Healthcare decision-makers need high quality evidence to inform policy and practice decisions. Systematic reviews of randomized controlled trials (RCTs), including meta- analyses of study effects, are considered one of the highest forms of evidence to inform such decisions. Most applications of systematic reviews and meta-analyses are based on a standardized cannon of methods that seek to collect, abstract, assess, and synthesize evidence from primary studies to produce a comprehensive and unbiased summary of the evidence. While useful, standard synthesis methods tend to assume simple data structures (e.g., two-arm comparison of a single intervention vs. a similar control evaluated in a parallel individual randomized design) and some practices (e.g., author contact) may not always be supported by empirical evidence. Complex interventions are of increasing focus in healthcare and public health and pose challenges to the standard methods of systematic review and meta-analysis. While different definitions of complex interventions have been proposed, most definitions assume: i) multiple intervention ‘components’ that may or may not interact with each other to increase or decrease observed intervention effects and ii) effect modification by study-specific characteristics (e.g., healthcare setting, patient population). At least three challenges may result from this complexity. First, reviewers will likely have to contact authors for additional information about intervention components and contextual factors that may operate as effect modifiers. Unfortunately, evidence supporting optimal strategies for achieving response from author contact is lacking. Second, complex interventions are often evaluated using a cluster randomized trial (CRT) design that randomize units of patients to different healthcare/health policy interventions. Analyses from CRTs that are not adjusted for the clustering effect are said to have unit of analysis errors, which if incorporated in meta-analyses could lead to biased summary estimates and overly precise confidence intervals (CIs). Methods for reviewers to appropriately appraise abstract evidence from CRTs are limited. Thirdly, standard meta-analyses estimate an overall effect of a singular ‘complex intervention’. Such analyses answer the question “Do complex interventions as a whole lead to a difference in observed outcomes?” and tend to exhibit high statistical heterogeneity since variation in intervention components and effect modifiers are not accounted for. Hierarchical multivariate meta-regression models have been proposed as an alternative synthesis approach for complex interventions to better account for observed heterogeneity and answer the question decision-makers are really interested in; that is “What component(s) (or combination of components) work and under what conditions?”. Hierarchical multivariate meta-regression models however have yet to be applied in the review of complex healthcare interventions. The overall aim of my doctoral research was to explore the utility of three methodological approaches to address these challenges and optimize the synthesis of complex interventions using a large systematic review of diabetes quality improvement interventions as a case study. The first objective of this thesis was to do an RCT evaluation of the effect of telephone call versus repeated email contact of non-responding authors for additional study information on response rates and research costs. We found authors contacted by telephone call were more likely to complete requests for additional information (response rate 36.7% vs. 20.2%; adjusted odds ratio 2.26 [95% CI 1.10-4.76]) but the intervention took more time to deliver in total (20 vs. 10 hours over several months vs. one month) and was more expensive overall (approximately $505 vs. $253). The second objective of this thesis was to better account for evidence from CRTs and involved a descriptive study and a methodological study. The descriptive study described the proportion of studies with unit of analysis errors and the nature of the error (inappropriate analysis versus unclear or incomplete reporting). The methodological study investigated the utility of building a database of intracluster correlation coefficients (ICCs) and use of an ICC posterior predictive distribution model to correct unit of analysis errors identified in the descriptive study. We found that although trials often adjusted for the cluster effect (67% across outcomes; range 25%-81%), most did not report enough information to extract adjusted effect estimates required for meta-analysis (an average of 77% of studies with remaining unit of analysis errors across outcomes; range 42%-100%). We were able to construct a posterior predictive distribution of the ICC for most outcomes in our review using estimates of the ICC obtained from the descriptive study combined with external estimates and use these distributions to impute missing ICCs to correct unit of analysis errors. Finally, the third objective of this thesis was to illustrate the use of hierarchical multivariate meta-regression for quantitative synthesis when estimating the effects of complex interventions and exploring effect heterogeneity. Using an arm-based analysis of post-treatment means of one continuous outcome, we demonstrated that hierarchical multivariate meta-regression models can be used to estimate a ‘response surface’ that accounts for complex intervention multiple components and study characteristics, and these models can be used to infer estimates of component effects, interactions among components, and effect modification by study covariates. Collectively the results from this thesis suggest three methodological approaches (contacting authors by telephone, imputing missing ICCs using a predictive distribution, estimating complex intervention effects using a hierarchical multivariate meta-regression) can be used to optimize the processes of synthesizing complex interventions. Further work is needed to evaluate the impact of additional study-covariates on explaining residual heterogeneity and testing these methods in other reviews of complex interventions.

