We welcome thesis students with different Ba-Ma backgrounds like physics, medicine, statistics, economics...
Health related quality of life research
Health related quality of life (HRQoL) research is a major topic in many applied fields, including health economics, health services and psychology. In health economics and epidemiology it is used to describe the burden of disease in general populations, and to estimate the gains in Quality Adjusted Life Years (QALYs) achieved through medical interventions. These are used in turn in cost-effectiveness analyses to inform policy makers on how to prioritise between different interventions in health care. In order to estimate QALY gains from medical interventions, it is necessary to dispose of baseline HRQoL values in the general population, and to understand better which general characteristics explain the differences between respondents.
Master Theses MSc Statistics (Hasselt University)
- 'Assessing the health related quality of life in the general population.'
- 'Assessing the health related quality of life in Flanders and the possible factors influencing it.'
- 'Exploring the relationship between three scale maps for measuring health related quality of life (VAS, EQ-5D 3 levels, and EQ-5D 5 levels.'
Influence of environmental factors on eliciting acute myocardial infarction in Belgium.
Aim. Exposure to air pollution has been shown to be a trigger of acute myocardial infarction (AMI). As in those studies no correction was made for other meteorological factors, the present thesis wanted to evaluate the independent environmental triggers of AMI.
Methods. Weekly counts of AMI patients that underwent primary percutaneous coronary intervention (pPCI) in the period 2006-2009 in 32 Belgian PCI centres were extracted from the national PCI database. Those data were correlated with average weekly meteorological data obtained from daily measurements in 73 meteorological sites, equally distributed in Belgium. The following meteorological measures were investigated: air pollution expressed as particulate matter both less than 10µM (PM10) and less than 2.5µM (PM2.5), black smoke, temperature and relative humidity. Time-series and Poisson regression analysis were carried out to investigate the correlation between environmental changes and the incidence of AMI.
Results & Conclusions. In a global environmental model, low temperature is by far the most important environmental trigger for AMI, whereas air pollution has a negligible effect.
Flexible estimation of unrelated health care cost
The aim of the study was to compare the evolution of medical costs over time for persons with a known positive isolate of streptococcus pneumoniae with the control group. The data set was used from the database from national alliance of Christian Sickness Fund (NACSF). We performed an age specific analysis using flexible modeling techniques. In this thesis, we presented flexible modeling techniques that take the correlation among measurements of the same subject in to account. We considered parametric and semiparametric models.
Identifying determinants of life expectancy in the EU
Life expectancy is the expected number of years of life remaining at a given age. It is a very important measure, a good proxy for health status and also it appears among the indicators used to assess the development of a country. Life expectancy is also used in public policy planning. This report looks for the determinants of life expectancy at birth and at age 65 in the EU over the period 1999-2007. Due to the difference in health status between men and women their life expectancies were modelled separately. Potential determinants present in the data set used are numerous and therefore are more likely to be concerned by multicollinearity.
A statistical analysis of the general public¹s ethical preferences regarding controversial vaccination policy options
In western countries, many infectious diseases still cause a significant burden of disease, although a perfectly safe and effective vaccine is available for everyone. This means that by increasing vaccine uptake in a population, large improvements can be achieved in reducing the morbidity and mortality that is caused by viruses and bacteria. Public health authorities can increase vaccination coverage in a population in many ways (e.g. making it legally compulsory) but there are important ethical obstacles that prevent these policy measures. We did a survey to investigate how the general public thinks about these different ethical obstacles. The student statistically analysed the influence of different standard back-ground variables (e.g. age, education level, etc) on the ethical preferences (i.e. the answers to the scenarios). But also the influence of some less common variables on the scenarios were investigated: the respondent¹s health state, and his/her position on a psychological scale. The statistical analysis of this sample generated interesting information both for public health policy and for academic discussion.
Bi-level Selection in Longitudinal Analysis with Time Dependent Covariates
Penalization methods such as the Lasso (Least absolute shrinkage and selection operator) (Tibshirani 1996) have been used in a variety of contexts to automatically select relevant variables and enhance predictive performance in regression models. Examples are analysis of genetic data and feature selection in image processing. Recently the group Lasso, group bridge, and group MCP have been proposed to deal with group structure in the data. The aim of group Lasso is to select a priori defined groups of variables as a whole. The group bridge and group MCP, in contrast, can perform bi-level selection by encouraging sparse solutions at the group and individual variable levels. In this thesis, we consider the problem of time dependent covariates in longitudinal data analysis to select relevant variable as well as to select the correct lag for each variable.