Statistics for Evidence Based Medicine
A.Y. 2024/2025
Learning objectives
The general objective of the course is to develop the critical ability to evaluate the scientific evidence in support of the medical acts of prevention, diagnosis, treatment and rehabilitation with reference to the integration of anamnestic data, the physical examination and diagnostic tests, of the practical effectiveness of the therapy, and of the prognostic indicators.
In the training process of the physician, the acquisition of this ability contributes, in the ethical and deontological perspective, to the construction of a mental habitus capable of integrating, in the practice of daily practice, clinical knowledge deriving from direct personal experience, the values expressed by the assisted persons and the knowledge produced by good clinical and biomedical research. A habit that is now considered necessary for all doctors, and in particular for general practitioners, who are increasingly involved in health research and in the evaluation processes of the effectiveness of their professional practice.
As specific objectives, the course aims to develop the knowledge a practical skills for:
i) statistical methods that allow to describe and evaluate the different sources of variability;
ii) diagnostic tests and related measures of reliability and diagnostic relevance;
iii) basic principles of planning observational studies, interpretation of measures of disease occurrence and association between risk factors and disease occurrence;
iv) the relationships between statistics and the fundamentals of the inductive / deductive scientific method for empirical research with reference to the planning of observation and experimental studies and to the methods of statistical inference;
basic principles of ethics and the relationship with the methodology of biomedical research with specific reference to experimental studies of therapeutic efficacy.
In the training process of the physician, the acquisition of this ability contributes, in the ethical and deontological perspective, to the construction of a mental habitus capable of integrating, in the practice of daily practice, clinical knowledge deriving from direct personal experience, the values expressed by the assisted persons and the knowledge produced by good clinical and biomedical research. A habit that is now considered necessary for all doctors, and in particular for general practitioners, who are increasingly involved in health research and in the evaluation processes of the effectiveness of their professional practice.
As specific objectives, the course aims to develop the knowledge a practical skills for:
i) statistical methods that allow to describe and evaluate the different sources of variability;
ii) diagnostic tests and related measures of reliability and diagnostic relevance;
iii) basic principles of planning observational studies, interpretation of measures of disease occurrence and association between risk factors and disease occurrence;
iv) the relationships between statistics and the fundamentals of the inductive / deductive scientific method for empirical research with reference to the planning of observation and experimental studies and to the methods of statistical inference;
basic principles of ethics and the relationship with the methodology of biomedical research with specific reference to experimental studies of therapeutic efficacy.
Expected learning outcomes
The course aims to provide the methodological tools needed to learn, apply and evaluate, through critical analysis of the medical literature:
- the validity of anamnestic data and objective findings
- the concepts of reliability, accuracy and precision, repeatability and reproducibility of the measures;
- the usefulness of diagnostic tests and prognostic indexes;
- measures of disease occurrence and association with risk/benefit factors in epidemiology/clinical research;
- the identification of sources of imprecision and inaccuracy in epidemiological / clinical studies;
- the difference between studies based on observation and experimentation together with the main study designs in epidemiology / clinical research for the assessment of causal relationships, with reflection on the ethical aspects of biomedical research
- the effectiveness of therapies, rehabilitation practices, prevention programs, reported by observational and experimental studies.
- the validity of anamnestic data and objective findings
- the concepts of reliability, accuracy and precision, repeatability and reproducibility of the measures;
- the usefulness of diagnostic tests and prognostic indexes;
- measures of disease occurrence and association with risk/benefit factors in epidemiology/clinical research;
- the identification of sources of imprecision and inaccuracy in epidemiological / clinical studies;
- the difference between studies based on observation and experimentation together with the main study designs in epidemiology / clinical research for the assessment of causal relationships, with reflection on the ethical aspects of biomedical research
- the effectiveness of therapies, rehabilitation practices, prevention programs, reported by observational and experimental studies.
Lesson period: First semester
Assessment methods: Esame
Assessment result: voto verbalizzato in trentesimi
Single course
This course cannot be attended as a single course. Please check our list of single courses to find the ones available for enrolment.
Course syllabus and organization
Linea Policlinico
Course syllabus
Introduction
Presentation of the course: objectives and methods of development.