A l'adolescence, les patients porteurs de maladie chronique depuis l'enfance passent une étape de transition des soins, qui implique la sortie des services pédiatriques et l'entrée dans les services pour adultes. Cette étape pose de nombreux défis et elle est reconnue comme une période à risque de rupture de soins. Pour éviter ces situations, des recommandations internationales et des programmes variés ont été proposées. En France, il n'existe pas de recommandation générique pour la transition. Notre objectif était de proposer un programme générique, pertinent et faisable pour la transition des jeunes porteurs de maladies chroniques variées dans un grand nombre de services en France. Pour cela nous avons employé une méthode mixte : 1) Revue systématique des études interventionnelles au niveau international ; 2) Enquête qualitative des pratiques de transition existantes dans un ensemble de services hospitaliers d'Ile de France ; 3) Enquête de consensus national pour l'élaboration d'un programme de transition pertinent et faisable par la méthode Delphi. Nous avons observé que l'existence de programmes de transition était très dépendante des services : certains ne proposant pratiquement aucune adaptation de la prise en charge, d'autres proposant des programmes très structurés. Globalement, les programmes sont peu spécifiques des pathologies. A l'issue de ce travail de thèse, nous avons proposé un programme générique qui peut être considéré comme un tronc commun. Il pourra notamment permettre aux professionnels impliqués dans le suivi de pathologies peu étudiées dans le contexte de la transition de proposer des pratiques essentielles et minimales aux patients concernés. / At the end of their adolescence, patients with chronic disease since childhood go through a phase of healthcare transition, which involves the discharge of pediatric services and the entry into adult care services. This step poses many challenges and is recognised as a high risk period of discontinuity in care pathway. To avoid these situations, international recommendations and various programs have been proposed. In France, there is no generic recommendation for transition to adult care.The final goal of this thesis work was to propose a generic, relevant and feasible program for the transition of young people with varied chronic diseases into a large number of services in France.For this purpose, we used a mixed method research including: 1) Systematic review of intervention studies at the international level; 2) Qualitative study of existing transition practices in a set of hospital services in the Ile de France region (including Paris and its surroundings); 3) National consensus survey for the development of a relevant and feasible transition program using the Delphi method.We observed that the existence of transition programs was highly dependent on services: some offering almost no adaptation of care during this process, others offering highly structured programs. Overall, the programs were not really specific to pathologies.At the end of this thesis work, we proposed a generic program which can be considered as a common core of transition practices. Among other thing, it will allow professionals involved in the follow-up of pathologies that are under-studied in the context of the transition to propose essential and minimal practices to their patients.

Background: There is a great need for effective treatments for mental health problems. Randomised controlled trials are the gold standard for evaluating treatments, however recruitment into trials is challenging, highlighting a clear need for evidence-based recruitment strategies. This thesis aimed to systematically develop a recruitment intervention and evaluate its effectiveness for improving the recruitment of participants into mental health trials. Methods: A mixed-methods approach, adopting the Medical Research Council’s complex interventions framework: 1) a systematic review to identify the evidence base and describe the factors affecting recruitment into depression trials; 2) a qualitative study to understand patients’ decision-making process in declining to enrol in a depression trial; 3) development of a recruitment intervention, using Participatory Design methods; and 4) evaluation of the recruitment intervention, using a randomised controlled trial, embedded in an ongoing mental health trial (the EQUIP trial). The primary outcome was the proportion of participants enrolled in EQUIP. Results: From the systematic review, a conceptual framework of factors influencing the decision to participate was developed, which highlighted that the decision to enrol involves a judgement between risk and reward. Findings suggested that patient and public involvement in research (PPIR) might be advertised to potential participants to reduce such perceived risk. The qualitative study found positive views of trials. Interviewees’ decision making resembled a four-stage process; in each stage they either decided to decline or progressed to the next stage. In Stage 1, those with an established position of declining trials opted out – they are termed ‘prior decliners’. In Stage 2, those who opted out after judging themselves ineligible are termed ‘self-excluders’. In Stage 3, those who decided they did not need the trial therapy and opted out are termed ‘treatment decliners’. In Stage 4, those who opted out after judging that disadvantages outweighed advantages are termed ‘trial decliners’. While ‘prior decliners’ are unlikely to respond to trial recruitment initiatives, the factors leading others to decline are amenable to amelioration as they do not arise from a rejection of trials. We recruited a host mental health trial (EQUIP), and worked with key stakeholders, including mental health service users and carers, to develop an intervention using a leaflet to advertise the nature and function of the PPIR in EQUIP to potential trial participants. 34 community mental health teams were randomised and 8182 patients invited. For the primary outcome, 4% of patients in the PPIR group were enrolled versus 5.3% of the control group. The intervention was not effective for improving recruitment rates (adjusted OR= 0.75, 95% CI= 0.53 to 1.07, p=0.113). Conclusions: This thesis reports the largest ever trial to evaluate the impact of a recruitment intervention. It also reports the largest trial of a PPIR intervention and makes a contribution to the evidence base on trial recruitment as well as to that assessing the impact of PPIR. Two further embedded trials are underway to evaluate the effectiveness of different versions of the recruitment intervention in different trial contexts and patient populations. This will also allow the results to be pooled to generate a more precise estimate of effect; to evaluate the impact of the intervention on trial retention; and to explore patient experiences of receiving the intervention.

