This study was a population-based retrospective cohort of all singleton births in British Columbia for the years 2001 to 2006. The purpose of this dissertation is to examine how social and physical environment factors influence the risk of adverse pregnancy outcomes and whether they interact with each other or with maternal characteristics to modify disease risk. The main environmental factors examined include ambient particulate air pollution (PM2.5), neighbourhood socioeconomic status (SES), neighbourhood immigrant density, neighbourhood level of post-secondary education level and the urban-rural context. Census dissemination areas (DAs) were used as the neighbourhood spatial unit. The data (N=242,472) was extracted from the BC Perinatal Data Registry (BCPDR) from Perinatal Services BC (PSBC). The main perinatal outcomes investigated include birth weight and indicators of fetal growth restriction such as small-for-gestational age (SGA), term low birth weight (tLBW), and intrauterine growth restriction (IUGR), preterm birth (PTB) and gestational age, gestational diabetes mellitus (GDM) and gestational hypertension (GH). Collectively, this dissertation contributes to the perinatal epidemiological literature linking particulate air pollution and neighbourhood SES context to adverse pregnancy outcomes. Assumptions about the linear effect of PM2.5 and smoking on birth weight are challenged showing that the effects are most pronounced between low and average exposures and that the magnitude of their effect is modified by neighbourhood and individual-level characteristics. These results suggest that focusing exclusively on individual risk factors may have limited success in improving outcomes without addressing the contextual influences at the neighbourhood-level. This dissertation therefore also contributes to the public health, sociological and community-urban development literature demonstrating that context and place matters. / Graduate / 0766 / 0573 / 0768 / anderse@uvic.ca

Mapping of mortality rates has been a valuable public health tool. We describe novel Bayesian methods for constructing maps which do not depend on a post stratification of the estimated rates. We also construct posterior modal maps rather than posterior mean maps. Our methods are illustrated using mortality data from chronic obstructive pulmonary diseases (COPD) in the continental United States. Poisson regression models have attracted much attention in the scientific community for their superiority in modeling rare events (including mortality counts from COPD). Christiansen and Morris (JASA 1997) described a hierarchical Bayesian model for heterogeneous Poisson counts under the exchangeability assumption. We extend this model to include latent classes (groups of similar Poisson rates unknown to an investigator). Also, it is standard practice to construct maps using quantiles (e.g., quintiles) of the estimated mortality rates. For example, based on quintiles, the mortality rates are cut into 5 equal size groups, each containing $20\%$ of the data, and a different color is applied to each of them on the map. A potential problem is that, this method assumes an equal number of data in each group, but this is often not the case. The latent class model produces a method to construct maps without using quantiles, providing a more natural representation of the colors. Typically, for rare events, the posterior densities of the rates are skewed, making the posterior mean map inappropriate and inaccurate. Thus, although it is standard practice to present the posterior mean maps, we also develop a method to provide the joint posterior modal map (i.e., the map with the highest posterior probability over the ensemble). For the COPD data, collected 1988-1992 over 798 health service areas, we use Markov chain Monte Carlo methods to fit the model, and an output analysis is used to construct the new maps.

