Dissertations / Theses on the topic 'Randomised controlled trials'

Abstract Background The randomised-controlled trial (RCT) provides the best evidence for evaluating treatment effects and is accepted as a gold standard for clinical and regulatory decision making (1;2). One of the major challenges to the conduct of RCTs is the recruitment of adequate numbers of participants. Inadequate numbers reduce the power of a study to detect statistically significant treatment effects, and may cause delays, increased costs and failure to complete trials. The need for clinical trials in children has been increasingly recognised by the scientific community, resulting in increased demands for the inclusion of children in trials. For several reasons, recruiting children to trials is more challenging than recruiting adults, as consent issues are more difficult because parents make decisions about trial participation on behalf of their child. Despite general professional and community support for paediatric clinical trials, parents and paediatricians express reluctance when their own child or patient is asked to participate. Although researchers working with children commonly experience difficulty with recruiting children to RCTs, little is known about this very important subject. The method by which potential participants are approached for trial participation, the influence of their health care provider and the attitude of potential participants (or their parents, in the case of children), are critical to the understanding of the decision making process for trial participation. This thesis is one of the first major attempts to explore the issues surrounding the recruitment of children to RCTs, and is divided into four studies which address these issues. Methods Recruitment strategies used to encourage participation in randomised controlled trials (systematic review) Eligible experimental and observational studies comparing methods of recruiting participants for RCTs were identified after a comprehensive search of Medline, Embase, the Cochrane Library and reference lists. Independent data extractions were completed by two reviewers who assessed the studies for eligibility and methodological quality. Outcome measures were consent rates, proportion enrolled by each method and cost of recruitment per participant. Summary estimators of effects were calculated using a random effects model and expressed as relative risk with 95% confidence intervals. Heterogeneity was analysed using the Q statistic. Paediatricians� attitudes to children�s participation in randomised controlled trials (focus group research) Qualitative analysis of focus group discussions involving 16 paediatricians and 5 trainees from a paediatric teaching hospital in Sydney was undertaken. Doctors varied in occupation, experience, research activity, age, gender, ethnicity and parenthood experience. A professional facilitator conducted the semi-structured group discussions. Recruitment ceased when informational redundancy was reached, after 4 focus groups involving 21 participants. The transcribed audiotapes were analysed by theme linkage using the constant comparative method. Australian paediatricians� and adult physicians� attitudes to randomised controlled trials (survey) A 44-item questionnaire was sent to 250 paediatricians and 250 adult physicians randomly selected from the membership list of the Royal Australasian College of Physicians. Questions assessing doctors� treatment philosophies and attitudes to trials were compared with demographic and practice variables. Parents� attitudes to children�s participation in randomised controlled trials (focus group research) Qualitative analysis of focus group discussions involving 33 parents from 5 different settings (representing parents of children with a life threatening, chronic or acute illness, with experience in trials and of healthy children) was undertaken. Parents varied in age, gender, ethnicity, level of education, research experience and their child�s health status. The transcribed discussions were analysed by theme linkage using the constant comparative method. Results Recruitment strategies used to encourage participation in randomised controlled trials (systematic review) Fifty papers were included (out of 8602 titles and abstracts searched) which described 8 RCTs, 2 quasi RCTs, 13 prospective cohort studies, 30 retrospective cohort studies and 2 before-after studies. These studies assessed how over 4 million people were approached for RCT participation using 87 different recruitment strategies, with 103,406 people enrolling in RCTs. Health care provider (HCP) referrals had the highest participant consent rates at the time of exposure to trial information (HCP referral versus target mailing: relative risk (RR) 1.84 (95% confidence interval (95%CI) 1.08, 3.13)). They also had the highest consent rates when potential participants respond to the recruitment material by further enquiry about the trial (HCP referral versus community presentation: RR 1.37 (1.06; 1.78); HCP referral versus worksite approach: RR 25.20 (20.19, 31.45); HCP referral versus general community approach: RR 2.53 (0.46, 14.05); HCP referral versus mailing: RR 3.29 (1.26, 8.60); HCP referral versus media: RR 2.66 (1.31, 5.41)). However, by the time potential participants attend eligibility assessment for trial participation, no difference in consent rates could be distinguished by method of recruitment. Higher proportions of study participants were recruited by methods that exposed larger numbers of potential candidates to trial information (despite their lower consent rates). The stated recruitment cost ranged from US$0 to $1108 per participant, with mailing being the most cost-effective method and community methods (such as community presentations, pamphlets and posters displayed at community sites) the least effective. Paediatricians� attitudes to children�s participation in randomised controlled trials (focus group research) From the focus group discussions, paediatricians thought parents balanced perceived gains and risks when deciding about trial participation. They also believed the child�s condition and parents� health beliefs and personal attributes influenced parents� decisions. Other factors thought to be important by paediatricians were the doctors� beliefs and their relationship with the investigators. Paediatricians perceived gains for trial participation including professional benefits for themselves, improved patient care, convenience for the families and themselves and scientific advancement. Perceived risks included inconvenience, inadequate resources and potential harms to the patient and the doctor-patient relationship. Paediatricians with previous research experience were most knowledgeable about RCTs and perceived greatest gains from trial participation. Paediatricians� personal treatment preferences hindered trial support. Australian paediatricians� and adult physicians� attitudes to randomised controlled trials (survey) Response rate from the paediatricians� and adult physicians� survey was 60% (300/500). Australian paediatricians and adult physicians are very similar in their treatment philosophies, and are clinician-oriented rather than research-oriented in their attitudes, with primary allegiance to their patients and preference for selecting treatment rather than referring for trial participation in the face of treatment uncertainty. Professional activities are clinically focused, with limited time assigned for research. Australian doctors perceive little reward for trial participation and claim that the opinions of referring doctors regarding RCTs does not influence them. Predictors of favourable attitudes to trial participation from the survey were time allocation for research, a history of referring patients to trials in the past and younger age (all p values less than 0.0001). Parents� attitudes to children�s participation in randomised controlled trials (focus group research) When parents were interviewed, they acknowledged balancing risks and benefits when deciding about trial participation for their child. Perceived benefits include the offer of hope, better care of their child, the opportunity to access new treatments, healthcare professionals and health information, meeting others in similar circumstances and helping others. Perceived risks include potential side effects, being randomised to ineffective treatments and the inconvenience of participation. The decision for trial participation is also influenced by parental factors (parents� knowledge, beliefs and emotional response), child factors (the child�s health status and preference about participation), trial factors (the use of placebos and the uncertainties of research) and doctor factors (doctor�s recommendations and communication of trial information). Conclusions There are many challenges to the successful conduct of RCTs. Ways of addressing these include: using effective methods of recruiting potential study participants (such as mailing of recruitment material to potential participants) and abandoning ineffective strategies (such as community methods): fostering greater willingness for trial participation by addressing parents� and paediatricians� concerns including enhancing communication between researchers, paediatricians and parents, and improving the gains-hazard balance (by increasing incentives while decreasing inconveniences); and reforming in the health care system to raise the priority placed on clinical research by restructuring clinical research in a clinically predominant workplace and with a clinically predominant workforce. The findings from this study have implications for researchers planning RCTs for children in the future. Careful consideration of the above will enhance RCTs participation for children improving efficiency, lowering costs and ultimately improving the future health care of children.

Factorial randomised controlled trials (RCTs) evaluate two or more interventions simultaneously, enabling assessment of interactions between treatments. This thesis presents literature reviews, methodological reviews, simulation studies and applied case studies that explore methods for assessing cost-effectiveness based on factorial RCTs. My systematic review suggests that factorial RCTs account for around 3% of trial-based economic evaluations, although there is currently no guidance or methodological work indicating the most appropriate methods. Around 40% of published studies assumed no interaction between treatments and many were poorly-reported. Various mechanisms are likely to produce large interactions within economic endpoints such as costs, quality-adjusted life-years (QALYs) and net benefits. Failing to take account of interactions can introduce bias and prevent efficient allocation of healthcare resources. I developed the opportunity cost of ignoring interactions as a measure of the implications of this bias. However, allowing for small, chance interactions is inefficient, potentially leading to over-investment in research if trial-based evaluations are used to inform decisions about subsequent research. Nonetheless, analyses on simulated trial data suggest that the opportunity cost of adopting a treatment that will not maximise health gains from the healthcare budget is minimised by including all interactions regardless of magnitude or statistical significance. Different approaches for conducting economic evaluations of factorial RCTs (including regression techniques, extrapolation using patient-level simulation, and considering different components of net benefit separately) are evaluated within three applied studies, including both full and partial factorials with 2x2 and 2x2x2 designs. I demonstrate that within both trial-based and model-based economic evaluation, efficient allocation of healthcare resources requires consideration of interactions between treatments, and joint decisions about interacting treatments based on incremental cost-effectiveness evaluated “inside-the-table” on a natural scale. I make recommendations for the design, analysis and reporting of factorial trial-based economic evaluations based on the results of this thesis.

Recruitment to randomised controlled trials is known to be difficult. Poor recruitment has several adverse consequences. It affects the validity of study findings, is a common cause of trial extensions and may result in premature termination of trials, which is a huge loss in terms of invested funds, resources and lost knowledge. Non-completion or delayed completion of studies maintains the uncertainty about the efficacy and safety of interventions, thereby delaying or preventing the use of effective interventions and prolonging the use of ineffective or potentially harmful treatments. Recruitment of children to randomised controlled trials is thought to be more challenging due to the vulnerability of the population and the fact that consent is provided by another person usually parents. This thesis aims to review the recruitment performance, i.e. comparison of achieved to anticipated recruitment, of randomised controlled trials with children and identify the factors associated with good or poor recruitment. We undertook a pilot systematic review of recruitment and retention in randomised controlled trials with children, in published literature, and found that few studies report recruitment information but those that do, report very high rates of percentage total recruitment achieved (%TR) and consent. It was not possible to obtain unbiased estimates of recruitment performance and consent rate due to the likelihood of selective reporting and/or non-publication of trials with unsuccessful recruitment. We subsequently conducted a review of recruitment of children to randomised controlled trials in the National Institute of Health Research (NIHR) Clinical Research Network (CRN) portfolio and found that under-recruitment and delayed recruitment are common problems in paediatric trials. Having a trial manager or coordinator was found to be significantly associated with successful recruitment. Other factors such as being an IMP (Investigational Medicinal Product) vs. non-IMP trial, trial of acute vs. chronic illness, having CTU (Clinical Trials Unit) involvement, pilot/feasibility study and additional trial demands had no statistically significant association with recruitment success. Since recruitment to a clinical trial can be affected by a number of internal and external factors, we conducted a survey with the clinical teams of a multi-centre randomised controlled trial with children, the MAGNETIC trial, to understand the various facilitators and barriers to recruitment. In order to identify the facilitators and barriers to recruitment and establish the recruitment experience of clinical teams in a systematic manner, we developed an evidence based recruitment survey tool. The survey tool is an online questionnaire that presents a comprehensive evidence based list of facilitators and barriers and free text space for responders to record the strategies applied to overcome these barriers and suggestions for change in organisation of trials to boost recruitment. The survey of clinical teams recruiting to the MAGNETIC trial found that a motivated clinical team with effective communication skills, effective coordination between study team members at site and between sites and CTU, trial management support, research experience of PI, presence of a research nurse and availability of a designated research team were imperative for trial recruitment success. Heavy clinical workload, shift patterns of work, lack of trained staff particularly out of hours, GCP training, local clinical arrangements and parental anxiety about the safety of experimental treatment were recognised as important barriers to recruitment. A trial specific barrier was difficulty faced by the clinicians in seeking consent from the parents of an acutely ill child in the emergency setting and suggestions were made for consideration of deferred consent. We concluded that recruitment to randomised controlled trials with children is challenging and poor recruitment and recruitment delays are a common problem. Reporting of recruitment and consent in paediatric trials is poor and needs improvement. Presence of a dedicated trial manager is significantly associated with successful recruitment and the various generic and trial specific facilitators and barriers to recruitment that have been identified can be used by trialists in planning and conducting future clinical trials with children.

