Background: The use of meta-analysis to combine the results of observational studies is controversial. Despite its common use, methodological work in this area is lacking. Because of the diversity of study designs, exposure and outcome measures, and differential adjustment for confounding variables, the identification of sources of heterogeneity among study effect estimates is particularly important when combining data from observational studies. This thesis presents the results of a study that examines the relative ability of individual participant data (IPD) meta-analysis (which was considered a “gold standard”) and traditional aggregate data (AD) meta-analysis to identify sources of heterogeneity among studies examining mechanical exposure and low back pain (LBP) in workers. Materials and Methods: A systematic literature search was conducted to identify relevant articles. The corresponding author of each article was contacted to request their individual-level data. Because the outcome definitions and exposure measures were not uniform across studies, two sub-studies were conducted 1) to identify sets of outcome definitions that could be combined in a meta-analysis and 2) to develop methods to translate mechanical exposure onto a common metric. IPD analyses were conducted using generalized estimating equation (GEE) regression to identify variables that acted as strong confounders and effect modifiers. Traditional AD meta-analysis was also conducted and potential sources of heterogeneity were tested using meta-regression. Key Findings: (1) Overall, we found an association between both forces and postures on LBP, although the magnitude varied depending on the exposure-outcome combination. Among the outcomes, the ORs tended to be highest for sick leave due to LBP. (2) There was very little evidence of strong confounders in the relationship between mechanical exposure and LBP; thus differential adjustment for confounders in studies would not likely be an important source of heterogeneity in an AD meta-analysis. (3) AD meta-analysis was able to identify the same study-level effect modifiers as IPD meta-analysis, but did not consistently identify individual-level effect modifiers. Both individual-level characteristics (older age and being male), and study-level characteristics (population-based studies and self-reported mechanical exposure), were associated with an increased OR for many of the LBP outcome and mechanical exposure combinations. Conclusion: AD meta-analysis is likely sufficient to detect heterogeneity for study-level factors but is not sufficient to identify individual-level effect modifiers. When the primary source of evidence in a research area is observational studies and when there is controversy despite several systematic reviews, IPD meta-analysis can be used to better understand sources of heterogeneity and provide context

Background: The use of meta-analysis to combine the results of observational studies is controversial. Despite its common use, methodological work in this area is lacking. Because of the diversity of study designs, exposure and outcome measures, and differential adjustment for confounding variables, the identification of sources of heterogeneity among study effect estimates is particularly important when combining data from observational studies. This thesis presents the results of a study that examines the relative ability of individual participant data (IPD) meta-analysis (which was considered a “gold standard”) and traditional aggregate data (AD) meta-analysis to identify sources of heterogeneity among studies examining mechanical exposure and low back pain (LBP) in workers. Materials and Methods: A systematic literature search was conducted to identify relevant articles. The corresponding author of each article was contacted to request their individual-level data. Because the outcome definitions and exposure measures were not uniform across studies, two sub-studies were conducted 1) to identify sets of outcome definitions that could be combined in a meta-analysis and 2) to develop methods to translate mechanical exposure onto a common metric. IPD analyses were conducted using generalized estimating equation (GEE) regression to identify variables that acted as strong confounders and effect modifiers. Traditional AD meta-analysis was also conducted and potential sources of heterogeneity were tested using meta-regression. Key Findings: (1) Overall, we found an association between both forces and postures on LBP, although the magnitude varied depending on the exposure-outcome combination. Among the outcomes, the ORs tended to be highest for sick leave due to LBP. (2) There was very little evidence of strong confounders in the relationship between mechanical exposure and LBP; thus differential adjustment for confounders in studies would not likely be an important source of heterogeneity in an AD meta-analysis. (3) AD meta-analysis was able to identify the same study-level effect modifiers as IPD meta-analysis, but did not consistently identify individual-level effect modifiers. Both individual-level characteristics (older age and being male), and study-level characteristics (population-based studies and self-reported mechanical exposure), were associated with an increased OR for many of the LBP outcome and mechanical exposure combinations. Conclusion: AD meta-analysis is likely sufficient to detect heterogeneity for study-level factors but is not sufficient to identify individual-level effect modifiers. When the primary source of evidence in a research area is observational studies and when there is controversy despite several systematic reviews, IPD meta-analysis can be used to better understand sources of heterogeneity and provide context

