Indiana University-Purdue University Indianapolis (IUPUI) / Cluster randomization is frequently used in clinical trials for convenience of inter ventional implementation and for reducing the risk of contamination. The opera tional convenience of cluster randomized trials, however, is gained at the expense of reduced analytical power. Compared to individually randomized studies, cluster randomized trials often have a much-reduced power. In this dissertation, I consider ways of enhancing analytical power with historical trial data. Specifically, I introduce a hierarchical Bayesian model that is designed to incorporate available information from previous trials of the same or similar interventions. Operationally, the amount of information gained from the previous trials is determined by a Kullback-Leibler divergence measure that quantifies the similarity, or lack thereof, between the histor ical and current trial data. More weight is given to the historical data if they more closely resemble the current trial data. Along this line, I examine the Type I error rates and analytical power associated with the proposed method, in comparison with the existing methods without utilizing the ancillary historical information. Similarly, to design a cluster randomized trial, one could estimate the power by simulating trial data and comparing them with the historical data from the published studies. Data analytical and power simulation methods are developed for more general situations of cluster randomized trials, with multiple arms and multiple types of data following the exponential family of distributions. An R package is developed for practical use of the methods in data analysis and trial design.

In cluster randomized trials (CRTs), groups of participants are randomized as opposed to individual participants. This design is often chosen to minimize treatment arm contamination or to enhance compliance among participants. In CRTs, we cannot assume independence among individuals within the same cluster because of their similarity, which leads to decreased statistical power compared to individually randomized trials. The intracluster correlation coefficient (ICC) is crucial in the design and analysis of CRTs, and measures the proportion of total variance due to clustering. Missing data is a common problem in CRTs and should be accommodated with appropriate statistical techniques because they can compromise the advantages created by randomization and are a potential source of bias. In three papers, I investigate statistical approaches for handling missing data in CRTs. In the first paper, I carry out a systematic review evaluating current practice of handling missing data in CRTs. The results show high rates of missing data in the majority of CRTs, yet handling of missing data remains suboptimal. Fourteen (16%) of the 86 reviewed trials reported carrying out a sensitivity analysis for missing data. Despite suggestions to weaken the missing data assumption from the primary analysis, only five of the trials weakened the assumption. None of the trials reported using missing not at random (MNAR) models. Due to the low proportion of CRTs reporting an appropriate sensitivity analysis for missing data, the second paper aims to facilitate performing a sensitivity analysis for missing data in CRTs by extending the pattern mixture approach for missing clustered data under the MNAR assumption. I implement multilevel multiple imputation (MI) in order to account for the hierarchical structure found in CRTs, and multiply imputed values by a sensitivity parameter, k, to examine parameters of interest under different missing data assumptions. The simulation results show that estimates of parameters of interest in CRTs can vary widely under different missing data assumptions. A high proportion of missing data can occur among CRTs because missing data can be found at the individual level as well as the cluster level. In the third paper, I use a simulation study to compare missing data strategies to handle missing cluster level covariates, including the linear mixed effects model, single imputation, single level MI ignoring clustering, MI incorporating clusters as fixed effects, and MI at the cluster level using aggregated data. The results show that when the ICC is small (ICC ≤ 0.1) and the proportion of missing data is low (≤ 25\%), the mixed model generates unbiased estimates of regression coefficients and ICC. When the ICC is higher (ICC > 0.1), MI at the cluster level using aggregated data performs well for missing cluster level covariates, though caution should be taken if the percentage of missing data is high.

It is well known that incorporating auxiliary covariates in the analysis of randomized clinical trials (RCTs) can increase efficiency. Questions still remain regarding how to flexibly incorporate baseline covariates while maintaining valid inference. Recent methodological advances that use semiparametric theory to develop covariate-adjusted inference for RCTs have focused on independent outcomes. In biomedical research, however, cluster randomized trials and longitudinal studies, characterized by correlated responses, are commonly used. We develop methods that flexibly incorporate baseline covariates for efficiency improvement in randomized studies with correlated outcomes. In Chapter 1, we show how augmented estimators may be used for cluster randomized trials, in which treatments are assigned to groups of individuals. We demonstrate the potential for imbalance correction and efficiency improvement through consideration of both cluster- and individual-level covariates. To improve small-sample estimation, we consider several variance adjustments. We evaluate this approach for continuous and binary outcomes through simulation and apply it to the Young Citizens study, a cluster randomized trial of a community behavioral intervention for HIV prevention in Tanzania. Chapter 2 builds upon the previous chapter by deriving semiparametric locally efficient estimators of marginal mean treatment effects when outcomes are correlated. Estimating equations are determined by the efficient score under a mean model for marginal effects when data contain baseline covariates and exhibit correlation. Locally efficient estimators are implemented for longitudinal data with continuous outcomes and clustered data with binary outcomes. Methods are illustrated through application to AIDS Clinical Trial Group Study 398, a longitudinal randomized study that compared various protease inhibitors in HIV-positive subjects. In Chapter 3, we empirically evaluate several covariate-adjusted tests of intervention effects when baseline covariates are selected adaptively and the number of randomized units is small. We demonstrate that randomization inference preserves type I error under model selection while tests based on asymptotic theory break down. Additionally, we show that covariate adjustment typically increases power, except at extremely small sample sizes using liberal selection procedures. Properties of covariate-adjusted tests are explored for independent and multivariate outcomes. We revisit Young Citizens to provide further insight into the performance of various methods in small-sample settings.

acase@tulane.edu

No description available.