La cartographie du risque en épidémiologie permet de mettre en évidence des régionshomogènes en terme du risque afin de mieux comprendre l’étiologie des maladies. Nousabordons la cartographie automatique d’unités géographiques en classes de risque commeun problème de classification à l’aide de modèles de Markov cachés discrets et de modèlesde mélange de Poisson. Le modèle de Markov caché proposé est une variante du modèle dePotts, où le paramètre d’interaction dépend des classes de risque.Afin d’estimer les paramètres du modèle, nous utilisons l’algorithme EM combiné à une approche variationnelle champ-moyen. Cette approche nous permet d’appliquer l’algorithmeEM dans un cadre spatial et présente une alternative efficace aux méthodes d’estimation deMonte Carlo par chaîne de Markov (MCMC).Nous abordons également les problèmes d’initialisation, spécialement quand les taux de risquesont petits (cas des maladies animales). Nous proposons une nouvelle stratégie d’initialisationappropriée aux modèles de mélange de Poisson quand les classes sont mal séparées. Pourillustrer ces solutions proposées, nous présentons des résultats d’application sur des jeux dedonnées épidémiologiques animales fournis par l’INRA. / The analysis of the geographical variations of a disease and their representation on a mapis an important step in epidemiology. The goal is to identify homogeneous regions in termsof disease risk and to gain better insights into the mechanisms underlying the spread of thedisease. We recast the disease mapping issue of automatically classifying geographical unitsinto risk classes as a clustering task using a discrete hidden Markov model and Poisson classdependent distributions. The designed hidden Markov prior is non standard and consists of avariation of the Potts model where the interaction parameter can depend on the risk classes.The model parameters are estimated using an EM algorithm and the mean field approximation. This provides a way to face the intractability of the standard EM in this spatial context,with a computationally efficient alternative to more intensive simulation based Monte CarloMarkov Chain (MCMC) procedures.We then focus on the issue of dealing with very low risk values and small numbers of observedcases and population sizes. We address the problem of finding good initial parameter values inthis context and develop a new initialization strategy appropriate for spatial Poisson mixturesin the case of not so well separated classes as encountered in animal disease risk analysis.We illustrate the performance of the proposed methodology on some animal epidemiologicaldatasets provided by INRA.

Spatially-explicit information is essential for planning and implementing interventions against vector-borne diseases. This is also true for African trypanosomoses, a group of diseases of both humans and animals caused by protozoa of the Genus Trypanosoma, and transmitted by tsetse flies (Genus Glossina).<p>In this thesis the knowledge gaps and the requirements for an evidence-based decision making in the field of tsetse and trypanosomoses are identified, with a focus on georeferenced data and Geographic Information Systems (GIS). Datasets, tools and analyses are presented that aim to fill some of the identified knowledge gaps.<p>For the human form of the disease, also known as sleeping sickness, case detection and treatment are the mainstay of control, so that accurate knowledge of the geographic distribution of infections is paramount. In this study, an Atlas was developed that provides village-level information on the reported occurrence of sleeping sickness. The geodatabase underpinning the Atlas also includes the results of active screening activities, even when no cases were detected. The Atlas enables epidemiological maps to be generated at a range of scales, from local to global, thus providing evidence for strategic and technical decision making.<p>In the field of animal trypanosomosis control, also known as nagana, much emphasis has recently been placed on the vector. Accurate delineation of tsetse habitat appears as an essential component of ongoing and upcoming interventions against tsetse. The present study focused on land cover datasets and tsetse habitat. The suitability for tsetse of standardized land cover classes was explored at continental, regional and national level, using a combination of inductive and deductive approaches. The land cover classes most suitable for tsetse were identified and described, and tailored datasets were derived.<p>The suite of datasets, methodologies and tools presented in this thesis provides evidence for informed planning and implementation of interventions against African trypanosomoses at a range of spatial scales. / Doctorat en Sciences agronomiques et ingénierie biologique / info:eu-repo/semantics/nonPublished