<p>The thesis presents three papers discussing some of the methodological issues regarding studies investigating complex community-based cardiovascular health interventions. All three studies involved the Cardiovascular Health Awareness Program (CHAP), a standardised blood pressure and risk factor assessment and educational sessions held in pharmacies or other locally accessible areas in small to mid-sized communities in Ontario, Canada.</p> <p>The first paper reviews the literature and proposes a guide on how to develop a theoretical framework for complex community-based interventions using CHAP as an example. The paper describes a stepwise process of developing a theoretical framework including challenges encountered and strategies employed to overcome them.</p> <p>The second paper presents how recently published randomized controlled trials evaluating complex community-based cardiovascular health interventions monitored and reported implementation fidelity based on a structured review of the published articles and a survey of their primary authors. The results showed that fidelity reporting of included studies was better than those described in previous reviews. Fidelity was verified through self-reports by implementers and supervision by researchers. Strategies described to standardize intervention delivery were through training of implementers and use of implementation guides. The authors’ survey results were consistent with the review results though there were some gaps which could be improved to strengthen fidelity reporting.</p> <p>A data analysis issue with studies investigating complex community-based interventions is that outcomes can be affected by factors from multiple levels. The third paper explores the association of individual, partnership, and community-related factors with CHAP participants’ use of health-related community resources and cardiovascular risk behaviours. This was a cross-sectional analysis of an on-going cohort study. The results showed that individual factors (age and self-efficacy) had the most consistent association with the outcomes. Community and partnership level variables showed less consistent association with the outcome. Methodological and analytical challenges were presented.</p> / Doctor of Philosophy (PhD)

Healthcare professionals such as speech and language therapists are expected to change their practice throughout their career. However, from a practice perspective, there is a lack of knowledge around what practice change is, what it really takes, and why there are different trajectories. Consequently, therapists, managers and commissioners lack empirical evidence on which to base decisions about enabling practice change. In addition, intervention researchers lack basic sociological research around implementation that could inform their research designs, reporting and impact. This case-based sociological inquiry, underpinned by critical realist assumptions, was designed to address this knowledge gap. It includes a two-stage qualitative synthesis of 53 (then 16) studies where speech and language therapists explained the work of their practice in depth, and a primary qualitative study focused on one professional jurisdiction, children with speech sound difficulties (SSD). Forty two speech and language therapists from three NHS areas and independent practice in Scotland participated in individual interviews or self-organised pairs or focus groups to discuss in depth how and why they had changed their practice with these children. A variety of comparative methods were used to detail, understand and explain this particular aspect of the social world. The resulting theory of SSD practice change comprises six configured cases of practice change (Transforming; Redistributing; Venturing; Personalising; Delegating; Refining) emerging from an evolving and modifiable practice context. The work that had happened across four key aspects of this context (Intervention; Candidacy; Caseload; Service) explained what made each case possible, and how practice had come to be one way rather than another. Among its practical applications, the theory could help services plan more realistic practice change. In addition, the inductively developed layered model of SSD intervention change has the potential to contribute to speech and language therapy education as well as methodological discussions around complex interventions.