# Epidemiology, 7.5 ECTS

Second level## Facts

No available facts**Course code**MT7008

## Syllabus

## Description

Epidemiology is the study of relationships between health and various characteristics of individuals in populations. The aim is to understand how individual differences in exposures (life-style, demographics, genes etc.) might explain patterns of disease distributions across populations. Since information on risk factors is rarely available for all individuals in a population, statistical tools are used to select individuals that are studied in detail (sampling). Issues of statistical design, analysis and interpretation are crucial in order to make valid and efficient statements about the larger population from which the sample is drawn.

In observational epidemiology the data do not originate from carefully designed and planned experiments, but rather from the passive observation of individuals in ‘free living conditions’. Several methods in ‘classical’ statistical theory are applicable to data from controlled experiments, i.e in situations where randomization is a key for obtaining valid comparisons. In contrast, associations in observational epidemiology do not in general imply causation. In order to obtain useful inferences the design, analysis and interpretation needs to pay careful attention to the control for factors that may confound the association between exposures and responses. Additional pitfalls include the potential for selection bias and information bias (bias due to measurement error).

Epidemiological studies are special also since the response variable – disease status – is typically binary. Statistical methods for continuous responses (comparison of means and use of regression models) need to be extended to the context of binary responses. An additional layer of complexity is induced by dynamics over time and the need for appropriate handling of individuals who are censored from observation through e.g. death or end of follow-up.

The aim of the course is to give an overview of the statistical methods used in epidemiology, the rationale behind them, their statistical properties and inter-relations as well as potential pitfalls in their use. Particular emphasis will be given to logistic regression models for retrospectively sampled data (case-control studies), conditional logistic regression models for matched case-control data and Poisson and Cox regression models for prospectively sampled time-to-event data (cohort studies).

## Area of interests: Information only in Swedish

## Subject

### Mathematical Statistics

Mathematical statistics is the subject within applied mathematics that describes and analyses random events.

The foundation is the mathematical probability theory, which goes back to the 17th century, but has in its modern form developed mostly during the 20th century. Probability theory is also the foundation of statistical theory on how to draw conclusions from data with random features. The rise of computer technology has also significantly contributed to broadening the field of applications within mathematical statistics. Today is mathematical statistics one of the most powerful tools in applied mathematics.

Examples that apply mathematical statistics to a large extent are the insurance and finance sectors, biological and medical research (biostatistics) and industrial applications like telecommunications and quality control.

The department cooperates closely both in research and teaching with the pharmaceutical industry, medical institutions and banking and insurance companies. Depending on direction of studies, a student with an exam in mathematical statistics has a good chance to get employment in these sectors.

Mathematical statistics can be studied in elective courses, as a major in a Bachelor’s or Master’s degree or as a complementary subject. Courses cover both theory and applications in various areas. Mathematical statistics is included in the Bachelor’s programmes in Mathematics, Biomathematics and Mathematics and Economics and a major in the Master’s programmes in Mathematical statistics, Biostatistics, Finance mathematics and Finance and Actuarial science.