Unplanned changes to the research plan of a randomised controlled trial (RCT) may be a necessary response to unforeseen circumstances. Such changes can help ensure the value and relevance of a trial, but also have the potential to introduce a bias if performed inappropriately. This thesis addresses three broad aims. The first was to highlight the methodological implications of approaches to making various unplanned changes through a series of simulation studies. The second was to estimate the prevalence of such changes to a contemporary sample of published RCTs registered with the Australian New Zealand Clinical Trial Registry (ANZCTR). The third was to assess, and extend as applicable, recommendations for ensuring such changes are performed appropriately and are well documented. The simulation studies were performed using data from a large RCT of statin therapy for secondary prevention. The first of these demonstrated how unplanned changes (to various aspects of the design such as the primary endpoint) risks introducing a bias if undertaken with knowledge of treatment allocation, but not if made without knowledge of treatment allocation. The second investigated strategies for reacting to an observed imbalance on baseline prognostic factors in an RCT. It demonstrated that: continuing with original plans of unadjusted analyses, provided valid p-values irrespective of the direction of the prognostic imbalance, and any decisions informed by knowledge of the direction of the prognostic covariate imbalance were prone to bias. The third simulation study investigated strategies for reacting to a lower-than-expected event rate. It showed that switching to an expanded composite endpoint in response to a low pooled event rate does not inflate the type 1 error rate (and is likely to improve the statistical power, provided the expanded composite is sound). A sample of 181 RCTs that were registered with the ANZCTR and had published a primary result paper were assessed for changes to their research plans, and whether these changes were legitimate or not (that is, made with or without knowledge of treatment allocation). A full audit was conducted on trials with accessible protocols (N=124) and a limited central review on trials without accessible protocols (N=57). The primary results publication was cross-checked with the protocol documents across six methodological aspects for the full audit, and trialists were contacted to resolve queries as necessary. The publication was checked against the registry record for the limited central review, over a subset of three methodological aspects. A key finding of this study was that it was often not possible to reliably assess whether changes had been made or whether changes were made in a blinded manner, based on review of documentation alone. After clarification was sought from the participating trialists, changes were found to be relatively common but typically made in a blinded manner. Improvements to the way unplanned changes to RCT research plans are documented are needed. A set of recommendations to supplement existing guidelines relating to the documentation of RCT research plans was developed. One key recommendation was that trialists should adopt and implement the principles of the recently developed CONSERVE 2021 Statement to appropriately perform and document substantive changes. A second was that strategies that oblige trialists to upload full protocols (and protocol amendments) to clinical trial registries warrant further investigation and further consideration should be given to including all key methodological aspects in registry records. The work undertaken provides valuable guidance for trialists (and stakeholders) on how to make and document a methodologically justifiable change to a planned RCT design/analysis and highlights the circumstances under which a change can lead to bias. This is important for ensuring that methodically unsound changes to RCTs are recognised and avoided, and that RCTs that have undergone methodically sound changes are not incorrectly dismissed as potentially biased.

The occurrence, severity, and duration of patient adverse events are routinely recorded during randomised clinical trials. This data is used by a trial's Data Monitoring Committee to make decisions regarding the safety of a treatment and may lead to the alteration or discontinuation of a trial if real safety issues are detected. There are many different types of adverse event and the statistical analysis of this data, particularly with regard to hypothesis testing, must take into account potential multiple comparison issues. Unadjusted hypothesis tests may lead to large numbers of false positive results, but simple adjustments are generally too conservative. In addition, the anticipated effect sizes of adverse events in clinical trials are generally small and consequently the power to detect such effects is low. A number of recent classical and Bayesian methods, which use groupings of adverse events, have been proposed to address this problem. We illustrate and compare a number of these approaches, and investigate if their use of a common underlying model, which involves groupings of adverse events by body-system or System Organ Class, is useful in detecting adverse events associated with treatments. For data where this type of grouped approach is appropriate, the methods considered are shown to correctly flag more adverse event effects than standard approaches, while maintaining control of the overall error rate. While controlling for multiple types of adverse event, these proposed methods do not take into account event timings or patient exposure time, and are more suited to end of trial analysis. In order to address the desire for the early detection of safety issues in clinical trials a number of Bayesian methods are introduced to analyse the accumulation of adverse events as the trial progresses, taking into account event timing, patient time in study, and body-system. These methods are suitable for use at interim trial safety analyses. The models which performed best were those that had a common body-system dependence over the duration of the trial.

To evaluate the methodological quality of randomised trials in recently published articles and to examine the associations between methodological quality and bias, three related investigations were undertaken. First, to ensure the development of useful measures for the adequacy of randomisation, approaches to allocation were assessed as reported in 206 parallel group trials published in recent volumes of journals of obstetrics and gynaecology. Next, a study was conducted of associations between methodological quality and treatment effects. The material analysed came from 250 trials in 33 meta-analyses on pregnancy and childbirth topics. Finally, the reported approaches to blinding and handling of exclusions were assessed from a random sample of 110 of the 206 previously identified reports. In the 206 published trials, 77% reported either inadequately or unclearly concealed treatment allocation. Additional analyses suggest that non-random manipulation of comparison groups may have occurred. In the next study. compared with trials in which authors reported adequately concealed treatment allocation, trials in which authors reported inadequately or unclearly concealed allocation yielded larger estimates of treatment effects (p < 0.001). Odds ratios were distorted by 41% and 33%, respectively. Those associations likely represent bias and are particularly disconcerting in light of the results above from recently published trials. Lack of double-blinding in trials was also associated with larger treatment benefits. However, trials in which authors reported excluding 2 participants after randomisation were not associated with larger treatment effects. That lack of association appeared to be due to incomplete reporting. 3 The analysis of 110 recently published trials also supported the findings that some of the trials not reporting exclusions may actually have had exclusions. In practice, that incomplete reporting could lead to misinterpretations of trial quality. Moreover, only about half the trials that could have double-blinded actually did so. When investigators attempted double-blinding, only 16% provided any written assurances of successfully implementing blinding and only 6% tested its efficacy.

The study had three components: (i) an overview of the literature on participant involvement in RCTs; (ii) a meta-ethnography focusing on factors impacting on participant recruitment to RCTs; and (iii) an embedded qualitative study investigating recruitment and participation in an ongoing UK multicentre trial. Non participant observation of trial recruitment consultations and in-depth interviews were conducted with patients invited to participate in the trial from two recruitment sites. Decisions about trial participation took part in a broad personal context, such as perceptions of treatment, levels of symptoms control, and prior experience of interventions. Views about the trial procedures, treatment interventions, and impressions of recruiting staff were also found to be salient. Although altruism was identified as a factor impacting on trial participation, this tended to be ‘conditional altruism’, dependent on expectations of personally benefiting from participating in a trial. Perceptions of benefiting personally from a trial were linked to the trial intervention(s) and also towards the trial process. Despite agreeing to be randomised it was apparent that some patients harboured a ‘preference’ for a treatment.  However, people subsequently seemed able to adjust to their allocated treatment, although there were misconceptions about how treatment decisions had been reached. Perceptions of trial involvement, and having directly benefited from trial participation seemed particularly dependent on which treatment people were allocated to. Crucially, the findings have emphasised that trials are not simply an experimental tool operating in a vacuum, independent of the view and experiences of participants; and offer rich insights, which are likely to benefit people recruiting in future trials, and for future trial participants.

Randomised controlled trials are considered the gold standard study design, as random treatment assignment provides balance in prognosis between treatment arms and protects against selection bias. When trials are subject to departures from randomised treatment, however, simple but naïve statistical methods that purport to estimate treatment efficacy, such as per protocol or as treated analyses, fail to respect this randomisation balance and typically introduce selection bias. This bias occurs because departure from randomised treatment is often clinically indicated, resulting in systematic differences between patients who do and do not adhere to their assigned intervention. There exist more appropriate statistical methods to adjust for departure from randomised treatment but, as demonstrated by a review of published trials, these are rarely employed, primarily due to their complexity and unfamiliarity. The focus of this research has been to explore, explain, demonstrate and compare the use of causal methodologies in the analysis of trials, in order to increase the accessibility and comprehensibility by non-specialist analysts of the available, but somewhat technical, statistical methods to adjust for treatment deviations. An overview of such methods is presented, intended as an aid to researchers new to the field of causal inference, with an emphasis on practical considerations necessary to ensure appropriate implementation of techniques, and complemented by a number of guidance tools summarising the necessary clinical and statistical considerations when carrying out such analyses. Practical demonstrations of causal analysis techniques are then presented, with existing methods extended and adapted to allow for complexities arising from the trial scenarios. A particular application from epilepsy demonstrates the impact of various statistical factors when adjusting for skewed time-varying confounders and different reasons for treatment changes on a complicated time to event outcome, including choice of model (pooled logistic regression versus Cox models for inverse probability of censoring weighting methods, compared with a rank-preserving structural failure time model), time interval (for creating panel data for time-varying confounders and outcome), confidence interval estimation method (standard versus bootstrapped) and the considerations regarding use of spline variables to estimate underlying risk in pooled logistic regression. In this example, the structural failure time model is severely limited by its restriction on the types of treatment changes that can be adjusted for; as such, the majority of treatment changes are necessarily censored, introducing bias similar to that in a per protocol analysis. With inverse probability weighting adjustment, as more treatment changes and confounders are accounted for, treatment effects are observed to move further away from the null. Generally, Cox models seemed to be more susceptible to changes in modelling factors (confidence interval estimation, time interval and confounder adjustment) and displayed greater fluctuations in treatment effect than corresponding pooled logistic regression models. This apparent greater stability of logistic regression, even when subject to severe overfitting, represents a major advantage over Cox modelling in this context, countering the inherent complications relating to the fitting of spline variables. This novel application of complex methods in a complicated trial scenario provides a useful example for discussion of typical analysis issues and limitations, as it addresses challenges that are likely to be common in trials featuring problems with nonadherence. Recommendations are provided for analysts when considering which of these analysis methods should be applied in a given trial setting.

