowledge in relation to practitioners’ attitudes towards young people who self-harm, providing as they do an insight into how young peoples’ immaturity and diminished agency, contribute to the framing of young people as vulnerable, thus their self-harming behaviour is less adversely judged. A negotiated order perspective remains a relevant lens through which to analyse and explore the organisation of hospital services and specifically the work of the A&E department; the findings of the research presented in this thesis have revealed how young people who self-harm, through both their actual and perceived behaviours, disrupt the organisation of children’s accident and emergency care, thereby distorting its ‘shape’. The ambiguity of adolescence as a life-stage is reflected in the attitudes and perceptions of the study participants and is also reflected in health policy and guidelines, which is particularly exemplified by inconsistency in how the emergency care needs of young people between the ages of 16– 18 years generally, and young people who self-harm specifically, are addressed. This inconsistency and ambiguity in turn serves to impede young people’s progress through emergency services following an episode of self-harm.

Aims: In this thesis I present findings from a series of linked studies examining the process and outcomes of a pragmatic RCT of brief intervention for alcohol misuse among people presenting to an emergency department (ED). The studies focus on screening, uptake of brief intervention and its impact on re-attendance to the department. Methods: After examining the acceptability and validity of the screening tool to be used in the trial, I tested the impact of health consequences feedback on uptake of brief advice and conducted a secondary analysis of trial data to examine the impact of the intervention on alcohol use and reattendance to the department in sub-groups of patients. Results: The Paddington Alcohol Test provides an acceptable and valid method for screening for hazardous drinking among people attending an ED. Provision of health consequences feedback increased the proportion of hazardous drinkers willing to accept advice. In addition to reducing levels of alcohol consumption, brief intervention decreased the likelihood of a re-attendance to the ED. Sensitivity analyses based on treatment allocation or excluding participants erroneously randomised did not alter these results. Older drinkers and those who stated that their attendance at the ED was related to their use of alcohol were more likely to attend an appointment with an Alcohol Health Worker; however attendance did not significantly reduce levels of consumption. Conclusions: Brief alcohol intervention delivered in an ED reduces levels of alcohol consumption for hazardous and heavy drinkers and reduces reattendance at the department. The provision of health consequences feedback increases the proportion of hazardous drinkers willing to accept advice. While only a minority of those who are offered an appointment with an Alcohol Health Worker in this setting attend it, attendance at an appointment did not appear to have an impact on the main study outcomes.

This thesis looks to use a variety of statistical methods to investigate the performance of hospitals for patients who receive trauma care. Our main motivation comes from data provided by the Trauma Audit Research Network (TARN). Trauma centre performance as measured by the Ws score of Hollis et al. (1995) was formulated from the TRISS (Trauma Injury Severity Score) methodology and is part of an ongoing study to improve trauma care across the UK. However, ring institutional performance and ranking hospitals is a difficult process, particularly in the presence of missing data and for hospitals with few trauma patient records. There are also many political issues arising when using ranking and compiling league tables, see Goldstein and Spiegelhalter (1996). One way to ensure that our assessment of institutional performance is as precise as possible is to make sure that we have quality data, i.e. complete and that our predictions are accurate.

Eight patients (who were admitted to the accident and emergency department of a large acute hospital) were recruited into the study over the course of one year. Data were collected using three data collection methods: interviews with patients and health professionals; observations of ward rounds and nurse handovers; and documentary analysis of nursing and medical notes. The data collection period for each case ranged from between 24 hours to three weeks. Results from the study demonstrated that there are many less well defined factors that influence emergency admission to hospital and delivery of care. These factors fell into one of three categories: individual patient characteristics and needs, delivery of care issues and the context and politics of care delivery. It can be concluded that, the needs of patients with chronic conditions and / or symptoms are not being addressed in a timely fashion. This is due in part, to the poor responsiveness of healthcare services in the community and the patient's lack of awareness of these services. Care within hospital is still delivered in a routine and ritualistic fashion with little regard given to the patient's individual requirements and knowledge of their condition. Health professional decision-making is influenced as much by their own perceptions, experience, uncertainty and access to diagnostic tests as the patient's clinical signs. There is currently no single person responsible for keeping the process of care moving within the hospital, with the result that care stalls and patients are detained in hospital longer than necessary.

Introduction. Trauma scoring models form an important part in trauma audit. Limitations of the TRISS model however has led investigators to search for new models which more accurately predict trauma deaths. Logistic regression has played an integral role in the development of such models in view of the fact that most investigators in this field use a dichotomous dependent variable (death/survival). Limitations in the methods used to evaluate new models currently makes it difficult to assess their true worth. Aims. The aim of this thesis was to evaluate current methods of model validation and also to evaluate alternative options. Methods and Results. A data set of 7069 complete trauma cases from the Los Angeles Trauma Registry was used for the study. Five goodness of fit tests were studied and were compared to the Hosmer Lemeshow test. The Copas test was found to be the most viable alternative to the Hosmer Lemeshow test. The two main methods of validating a prognostic model (i.e. data splitting and cross-validation) were evaluated using a series of simulation studies. Both methods were found to be unreliable with regard to their ability to reduce over-fitting of a prognostic model. Bootstrapping was also evaluated as a means of generating confidence intervals for the Hosmer Lemeshow test and the Copas test. Two models were used to assess the accuracy of the percentile method. Accurate bootstrap confidence intervals were developed for the Copas test but not for the Hosmer Lemeshow test. Conclusions. The Copas test with bootstrap confidence intervals provides a superior approach to validating a trauma model compared to the currently employed methods.