The course of history of medicine will consider some significant turning points of the past of medicine to understand the present, explaining the process of transmission of mainly Western medical knowledge. Particular emphasis will be placed on the evolution of the concept of disease by analyzing in detail some basic conceptual articulations that are found at the origins of contemporary biomedicine.
These conceptual coordinates are proposed in order to provide students with the ability to perceive medicine and health, in all its ramifications, as a process historically changeable, to make them critically aware of the special dimension of knowledge and medical practice in the history and in the contemporary age.
Multiple choice test
MEDICAL STATISTICS COURSE
The students will learn to recognize and statistically control the random variability intrinsic in the clinical and biological observations. This allows them to appropriately use the information they will collect in their future diagnostic and therapeutic practice. The knowledge acquired will also allow them to critically appraise the published results of the current medical research.
The students will learn to:
1. Read, interpret, produce tabular or graphical statistical descriptions of medical data
2. Interpret and generate probabilistic statements regarding medical and epidemiological hypotheses.
3. Evaluate and state the precision of the estimates of therapeutic or risk effects, investigated by themselves or others.
4. Understand and formulate simple statistical models, to explain the relationships between several prognostic or therapeutic variables and clinical outcome.
1. The statistical regularity detectable in stochastic experiments: the law of large numbers
2. Methods to efficiently synthesize large numbers of medical observations: frequency distributions, histograms and other graphs, position and variability indices.
3. The probability concept and probabilistic models useful to describe biological variables.
4. From the description of the observations to the inferential medical statements: the statistical test.
5. Evaluating the error probability of epidemiological conclusions: the p-value.
6. Tests for different sorts of medical observations: t-test, chi square test, F-tests, and others.
7. The danger of using too many statistical tests: Bonferroni and other corrections
8. Using a sample to guess a characteristic of the whole patient population: point and interval estimation, its confidence level.
9. The power of a study to detect a presumed therapeutic or harmful effect - the computation of the wanted sample size.
10. Statistical models for estimation or prediction; an example: the linear regression, its representation and epidemiological interpretation
11. Building richer models - the general linear model - its use to eliminate the confounding of prognostic variables.
The theoretical coverage of all topics will always be supplemented by substantial practical applications. These will entail statistical analyses of real medical data, performed using the statistical language "R”. If the logistics of the course will allow it, every student will individually carry out the relevant computation on a PC, under the teacher's supervision
Assessment and exams
The students will be asked to answer 10 open ended questions concerning the topics of the course.
Betty R. Kirkwood and Jonathan A. C. Sterne: Essential Medical Statistics - Second Edition. Blackwell Science, Oxford (UK) 2003.
Lectures (2 hours each)
1. Statistics: the scientific language of Medicine
2. Univariate descriptive Statistics
3. Bivariate descriptive Statistics
4. Observed cumulative distribution function and quantiles
5. The law of large numbers: introduction to probability
6. Axioms of probability & Bayes theorem
7. Sensitivity, Specificity and ROC curve of a diagnostic test
8. The Binomial probability model
9. The Gaussian probability model
10. The central limit theorem
11. Elements of point estimation
12. Introduction to confidence intervals
13. Application of confidence intervals to real medical data sets
14. The risk difference, the relative risk, the odds ratio
15. Introduction to the statistical test: the P-value
16. Tests of means, proportions and independence of categorical variables
17. Power of statistical tests, width of confidence intervals and sample size
18. Introduction to statistical models: the simple linear regression
19. Multivariable models: the multiple regression
20. Countering the confounding by multivariable models
- Teacher: MARIO ANGELO COMELLI