A sample size estimate for a clinical trial is an important issue as incorrectly estimating it could have both ethical and financial implications for the trial. Calculating the required sample size for a trial with a continuous outcome requires an estimate of the population variance. A pilot trial can be used to get an estimate of the population variance. However, pilot trials are often small and may give imprecise estimates; adjustment methods are discussed which allow for this imprecision. Theoretical minimum values for the overall trial sample size when using an adjustment method to design the main trial after an external pilot trial are provided. Using the results recommendations for external pilot trial sample size are presented which aim to minimise the overall trial sample size. It was found that the optimal pilot trial sample size increases with the size of the main trial, therefore stepped rules of thumb are proposed. For a 90% powered main trial this method indicates that the sample size for a two-armed pilot trial to minimise the overall sample size should be 150, 50, 30 and 20 for standardised effect sizes (δ) of δ < 0.1, 0.1 ≤ δ < 0.3, 0.3 ≤ δ < 0.7 and δ ≥ 0.7 respectively. The work is extended to allow for unequal cost per patient between the two trials. The results show that when the pilot trial is less expensive per patient than the main trial the optimal pilot trial sample size increases, giving more precision for the variance estimate and a relatively small main trial. The opposite is true when the main trial is less expensive than the pilot trial. For a 90% powered main trial this method indicates that the sample size for a two-armed pilot trial to minimise the overall sample size should be between 40- 260, 20-80, 20-40 and 20-30 dependent on the relative cost of the pilot and main trial per participant for standardised effect sizes (δ) of δ < 0.1, 0.1 ≤ δ < 0.3, 0.3 ≤ δ < 0.7 and δ ≥ 0.7 respectively. For internal pilot trials it is shown that the restricted sample size recalculation procedure raises the average sample size and power of the main trial. Aiming to minimise the overall trial sample size, it was found that the optimal pilot trial sample size rises as the main trial size increases. The work presented aims to help researchers choose sample sizes for pilot trials and to assess the impact selected methods have on the power and required sample size of the subsequent main trial.

Aims and objectives. For a particular randomised controlled trial, it is often useful to retrieve associated siblings - qualitative research, process and economic evaluations done alongside the randomised controlled trial (RCT). This thesis examines both the effectiveness and efficiency of search strategies, and the productivity of different databases, in retrieving sibling studies for an RCT. Methods. Five seed studies from different clinical areas were selected. A range of Boolean searches with simple subject term combinations with authors’ names, together with citation and similarity search strategies, were applied, on different databases that had different subject coverage and interests. Specialised search filters were combined with the simple search strategy and tested. The retrieval performances of the simple and sophisticated search strategies on PubMed were tested and compared using one of the seed studies as a case study. Recall, precision and odds estimators were used for all retrieval tests. Non-parametric statistical tests were used to test a set of hypotheses that set out to explore relationships underlying retrieval performance. Results. Neither one particular search strategy nor one database was an overall winner. The simple author-subject search provided a good recall with a readable retrieval size. The recall varied among seed studies and different databases. Search filters provided good recall for retrieving specific types of sibling, especially the qualitative filter. PubMed related articles strategy provided a good performance for some seeds, but not as good overall as the simple author-subject searching. Combining a similarity search with simple author-subject search provided complementary retrieval performance and therefore yields an optimal performance. Citing search did not perform well in terms of retrieving sibling studies. The simple author-subject search shows performance consistency, being the best search strategy among other strategies for all seed studies in terms of recall and precision. WoK and SCOPUS were the best databases for retrieving sibling studies. Conclusions. Simple author-subject search, especially when searching multiple databases, can yield an optimal performance in retrieving sibling studies.

Smokers attend preferentially to smoking-related cues in the environment, known as attentional bias. Evidence suggests that attentional bias is related to craving and relapse. Attentional retraining (AR) procedures have been used in laboratory studies to modify attentional bias and processes related to drug use, but investigations on the clinical value of AR in addiction are scarce. This thesis reports on two randomised controlled trials investigating the efficacy of AR with modified visual probe tasks in smokers. The first study explored the effects of varying the length of AR on attentional bias, craving, mood and withdrawal in current smokers. No retraining effects were observed after either a short, medium or long block of AR. The second study explored the efficacy of AR on attentional bias and smoking cessation outcomes in treatment-seeking smokers. While AR procedures were feasible to deliver within smoking cessation clinics, the intervention did not significantly reduce attentional bias, craving, withdrawal symptoms or the likelihood of relapse. These results and the literature in general show that there is no clear association between attentional bias and craving and relapse. Current AR procedures are not effective in smokers and should not be used in smoking cessation treatments, as they currently stand.

‘The capability approach is a broad, normative framework for the evaluation of well-being’(p.94)[1], which has attracted growing interest in health and health economics research. A broader measure of well-being may more accurately capture the effects of some interventions, than traditional health-related quality of life measures. The ICECAP-A and ICECAP-O are two measures of a person’s well-being, with a theoretical grounding in the capability approach, designed for use in health and social care research. This thesis reports qualitative and quantitative investigations into the validity and responsiveness of the ICECAP measures. A methodological review of existing validation studies was completed. Seventeen semi-structured interviews with health research professionals were carried out and an iterative, constant comparative, thematic analysis was completed to assess the content validity of the ICECAP-A. The construct validity and responsiveness of the measures were assessed using two randomised controlled trials: the BEEP trial (ISRCTN 93634563) and the Past BP trial (ISRCTN 29062286). Qualitative and quantitative results provide positive indications of validity. The qualitative work showed that research professionals viewed the ICECAP-A as a relevant and feasible measure for use in health research. The quantitative results confirmed the majority of a priori hypotheses in the validity analyses, while longitudinal data provided evidence that the measures are responsive to self-reported changes in health status. In conclusion, this thesis reports the first assessment of validity in a randomised controlled trial setting and the first analysis of responsiveness. While further testing of the ICECAP measures is required, results indicate that the measures are appropriate for use in health research.

This thesis aims to extend the existing knowledge and enhance the methodological quality of future stepped-wedge cluster randomised trial (SW-CRTs). A systematic review of published SW-CRTs shows that pre-trial sample sizes calculations display a poor standard of reporting, with little adherence to published guidelines. The methodological rigor is often substandard, with inappropriate methods often used to determine sample size. In SW-CRTs, it is assumed that the correlation between observations is independent of the timing of them. We test the validity of this assumption by outlining a method to estimate the within-period and inter-period correlation. A case study illustrates what these correlations may look like in practice. The impact of varying cluster size in a SW-CRT is then demonstrated by comparing a design with unequal cluster size to a design with equal cluster size. A simulation study provides evidence that the SW-CRT is affected less, on average, than a P-CRT by varying cluster size. However, the potential power in a SW-CRT with unequal cluster sizes is extremely variable. A practical method for estimated power in a SW-CRT with varying cluster size is then illustrated through a Stata function.

Stroke is a major cause of death and disability in the UK. Few treatments exist and those that do, such as thrombolysis (‘clot-busting’ treatment) must be given urgently and are not risk-free. Large scale randomised controlled trials are crucial for the development of safe, effective, acute interventions, but progress has been limited, ostensibly due to ethical and regulatory difficulties. Theoretical work in this area has focussed primarily upon the requirement for prospective informed consent, but has also considered potential conflicts of interests inherent in the dual role of clinicianresearchers, and the notion that research and clinical practice are, can be, and should be conducted separately. Empirical evidence on this topic is lacking. By providing such evidence, this study examines claims made in the literature regarding the difficulties encountered or perceived in conducting emergency research. It also explores whether, how, and to what effect, the distinction between research and clinical activity advocated in the bioethical literature is maintained. Methods Ethnographic methods were employed, including participant observation, semistructured interviews, and audio-recording of research consent interactions in an acute stroke unit. Data were analysed drawing upon constant comparative and framework methods. Results and conclusion Whilst providing empirical evidence supporting some of the theoretical and conceptual literature, the data also furnish a detailed account of pragmatic issues encountered and managed daily by healthcare professionals in the acute stroke environment. Whilst attempts were made at the study site to separate, at least in part, clinical and research activity, it was observed that absolute separation of clinical activities is neither attainable, sustainable, nor desirable. Placement of research nurses within the clinical environment may promote transparency and greater understanding of their role, whilst simultaneously demystifying research concepts. Ultimately this may promote closer working relationships, contributing to enhanced recruitment, retention and management of research participants.

Randomised controlled trials (RCTs) play a fundamental role in the development and marketing activities of pharmaceutical companies. They are the primary means of evaluating the tolerability, safety and efficacy of a drug, and for providing information relevant for pricing and reimbursement decisions and clinical decision-making. RCTs require a substantial investment by pharmaceutical companies and the financial consequences of poorly or sub-optimally designed trials are potentially substantial. Revenue does not materialise unless a licence to market a product is granted and sales may be restricted if a trial fails to provide evidence of sufficient strength or relevance for those involved in product adoption decisions. From a pharmaceutical company's perspective, the value of RCTs can therefore be judged on the contribution they make to the performance of a drug in the market and hence on their contribution to the performance of the firm. Consequently the design choices made in the planning of RCTs are effectively investment appraisal decisions. However, the application of investment appraisal techniques to RCT design has not previously been proposed. The purpose of this thesis is to consider how private sector investment appraisal methods might be applied to RCT design decision-making and to explore aspects of the practicalities of application. A general investment appraisal model is presented and its application to determine profit maximising RCT designs is illustrated. Considering the cost side of the investment appraisal equation, it is shown how decision-makers' requirements for cost-effectiveness evidence derived from trials could have a significant impact on the major determinants of cost (sample size and study duration) depending on their specific preferences for evidence defined over key components of RCT design. Considering the revenue side of the investment appraisal equation, it is shown how discrete choice analysis could be used to incorporate decision-makers' preferences for RCT designs into the planning of studies. Specifically, it is shown how the predicted probabilities derived from the application of this technique could be used within an investment appraisal framework. Directions for future research into the application of investment appraisal to RCT design are proposed.

