Module overview
This module focuses on the application of statistical methods specially developed for epidemiological study data. Topics include the basic disease occurrence measures of prevalence and incidence with their role in surveillance including standardization, Mantel-Haenszel estimation of various effect measures including the risk ratio and risk difference for cohort studies and the odds ratio for case-control studies as well as Poisson and logistic regression to adjust for potential confounders simultaneously. The module also includes elements of time-to-event analysis including Kaplan-Meier estimation and Cox' proportional hazards model for confounder adjustment. Finally, basic concepts of statistical methods for meta-analysis will be introduced. The module includes a mixture of lectures and practical workshops using the software STATA.
Aims and Objectives
Learning Outcomes
Knowledge and Understanding
Having successfully completed this module, you will be able to demonstrate knowledge and understanding of:
- Perform elementary descriptive time-to-event analysis including the Kaplan-Meier estimate of the survivor function and nonparametric tests for comparing two survivor distributions (log-rank test).
- Demonstrate an understanding of the basic concepts and application of statistical estimation, hypothesis tests and inference to epidemiological data, in particular in the context when adjusting for confounding and effect modification variables.
- Use the STATA software to perform all of the above
- Model complex study data including several exposures and confounders using logistic regression, Poisson and log-linear regression.
- Perform regression models on time-to-event outcomes (Cox’ proportional hazard’s model).
- Analyse study data of various types including Mantel-Haenszel estimation with hypothesis testing and confidence interval estimation.
- Undertake basic statistical meta-analysis.
Subject Specific Intellectual and Research Skills
Having successfully completed this module you will be able to:
- Analyse (using STATA) epidemiological study data with the appropriate statistical tools including Mantel-Haenszel estimation and regression modelling.
- To read, understand and critically appraise published epidemiological research.
- Identify the appropriate statistical tools for a given epidemiological study with a specific design such cross-sectional, cohort or case-control (matched or unmatched).
- Identify the appropriate statistical models given for a given epidemiological study.
- Identify the right tools in STATA to analyse a given epidemiological data set including the interpretation on the various output coefficients and tests provided by STATA
Transferable and Generic Skills
Having successfully completed this module you will be able to:
- Develop your own epidemiologic research in design, data collection and analysis.
- Ability to use STATA for epidemiological analysis
- Be able to discuss modern quantitative strategies in epidemiological research.
- Read critically empirical based research literature.
Syllabus
• Introducing STATA for epidemiologists.
• Inferential concepts including confidence intervals and hypothesis tests with focus on adjusted measures (MHE).
• Confounding, effect modification and Mantel-Haenszel estimation.
• Regression and logistic regression.
• Log-linear modelling and Poisson regression.
• Time-to-event modelling, survival function, hazard function, censoring.
• Kaplan-Meier estimation and nonparametric group tests.
• Cox’ regression model
• Basic statistics for meta-analysis
Learning and Teaching
Teaching and learning methods
A variety of methods will be used including lectures, active participatory methods, case studies of epidemiology in practice, practical exercises using STATA, guided reading, group study and individual study.
| Type | Hours |
|---|---|
| Teaching | 30 |
| Independent Study | 120 |
| Total study time | 150 |
Resources & Reading list
Textbooks
Jewell NP (2004). Statistics for epidemiology. London: Chapman & Hall/CRC Press.
Clayton D, Hills M (1993). Statistical models in epidemiology. Oxford: Oxford Science Publications.
Woodward M (1999). Epidemiology: Study desig and data analysis. London: Chapman&Hall/CRC.
Assessment
Assessment strategy
The assessment is summative, and at an individual level. The pass mark for the module is 50%. If you have failed the module, you will have the opportunity to submit work at the next referral (re-sit) opportunity. On passing your referrals, your final module mark will be capped at 50%.Formative
This is how we’ll give you feedback as you are learning. It is not a formal test or exam.
Class practicals
- Assessment Type: Formative
- Feedback:
- Final Assessment: No
- Group Work: No
Summative
This is how we’ll formally assess what you have learned in this module.
| Method | Percentage contribution |
|---|---|
| Coursework | 100% |
Referral
This is how we’ll assess you if you don’t meet the criteria to pass this module.
| Method | Percentage contribution |
|---|---|
| Coursework | 100% |
Repeat Information
Repeat type: Internal & External