The thesis investigated the influence of centre selection on the generalisability across locations of trial-based economic evaluations. A novel methodology to assess and enhance the generalisability of trial findings was demonstrated using the comparison between wound-edge protection devices (WEPDs) and standard care to reduce surgical site infection (SSI) after open abdominal surgery as a case study. A systematic review and a preliminary economic model suggested that WEPDs may be effective and cost-effective in reducing SSI compared to standard care, although the methodological quality of available studies was poor. ROSSINI was a high quality multi-centre randomised controlled trial (RCT) which demonstrated that WEPDs are unlikely to be effective or cost-effective, so their routine use cannot be recommended. The impact of centre selection on trial results was then investigated using ROSSINI as a case study. Mixed methods research demonstrated that most RCTs do not enrol centres so as to ensure a representative sample at jurisdiction level. The Generalisability index (Gix) was introduced as the basis of a novel methodology to assess generalisability, which was demonstrated using simulation methods and ROSSINI data. The results suggested that the characteristics of the sample of participating centres can significantly affect RCT clinical and cost-effectiveness estimates.

Background: With increasing numbers of people with a learning disability (LD) choosing to become parents (Booth and Booth, 1994) it is important the right support is provided to enable them to parent effectively (Macintyre and Stewart, 2011). Materials and Methods: This study used semi-structured interviews with four women with a LD who received doula support prenatally, during labour and postnatally. The women were interviewed during prenatal and postnatal support periods. The doulas were interviewed about their experience of supporting a woman with a LD towards the end of the postnatal support period. Results: Interview transcripts were analysed using Interpretive Phenomenological Analysis (IPA). Themes were identified from each of the ten interviews, before analysis of themes for the same support phase were analysed together for mothers and doulas, allowing discussion of similarities and differences. Conclusions: The findings show that prenatally the women considered the doula to be a reliable source of information about pregnancy and birth. Each mother perceived doula support as a means of keeping her child in her care. Postnatally, mothers described a trusting relationship with their doula, which enabled them to make informed choices. Doulas described the need to adapt their work to meet the needs of parents with LD. The experience of working alongside statutory agencies was perceived as potentially challenging, and an important area for supervision.

Recent health care reforms within Saudi Arabia have advocated use of economic evaluation in health care decision making. Little research has, however, considered the use of economic evaluation to set priorities in rentier state settings. This thesis explores the nature of the rentier state and the basis of health care priority setting, and conducts a systematic review of the use of economic evaluation in priority setting. The thesis uses in-depth qualitative research to explore health care priority setting and use of economic evaluation in Saudi Arabia. Qualitative data comprised 22 in-depth interviews with decision makers at the national and district levels, 3 focus groups, and one meeting observation. Data collection and analysis were conducted iteratively using constant comparison. Findings show that contextual factors have a great influence on the decision making process and that the use of economic evaluation is still very limited. There appeared to be two types of barriers to the use of economic evaluation: decision context-related barriers and barriers relating to the production of economic evaluation data. Incorporating economic evaluation into the health care decision making process in Saudi Arabia is proving to be complex and contextual factors have more influence on priority decisions than economic evaluation.

Conducting economic evaluations alongside multinational studies presents a range of diverse challenges which have contributed to a lack of consensus on how they should be approached particularly because of the difficulties of resolving between country differences. This thesis examines the implications of conducting economic evaluation alongside multinational studies and (i) explores different approaches to obtaining unit costs; (ii) investigates the impact of using different tariffs to value EQ-5D health state descriptions; and (iii) provides a systematic comparison of the pooled and split approaches to economic evaluation alongside multinational trials. This study documents challenges that have been reported in published studies and makes recommendations to help researchers undertake economic evaluations alongside multinational studies. Results indicate that the main challenge related to dealing with the differences between countries. Collecting unit cost data in all participating countries proved a difficult task, but was most effectively done by collaborating/direct contact with project partners and researchers/health economists from participating countries. Applying different EQ-5D value sets within the context of multinational trials did not make a difference to the conclusions in most cases. However, it is recommended that results from various tariffs are compared within sensitivity analysis. This study also showed that the choice of whether to pool or split the data can lead to different conclusions and recommendations about the cost-effectiveness of interventions. The culmination of this work is a 10 point checklist to guide good practice in the design, conduct and analysis of multinational economic evaluation studies and also highlights many areas where further research is needed. The work provides researchers, policy makers and stakeholders with additional insight into the economic analysis of multinational studies.