OBJECTIVE: This study examined participants’ opinions and beliefs about Randomised Controlled Trials (RCTs) in an intellectual disability context. BACKGROUND: RCTs in this field require co-operation from various stakeholders, including carers and professionals from a variety of disciplines. However, previous research indicates that local stakeholders may have negative views regarding RCTs in this population, and that it may be difficult for researchers to gain access to participants. This is compounded by the potential problems surrounding communication with a proportion of the service users. METHOD: The present study builds upon an RCT for a behaviour therapy intervention for people with intellectual disability, which was situated within community based services in one county of South East England. Fifty-one individuals were interviewed; 11 paid carers, 7 family carers, 6 adults with mild intellectual disability, and 27 professionals from health and social care services. The interviews elicited opinions, beliefs and decision-making processes relating to stakeholder experiences of the RCT. Data was analysed through coding emergent categories into a framework, which evolved throughout the analysis. RESULTS: The data revealed that opinions about RCTs were shaped by several concerns. The most important of these included the following; continued ability to access interventions, the ethical concerns surrounding randomisation, perceptions of limited financial resources, and problems involving communication and consent. DISCUSSION: RCTs are ubiquitous in clinical research, including psychiatry. However, they present difficulties for researchers and participants in the field of intellectual disability. Good communication with all stakeholders is essential to ensure the successful conduct of an RCT. This study provides information for academics and clinicians who plan to conduct future research and RCTs with people who have intellectual disability. The findings may be used in future to develop appropriate strategies to assist with recruitment for RCTs in intellectual disability, and to increase stakeholders’ acceptance of the procedure.

Introduction and background Recruitment to cancer clinical trials needs to be improved, as does patient understanding about clinical trials, to enable patients to make an informed choice about whether or not to take part. The main reason that clinically eligible patients do not take part in clinical trials is because they refuse; poor understanding of the research has been associated with patient refusal. Audiovisual patient information (AVPI) has been shown to improve knowledge/understanding in various areas of practice but there is limited information about its effect in the cancer clinical trial setting, particularly in relation to recruitment rates. Understanding the research is necessary for informed consent, and it was hypothesised that if patient understanding about clinical trials was increased with AVPI, then this could result in a reduction in the number of patients refusing clinical trials, and therefore provide an ethical approach to improving recruitment. This study aimed to test the impact of an audiovisual patient information intervention on recruitment to randomised cancer clinical trials (refusal rates), patient understanding of the information given, and levels of anxiety. Reasons for patients’ decisions about trial participation were also assessed. Method An AVPI intervention was developed that aimed to address the common misconceptions associated with randomisation and clinical equipoise, as well as improve patient understanding generally of randomised cancer trials, and of other core clinical trial informational requirements, such as voluntariness. Patients were randomised to receive either AVPI in addition to the standard trial-specific written information, or the written information alone. A new questionnaire was developed to assess patient understanding (also referred to as knowledge) in the randomised trial setting and, following testing with patients and research nurses, this was shown to be reliable and valid. Patients completed self-report questionnaires to assess their understanding (new knowledge questionnaire) and anxiety (Spielberger State-Trait Anxiety Inventory), at baseline and after they had made their decision about clinical trial entry, when their perceptions of the intervention, as well as factors contributing to their decision were also determined (this tool incorporated Jenkins and Fallowfield’s (2005) questionnaire which assessed reasons for accepting and declining randomised cancer trials). Results A total of 173 patients with breast cancer (65%), colorectal cancer (32%) and lung cancer (3%) were entered into the main study. The median age was 60 (range 37-92 years). There was no difference in clinical trial recruitment rates between the two groups: 72.1% in the AVPI group and 75.9% in the standard information group. The estimated odds ratio for refusal (intervention/no intervention) was 1.19 (95% ci 0.55-2.58, p=0.661). Knowledge scores increased more in the intervention group compared to the standard group (U= 2029, p=0.0072). The change in anxiety score between the arms was also statistically significant (p=0.011) with anxiety improving in the intervention arm more than in the no-intervention arm. The estimated difference in the median anxiety change score between the groups is –4.6 (95% ci –7.0 to –2.0). Clinical trial entry was not influenced by tumour type, stage of cancer, age, educational qualifications or previous research experience, however, there was a modest association with deprivation status (p=0.046) where more affluent patients were the least likely to consent to a trial. Educational qualifications and stage of cancer were independently associated with knowledge: patients who were better educated had higher levels of knowledge about randomised trials, and patients who had limited stage of cancer had higher baseline knowledge than patients with advanced cancer. Acceptability of the intervention was high with 93% of those who watched it finding it useful, and 42% stating that it made them want to take part in the clinical trial. Personal benefit and altruism were key motivating factors for clinical trial participation, with reasons for refusal being less clear. Discussion and conclusions Although the potential for AVPI to increase clinical trial recruitment rates was highlighted in the literature, in this study, AVPI was not shown to have any effect on refusal rates to randomised cancer trials. However, by improving patient understanding prior to decision making, AVPI was shown to be a useful addition to the consent process for randomised cancer trials. AVPI addresses the fundamental ethical challenges of informed consent by improving patient understanding, and supports the ethical framework integral to Faden and Beauchamp’s (1986) theory of informed consent. The new knowledge questionnaire was shown to be a sensitive and effective instrument for measuring understanding of randomised clinical trials in the cancer setting, although it would benefit from further testing. The AVPI appears to reduce anxiety at the decision making time point and has been shown to be an acceptable medium for patients. This study confirms existing findings from studies assessing factors affecting decision making, with personal benefit and altruism being key motivating factors, and reasons for refusal being less clear. The need for further qualitative work in this area is highlighted to gain a deeper understanding of what is important to patients, in terms of why they refuse clinical trial participation. Implications for practice and further research Several implications for practice have been identified, including using AVPI as part of the standard information package for patients considering randomised cancer trials, and focussing on patient and staff education in this area. The knowledge questionnaire could be introduced to routine practice as a tool to determine patient understanding prior to decision making, allowing clinicians the opportunity to correct any misconceptions prior to consent. Further research focussing on AVPI specific to individual trials would be helpful, to determine if a more customised approach would be of benefit in terms of clinical trial recruitment. The importance of studying other aspects of the consent process such as the interaction between the clinician and the patient, in addition to more detailed exploration of the factors affecting patients’ decisions were highlighted.

Thesis (Ph.D.) - University of Glasgow, 2008.
Ph.D. thesis submitted to the Department of Statistics, Faculty of Information and Mathematical Sciences, University of Glasgow, 2008. Includes bibliographical references. Print version also available.

China is an emerging force in undertaking randomised clinical trials. The quality of trials from China may affect not only its own substantial population but also potentially contribute to health policy throughout the world. However, little is known about the quality of clinical trials conducted there. In this thesis, I will evaluate the quality of published Chinese randomised controlled trials (RCTs) by comparing them with Indian RCTs as well as a set of ‘gold standard’ trials reported in leading European and North American journals. I will also describe and contrast the quality and biases within Chinese RCTs. I then explore the reasons for these differences from the point of view of the major RCT stakeholders: Chinese clinicians and patients. The potential influences from Chinese traditional culture is also evaluated. Chinese medical journal editors need to undertake more training on the reporting of RCTs; all medical societies should take more concern about doctors’ research work; the Chinese public media should help the general population to understand more about RCT principles.

Background: Glaucoma is a chronic, progressive eye disease and the second cause of blindness in the world. To measure the patients’ perspective in randomised controlled trials (RCTs), patient-reported outcome measures (PROMs) are increasingly being used. However, the use of PROMs in glaucoma trials is low suggesting there may be a reluctance to use PROMs. Objectives: To explore three methodological challenges of using PROMs in RCTs in glaucoma: 1) PROM selection; 2) characterising glaucoma severity; and 3) interpreting PROM scores in terms of minimal important difference (MID). Methods: Vision PROMs used in glaucoma studies were identified and content validated using a systematic review approach and categorised by a new PROM taxonomy. Existing visual field staging systems (VFSSs) based on standard automated perimetry were systematically identified and quality assessed with a new tool developed for this review using a consensus method. The performance of four high quality visual field staging systems were evaluated and referenced against an experienced ophthalmologist in a diagnostic test accuracy study. A pilot study using the social comparison approach was undertaken to test the feasibility of an anchor-based approach in determining the MID of a vision PROM in a glaucoma population. Results: Thirty-three vision PROMs were identified and categorised, according to content into impairment, disability, status and satisfaction measures. Twenty-three VFSSs were identified but evaluation of quality assessment, particularly performance, was affected by poor VFSS reporting. The diagnostic accuracy study demonstrated suboptimal performance of the four highest quality staging systems. The pilot study to determine the MID for a vision PROM found the social comparison method to be a feasible approach in a glaucoma population. Conclusion: This thesis demonstrated how to select a PROM and identified difficulties with characterising glaucoma severity. Future research needs include development of robust methods for characterising glaucoma severity and full scale evaluation of MIDs in PROMs in glaucoma.

Most drugs come with unwanted, and perhaps harmful, side-effects. Depending on the size of the treatment benefit such harms may be tolerable. In acute stroke, treatment with aspirin and treatment with alteplase have both proven to be effective in reducing the odds of death or dependency in follow-up. However, in both cases, treated patients are subject to a greater risk of haemorrhage – a serious side-effect which could result in early death or greater dependency. Current treatment licenses are restricted so as to avoid treating those with certain traits or risk factors associated with bleeding. It is plausible however that a weighted combination of all these factors would achieve better discrimination than an informal assessment of each individual risk factor. This has the potential to help target treatment to those most likely to benefit and avoid treating those at greater risk from harm. This thesis will therefore: (i) explore how predictions of harm and benefit are currently made; (ii) seek to make improvements by adopting more rigorous methodological approaches in model development; and (iii) investigate how the predicted risk of harm and treatment benefit could be used to strike an optimal balance. Statistical prediction is not an exact science. Before clinical utility can be established it is essential that the performance of any prediction method be assessed at the point of application. A prediction method must attain certain desirable properties to be of any use, namely: good discrimination – which quantifies how well the prediction method can separate events from non-events; and good calibration – which measures how close the obtained predicted risks match the observed. A comparison of informal predictions made by clinicians and formal predictions made by clinical prediction models is presented using a prospective observational study of stroke patients seen at a single centre hospital in Edinburgh. These results suggest that both prediction methods achieve similar discrimination. A stratified framework based on predicted risks obtained from clinical prediction models is considered using data from large randomised trials. First, with three of the largest aspirin trials it is shown that there is no evidence to suggest that the benefit of aspirin on reducing six month death or dependency varies with the predicted risk of benefit or with the predicted risk of harm. Second, using data from the third International Stroke Trial (IST3) a similar question is posed of the effect of alteplase and the predicted risk of symptomatic intracranial haemorrhage. It was found that this relationship corresponded strongly with the relationship associated with stratifying patients according to their predicted risk of death or dependency in the absence of treatment: those at the highest predicted risk from either event stand to experience the largest absolute benefit from alteplase with no indication of harm amongst those at lower predicted risk. It is concluded that prediction models for harmful side-effects based on simple clinical variables measured at baseline in randomised trials appear to offer little use in targeting treatments. Better separation between harmful events like bleeding and overall poor outcomes is required. This may be possible through the identification of novel (bio)markers unique to haemorrhage post treatment.