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.

<p>Correlated data with complex association structures arise from longitudinal studies and cluster randomized trials. However, some methodological challenges in the design and analysis of such studies or trials have not been overcome. In this thesis, we address three of the challenges: 1) <em>Power analysis for population based longitudinal study investigating gene-environment interaction effects on chronic disease:</em> For longitudinal studies with interest in investigating the gene-environment interaction in disease susceptibility and progression, rigorous statistical power estimation is crucial to ensure that such studies are scientifically useful and cost-effective since human genome epidemiology is expensive. However conventional sample size calculations for longitudinal study can seriously overestimate the statistical power due to overlooking the measurement error, unmeasured etiological determinants, and competing events that can impede the occurrence of the event of interest. 2) <em>Comparing the performance of different multiple imputation strategies for missing binary outcomes in cluster randomized trials</em>: Though researchers have proposed various strategies to handle missing binary outcome in cluster randomized trials (CRTs), comprehensive guidelines on the selection of the most appropriate or optimal strategy are not available in the literature. 3) <em>Comparison of population-averaged and cluster-specific models for the analysis of cluster randomized trials with missing binary outcome</em>: Both population-averaged and cluster-specific models are commonly used for analyzing binary outcomes in CRTs. However, little attention has been paid to their accuracy and efficiency when analyzing data with missing outcomes. The objective of this thesis is to provide researchers recommendations and guidance for future research in handling the above issues.</p> / Doctor of Philosophy (PhD)