The role of statistical science in the biomedical field
Scientific evidence
Characteristics of the various types of observational and experimental scientific studies: cohort, case-control, cross-sectional and experimental studies (randomized clinical studies). The 4 phases of experimental studies. Tools and methods for evaluating the effectiveness of an intervention, or exposure to a risk factor, in scientific studies.
Endpoints (qualitative and quantitative).
Data.
Information, variables and data.
Methods and tools for data collection.
Construction of a questionnaire for data collection.
Descriptive statistics
Descriptive statistics I: construction and reading of frequency tables. Frequencies (absolute and relative; cumulative) and proportions. One-way and two-way tables.
Descriptive statistics II: construction and reading of graphs. The most common graphs and charts.
Descriptive statistics III: measures of central tendency (mode, median and mean) and measures of dispersion (interquartile range and standard deviation).
Introduction to probability and diagnostic test accuracy
Uncertainty and probability. Introduction to probabilistic reasoning and basic probability rules.
Probabilistic models: the normal and the binomial model.
Biological variability
Systematic and random errors: accuracy and precision
Reference individual and reference ranges of biochemical parameters.
Basic measures of clinical epidemiology
Measures of occurrence: incidence and prevalence.
Measures for the assessment of the efficacy of treatments or for the assessment of the effect of potential risk factors: risk differences, relative risks, odds ratios, number needed to treat. Methods of calculation and interpretation.
Confounding factors in clinical and epidemiological studies: definition and statistical tools for control.
Diagnostic test accuracyMeasures for evaluating the accuracy of a diagnostic or screening test: sensitivity, specificity, predictive values and likelihood ratios. The concordance within and between operators.
Diagnostic test based on the measurement of a continuous parameter: the ROC curve.
Using the results of a diagnostic accuracy study in clinical practice: pre- and post-test probability of disease.
Sampling, random variability and statistical inference
Concept of population and sample. Sampling and sample studies.
Population parameters and sample estimates; sampling distributions and statistical inference. Probabilistic models: Gaussian, Student's t and chi-square distributions. Central limit theorem.
Sampling and uncertainty: the standard error of the estimate of a parameter. Interval estimate of parameters: confidence intervals.
The logic of statistical inference in observational and experimental scientific studies: the hypothesis test. Null hypothesis and alternative hypothesis. Type I and II error. Power of a study. The concept of p value.
Qualitative endpoints. Comparison of two proportions: confidence intervals and appropriate statistical tests (z-test, chi-square test, Fisher's exact test, and McNemar's test)
Quantitative endpoints. Comparison of means of two or more groups: confidence intervals and appropriate statistical tests (z-test, t-test, analysis of variance, non-parametric tests).
Difference between statistical significance and clinical relevance.
Methods for defining the number of patients needed in a clinical study (sample size).
Time-to-events. The Kaplan-Meier method and the log-rank test
Modelling approach to the analysis of clinical data.
Introduction to regression models. Independent (causes) and dependent (outcomes) variables.
When the outcome is a quantitative variable: linear regression. Interpretation of results.
When the outcome is a qualitative variable: analysis with logistic regression models. Interpretation of results.
When the outcome is time-to-event: survival analysis: the Cox model. Interpretation of results.
The concept of multiple regression analysis: strategies and meaning.
Presentation of the course: objectives and methods of development.
The role of statistical science in the biomedical field
Scientific evidence
Characteristics of the various types of observational and experimental scientific studies: cohort, case-control, cross-sectional and experimental studies (randomized clinical studies). The 4 phases of experimental studies. Tools and methods for evaluating the effectiveness of an intervention, or exposure to a risk factor, in scientific studies.
Endpoints (qualitative and quantitative).
Data.
Information, variables and data.
Methods and tools for data collection.
Construction of a questionnaire for data collection.
Descriptive statistics
Descriptive statistics I: construction and reading of frequency tables. Frequencies (absolute and relative; cumulative) and proportions. One-way and two-way tables.
Descriptive statistics II: construction and reading of graphs. The most common graphs and charts.
Descriptive statistics III: measures of central tendency (mode, median and mean) and measures of dispersion (interquartile range and standard deviation).
Introduction to probability and diagnostic test accuracy
Uncertainty and probability. Introduction to probabilistic reasoning and basic probability rules.
Probabilistic models: the normal and the binomial model.