Disease maps are powerful tools for depicting spatial variations in disease risk and its underlying drivers.  However, producing effective disease maps requires careful consideration of the statistical and spatial properties of the disease data. In fact, the choice of mapping method influences the resulting spatial pattern of the disease, as well as the understanding of its underlying population characteristics. New developments in mapping methods and software in addition to continuing improvements in data quality and quantity are requiring map-makers to make a multitude of decisions before a map of disease burdens can be created. The impact of such decisions on a map, including the choice of appropriate mapping method, not been addressed adequately in the literature. This research demonstrates how choice of mapping method and associated parameters influence the spatial pattern of disease. We use four different disease-mapping methods – unsmoothed choropleth maps, smoothed choropleth maps produced using the headbanging method, smoothed kernel density maps, and smoothed choropleth maps produced using spatial empirical Bayes methods and 5-years of zip code level HIV incidence (2007- 2011) data from Dallas and Tarrant Counties, Texas. For each map, the leading population characteristics and their relative importance with regards to HIV incidence is identified using a regression analysis of a CDC recommended list of socioeconomic determinants of HIV. Our results show that the choice of mapping method leads to different conclusions regarding the associations between HIV disease burden and the underlying demographic and socioeconomic characteristics. Thus, the choice of mapping method influences the patterns of disease we see or fail to see. Accurate depiction of areas of high disease burden is important for developing and targeting appropriate public health interventions.

This thesis addresses three interrelated challenges of disease mapping and contributes a new approach for improving visualization of disease burdens to enhance disease surveillance systems. First, it determines an appropriate threshold choice (smoothing parameter) for the adaptive kernel density estimation (KDE) in disease mapping. The results show that the appropriate threshold value depends on the characteristics of data, and bandwidth selector algorithms can be used to guide such decisions about mapping parameters. Similar approaches are recommended for map-makers who are faced with decisions about choosing threshold values for their own data. This can facilitate threshold selection. Second, the study evaluates the relative performance of the adaptive KDE and spatial empirical Bayes for disease mapping. The results reveal that while the estimated rates at the state level computed from both methods are identical, those at the zip code level are slightly different. These findings indicate that using either the adaptive KDE or spatial empirical Bayes method to map disease in urban areas may provide identical rate estimates, but caution is necessary when mapping diseases in non-urban (sparsely populated) areas. This study contributes insights on the relative performance in terms of accuracy of visual representation and associated limitations. Lastly, the study contributes a new approach for delimiting spatial units of disease risk using straightforward statistical and spatial methods and social determinants of health. The results show that the neighborhood risk map not only helps in geographically targeting where but also in tailoring interventions in those areas to those high risk populations. Moreover, when health data is limited, the neighborhood risk map alone is adequate for identifying where and which populations are at risk. These findings will benefit public health tasks of planning and targeting appropriate intervention even in areas with limited and poor-quality health data. This study not only fills the identified gaps of knowledge in disease mapping but also has a wide range of broader impacts. The findings of this study improve and enhance the use of the adaptive KDE method in health research, provide better awareness and understanding of disease mapping methods, and offer an alternative method to identify populations at risk in areas with limited health data. Overall, these findings will benefit public health practitioners and health researchers as well as enhance disease surveillance systems.

Genome-wide association studies have greatly improved our understanding of the contribution of common variants to the genetic architecture of complex traits. However, two major limitations have been highlighted. First, common variant associations typically do not identify the causal variant and/or the gene that it is exerting its effect on to influence a trait. Second, common variant associations usually consist of variants with small effects. As a consequence, it is more challenging to harness their translational impact. Association studies of rare variants and complex traits may be able to help address these limitations. Empirical population genetic data shows that deleterious variants are rare. More specifically, there is a very strong depletion of common protein truncating variants (PTVs, commonly referred to as loss-of-function variants) in the genome, a group of variants that have been shown to have large effect on gene function, are enriched for severe disease-causing mutations, but in other instances may actually be protective against disease. This thesis is divided into three parts dedicated to the study of protein truncating variants, their medical relevance, and their functional consequences. First, I present statistical, bioinformatic, and computational methods developed for the study of protein truncating variants and their association to complex traits, and their functional consequences. Second, I present application of the methods to a number of case-control and quantitative trait studies discovering new variants and genes associated to breast and ovarian cancer, type 1 diabetes, lipids, and metabolic traits measured with NMR spectroscopy. Third, I present work on improving annotation of protein truncating variants by studying their functional consequences. Taken together, these results highlight the utility of interrogating protein truncating variants in medical and functional genomic studies.