Learning objectives
to know how to set up a statistical analysis by taking a small clinical experimentation example; know how to draw conclusions using appropriate statistical tests
Prerequisites
none
Course unit content
- Statistical universe
- Types of measurements, test applicability
- The biological sample: extraction and sample size
- Randomisation
- Frequency distributions: Histograms
- Normal Gaussian distribution
- Probability
DESCRIPTIVE STATISTICS
- Graphical representation
- mean, variance, standard deviation and standard error
- Confidence limits of the mean and relative clinical significance
- Percentiles
PARAMETRIC STATISTICS
- The null hypothesis ( Ho ), types of error
- Comparison of means : Student t test, analysis of the variance,
- Percentage comparison analysis: contingency tables, Chi-squared test
- Fisher’s exact test
- Correlation: linear regression
NON PARAMETRIC STATISTICS
- Subjective assessment scales in clinical medicine
- Limits and advantages of non parametric tests
CLINICAL RESEARCH METHODOLOGY
- Definition of normal and pathological parameters
- Clinical experimentation
- Clinical trials: blind and double blind
- Control, placebo, follow-up
- Informed consent
Full programme
Introduction: medical statistics and related disciplines. Logic and
statistical planning. Overview of combinatorial analysis: permutations,
arrangements, combinations. Applications. Overview of probability
calculations: simple and compound probability, Bayes theorem.
Odds. Odds ratios. Likelihood ratios. applications.
Probability distributions : binomial distribution, Poisson distribution,
normal and standard normal distribution. Tables and their use.
Summarising data. Units of measure. Measurements of position, order
and variation. Indices of central tendency, mean median, mode.
Indices of variability, variance, standard deviation, CV. Percentiles and
their use.
General principles of statistical inference. Sampling distribution.
Hypothesis and hypothesis testing. Type 1 and type 2 error. Power of a
test and operating curve.
Power analysis and sample size determination.
Parametric test : Student t-test, ANOVA with 1 and 2 classification
criteria.
Non-parametric test: Wilcoxon test, Mann-Whitney test, Kruskal-Wallis
test, Friedman test, median test, Chi-square test, Fisher exact test.
Linear regression and correlation. Multiple regression. Logistic regression.
Computer exercises with the software "R" and Epi Info.
Bibliography
STANTON A. GLANTZ: Statistica per discipline Bio-mediche, McGraw Hill
L. LISON: Statistica applicata alla biologia sperimentale, Casa Ed. Ambrosiana
Introduzione ai sistemi informatici (Terza edizione)
D. Sciuto, G. Buonanno, W. Fornaciari, L. Mari,
Mc Graw Hill, 2004
Teaching methods
classroom lectures
Assessment methods and criteria
written and oral examination
Other information
- - -