Clostridium difficile infection (CDI) prevention is of high priority. We reviewed clinical practice guidelines (CPGs), and conducted an individual participant data meta-analysis (IPMDA) of randomized controlled trials (RCTs) to assess effectiveness and safety of probiotic prophylaxis. For CPGs, we rated quality, summarized recommendations with their strength and author-reported evidence, then re-evaluated evidence. For the IPDMA, we pooled RCTs investigating probiotics versus control for CDI prevention among antibiotic consumers, using generalized linear mixed models. Our outcomes were CDI and serious adverse events (SAEs). We adjusted for age, sex, hospitalization status, and exposure to high risk antibiotics. We assessed study risk of bias and confidence in estimates of effect. Five international guidelines were evaluated, and all scored poorly for applicability, stakeholder involvement, and rigor of development. Recommendations were not always linked to evidence, and guideline authors were not transparent about how evidence limitations impacted their decisions. None of the guidelines recommended probiotics. Fourteen studies contributed data, with one pending. Probiotics reduced CDI among all studies and the adjusted model. No covariates were significantly associated with CDI. Subgroups suggested that high incidence did not affect probiotic effectiveness, and high-dose, multi-strain probiotics were more beneficial. Our estimate was robust to sensitivity analyses. Probiotics did not significantly affect SAE odds among all studies and the adjusted model. Increasing age was a significantly associated with SAEs. No SAEs were reportedly probiotics-related. For both outcomes, estimates were similar from data of obtained and not obtained studies. Confidence in estimates was moderate for both outcomes, due to low event rates. Current guidelines on CDI prevention did not adhere well to validated standards for development and reporting, most notably due to insufficient links between recommendations and supporting evidence. Our preliminary analysis suggests that probiotic prophylaxis is useful and safe for CDI prevention. / Thesis / Master of Science (MSc) / Clostridium difficile infection (CDI) is a common hospital-associated infection, and prevention is of high priority. We reviewed clinical practice guidelines on CDI prevention to summarize their recommendations, and assess the quality of guideline development and reporting. Furthermore, we analysed patient data from randomized clinical trials to obtain an overall estimate (meta-analysis) of whether using a novel strategy, probiotic prophylaxis, is effective and safe. The guidelines had several limitations, importantly that authors were not transparent about how recommendation were developed, and recommendations were not always linked to evidence. Although no guideline recommended using probiotics to prevent CDI, our advanced analysis of previously conducted trials suggested that it was an effective intervention, reducing infections by approximately 76%, and was not associated with differences in serious adverse events compared to participants not receiving probiotics. In summary, guidelines on CDI prevention should be improved, and probiotics may be considered as an additional strategy.

Background: Subjective cognitive decline (SCD) is recognized as a risk stage for Alzheimer’s disease (AD) and other dementias, but its prevalence is not well known. We aimed to use uniform criteria to better estimate SCD prevalence across international cohorts. Methods: We combined individual participant data for 16 cohorts from 15 countries (members of the COSMIC consortium) and used qualitative and quantitative (Item Response Theory/IRT) harmonization techniques to estimate SCD prevalence. Results: The sample comprised 39,387 cognitively unimpaired individuals above age 60. The prevalence of SCD across studies was around one quarter with both qualitative harmonization/QH (23.8%, 95%CI = 23.3–24.4%) and IRT (25.6%, 95%CI = 25.1–26.1%); however, prevalence estimates varied largely between studies (QH 6.1%, 95%CI = 5.1–7.0%, to 52.7%, 95%CI = 47.4–58.0%; IRT: 7.8%, 95%CI = 6.8–8.9%, to 52.7%, 95%CI = 47.4–58.0%). Across studies, SCD prevalence was higher in men than women, in lower levels of education, in Asian and Black African people compared to White people, in lower- and middle-income countries compared to high-income countries, and in studies conducted in later decades. Conclusions: SCD is frequent in old age. Having a quarter of older individuals with SCD warrants further investigation of its significance, as a risk stage for AD and other dementias, and of ways to help individuals with SCD who seek medical advice. Moreover, a standardized instrument to measure SCD is needed to overcome the measurement variability currently dominant in the field.