La santé publique fondée sur des données probantes doit être basée sur les meilleures preuves disponibles pour prendre des décisions éclairées, afin de mettre en place des interventions dirigées vers le maintien et l’amélioration de la santé, ainsi que vers le bien-être de toute la population. Les essais contrôlés randomisés (ECR) sont souvent utilisés en recherche clinique pour tester les effets d’un médicament, d’une thérapie ou d’une intervention sur un groupe expérimental qui bénéficiera de l’intervention, en le comparant à un groupe contrôle qui recevra un placebo ou aucun traitement. Bien que le débat persiste, les essais randomisés constituent une source importante et, apparemment, de haute qualité pour évaluer l’efficacité des interventions en santé. Dû à de multiples facteurs, les essais randomisés en grappes (ERG) sont largement utilisés pour évaluer la prestation des services de santé et des interventions en santé publique. Dans ce type d’essai, ce ne sont plus des individus qui sont randomisés, mais des groupes d’individus tels que les familles, les médecins, les villages qui vont recevoir l’intervention. Ces interventions peuvent varier pendant la mise en œuvre en raison de divers facteurs liés à la conception de l’intervention, aux participants, aux intervenants ainsi qu’aux facteurs du contexte qui influencent les résultats. Ces facteurs doivent être pris en compte au moment de l’évaluation, et avant la réplication dans d’autres contextes. L’évaluation de la fidélité de la mise en œuvre, outil clé de l’évaluation du processus et élément essentiel du processus de mise à l’échelle, vise à mesurer le degré selon lequel une intervention a été implantée telle que conçue par les concepteurs. Cette thèse a comme objectif principal examiner la fidélité de la mise en œuvre des interventions en santé publique dans le cadre des essais randomisés en grappes, pour savoir si les interventions mises en place sous un modèle contrôlé doivent prendre en compte ce type d’évaluation pour renforcer ces résultats et faciliter leur réplication à grande échelle. Cette thèse comporte deux volets : une revue systématique et une étude de cas unique à trois unités d’analyse selon une approche mixte concomitante. Le premier article évalue la pratique de la fidélité de la mise en œuvre des interventions en santé publique dans le cadre des essais randomisés en grappes des études publiées qui ont été identifiées et incluses dans la révision systématique. La révision systématique met en lumière que les interventions mises en place sous ce modèle ne tiennent pas compte de cette évaluation de façon systématique, que la façon de la faire est très hétérogène, et que l’évaluation n’est pas bien documentée. Les deuxième et troisième articles sont les résultats de recherche de l’évaluation d’une intervention, à travers une étude de cas comme méthode de recherche, qui a été menée, dans un premier temps, pour examiner la plausibilité de la théorie de l’intervention, et, dans un deuxième temps, pour évaluer leur fidélité de la mise en œuvre et leur acceptabilité auprès des participants dans le but de l’améliorer, si nécessaire, avant sa mise en place à grande échelle. L’évaluation de l’intervention met en lumière plusieurs aspects. D’abord, la théorie sous-jacente et le modèle de l’intervention évaluée sont bien conçus pour parvenir aux résultats visés. L’évaluation fournit des points clés et des actions à prendre en considération, pendant le développement des interventions, pour servir les communautés difficiles à atteindre, et pour améliorer les résultats en matière de santé. Ensuite, les résultats ont démontré une fidélité de mise en œuvre élevée. La clarté de la théorie de l'intervention, la motivation et l'engagement des intervenants, ainsi que les réunions périodiques des superviseurs avec les intervenants-terrain expliquent largement le haut niveau de fidélité obtenu. Des facteurs contextuels tels que la distance géographique, l'accès à un téléphone portable, le niveau d'éducation et les normes de genre ont contribué à l'hétérogénéité de la participation du groupe cible de l’intervention. Finalement, cette évaluation souligne que la plateforme mobile combinée à la mobilisation communautaire, composantes clés de l’intervention, ont été bien accueillies par les participants, et pourraient être mis en place à grand échelle. Cette thèse contribue au développement des connaissances sur le plan méthodologique concernant l’évaluation de la fidélité de la mise en œuvre des interventions en santé publique en mettant en relief des lacunes dans ce domaine, et en suggérant un outil pour faire avancer cette pratique évaluative. Cette thèse participe également au renforcement de la recherche dans les sciences de l’implémentation, et apporte sur le plan empirique des éléments clés essentiels pour évaluer la fidélité de la mise en œuvre de ce type d’intervention à l’aide des essais randomisés en grappes, évaluation de cette fidélité qui est l’objet de cette recherche doctorale. / Evidence-based public health should be based on the best available evidence to make informed decisions and to implement interventions aimed at maintaining and improving the health and well-being of all people. Randomized controlled trials (RCTs) are often used in clinical research to test the effects of a drug, therapy, or intervention on an experimental group that may benefit from the intervention, comparing it to a control group that received either a placebo or no intervention treatment. Although the debate persists, randomized controlled trials are an important and objectively high quality method for evaluating the effectiveness of health interventions. Due to multiple factors, cluster randomized trials (CRTs) are widely used to assess the delivery of health services and public health interventions. In this type of trial, it is no longer individuals who are randomized, but groups of individuals such as families, doctors, and village communities who receive the intervention. These interventions may differ during implementation as a result of various factors related to the complexity of the intervention design, context, participants, and stakeholders involved. These factors should be considered at the time of assessment and before replication in other contexts. Implementation fidelity assessment, a key tool in process evaluation, examines study processes to assess the extent to which the intervention was carried out as originally intended. The fidelity of implementation is an essential part of the scale-up process. This thesis aimed to examine the fidelity of implementation of public health interventions in the context of cluster randomized trials, to determine whether the interventions implemented under a controlled model should consider this type of evaluation to strengthen their results and facilitate their replication on a large scale. This thesis has two parts: a systematic review and a single case study with three units of analysis using a mixed triangulated approach. The first article assessed the implementation fidelity of public health interventions in the context of cluster randomized trials. The systematic review highlighted the finding that public health interventions implemented under this model did not systematically consider this type of evaluation, that the way of doing it was very heterogeneous, and that the evaluation was not adequately documented. The second and third articles were the research findings of the evaluation of an intervention, using a case study as the research method, that was conducted to first examine the plausibility of the intervention theory and to better understand the design and context of the intervention being evaluated, and second, to evaluate implementation fidelity and its acceptability among the participants with the aim of making improvements (if necessary) before large-scale replication. The evaluation of the case study highlighted several key findings. First, the results of the evaluation reflected that the underlying theory and model of the public health intervention were well designed to achieve the desired results. The evaluation provided key points and actions to consider during intervention development to serve hard-to-reach communities and improve health outcomes. Further, it was shown that the results demonstrated a high degree of implementation fidelity. The clarity of the theory of the intervention, the motivation and commitment of the stakeholders as well as the periodic meetings of supervisors with the field team largely explained the high level of fidelity obtained. Contextual factors such as geographical distance to the intervention, access to a mobile phone, level of education, and gender norms contributed to the heterogeneity of the participation of the intervention target group. Finally, this evaluation underlined the finding that the mobile platform coupled with community mobilization, both key components of the intervention, were well received by the participants and may be an effective means of improving health knowledge and changing health-related behaviors. This thesis contributes to the development of methodological knowledge concerning the evaluation of the fidelity of implementation of public health interventions by identifying gaps in this field, and by suggesting a tool that facilitates advancing this evaluation practice. This thesis also contributes to the strengthening of research in implementation sciences, and empirically provides key elements essential to assess the fidelity of the implementation of this type of intervention using CRT studies and evaluation of this fidelity, which is the subject of this doctoral research.