Switch to low bandwidth version Close

Continuing Professional Development: MSc Programme Modules (London-based)

STATISTICAL METHODS IN EPIDEMIOLOGY (2402)

ORGANISERS: Professor Simon Cousens, Dr Katherine Fielding

11 January 2012 to 10 February 2012 (Wednesdays 2 pm to Fridays 5pm)


AIM
To equip students with the skills needed to analyse and interpret data from cohort, case-control and cross-sectional studies, using cross-tabulation, stratification and regression models.


OBJECTIVES
By the end of this module students should be able to:

  1. explain the key statistical and epidemiological concepts which underlie the analysis of epidemiological data;
  2. perform analyses of data arising from epidemiological studies, using appropriate computer software (the software used throughout will be STATA);
  3. investigate confounding and effect modification (interaction) in epidemiological data;
  4. interpret appropriately the results of these analyses, taking into account study design issues;
  5. write a clear report presenting and interpreting the results of an analysis of epidemiological data.


CONSTITUENCY
This module is primarily intended for students who have attended the Term 1 modules in Statistics for EPH (2021) and Extended Epidemiology (2007), and are familiar with STATA and who wish to acquire further skills in the analysis and interpretation of epidemiological studies.  Students need to have a good grasp of this Term 1 material in order to benefit from this module.  In particular, students should be familiar with the three major epidemiological study designs, with the concepts of confounding and effect modification/ interaction, with the interpretation of statistical tests and confidence intervals, and with the basic data handling commands in STATA.  Students wishing to take the Advanced Statistical Methods in Epidemiology (2412) module in Term 3 need to take this course.  The module Design & Analysis of Epidemiological Studies (2417) is more appropriate for students who do not need detailed knowledge of data analysis techniques.


CONCEPTUAL OUTLINE 
This module is composed of four blocks:

  1. Cohort studies: analysis of rates using stratification to investigate confounding and interaction; simple survival analysis (life tables and Kaplan-Meier).  Introduction to Poisson and Cox regression. 
  2. Case-control studies: design issues including matching; analysis of studies using stratification to investigate confounding and interaction.
  3. Likelihood theory.
  4. Logistic regression for the analysis of case-control, cross-sectional and fixed cohort studies.

The emphasis is on the practical application of methods, with a brief introduction to likelihood theory, which provides the theoretical basis for most of the statistical methods covered in the module.


TEACHING STRATEGY 
Teaching consists mostly of lectures followed by computer practical sessions.  Methods are illustrated using data drawn from research work of staff in the Departments of Epidemiology & Population Health, and Infectious and Tropical Diseases.  These include both industrialised and developing country studies. The computer package STATA is used extensively.


LEARNING TIME 
The module is made up of 150 Notional Learning Hours – 38.5 hours contact time, 1.5 hours directed self-study, 70 hours self-directed learning, and 40 hours assessment, review and revision.


ASSESSMENT
Students will analyse an epidemiological dataset. They will each write a brief report describing their approach to the analysis and presenting and interpreting their results.

FEE

£1,600 including access to LSHTM library and learning resources, study materials and assessment.

Back to top