Central to the goals of public health are obtaining and interpreting timely and relevant information for the benefit of humanity. In this dissertation, we propose methods to monitor and assess the spread HIV in a more rapid manner, as well as to improve decisions regarding patient treatment options. In Chapter 1, we propose a method, extending the previously proposed dual-testing algorithm and augmented cross-sectional design, for estimating the HIV incidence rate in a particular community. Compared to existing methods, our proposed estimator allows for shorter follow-up time and does not require estimation of the mean window period, a crucial, but often unknown, parameter. The estimator performs well in a wide range of simulation settings. We discuss when this estimator would be expected to perform well and offer design considerations for the implementation of such a study. Chapters 2 and 3 are concerned with obtaining a more complete understanding of the impact of treatment in randomized clinical trials in which multiple patient outcomes are recorded. Chapter 2 provides an illustration of methods that may be used to address concerns of both risk-benefit analysis and personalized medicine simultaneously, with a goal of successfully identifying patients who will be ideal candidates for future treatment. Riskbenefit analysis is intended to address the multivariate nature of patient outcomes, while “personalized medicine” is concerned with patient heterogeneity, both of which complicate the determination of a treatment’s usefulness. A third complicating factor is the duration of treatment use. Chapter 3 features proposed methods for assessing the impact of treatment as a function of time, as well as methods for summarizing the impact of treatment across a range of follow-up times. Chapter 4 addresses the issue of meta-analysis, a commonly used tool for combining information for multiple independent studies, primarily for the purpose of answering a clinical question not suitably addressed by any one single study. This approach has proven highly useful and attractive in recent years, but often relies on parametric assumptions that cannot be verified. We propose a non-parametric approach to meta-analysis, valid in a wider range of scenarios, minimizing concerns over compromised validity.

L’objectif de cette thèse était de développer un cadre conceptuel et méthodologique permettant d’évaluer l’impact global de l’alimentation sur la santé des consommateurs, en prenant en compte les dimensions microbiologiques, chimiques et nutritionnelles. Cette méthodologie a été développée à l’aide d’un cas d’étude portant sur l’alimentation des nourrissons (lait maternel et formules infantiles), incluant les facteurs suivants : C. sakazakii, Cryptosporidium, arsenic, polychlorobiphényles de type dioxine et acide docosahexaénoïque. Cinq modèles mathématiques probabilistes ont été développés pour quantifier les risques / bénéfices associés à chaque facteur. Ils ont été ensuite harmonisés, quand cela a été possible, à l’aide d’un indicateur commun de santé publique, le DALY. Les résultats ont été obtenus par simulation de Monte Carlo de second ordre afin de quantifier séparément l’incertitude et la variabilité. Les techniques probabilistes ont permis de prendre en compte d’une part la variabilité inhérente à la biologie (hétérogénéité entre individus d’une même population) et d’autre part l’incertitude liée au manque de connaissances et de données. De plus, la séparation de la variabilité et de l’incertitude a consolidé l’évaluation, permettant une interprétation plus cohérente des résultats et donc fournissant des informations plus complètes aux décisionnaires. La méthode mise en oeuvre dans ce travail de thèse pourra servir de base pour d’autres cas d’études et pourra aussi être utilisée pour continuer le développement méthodologique de l’évaluation risque-bénéfice. Cette démarche s’inscrit dans une approche plus générale d’analyse multi-critères des systèmes agronomiques et alimentaires / The objective of the present PhD project was to develop a conceptual and methodological framework to assess the overall impact of food consumed on human health, including microbiological, chemical and nutritional dimensions. This methodology was developed using a case study on infant milk-based diet (breast milk and infant formulas) taking into account the following selected factors: C. sakazakii, Cryptosporidium, arsenic, dioxin like polychlorinated biphenyls and docosahexaenoic acid. Five probabilistics mathematicals models were developed to quantify risks / benefits associated with these factors. When possible, they were harmonized using a common public health indicator, the DALY. Results were obtained by second-order Monte Carlo simulations in order to quantify separately the uncertainty and the variability. Probabilistic techniques enabled to take into account on the one hand the biology related to variability (heterogeneity between individuals of the same population) and on the other hand the uncertainty linked to the lack of knowledge and data. In addition, separation of variability and uncertainty strengthened the evaluation by enabling a more accurated interpretation of results and by providing more comprehensive information for policy makers. The method used in this PhD thesis can be considered as a robust basis for other case studies and can be also used to continue methodological development in risk-benefit assessment. This approach is also part of a broader area: the multi-criteria decision analysis of agronomic and food systems