Biological variability
Systematic and random errors: accuracy and precision
Reference individual and reference ranges of biochemical parameters.
Basic measures of clinical epidemiology
Measures of occurrence: incidence and prevalence.
Measures for the assessment of the efficacy of treatments or for the assessment of the effect of potential risk factors: risk differences, relative risks, odds ratios, number needed to treat. Methods of calculation and interpretation.
Confounding factors in clinical and epidemiological studies: definition and statistical tools for control.
Diagnostic test accuracyMeasures for evaluating the accuracy of a diagnostic or screening test: sensitivity, specificity, predictive values and likelihood ratios. The concordance within and between operators.
Diagnostic test based on the measurement of a continuous parameter: the ROC curve.
Using the results of a diagnostic accuracy study in clinical practice: pre- and post-test probability of disease.
Sampling, random variability and statistical inference
Concept of population and sample. Sampling and sample studies.
Population parameters and sample estimates; sampling distributions and statistical inference. Probabilistic models: Gaussian, Student's t and chi-square distributions. Central limit theorem.
Sampling and uncertainty: the standard error of the estimate of a parameter. Interval estimate of parameters: confidence intervals.
The logic of statistical inference in observational and experimental scientific studies: the hypothesis test. Null hypothesis and alternative hypothesis. Type I and II error. Power of a study. The concept of p value.
Qualitative endpoints. Comparison of two proportions: confidence intervals and appropriate statistical tests (z-test, chi-square test, Fisher's exact test, and McNemar's test)
Quantitative endpoints. Comparison of means of two or more groups: confidence intervals and appropriate statistical tests (z-test, t-test, analysis of variance, non-parametric tests).
Difference between statistical significance and clinical relevance.
Methods for defining the number of patients needed in a clinical study (sample size).
Time-to-events. The Kaplan-Meier method and the log-rank test
Modelling approach to the analysis of clinical data.
Introduction to regression models. Independent (causes) and dependent (outcomes) variables.
When the outcome is a quantitative variable: linear regression. Interpretation of results.
When the outcome is a qualitative variable: analysis with logistic regression models. Interpretation of results.
When the outcome is time-to-event: survival analysis: the Cox model. Interpretation of results.
The concept of multiple regression analysis: strategies and meaning.
Prerequisites for admission
The course has no specific prerequisites
Teaching methods
The course consists of 6 ECTS. In addition, one extra ECTS is dedicated to practical skills activity.
Formal teaching: lectures with the use of slides and with the active participation of students.
Blended learning will be used with students working, individually or in groups, on some specific topics and a feedback will be given in the form of classroom discussion.
Formal teaching: lectures with the use of slides and with the active participation of students.
Blended learning will be used with students working, individually or in groups, on some specific topics and a feedback will be given in the form of classroom discussion.
Teaching Resources
Diapositive PowerPoint utilizzate a lezione ed eventuale materiale didattico aggiuntivo fornito dal docente resi disponibili sul sito Ariel del corso.
Bibliografia consigliata:
M. Bland Statistica medica. Maggioli Editore. 2019
W. Daniel Biostatistica. Concetti di base per l'analisi statistica delle scienze dell'area medico-sanitaria. Edises. 2019
M. Pagano K. Gauvreau Biostatistica. Idelson-Gnocchi 2003
D.G. Altman. Practical statistics for medical research. Chapman and Hall, London. 1991.
Casazza, G., Costantino G. Introduzione all'approccio critico alla decisione clinica. Milano: Milano University Press, 2024. ISBN 979-12-55101-02-4 (print). Per scaricare gratuitamente il volume: https://doi.org/10.54103/milanoup.164
Bibliografia consigliata:
M. Bland Statistica medica. Maggioli Editore. 2019
W. Daniel Biostatistica. Concetti di base per l'analisi statistica delle scienze dell'area medico-sanitaria. Edises. 2019
M. Pagano K. Gauvreau Biostatistica. Idelson-Gnocchi 2003
D.G. Altman. Practical statistics for medical research. Chapman and Hall, London. 1991.