This thesis presents a Realist Evaluation of the Brook Traffic Light Tool (TLT) which supports professionals in reducing harmful sexual behaviours (HSB) in children and young people (CYP). This is important as the numbers of HSB in CYP are rising, and no national strategy exists in the UK. The conceptual element consisted of developing a hybrid methodological approach combining: Realist Evaluation; The Kirkpatrick Model; evidence based policy, policy transfer and policy success perspectives. The empirical element involved evaluating the implementation of the TLT across Cornwall, with data collected in a three phase mixed method approach: a questionnaire with 436 responses; 60 detailed questionnaires and 13 interviews. The context, mechanisms and outcomes that emerged were complex. Conceptually, questions were raised about empirical research that underpins the TLT, with UK transfer and policy success criteria presenting a mixed picture of success. In Cornwall the TLT met the majority of its outcomes. Direct impact upon the behaviour of CYP was difficult to ascertain and categorisation anomalies were found when behaviours were not explicitly outlined in the TLT. The study found that a range of professionals were encountering both harmful and healthy sexual behaviours, yet comparison to national figures proved problematic as no baseline exists.

Exposure to air pollution can cause adverse health effects, may also adversely affect the central nervous system and affect cognitive performance. Epidemiological studies depend on central site monitors as surrogates to assess personal exposure to air pollution, which can be inaccurate because they do not assess personal exposure in a variety of activities and microenvironments. This thesis aims to assess the level of misclassification in data from central site monitors by using portable modern sensors with high temporal resolution to characterize personal inhaled doses of BC, PM2.5, and UFP, and compare the measurements with surrogate exposure metrics. It also seeks to identify contributing activities and sources associated with the highest concentrations of the three pollutants, and to determine the contribution of these activities and microenvironments to personal exposure, and to study the impact of short-term exposure to air pollution on cognitive function. The first finding is that central site monitors are not a good surrogate for personal exposure. Secondly, travelling in vehicles is linked to the highest concentrations of the three pollutants, while other outdoors activities and outdoors commuting are linked to the highest concentrations of BC and PM2.5, cooking is linked to the highest concentrations of UFP, and activities and time spent indoors are the highest contributors to personal exposure. Thirdly, the results provide strong evidence that short-term exposure to PM2.5 from candle burning and commuting has an adverse effect on cognitive performance.

Importance of home healthcare is growing rapidly since populations of developed and even developing countries are getting older quickly and the number of hospitals, retirement homes, and medical staff do not increase at the same rate. We present Scenario Based Approach (SBA) for the Home Healthcare Nurse Scheduling Problem. In this problem, arrivals of patients are dynamic and acceptance and appointment time decisions have to be made as soon as patients arrive. The primary objective is to maximise the average number of daily visits. For the sake of service continuity, patients have to be visited at the same days and times each week during their service horizon. SBA is basically a simulation procedure based on generating several scenarios and scheduling new customers with a simple but fast heuristic. Then results are analysed to decide whether to accept the new patient and at which appointment day/time. First, two different versions of SBA, Daily and Weekly SBA are developed and analysed for a single nurse. We compare Daily SBA to two greedy heuristics from the literature, distance and capacity based, and computational studies show that Daily SBA makes significant improvements compared to these other two methods for a single nurse. Next, we extend SBA for a multi-nurse case. SBA is compared to a greedy heuristic under different conditions such as same depot case where nurses start their visits from and return to same place, clustered service area, and nurses with different qualification level. SBA gives superior results under all experiment conditions compared to the greedy heuristic.

Healthcare organisations have struggled to improve safety. Over the last 20 years rates of patient harm have remained at around 10%, despite implementation of various improvement initiatives linked to the patient safety movement. This study identifies key conditions perpetuating this safety ‘implementation gap’: 1) a hierarchical challenge and 2) second victim phenomenon. Employee voice is adopted as a sensitising concept. The aim is to identify conditions which moderate the hierarchical challenge, encouraging the enactment of voice, leading to prevention of further medical errors. This PhD’s original contribution to knowledge is: second victims are key actors in attenuating hierarchical barriers through enactment of positively valenced practices. Adopting a practice-based approach identifying medical errors as break-downs in professional practice, three cases of serious medical error at a single NHS Trust are chosen for comparative analysis. Data collection includes over 100 hours of observations, 50 interviews, and review of 35 documents. Evidence for a climate of silence was found in each case stemming from: a hierarchical culture, blame culture, and futility of voice. Acquiescent or defensive silence contributed directly to each serious medical error. Second victims were found in each case, generally experiencing guilt, shame, anger, and compassion. The recovery trajectory of these second victims varied, with one ‘thriving’ while others ‘survived’ or ‘dropped out’. Positively valenced practice changes, which set the conditions for voice, were enacted by affectively charged individuals, either the second victim themselves, or through a process of emotional contagion, their colleagues. These conditions for voice included: setting expectations for voice, management engendering voice, closer adherence to policy and standard operating procedures, and a reinvigorated sentiment of care. These changes led to development of a voice climate encouraging the enactment of both defensive and prosocial voice. A safety incident model of voice for second victims was developed and transferability discussed.

