28172 - Biostatistics

Course Unit Page

Academic Year 2018/2019

Learning outcomes

By the end of the course the students know the basic concepts and the statistical methods for the analysis of problems in the biomedical sciences. In particular students should be able: -to calculate and interpret the principal epidemiological measurements in various types of epidemiological studies. -to fit and interpret generalized linear models applied to biomedical data. -to apply the methods and models for the analysis of survival data.

Course contents

  • Epidemiology, public health and evidence-based medicine

    • The basics of epidemiology: concepts, definitions and measures. Main areas of application of biostatistics: from disease occurrence to causal effects. Risk, incidence rate and prevalence.

    • Relations between measures of occurrence. Exposure and population at risk. Effect measures: attributable risk, rate and fraction. Analysis of binary data: relative risks and odds ratios.

    • Principles and practice of evidence-based medicine. Levels of evidence and the Cochrane collaboration. Evaluating the efficacy, effectiveness and efficiency of health care.

    • R Labs: basics for data management and biostatistical analysis. Packages and data sources. Recoding variables for tabular outputs. Producing graphical outputs.

  • Types of epidemiologic studies: methodology and biostatistical tools

    • Random error and the role of statistics: confidence intervals and test of hypothesis. Types of epidemiologic studies: cohort, case-control, cross-sectional, randomized studies. Number needed to treat. Sources of bias and confounding.

    • Epidemiology in clinical settings: diagnosis and screening. Sensitivity, specificity, predictive values, likelihood ratios, the ROC curve.

    • Model building strategies: general guidelines and how to control for confounding and assess interactions. Applications using logistic regression.

    • R Labs: computing risk estimates, confidence intervals, p values and summary measures for different type of studies.

  • Survival analysis

    • Life tables for follow-up studies, parametric and non parametric methods. The Kaplan Meier product-limit estimate of survival. Survivor and hazard function.

    • Treatment of missing values, lost to follow up and intention to treat analysis. Comparison of two groups (log-rank, Peto and Wilcoxon tests) and three or more groups of survival data.

    • The Cox Proportional Hazards model: theory and practice

    • R Labs. Applied survival analysis using R

  • How to read a scientific paper

    • Biostatistics in practice: how to read/write scientific papers and technical reports. Criteria for quality appraisal.

    • Appraisal Labs: group work

Readings/Bibliography

Main References

  • Greenalgh T, How to Read a Paper: The Basics of Evidence-Based Medicine. 5th Edition, Wiley-BMJ Books 2014.

  • Rothman KJ, Epidemiology: an introduction, 2nd Edition, Oxford University Press 2012.

  • Kleinbaum DG, M Klein, Survival analysis. A self learning Text. 3rd Edition. Springer 2012.

Selected sections
  • Kleinbaum DG, Logistic Regression. A self learning text. 3rd Edition, Springer, 2010.

  • Agresti A, An introduction to categorical data analysis, 2nd Edition, Wiley 2007.

Papers (provisional list)

  • Isaacs D, Fitzgerald D. Seven alternatives to evidence based medicine. BMJ. 1999 Dec 18-25;319(7225):1618. Available at: http://www.bmj.com/content/319/7225/1618.long

  • Schmidt CO, Kohlmann T. When to use the odds ratio or the relative risk?Int J Public Health. 2008;53(3):165-7.

  • Nicolucci A et al. A comprehensive assessment of the avoidability of long-term complications of diabetes. A case-control study. SID-AMD Italian Study Group for the Implementation of the St. Vincent Declaration. Diabetes Care. 1996 Sep;19(9):927-33.

Teaching methods

  • Lectures with open discussion
  • Computer labs
  • Selected readings

Assessment methods

The exam will evaluate the achievements of the student through a direct assessment of the following learning outcomes:

  • understanding of key research questions addressed by studies in the different areas of epidemiology and evidence-based medicine

  • ability to identify the most suitable type of studies, materials and methods required to carry out the analysis and interpretation of health-related data

  • ability to interpret the results of a scientific publication in any of the relevant research areas

The exam will consist of an in depth oral assessment concerning the ability to understand a problem and conduct a biostatistical analysis in any of the above fields of interest. The student will be presented with a short publication, whose strengths and weaknesses will need to be rapidly assessed, together with an explanation of the statistical methods and a sound interpretation of the final results. The student should be able to respond to questions regarding the choice of methods in different situations frequently occurring in biostatistics, including aspects related to data acquisition and statistical processing using ad hoc software.

Teaching tools

Teacher's notes

Office hours

See the website of Fabrizio Carinci