Concern with potential toxicity due to the widespread use of unlicensed and off label drugs in children has led to regulatory changes aimed to strengthen the evidence base for paediatric drugs. This thesis examines paediatric randomised controlled trials (RCTs), the highest level of evidence, and assesses them in relation to global child health. A systematic review was performed using validated methods to search three major databases for paediatric RCTs published in 2007. More than 600 RCTs were identified involving more than 100,000 children. The RCTs appear to study the appropriate clinical areas however few studies involved neonates. The RCTs also seem to be of good methodological quality with a mean Jadad score of 3.22. The reporting of RCTs that involve both adults and children needs to be improved to add to the evidence base of paediatric medicines. More attention is also needed on the reporting of safety information from the RCTs to provide useful toxicity data. Although severe and moderate ADRs were seen in 25% of the RCTs, few RCTs (12%) established safety monitoring committees (SMCs). SMCs are vital to ensure patients in paediatric RCTs are protected from toxicity. The burden of childhood disease is heaviest in low and middle income countries (LMIC). A minority of the RCTs were performed in LMIC, although they are increasingly globalised. RCTs conducted in LMIC appear to have lower methodological quality, and reported less well on ethical approval and adverse events. In conclusion high quality, ethical paediatric RCTs should add to the evidence base for paediatric medicines. However they should correspond with the health needs of children on a global basis.

The majority of research on the placebo effect has focused on beneficial effects in patients or participants told to expect an active treatment, but who are actually given a placebo. Two important and relatively understudied aspects of the placebo effect are the extent to which expectancies influence outcomes in double-blind randomised placebo-controlled trials (RCTs) and whether the placebo effect contributes to treatment side effects. The current project investigated these two issues in both clinical and experimental settings. The first study involved reanalysing a double-blind RCT of naltrexone and acamprosate for alcohol dependence based on whether participants believed they had been allocated to receive active treatment or placebo (perceived treatment). The second study extended on this by developing an experimental model for these effects using dummy (placebo only) double-blind RCTs for cognitive performance. This allowed for the manipulation of observable changes in the form of false feedback. The third study investigated whether warning participants about side effects increases their occurrence, frequency, and/or severity in three dummy trials for sleep difficulty in healthy volunteers. The final study complemented this by examining whether first time chemotherapy patients’ expectancies for nausea were associated with their post-chemotherapy nausea. The studies on perceived treatment in double-blind RCTs indicated that participants’ beliefs about their treatment allocation can influence their actual treatment outcomes via the placebo effect and that these beliefs are affected by the feedback they receive about their performance. The studies on placebo-induced side effects indicated that the placebo effect may contribute to treatment side effects but that this effect is generally likely to be small. These findings confirm that the placebo effect can influence treatment outcomes and emphasise the importance of considering patient expectancies when delivering medical treatment. They also highlight some general limitations associated with research on the placebo effect, which include, whether conveying uncertainty undermines the placebo effect and whether measuring or manipulating expectancies is the best way to evaluate the placebo effect.

Evidence-based policing (EBP) has emerged as a key strand of police innovation since Sherman’s (1998) Police Foundation lecture. However, for others EBP raises as many questions as answers. One of the most contentious areas is the role advocated for randomised controlled trials in testing practice and developing knowledge to support EBP. RCTs are controversial with some scholars who argue that policing is not comparable to medicine and that RCTs are unable to reflect the complexity of the police role and context. Even those who advocate the use of RCTs recognise that there are significant challenges in achieving the high dosage and high fidelity that a successful experiment requires. This dissertation responds to these challenges by analysing the completed randomised controlled trials in policing and using a case study, Operation Turning Point, to identify the factors that may contribute to the conduct and management of police field trials with high levels of treatment integrity. In the introduction, Chapter 1, the approach is set out, framed around grounded theory, to be developed in four, linked, chapters. Chapter 2 is focused on understanding treatment integrity in RCTs involving the police: A search for police RCTs is produced 122 Police RCTs completed and reported by 2016. The levels of treatment integrity are analysed. 78 of the 122 RCTs exceeded a 60% threshold, with 49 being above 90%. In Chapter 3, a “novice theory” is developed and tested as an explanation for levels of treatment integrity in police randomised controlled trials: Analysis of the 122 RCTs suggests that “novice theory” can provide an explanation for the general patterns of treatment integrity. Further detailed analysis suggested that there are, however, other factors which may be important in determining the treatment integrity. These are developed in Chapter 4, which centres on a case study of Operation Turning Point. Using published case studies and an analysis of juvenile justice RCTs, a potential framework of operational factors is developed that appear to be important in effective conduct and management. The Turning Point case study is used to develop and expand on those operational factors. Finally, taking the two together, the analysis concluded that, beyond the operational factors, there were some more strategic, “protective factors” that were also critical. These are developed in Chapter 5, by using the coding and analysis of interviews with a sample of key staff involved in Turning Point Our analysis suggests that novice theory needs to be understood in the context of both the operational and protective factors that we have identified. Taken together these findings indicate the potential advantages of building institutional frameworks in which the development of practitioners and researchers and the conduct and management of experimental research could be brought closer together. We conclude with ten recommendations designed to improve the treatment integrity of police RCTs.

The purpose of this research study was to improve the nursing care of intravenous catheters by providing evidence on the effects of prolonged duration of intravenous administration set use. Intravenous therapy is a vital part of modern health care. However, its invasive nature can result in infection, with high associated morbidity and mortality. The highest infection rates are displayed in intensive care patients with central venous catheters. The duration of intravenous administration set use may have an impact on infection rates,however the current practice usage and the optimum duration of use is unknown. Previous studies of central venous catheters have reported equal infection rates with 1 to 4 days of administration set use; however few patients have been evaluated with administration sets used beyond this time. Previous research has been limited by the inadequacy of available definitions for Catheter-Related Infection. A prospective, randomised, controlled clinical trial was performed to assess the infection risk of using administration sets for prolonged periods. In the developmental phase prior to the clinical trial; definitions of Catheter-Related Bloodstream Infection (CRBSI) were developed; a nursing practice survey was undertaken to establish the current duration of administration set use; and laboratory experiments were executed to assess the impact of prolonged use on administration set physical integrity and performance. Central venous catheters were randomised to have their administration sets used for 4 days (n = 203) or 7 days (n = 201). Percutaneous central venous catheters were enrolled into the study from two adult intensive care units at a metropolitan, tertiary-referral, teaching hospital. Catheters were multiple-lumen, chlorhexidine-gluconate and silver-sulphadiazine coated lines, both inserted and removed in the intensive care unit. Catheters were cultured for microbial colonisation on removal using the Maki roll-plate technique. Patients were assessed for CRBSI using the developed definitions consisting of categories: definite, probable (type I and II), possible and absent. Prior to the clinical trial, a practice survey questionnaire was administered, and laboratory experimentation was performed. Normality of distribution for continuous variables was assessed using the Kolmogorov- Smirnov statistic. The distribution between groups of variables considered risk factors for Catheter-Related Infection were tested to assess for bias using Chi-square and T-test. Logistic regression modelling was performed to analyse the influence of potentially confounding variables. The incidence of catheter colonisation and CRBSI was tested between groups using Kaplan-Meier survival curve with Log-rank test. Paired T-tests were performed to test for difference in programmed and delivered volumes of administration sets. A general linear model (ANOVA)± a Scheffe post hoc test to isolate difference was fitted to the standardised values of delivered volumes to determine the effects of day of measurement and volume delivery rate on the accuracy of volume delivery. There were 10 colonised tips in the intervention group and 19 in the control group. This difference was not statistically significant (Kaplan Meier survival analysis, Log Rank = 0.87, df = 1, p = 0.35). There were 3 cases of CRBSI per group and the difference in survival from CRBSI was not statistically significant (Kaplan Meier with Log Rank test, p = 0.86). The pre-clinical trial phases of the research programme established that current clinical practice was 3 to 7-day use of administration sets; that administration sets were physically intact and delivered clinically accurate volumes after 7 days of use; and developed useful definitions of CRBSI. Prolonged intravenous administration set use of 7 days was found to have no significant impact on patient infection indicators or physical performance of the sets. This finding is congruent with previous research and trends in current clinical practice. In conclusion, the research findings support the use of intravenous administration sets for 7 days.

The purpose of this research study was to improve the nursing care of intravenous catheters by providing evidence on the effects of prolonged duration of intravenous administration set use. Intravenous therapy is a vital part of modern health care. However, its invasive nature can result in infection, with high associated morbidity and mortality. The highest infection rates are displayed in intensive care patients with central venous catheters. The duration of intravenous administration set use may have an impact on infection rates,however the current practice usage and the optimum duration of use is unknown. Previous studies of central venous catheters have reported equal infection rates with 1 to 4 days of administration set use; however few patients have been evaluated with administration sets used beyond this time. Previous research has been limited by the inadequacy of available definitions for Catheter-Related Infection. A prospective, randomised, controlled clinical trial was performed to assess the infection risk of using administration sets for prolonged periods. In the developmental phase prior to the clinical trial; definitions of Catheter-Related Bloodstream Infection (CRBSI) were developed; a nursing practice survey was undertaken to establish the current duration of administration set use; and laboratory experiments were executed to assess the impact of prolonged use on administration set physical integrity and performance. Central venous catheters were randomised to have their administration sets used for 4 days (n = 203) or 7 days (n = 201). Percutaneous central venous catheters were enrolled into the study from two adult intensive care units at a metropolitan, tertiary-referral, teaching hospital. Catheters were multiple-lumen, chlorhexidine-gluconate and silver-sulphadiazine coated lines, both inserted and removed in the intensive care unit. Catheters were cultured for microbial colonisation on removal using the Maki roll-plate technique. Patients were assessed for CRBSI using the developed definitions consisting of categories: definite, probable (type I and II), possible and absent. Prior to the clinical trial, a practice survey questionnaire was administered, and laboratory experimentation was performed. Normality of distribution for continuous variables was assessed using the Kolmogorov- Smirnov statistic. The distribution between groups of variables considered risk factors for Catheter-Related Infection were tested to assess for bias using Chi-square and T-test. Logistic regression modelling was performed to analyse the influence of potentially confounding variables. The incidence of catheter colonisation and CRBSI was tested between groups using Kaplan-Meier survival curve with Log-rank test. Paired T-tests were performed to test for difference in programmed and delivered volumes of administration sets. A general linear model (ANOVA)± a Scheffe post hoc test to isolate difference was fitted to the standardised values of delivered volumes to determine the effects of day of measurement and volume delivery rate on the accuracy of volume delivery. There were 10 colonised tips in the intervention group and 19 in the control group. This difference was not statistically significant (Kaplan Meier survival analysis, Log Rank = 0.87, df = 1, p = 0.35). There were 3 cases of CRBSI per group and the difference in survival from CRBSI was not statistically significant (Kaplan Meier with Log Rank test, p = 0.86). The pre-clinical trial phases of the research programme established that current clinical practice was 3 to 7-day use of administration sets; that administration sets were physically intact and delivered clinically accurate volumes after 7 days of use; and developed useful definitions of CRBSI. Prolonged intravenous administration set use of 7 days was found to have no significant impact on patient infection indicators or physical performance of the sets. This finding is congruent with previous research and trends in current clinical practice. In conclusion, the research findings support the use of intravenous administration sets for 7 days.