Background The world’s ageing population is frequently cited as presenting economic and social burdens and critical challenges to healthcare systems, with falls and mental health featuring as two of the biggest burdens. Whilst physical activity is recommended for older adults to optimise functional independence and guidelines for the intensity, duration and frequency of physical activity exist, there are few guidelines on the specific types of exercises that are recommended. As a physical activity that incorporates the recommended aspects of aerobic, strengthening and balance work, the aim of this study was to explore the influence of ballroom dancing on the physical, mental and social health of community-dwelling older adults. Method A qualitative-dominant, concurrent mixed-methods design was utilised. Ethical approval was granted via the University of Huddersfield’s research ethics panel. Participants were community-dwelling older adults (over 55 years of age) who were recruited from local ballroom dancing classes in West Yorkshire. They participated in ballroom dancing classes for at least one hour per week for 12 months and were tested during this period at baseline, and after 3, 6, 9 and 12 months. Demographic details were collected at baseline and the incidence of falls, changes to medical history or medications and levels of exercise were recorded at each data collection point. Quantitative measures of physical function were assessed using the Functional Reach Test (FR test), Timed-Up-and-Go Test (TUGT), the Four Square Step Test (FSST) and Tinetti’s test. In addition, balance was specifically assessed using a Biodex Balance System SD (BBS). The Falls Efficacy Scale-International (FES-I) was completed alongside the Clinical Outcomes in Routine Examination-General Population (CORE-GP) for well-being. The qualitative aspect of the study used semi-structured interviews at baseline, 6 and 12 months. Forty-one interviews were performed with participants individually or in dancing pairs. Qualitative data were managed and analysed using the Framework Analysis approach. Results Of the 26 older adult ballroom dancers recruited to the study who consented to participate, 23 (10 male, 13 female) completed the 12-months of data collection (mean age 66.5 years (SD 5.96 years; age range 58-83 years)). Using a Framework Analysis approach, 4 key themes were identified amongst the sample of older adults. ’Active ageing’ considers physical health perceptions, the acceptance, adaptation and frustrations of ageing and maintaining an active body and mind. The theme ‘class commitment and congruence’ highlights factors that were deemed important to the success of a dancing class and would encourage participants to keep dancing. The ‘Social dance community’ theme considers the influence and dynamics of dancing partnerships and the importance of building social networks. Finally, ‘enjoyment’ acknowledges the role of ballroom dancing for well-being. This study demonstrated a low attrition rate, no adverse effects due to ballroom dancing and a low falls risk in this sample of older adults. The primary outcome was assessment of functional changes between baseline and 12 months, with the 3-monthly intervals serving as interim data collection points. Over the 12-month period, whilst not expected in a feasibility study, there were no significant changes in the FR test, TUGT, FSST or Tinetti’s test. For the CORE-GP scale 97.5% of recordings were within the ‘healthy’ well-being score range, and there was a 10% decrease in points-score for the FES-I, indicating a possible substantive finding for clinical practice. The BBS tests demonstrated women had a statistically significant better level of postural stability then men, and as a group, their scores on the FR test were significantly lower than normalised scores for their age group. Discussion The findings suggest that ballroom dancing should be promoted by health professionals as a socially inclusive,safe physical activity that provides considerable pleasure. In addition to the current literature, the ‘pleasure of practice’and the ‘pleasure of community’are two types of pleasure suggested to be associated with ballroom dancing. Ballroom dancing also enhances opportunities for physical, psychological and social resilience in older adults by resisting age-related physical decline, assisting with acceptance of the ageing body, providing a strong sense of enjoyment and well-being and helping to form strong social connections, which are all important factors for resilient ageing. It is recommended that future comparative studies are of an adequate power to detect group differences that may exist in quantitative outcome measures, such as assessing balance with the BBS. A control group should be used for comparison with the ballroom dancing intervention group, with the intervention group all starting ballroom dancing at baseline. Given the findings of this study that the participants appeared to be high functioning when performing physical tests, it is recommended that alternative outcome measures are sought that will discriminate sufficiently amongst community-dwelling older adults.

Organisational development claims to improve an organisation’s functioning through enhancing its members’ performance. OD achieves this through deploying talk and text to trigger ideational change and so produce new sense-making and behaviours in organisational members. This thesis makes this explicit through a realist evaluation of three cases of OD practice. OD literature highlights inconsistent outcomes, failure to produce transformational change and an absence of studies exploring the means by which interventions actually succeed or fail to produce change. Although the influence of context is regarded as a source of explanation for these shortcomings, OD has not resolved the issue of how to theorise and integrate it into practice. This thesis addresses these shortcomings through the proposal of an alternative theory of change upon which the field’s theory and practice could be based. Neo-Durkheimian institutional theory articulates an inter-relationship between ideation and institution. It points to ways of developing culturally-specific OD practice. It provides an explanation for the need for different interventions for transformational and transactional change, and for the success of OD interventions within different cultural forms. NDIT’s potential contribution to a richer understanding and explanation of OD is highlighted through an NDIT-driven realist evaluation of three OD cases.