Rationale: Cigarette cravings are one of the most important clinical phenomena in tobacco addiction. A wide range of studies and research designs may help to increase understanding of the relationship between physical activity (PA) and cigarette cravings. Aims: (i) To investigate the acute effects of walking and isometric exercise on cigarette cravings, withdrawal, and attentional bias among temporarily abstaining smokers. (ii) To quantify the effects of short bouts of PA on cigarette cravings among temporarily abstaining smokers. (iii) To examine who most benefits from PA, whether changes in affect mediate these effects, and whether a specific attribute of PA is associated with cravings. (iv) To investigate whether any association between habitual PA and cravings in smokers could be found. Methods: A randomised controlled crossover trial with three arms addressed aim (i). A systematic review of literature and individual participant data meta-analysis using hierarchical modelling addressed aims (ii) and (iii). Aim (iv) was achieved by using linear regression modelling of cross-sectional data from a smoking cessation study. Results: No difference in cravings, withdrawal, and attentional bias between walking and isometric exercise versus control was found. Bouts of PA decreased cigarette cravings by approximately 30%. Moderate intensity PA provided increased benefit when compared with light intensity, whereas vigorous intensity did not confer additional benefits compared with moderate intensity PA. Also bouts of medium (10 minutes) and longer duration (≥15minutes) appeared to be more effective than short duration (≤ 5 min). No moderators and mediators of this association were identified. Habitual moderate intensity PA was the strongest predictor of cigarette cravings in smokers, MPSS was an additional predictor and alcohol consumption moderated the effects of habitual PA on cravings. Conclusion: Moderate intensity PA could be recommended to smokers to help decrease cigarette cravings.

Le bénéfice de survie restreint à un horizon temporel donné a été proposé comme mesure alternative aux mesures relatives de l’effet d’un traitement notamment dans le cas de non proportionnalité des risques de décès. Le bénéfice de survie restreint correspond à la différence des survies moyennes entre deux bras de traitement et s’exprime en nombre d’années de vie gagnées. Dans la littérature, cette mesure est présentée comme plus intuitive que le hazard ratio et plusieurs auteurs ont proposé son utilisation pour le design et l’analyse d’un essai clinique. Toutefois, ce n’est pas actuellement la mesure qui est utilisée de manière courante dans les essais randomisés. Cette mesure s’applique quelle que soit la distribution des temps de survie et est adaptée si l’hypothèse des risques proportionnels n’est pas respectée. De plus, le bénéfice de survie restreint peut être utilisé en évaluation médico-économique où la mesure d’un effet absolu est nécessaire (nombre d’années de vie gagnées pondérées ou non par la qualité de vie). Si l’on souhaite estimer le bénéfice de survie restreint à partir d’une méta-analyse sur données individuelles, se pose alors la question de prendre en compte l’effet essai dû à la structure hiérarchique des données. L’objectif de cette thèse était de comparer des méthodes statistiques d’estimation du bénéfice de survie restreint à partir de données individuelles d’une méta-analyse d’essais cliniques randomisés. Le point de départ a été une étude de cas (étude coût-efficacité) réalisée à partir des données de la Meta-Analysis of Radiotherapy in Lung Cancer. Cette étude a montré que les cinq méthodes d’estimation étudiées conduisaient à des estimations différentes du bénéfice de survie et de son intervalle de confiance. Le choix de la méthode d’estimation avait également un impact sur les résultats de l’analyse coût-efficacité. Un second travail a consisté à mener une étude de simulation pour mieux comprendre les propriétés des méthodes d’estimation considérées en termes de biais moyen et d’erreur-type. Enfin, la dernière partie de la thèse a mis en application les enseignements de cette étude de simulation au travers de trois méta-analyses sur données individuelles dans le cancer du nasopharynx et dans le cancer du poumon à petites cellules. / The survival benefit restricted up to a certain time horizon has been suggested as an alternative measure to the common relative measures used to estimate the treatment effect, especially in case of non-proportional hazards of death. The restricted survival benefit corresponds to the difference of the two restricted mean survival times estimated for each treatment arm, and is expressed in terms of life years gained. In the literature, this measure is considered as more intuitive than the hazard ratio and many authors have suggested its use for the design and the analysis of clinical trials. However, it is not currently the most used measure in randomized trials. This measure is valid under any distribution of the survival times and is adapted if the proportional hazards assumption does not hold. In addition, the restricted survival benefit can be used in medico-economic evaluation where an absolute measure of the treatment effect is needed (number of [quality adjusted] life years gained). If one wants to estimate the restricted survival benefit from an individual participant data meta-analysis, there is a need to take into account the trial effect due to the hierarchical structure of the data. The aim of this thesis was to compare statistical methods to estimate the restricted survival benefit from an individual participant data meta-analysis of randomized trials. The starting point was a case study (cost-effectiveness analysis) using data from the Meta-Analysis of Radiotherapy in Lung Cancer. This study showed that the five investigated methods yielded different estimates for the restricted survival benefit and its confidence interval. The choice of a method to estimate the survival benefit also impacted on cost-effectiveness results. Our second project consisted in a simulation study to have a better understanding of the properties of the investigated methods in terms of bias and standard error. Finally, the last part of the thesis illustrated the lessons learned from the simulation study through three examples of individual participant data meta-analysis in nasopharynx cancer and in small cell lung cancer.