Background: Successful recruitment, adherence and retention are essential for randomised controlled trials (RCTs) to produce robust and meaningful findings. Studies exploring trial participation predominantly focus on the characteristics and views of participants and staff and often reveal contradictory findings. To date little is known about adherence and retention to RCTs. Aim: To further understanding of recruitment, adherence and retention to non-therapeutic paediatric RCTs, with particular emphasis on the role of social context. Methods: An ethnographic approach was taken using two RCTs as case studies. Participant observation (130 hours) was conducted on a clinical trials unit. Twenty-six trial staff and 56 parents who considered or had participated in the RCTs were interviewed and relevant documents collected. Data were analysed using the principles of thematic analysis. Results: Drawing on Bourdieu’s (1977; 1990) ’Theory of Practice’ and Titmuss’ (1970) ’The Gift Relationship’ it was evident that recruitment, adherence and retention were influenced by the values and beliefs of parents and staff and by the wider context in which the RCTs were conducted. Recruitment and adherence were influenced by the degree of concordance between the philosophies of the trials’ fields and those of the wider fields of parenting, infant feeding, medical research and allergy healthcare. Perceptions of personal and societal benefit were relevant to participation but, reflecting the philosophy of the parenting field, families often prioritised personal benefit.
Conclusion: Open and regular personal communication between parents and staff was particularly important for retention. Trials that maximise personal contact may have more success retaining participants. Comparing recruitment, adherence and retention between the two RCTs illuminated the relevance of the wider context for participation, particularly recruitment and adherence. Conducting a thorough assessment of the context in which an RCT will take place will allow potential barriers to participation to be identified before trial commencement.

Introduction: Oral mucositis is an inflammatory and frequently ulcerative side effect of cancer therapy, which has been identified by patients as the most debilitating side effect of their treatment. Mucositis is a dose limiting toxicity which exerts a substantial clinical and economic impact and negatively affects patient quality of life. The patient experience of mucositis is under-reported in the literature. To date, no interventions have been identified that have proven successful in the prevention of mucositis for patients receiving all types of therapy. Vitamin E has shown conflicting results in clinical trials. This thesis combines appraisal of the literature and empirical research,and uses lessons learned from previous studies together with the results of a feasibility study to identify a best practice model for future trials. Methods: The Cochrane risk of bias (ROB) instrument was used to assess the ROB in the studies included in the Cochrane prevention review. A sensitivity analysis was conducted after studies assessed at unclear or high risk of overall bias were excluded. A systematic review of assessment instruments was conducted which identified 50 instruments. Consideration of the appropriateness of these instruments for the use in a clinical trial for the prevention of mucositis was based on the practicality, comparability, and reproducibility, and the impact of these instruments on patients. Three of these instruments were chosen for use in a clinical trial of adults undergoing stem cell transplant. Finally, a feasibility study was designed, developed and conducted which investigated vitamin E for the prevention of mucositis in patients undergoing conditioning for bone marrow transplantation. Through lessons learned from previous studies, consultations with medical professional, the MHRA, ethics committee and suppliers, a protocol was developed for a double blind RCT. The process of gaining MHRA and ethical approval, and the repackaging of intervention and placebo products to meet MA-IMP requirements are described. Results: 130 articles were assessed for risk of bias. Only ten studies were assessed as being at low overall risk of bias. Blinding of outcome assessors and adequate allocation concealment were identified to be important considerations in the planning of future studies. Although only nine patients were recruited into the feasibility study, a number of issues affecting the design and conduct of future trials were identified. Recruitment in particular was identified to be problematic. Strategies for overcoming this problem in future trials were discussed. The methods of blinding and allocation concealment employed were found to be feasible for use in future trials. Expected adverse events patients undergoing stem cell transplantation were also reported. Conclusion: Further studies are required to investigate interventions for the prevention of mucositis. It is of upmost importance that these trials are rigorous in both their methodology and subsequent reporting in order to elicit the maximum benefit for patients taking part in clinical trials, and future patients undergoing therapy for cancer.

Background: The ethical basis of randomised controlled trials is equipoise, whether at the collective or individual level. Neonatal intensive care trials are therefore conducted in a context of clinical uncertainty as well as stress and trauma. The theoretical literature suggests that tensions exist in the trials situation between the aims of care and research. Objectives: To improve understanding of decisions that clinicians and parents make about neonatal trial collaboration, participation and non-participation. Methods: Semi-structured interviews were conducted with 30 neonatologists and 63 parents from 5 UK hospitals who were offered enrolment in the INNOVO and/or CANDA trials. Qualitative analysis was aided by ATLAS-ti. Results: The neonatologists' interviews suggested an intermediate level of equipoise. A therapeutic orientation operated for the INNOVO Trial but not for the CANDA Trial. Neonatologists often did not connect trial participation and trial-related postmortem pathology studies. Most parents made very rapid decisions about trial participation. Perception of risk was independent of the trial under consideration but associated with a slower decision-making process. The 'therapeutic misconception' was present for parents in both trials. Many supported contributing to research. For some of the bereaved parents, this extended to contribution to trial-related pathology studies. Parents who declined the CANDA Trial saw risks in the trial situation. Conclusions: Decisions were complex and multi-tiered. The boundaries between care and research were often unclear for neonatologists and parents. Clarification of the nature of decisions at the heart of clinical trials is needed, so that those associated with research might be willing collaborators and participants, fully cognisant of the activity in which they are engage

Missing data is a potential source of bias in the results of randomised controlled trials (RCTs), which can have a negative impact on guidance derived from them, and ultimately patient care. This thesis aims to improve the understanding, handling, analysis and reporting of missing data in patient reported outcome measures (PROMs) for RCTs. A review of the literature provided evidence of discrepancies between recommended methodology and current practice in the handling and reporting of missing data. Particularly, missed opportunities to minimise missing data, the use of inappropriate analytical methods and lack of sensitivity analyses were noted. Missing data patterns were examined and found to vary between PROMs as well as across RCTs. Separate analyses illustrated difficulties in predicting missing data, resulting in uncertainty about assumed underlying missing data mechanisms. Simulation work was used to assess the comparative performance of statistical approaches for handling missing available in standard statistical software. Multiple imputation (MI) at either the item, subscale or composite score level was considered for missing PROMs data at a single follow-up time point. The choice of an MI approach depended on a multitude of factors, with MI at the item level being more beneficial than its alternatives for high proportions of item missingness. The approaches performed similarly for high proportions of unit-nonresponse; however, convergence issues were observed for MI at the item level. Maximum likelihood (ML), MI and inverse probability weighting (IPW) were evaluated for handling missing longitudinal PROMs data. MI was less biased than ML when additional post-randomisation data were available, while IPW introduced more bias compared to both ML and MI. A case study was used to explore approaches to sensitivity analyses to assess the impact of missing data. It was found that trial results could be susceptible to varying assumptions about missing data, and the importance of interpreting the results in this context was reiterated. This thesis provides researchers with guidance for the handling and reporting of missing PROMs data in order to decrease bias arising from missing data in RCTs.