Casazza, G., Costantino G. Introduzione all'approccio critico alla decisione clinica. Milano: Milano University Press, 2024. ISBN 979-12-55101-02-4 (print). Per scaricare gratuitamente il volume: https://doi.org/10.54103/milanoup.164
Assessment methods and Criteria
Assessment of student learning will be based on a PC based written test that may consist of questions, exercises for calculating and/or interpreting the results reported in scientific papers (critical reading of tables, graphs and numerical results reported in the "Results" part of a scientific article). Questions and exercises can be formulated in the form of open questions or multiple-choice questions. Each question will be assigned a score, which will be attributed to the student in case of a correct answer. The final grade, expressed in /30, will be obtained from the sum of the scores obtained for the individual questions. During the test students will be allowed to use written or printed material (books, slides and notes) and a pocket calculator. The use of smartphones, tablets and laptops is not allowed. The results of the tests will be published on the educational website (Ariel) or will be communicated by email.
MED/01 - MEDICAL STATISTICS - University credits: 6
Lessons: 52 hours
: 20 hours
: 20 hours
Professor:
Casazza Giovanni
Shifts:
Turno
Professor:
Casazza GiovanniLinea San Donato
Course syllabus
Introduction
Presentation of the course: objectives and methods of development.
The role of statistical science in the biomedical field
Scientific evidence
Characteristics of the various types of observational and experimental scientific studies: cohort, case-control, cross-sectional and experimental studies (randomized clinical studies). The 4 phases of experimental studies. Tools and methods for evaluating the effectiveness of an intervention, or exposure to a risk factor, in scientific studies.
Endpoints (qualitative and quantitative).
Data.
Information, variables and data.
Methods and tools for data collection.
Construction of a questionnaire for data collection.
Descriptive statistics
Descriptive statistics I: construction and reading of frequency tables. Frequencies (absolute and relative; cumulative) and proportions. One-way and two-way tables.
Descriptive statistics II: construction and reading of graphs. The most common graphs and charts.
Descriptive statistics III: measures of central tendency (mode, median and mean) and measures of dispersion (interquartile range and standard deviation).
Introduction to probability and diagnostic test accuracy
Uncertainty and probability. Introduction to probabilistic reasoning and basic probability rules.
Probabilistic models: the normal and the binomial model.
Biological variability
Systematic and random errors: accuracy and precision
Reference individual and reference ranges of biochemical parameters.
Basic measures of clinical epidemiology
Measures of occurrence: incidence and prevalence.
Measures for the assessment of the efficacy of treatments or for the assessment of the effect of potential risk factors: risk differences, relative risks, odds ratios, number needed to treat. Methods of calculation and interpretation.
Confounding factors in clinical and epidemiological studies: definition and statistical tools for control.
Diagnostic test accuracyMeasures for evaluating the accuracy of a diagnostic or screening test: sensitivity, specificity, predictive values and likelihood ratios. The concordance within and between operators.
Diagnostic test based on the measurement of a continuous parameter: the ROC curve.
Using the results of a diagnostic accuracy study in clinical practice: pre- and post-test probability of disease.
Sampling, random variability and statistical inference
Concept of population and sample. Sampling and sample studies.
Population parameters and sample estimates; sampling distributions and statistical inference. Probabilistic models: Gaussian, Student's t and chi-square distributions. Central limit theorem.
Sampling and uncertainty: the standard error of the estimate of a parameter. Interval estimate of parameters: confidence intervals.
The logic of statistical inference in observational and experimental scientific studies: the hypothesis test. Null hypothesis and alternative hypothesis. Type I and II error. Power of a study. The concept of p value.
Qualitative endpoints. Comparison of two proportions: confidence intervals and appropriate statistical tests (z-test, chi-square test, Fisher's exact test, and McNemar's test)
Quantitative endpoints. Comparison of means of two or more groups: confidence intervals and appropriate statistical tests (z-test, t-test, analysis of variance, non-parametric tests).
Difference between statistical significance and clinical relevance.
Methods for defining the number of patients needed in a clinical study (sample size).
Time-to-events. The Kaplan-Meier method and the log-rank test
Modelling approach to the analysis of clinical data.
Introduction to regression models. Independent (causes) and dependent (outcomes) variables.
When the outcome is a quantitative variable: linear regression. Interpretation of results.
When the outcome is a qualitative variable: analysis with logistic regression models. Interpretation of results.
