The aim of the course "Epidemiological methods and clinical epidemiology" is to provide the theoretical and practical tools for assessing the frequency of diseases in human populations and the associated risk factors. The course therefore aims to provide expertise in epidemiology, bio-statistics and information technology applied to biomedical data analysis. At the end of the course, the student must demonstrate to have understood the main bio-statistical methods applied to epidemiology and how to apply these methods to analyse data using a statistical software.
The course is structured in theoretical frontal lessons (32h) and in practical lessons (24h) on the use of a statistical software (STATA) for the quantitative analysis of biomedical data.
1. Introduction to epidemiology
- Definition and key features
- Traditional classification of epidemiology
- John Snow and cholera outbreaks in London
2. Measures of occurrence
- Outcomes
- Prevalence
- Cumulative incidence
- Incidence rate
3. Measures of association and public health impact
- Determinants
- Epidemiological associations
- Attributable risk (AR) and AR%
- Relative risk (RR) and Odds ratio (OR)
- Effect modification
- Population attributable risk (PAR) and PAR%
4. Types of epidemiological studies
- Ecological studies
- Cross-sectional studies
- Cohort studies
- Case-control studies
- Experimental studies
5. Causal interpretation of an empirical association
- Statistical vs. causal associations
- Causal models in epidemiology
- Validity of a study (random error, bias, confounding)
- Types of bias
- Methods to control confounding
- Hill’s positive criteria for causality
6. Principles of inference
- Principles of sampling
- Point estimate and sampling distribution
- Confidence interval
- Hypothesis test
- Test of significance
7. Stratified analysis
- Effect modification vs. confounding
- Stratum-specific estimates
- Testing homogeneity
- Pooled estimate
- Testing the stratified null hypothesis of no association
8. Basic statistical models in epidemiological research
- Linear regression model
- Logistic regression model
9. Statistical methods for survival analysis
- Kaplan-Meier non-parametric estimator
- Cox regression model
| Author | Title | Publisher | Year | ISBN | Note |
| Marubini E, Valsecchi MG | Analysing Survival Data from Clinical Trials and Observational Studies | John Wiley & sons | 1995 | ||
| Pearce N | A short Introduction to Epidemiology (Edizione 2) | 2005 | https://vula.uct.ac.za/access/content/group/9c29ba04-b1ee-49b9-8c85-9a468b556ce2/DOH/Module%202%20(Bio_Epi)/Epidemiology/EPIDEMIOLOGY/Pearce.pdf | ||
| Hennekens CH, Buring JE | Epidemiology in Medicine | Lippincott Williams & Wilkins | 1987 | ||
| McCullagh P, Nelder JA | Generalized Linear Models (Edizione 2) | Chapman and Hall/CRC | 1989 | ||
| Rothman KJ, Greenland S, Lash TL | Modern Epidemiology (Edizione 3) | Lippincott Williams & Wilkins | 2008 | ||
| Glantz SA | Statistica per Discipline Biomediche (Edizione 6) | McGraw-Hill | 2007 | 9788838639258 |
The final test is a written exam in computer lab. The test is the same for attending and non-attending students.
The aim of the test is to verify the knowledge of all the topics discussed and the ability to solve a biomedical problem by analyzing health data using the statistical software STATA.
The commands, results and interpretation of the analysis are reported in written form. In addition, students have to answer some questions to ascertain the understanding of theory.
The final evaluation is expressed in thirtieths.