Abstract Low back pain is a significant public health problem in many countries of the world being one of the major causes of work absence and disability. Although the outlook for evidence-based management of low back pain has greatly improved over the past decades, many questions remain. Questions related to treatment options, underlying mechanisms of treatment effects and optimal assessment of low back pain have yet to be fully addressed by researchers. The broad aim of this thesis therefore was to contribute to a better understanding of the contemporary management of low back pain by performing studies in these key research areas. Most clinical practice guidelines recommend exercise as an effective treatment option for chronic low back pain. However the evidence for this recommendation comes from trials that are not placebo-controlled and so this may potentially provide biased estimates of the effects of exercise. Therefore a randomised controlled trial testing the effect of motor control exercise versus placebo in patients with chronic low back pain was conducted. Chapters 2 and 3 describe the trial protocol and the report of the trial respectively. A total of 154 patients with chronic low back pain were randomised to receive a motor control exercise program, or placebo (i.e. detuned short-wave therapy and detuned ultrasound therapy). Primary outcomes were pain, function, and the patient’s global impression of recovery measured at 2 months. The exercise intervention improved function and patient’s global impression of recovery, but not pain, at 2 months. The mean effect of exercise on function was 1.1 points (95%CI, 0.3 to 1.8), the mean effect on global impression of recovery was 1.5 points (95%CI, 0.4 to 2.5) and the mean effect on pain was 0.9 points (95%CI, - 0.01 to 1.8), all measured on 11 point scales. Secondary outcomes also favoured motor control exercise. This is the first study ever to demonstrate that motor control exercise is better than placebo for patients with chronic low back pain. Most of the treatment effects were maintained at 6 and 12 months follow-up. These results suggest that this intervention should be considered for patients with chronic low back pain in order to improve disability, function, and global impression of recovery, and to improve pain intensity in the long term, but not in the short term. Rehabilitative ultrasound imaging (RUSI) has been increasingly used by physiotherapists in order to identify impairments in motor control as well as to monitor progress of patients with low back pain. As with any other clinical measure it is important to know how reproducible the RUSI measures are, and although there are some reproducibility studies in the literature, no systematic review on this topic has been conducted. Therefore a systematic review was performed with the objective of assessing the reproducibility studies of RUSI for abdominal wall muscles (Chapter 4). Eligible studies were indentified via searches in CINAHL, EMBASE and MEDLINE with citation tracking via the Web of Science Index. A total of 21 studies were included. Due to heterogeneity of the studies’ designs, pooling the data for a meta-analysis was not possible. RUSI measures of thickness of abdominal wall muscles were found to be reliable. Few studies analysed the reliability for the measurement of thickness changes (reflecting the muscle activity) finding good to poor results. Evidence for the reproducibility of the difference in thickness changes over time (necessary to evaluate improvements in muscle activity with treatment) was not available. A limitation of the existing literature is that studies typically had suboptimal designs and analysis. The current evidence for the reproducibility of RUSI for measuring abdominal muscle activity is mainly based upon studies with suboptimal designs that included mostly healthy subjects, making generalisability to clinical settings uncertain. Some questions about the reproducibility of RUSI measures of abdominal wall muscles are still unanswered; this is mainly due to design issues, such as inadequate statistics, inadequate sampling and lack of control of sources of bias (e.g. blinding and absence of controlling for ordering effects). In addition the clinically important questions about the reproducibility of thickness changes (reflecting the muscle activity) and differences in thickness changes over time (reflecting the improvement or deterioration of muscle activity) have not been adequately investigated. Therefore a reproducibility study that aimed to answer these questions was performed (Chapter 5). Thirty-five patients seeking care for chronic low back pain participated in this study. RUSI measures were taken at baseline and eight weeks post-baseline. Replicate measures of thickness, thickness changes and differences in thickness changes over time were analysed. The reproducibility of static images (thickness) was excellent (ICC2,1 = 0.97, 95%CI = 0.96-0.97, Standard Error of the Measurement (SEM) = 0.04cm, Smallest Detectable Change (SDC) = 0.11cm), the reproducibility of thickness changes was moderate (ICC2,1 = 0.72, 95%CI 0.65-0.76 SEM = 15%, SDC 41%), while the reproducibility of differences in thickness changes over time was poor. Improvements in the test protocol should be undertaken in order to enhance the reproducibility of RUSI measures, especially for differences in thickness chang over time. Self-report outcome measures (questionnaires) are widely used by health care providers for measuring patient’s health status or treatment outcomes. Most of the questionnaires related to low back pain were developed in English and therefore their usefulness in non-English speaking countries is considerably limited. Cross-cultural adaptation and clinimetric testing are possibly the most efficient methods for solving this problem. Although there are many publications on the topic, a simple guide on how to perform a cross-cultural adaptation and clinimetric testing was not available. Therefore a “clinician-friendly” narrative review for Brazilian physical therapists (Chapter 6) was written. This review aimed firstly to explain the concepts and the relevance of cross-cultural adaptation and clinimetrics testing, secondly to summarise the current guidelines on the topic, thirdly to provide advice on how to choose a relevant questionnaire and finally how to evaluate the quality of an adapted questionnaire. Some examples of cross-cultural adaptations and clinimetrics testing of relevant low back pain questionnaires in the Brazilian-Portuguese language were also provided. Although the number of international versions of low back questionnaires is growing, to date it is unclear which questionnaires have been cross-culturally adapted and into which specific language. To answer these questions a systematic review was conducted in order to describe the available cross-cultural adaptations of low back pain self-report outcome measures and the clinimetric testing that has occurred for each adaptation (Chapter 7). Searches were performed in MEDLINE, EMBASE, CINALH and LILACS; these searches were supplemented with information from experts in the field of low back pain from 27 different countries to ensure that the results were comprehensive. Sixty-one adaptations were identified. While there are a large number of low back pain questionnaires available, very few have been adapted into other languages, particularly commonly spoken languages such as Mandarin, Hindi and Portuguese. The quality and comprehensiveness of clinimetric testing varied considerably, with the evaluation of reliability and construct validity most common. Further cross-cultural adaptation and clinimetric studies are clearly needed and special consideration must be given to study designs for clinimetric testing. The final aim of this thesis was to cross-culturally adapt self-report instruments relevant to the management of low back pain in Brazil. This was achieved by two independent studies. The first study (Chapter 8) aimed to cross-culturally adapt the Functional Rating Index (FRI) into Brazilian-Portuguese and to test the clinimetric properties of the FRI and also of an existing Brazilian-Portuguese version of the Roland Morris Disability Questionnaire (RMDQ) which was not fully evaluated in the original study. Both instruments were tested for internal consistency, reliability, construct validity, ceiling and floor effects and internal responsiveness in 140 chronic low back patients presenting for physiotherapy treatment in Brazil. Both instruments were considered reliable and valid for the measurement of disability in Brazilian-Portuguese speakers with low back pain, no ceiling or floor effects were detected, but the internal responsiveness of both instruments was considered small. The second study (Chapter 9) aimed to cross-culturally adapt the Patient-Specific Functional Scale (PSFS) and to perform a head-to-head comparison of the clinimetric properties of the PSFS, RMDQ and FRI. All instruments were tested for internal consistency, reliability, construct validity, ceiling and floor effects, internal and external responsiveness in 99 acute low back patients presenting for physiotherapy treatment in Brazil. In order to fully test the construct validity and external responsiveness, it was necessary to cross-culturally adapt the Pain Numerical Rating Scale and the Global Perceived Effect Scale. The results of this study demonstrate that the Brazilian-Portuguese versions of the RMDQ, FRI and PSFS have similar clinimetric properties to each other and to the original English versions; however the PSFS was the most responsive instrument. The results from the studies in Chapters 8 and 9 will benefit the understanding of low back pain by enabling international comparisons between studies conducted in Brazil and English speaking countries. In addition it will encourage researchers to include Brazilian- Portuguese speakers in their future clinical trials. Overall, the studies included in this thesis have provided an important contribution to the contemporary management of low back pain. Firstly the use of motor control exercise could be considered for patients with chronic low back pain as it produces improvements in global impression of recovery, function, disability and pain. Secondly RUSI measures of abdominal wall muscles in patients with low back pain were considered reproducible for the measurement of muscle activity, but not as an outcome measure to detect improvement/deterioration of muscle activity over the course of treatment. Thirdly just a few high-quality cross-cultural adaptations and clinimetrics testing for self-report outcome measures relevant to the management of low back pain are available, and clearly more studies in this area are needed. Finally the Brazilian-Portuguese versions of the Functional Rating Index, the Roland Morris Disability Questionnaire and the Patient-Specific Functional Scale have acceptable clinimetric properties and could be used in clinical practice as well as in research studies in Brazil.

PhD
Abstract Low back pain is a significant public health problem in many countries of the world being one of the major causes of work absence and disability. Although the outlook for evidence-based management of low back pain has greatly improved over the past decades, many questions remain. Questions related to treatment options, underlying mechanisms of treatment effects and optimal assessment of low back pain have yet to be fully addressed by researchers. The broad aim of this thesis therefore was to contribute to a better understanding of the contemporary management of low back pain by performing studies in these key research areas. Most clinical practice guidelines recommend exercise as an effective treatment option for chronic low back pain. However the evidence for this recommendation comes from trials that are not placebo-controlled and so this may potentially provide biased estimates of the effects of exercise. Therefore a randomised controlled trial testing the effect of motor control exercise versus placebo in patients with chronic low back pain was conducted. Chapters 2 and 3 describe the trial protocol and the report of the trial respectively. A total of 154 patients with chronic low back pain were randomised to receive a motor control exercise program, or placebo (i.e. detuned short-wave therapy and detuned ultrasound therapy). Primary outcomes were pain, function, and the patient’s global impression of recovery measured at 2 months. The exercise intervention improved function and patient’s global impression of recovery, but not pain, at 2 months. The mean effect of exercise on function was 1.1 points (95%CI, 0.3 to 1.8), the mean effect on global impression of recovery was 1.5 points (95%CI, 0.4 to 2.5) and the mean effect on pain was 0.9 points (95%CI, - 0.01 to 1.8), all measured on 11 point scales. Secondary outcomes also favoured motor control exercise. This is the first study ever to demonstrate that motor control exercise is better than placebo for patients with chronic low back pain. Most of the treatment effects were maintained at 6 and 12 months follow-up. These results suggest that this intervention should be considered for patients with chronic low back pain in order to improve disability, function, and global impression of recovery, and to improve pain intensity in the long term, but not in the short term. Rehabilitative ultrasound imaging (RUSI) has been increasingly used by physiotherapists in order to identify impairments in motor control as well as to monitor progress of patients with low back pain. As with any other clinical measure it is important to know how reproducible the RUSI measures are, and although there are some reproducibility studies in the literature, no systematic review on this topic has been conducted. Therefore a systematic review was performed with the objective of assessing the reproducibility studies of RUSI for abdominal wall muscles (Chapter 4). Eligible studies were indentified via searches in CINAHL, EMBASE and MEDLINE with citation tracking via the Web of Science Index. A total of 21 studies were included. Due to heterogeneity of the studies’ designs, pooling the data for a meta-analysis was not possible. RUSI measures of thickness of abdominal wall muscles were found to be reliable. Few studies analysed the reliability for the measurement of thickness changes (reflecting the muscle activity) finding good to poor results. Evidence for the reproducibility of the difference in thickness changes over time (necessary to evaluate improvements in muscle activity with treatment) was not available. A limitation of the existing literature is that studies typically had suboptimal designs and analysis. The current evidence for the reproducibility of RUSI for measuring abdominal muscle activity is mainly based upon studies with suboptimal designs that included mostly healthy subjects, making generalisability to clinical settings uncertain. Some questions about the reproducibility of RUSI measures of abdominal wall muscles are still unanswered; this is mainly due to design issues, such as inadequate statistics, inadequate sampling and lack of control of sources of bias (e.g. blinding and absence of controlling for ordering effects). In addition the clinically important questions about the reproducibility of thickness changes (reflecting the muscle activity) and differences in thickness changes over time (reflecting the improvement or deterioration of muscle activity) have not been adequately investigated. Therefore a reproducibility study that aimed to answer these questions was performed (Chapter 5). Thirty-five patients seeking care for chronic low back pain participated in this study. RUSI measures were taken at baseline and eight weeks post-baseline. Replicate measures of thickness, thickness changes and differences in thickness changes over time were analysed. The reproducibility of static images (thickness) was excellent (ICC2,1 = 0.97, 95%CI = 0.96-0.97, Standard Error of the Measurement (SEM) = 0.04cm, Smallest Detectable Change (SDC) = 0.11cm), the reproducibility of thickness changes was moderate (ICC2,1 = 0.72, 95%CI 0.65-0.76 SEM = 15%, SDC 41%), while the reproducibility of differences in thickness changes over time was poor. Improvements in the test protocol should be undertaken in order to enhance the reproducibility of RUSI measures, especially for differences in thickness chang over time. Self-report outcome measures (questionnaires) are widely used by health care providers for measuring patient’s health status or treatment outcomes. Most of the questionnaires related to low back pain were developed in English and therefore their usefulness in non-English speaking countries is considerably limited. Cross-cultural adaptation and clinimetric testing are possibly the most efficient methods for solving this problem. Although there are many publications on the topic, a simple guide on how to perform a cross-cultural adaptation and clinimetric testing was not available. Therefore a “clinician-friendly” narrative review for Brazilian physical therapists (Chapter 6) was written. This review aimed firstly to explain the concepts and the relevance of cross-cultural adaptation and clinimetrics testing, secondly to summarise the current guidelines on the topic, thirdly to provide advice on how to choose a relevant questionnaire and finally how to evaluate the quality of an adapted questionnaire. Some examples of cross-cultural adaptations and clinimetrics testing of relevant low back pain questionnaires in the Brazilian-Portuguese language were also provided. Although the number of international versions of low back questionnaires is growing, to date it is unclear which questionnaires have been cross-culturally adapted and into which specific language. To answer these questions a systematic review was conducted in order to describe the available cross-cultural adaptations of low back pain self-report outcome measures and the clinimetric testing that has occurred for each adaptation (Chapter 7). Searches were performed in MEDLINE, EMBASE, CINALH and LILACS; these searches were supplemented with information from experts in the field of low back pain from 27 different countries to ensure that the results were comprehensive. Sixty-one adaptations were identified. While there are a large number of low back pain questionnaires available, very few have been adapted into other languages, particularly commonly spoken languages such as Mandarin, Hindi and Portuguese. The quality and comprehensiveness of clinimetric testing varied considerably, with the evaluation of reliability and construct validity most common. Further cross-cultural adaptation and clinimetric studies are clearly needed and special consideration must be given to study designs for clinimetric testing. The final aim of this thesis was to cross-culturally adapt self-report instruments relevant to the management of low back pain in Brazil. This was achieved by two independent studies. The first study (Chapter 8) aimed to cross-culturally adapt the Functional Rating Index (FRI) into Brazilian-Portuguese and to test the clinimetric properties of the FRI and also of an existing Brazilian-Portuguese version of the Roland Morris Disability Questionnaire (RMDQ) which was not fully evaluated in the original study. Both instruments were tested for internal consistency, reliability, construct validity, ceiling and floor effects and internal responsiveness in 140 chronic low back patients presenting for physiotherapy treatment in Brazil. Both instruments were considered reliable and valid for the measurement of disability in Brazilian-Portuguese speakers with low back pain, no ceiling or floor effects were detected, but the internal responsiveness of both instruments was considered small. The second study (Chapter 9) aimed to cross-culturally adapt the Patient-Specific Functional Scale (PSFS) and to perform a head-to-head comparison of the clinimetric properties of the PSFS, RMDQ and FRI. All instruments were tested for internal consistency, reliability, construct validity, ceiling and floor effects, internal and external responsiveness in 99 acute low back patients presenting for physiotherapy treatment in Brazil. In order to fully test the construct validity and external responsiveness, it was necessary to cross-culturally adapt the Pain Numerical Rating Scale and the Global Perceived Effect Scale. The results of this study demonstrate that the Brazilian-Portuguese versions of the RMDQ, FRI and PSFS have similar clinimetric properties to each other and to the original English versions; however the PSFS was the most responsive instrument. The results from the studies in Chapters 8 and 9 will benefit the understanding of low back pain by enabling international comparisons between studies conducted in Brazil and English speaking countries. In addition it will encourage researchers to include Brazilian- Portuguese speakers in their future clinical trials. Overall, the studies included in this thesis have provided an important contribution to the contemporary management of low back pain. Firstly the use of motor control exercise could be considered for patients with chronic low back pain as it produces improvements in global impression of recovery, function, disability and pain. Secondly RUSI measures of abdominal wall muscles in patients with low back pain were considered reproducible for the measurement of muscle activity, but not as an outcome measure to detect improvement/deterioration of muscle activity over the course of treatment. Thirdly just a few high-quality cross-cultural adaptations and clinimetrics testing for self-report outcome measures relevant to the management of low back pain are available, and clearly more studies in this area are needed. Finally the Brazilian-Portuguese versions of the Functional Rating Index, the Roland Morris Disability Questionnaire and the Patient-Specific Functional Scale have acceptable clinimetric properties and could be used in clinical practice as well as in research studies in Brazil.

The current evidence base for the treatment of Posttraumatic Stress Disorder (PTSD) is based upon trauma-focused psychological therapy delivered on an individual, face-to-face basis with a therapist. Many barriers to accessing treatment exist, and if untreated, chronic PTSD can result in significant personal, occupational, social, financial, and health problems, reducing years and quality of life. In 2005, The National Institute for Health and Care Excellence recommended research into newly developed guided self-help (GSH) materials based on trauma-focused psychological interventions. Unlike other common mental health disorders, currently, there is no meta-analytical evidence available to support the implementation of GSH as a low intensity psychological intervention for PTSD. A metaanalysis of eight randomised controlled-trials, was conducted to review the effectiveness of trauma-focused GSH (TF-GSH) for adults with PTSD. These studies compared TF-GSH against active or passive control comparators, and seven of these studies delivered Internetbased interventions. Results show that at postassessment a large treatment effect is associated with TF-GSH in reducing symptoms of PTSD. A moderate effect size was found in favour of TF-GSH in reducing co-morbid symptoms of depression. The rate of dropouts from TF-GSH was comparable to current evidence-based treatments for PTSD. Sensitivity analyses revealed that the magnitude of effect remained when studies judged as at high risk of bias were removed. However, there was statistically significant and clinical heterogeneity present amongst studies, which could not be addressed with additional analysis due to the small number of studies included within the review. The quality of evidence was evaluated, as low and further research is required to increase confidence in estimating the treatment effect of TF-GSH for PTSD.

Enthusiasm for uvulopalatopharyngoplasty (UPPP) in the treatment of mild-to-moderate obstructive sleep apnea syndrome (OSAS) has declined in recent years, partly because of a lower success rate over time and partly because of adverse effects. In more severe cases, the patients are generally treated with nasal continuous positive airway pressure (CPAP). However, many patients do not satisfactorily tolerate CPAP as a result of frequent side-effects. Consequently, there is a need for an alternative treatment. Reports on the beneficial effects of mandibular advancement appliances in the treatment of mild-to-moderate OSA exist in the form of short–term evaluations.

One of the aims of the present thesis was to compare treatment effects with a mandibular advancement appliance and UPPP in patients with OSA with follow-up after one and four years. Ninety-five male patients with confirmed mild-to-moderate OSA (apnea index, AI >5 and <25) were randomised to treatment with a dental appliance or UPPP. Sleep studies were performed before and one and four years after intervention. According to the criteria for normalisation (AI<5 and apnea hypopnea index, AHI<10), 78% of the patients in the dental appliance group and 51% of the patients in the UPPP group had normalised after one year (p<0.05). Still after four years of treatment, 63% of the patients in the dental appliance group and 33% of the patients in the UPPP group were normalised. The dental appliance group had a higher normalisation rate than the UPPP group, but the efficacy was partly invalidated by the compliance rate of 62%.

Quality of life assessments in the dimensions of vitality, contentment and sleep improved in both groups at the one-year follow-up after treatment. There was no difference between the groups in terms of vitality and sleep. The UPPP group, however, reported a higher degree of contentment than the dental appliance group, even though the somnographic values were superior in the latter group.

Another aim was to conduct a randomised study to test the hypothesis that severe OSA patients will benefit from more pronounced mandibular advancement (MA) compared with a shorter advancement. Eighty-six males with severe OSA (AI>20) were randomly allocated to either 75% or 50% MA for a six-month treatment period. Treatment with a more pronounced mandibular advancement yielded a 20% higher normalisation rate than a shorter advancement. A mean normalisation rate of 45% was found for patients in this category with few side-effects, good patient satisfaction and a compliance of 92% after 6 months.

The overall conclusion is that dental appliance treatment is effective in patients with mild to moderate OSA and even for patients with severe OSA. The efficacy in terms of normalisation in patients with mild to moderate OSA was higher after the dental appliance treatment with a 50% degree of advancement than after the UPPP treatment. However, severe OSA patients might benefit from more pronounced advancement (75%) compared with a shorter degree of advancement (50%). QOL improved significantly after both dental appliance and UPPP treatment.

Pharmacovigilance plays an important role in monitoring adverse drug reactions (ADRs) resulting from an intervention related to medicinal products. Due to the frequencies and potentially serious consequences, ADRs pose a considerable economic and clinical burden. Patients with an underlying risk factor or established cardiovascular disease (CVD) are usually on long-term treatment, thus it is important to monitor the efficacy and safety of drugs prescribed. Modern antihypertensive drugs have been showed to effectively reduce high blood pressure (BP) hence prevents the development or complications of CVD in high-risk patients. However, there is evidence from clinical trials and observational studies suggesting the association between antihypertensive drugs and risk of cancer. Furthermore, these observations are inconsistent and the majority of clinical trials were directed towards cardiovascular outcomes. The thesis is divided into five main result chapters (4 to 8) based on the antihypertensive drug classes evaluated for risk of cancer in the systematic review and meta-analyses. Altogether, 90 randomised controlled trials (RCTs) enrolling 390,750 participants with an average follow-up of 3.5 years were included for qualitative and quantitative analysis. Angiotensin-converting enzyme inhibitor (ACEI) and risk of cancer: ACEI lowers BP through preventing the conversion of angiotensin 1 to angiotensin II by ACE in the renin-angiotensin system (RAS) pathway. In the present study, no significant association between ACEI and risk of cancer incidence or cancer-related death is reported. Factors such as tissue binding capacity, comparator used, clinical settings, age, and study duration do not affect the risk of cancer overall. Angiotensin receptor blockers (ARB) and risk of cancer: In the RAS pathway, ARB acts directly on the angiotensin type 1 (AT1) receptor to inhibit downstream signalling which results in downregulation of sympathetic activity and lowering of BP. The present meta-analysis has reported no association between ARB use and risk of cancer incidents or cancer-related mortality. Subgroup assessment indicates that valsartan has a cancer-protective effect, particularly against lung cancer. Patients' clinical settings, age, and study duration do not influence the risk of cancer in relation to ARB overall. Calcium channel blockers (CCB) and risk of cancer: As a class, CCBs are potent vasodilators and are recommended for use as first or second-line drugs in treating hypertension. This study has reported a marginally increased risk of cancer incidents (P=0.06) but not cancer-related death overall in relation to CCB use. DHP-CCB is associated with a 9% increased risk for cancer compared to controls (P=0.05). A positive relationship is also observed with older patients and in patients with longer exposure to CCB. Therefore, a properly designed further research into the risk of a specific type of cancer with use of DHP CCB is warranted to detect a safety signal. Beta-blockers (BB) and risk of cancer: Inhibition of stress mediators from activating beta-adrenoceptors has been proposed to be the underlying mechanism by which BB lower the risk of cancer. This study has found no evidence of an association between BB and the risk of cancer or cancer-related death. Factors such as cardioselectivity, treatment indication, age and study duration do not have an impact on cancer risk altogether. Thiazide diuretics (TZ) and risk of cancer: TZ induces diuresis at the distal convoluted tubule and a great number of studies had attempted to link TZ and risk of renal cancer. No evidence of an association between TZ and the risk of cancer or cancer mortality is reported in the present study. Chemical structure differences, clinical settings, age, and study duration does not significantly influence the risk of cancer in relation to TZ use. Strengths and limitation: The strengths of the systematic review and meta-analyses conducted in this thesis include; only RCTs were included, a comprehensive search strategy spanning over 60 years with no language restrictions, and a sufficiently large sample size of over 390,000 trial participants from various clinical settings with an average 3.5 years follow-up duration. Lack of individual-level data and non-standard reporting of cancer in RCTs are the main limitations. Future recommendations: All RCT evaluating drug intervention should pre-identify cancer as one of the study outcomes as part of drug safety monitoring.