When the outcome is time-to-event: survival analysis: the Cox model. Interpretation of results.
The concept of multiple regression analysis: strategies and meaning.
Presentation of the course: objectives and methods of development.
The role of statistical science in the biomedical field
Scientific evidence
Characteristics of the various types of observational and experimental scientific studies: cohort, case-control, cross-sectional and experimental studies (randomized clinical studies). The 4 phases of experimental studies. Tools and methods for evaluating the effectiveness of an intervention, or exposure to a risk factor, in scientific studies.
Endpoints (qualitative and quantitative).
Data.
Information, variables and data.
Methods and tools for data collection.
Construction of a questionnaire for data collection.
Descriptive statistics
Descriptive statistics I: construction and reading of frequency tables. Frequencies (absolute and relative; cumulative) and proportions. One-way and two-way tables.
Descriptive statistics II: construction and reading of graphs. The most common graphs and charts.
Descriptive statistics III: measures of central tendency (mode, median and mean) and measures of dispersion (interquartile range and standard deviation).
Introduction to probability and diagnostic test accuracy
Uncertainty and probability. Introduction to probabilistic reasoning and basic probability rules.
Probabilistic models: the normal and the binomial model.
Biological variability
Systematic and random errors: accuracy and precision
Reference individual and reference ranges of biochemical parameters.
Basic measures of clinical epidemiology
Measures of occurrence: incidence and prevalence.
Measures for the assessment of the efficacy of treatments or for the assessment of the effect of potential risk factors: risk differences, relative risks, odds ratios, number needed to treat. Methods of calculation and interpretation.
Confounding factors in clinical and epidemiological studies: definition and statistical tools for control.
Diagnostic test accuracyMeasures for evaluating the accuracy of a diagnostic or screening test: sensitivity, specificity, predictive values and likelihood ratios. The concordance within and between operators.
Diagnostic test based on the measurement of a continuous parameter: the ROC curve.
Using the results of a diagnostic accuracy study in clinical practice: pre- and post-test probability of disease.
Sampling, random variability and statistical inference
Concept of population and sample. Sampling and sample studies.
Population parameters and sample estimates; sampling distributions and statistical inference. Probabilistic models: Gaussian, Student's t and chi-square distributions. Central limit theorem.
Sampling and uncertainty: the standard error of the estimate of a parameter. Interval estimate of parameters: confidence intervals.
The logic of statistical inference in observational and experimental scientific studies: the hypothesis test. Null hypothesis and alternative hypothesis. Type I and II error. Power of a study. The concept of p value.
Qualitative endpoints. Comparison of two proportions: confidence intervals and appropriate statistical tests (z-test, chi-square test, Fisher's exact test, and McNemar's test)
Quantitative endpoints. Comparison of means of two or more groups: confidence intervals and appropriate statistical tests (z-test, t-test, analysis of variance, non-parametric tests).
Difference between statistical significance and clinical relevance.
Methods for defining the number of patients needed in a clinical study (sample size).
Time-to-events. The Kaplan-Meier method and the log-rank test
Modelling approach to the analysis of clinical data.
Introduction to regression models. Independent (causes) and dependent (outcomes) variables.
When the outcome is a quantitative variable: linear regression. Interpretation of results.
When the outcome is a qualitative variable: analysis with logistic regression models. Interpretation of results.
When the outcome is time-to-event: survival analysis: the Cox model. Interpretation of results.
The concept of multiple regression analysis: strategies and meaning.
Prerequisites for admission
The course has no specific prerequisites
Teaching methods
The course consists of 6 ECTS. In addition, one extra ECTS is dedicated to practical skills activity.
Formal teaching: lectures with the use of slides and with the active participation of students.
Blended learning will be used with students working, individually or in groups, on some specific topics and a feedback will be given in the form of classroom discussion.
Formal teaching: lectures with the use of slides and with the active participation of students.
Blended learning will be used with students working, individually or in groups, on some specific topics and a feedback will be given in the form of classroom discussion.
Teaching Resources
Diapositive PowerPoint utilizzate a lezione ed eventuale materiale didattico aggiuntivo fornito dal docente resi disponibili sul sito Ariel del corso.
Bibliografia consigliata:
M. Bland Statistica medica. Maggioli Editore. 2019
W. Daniel Biostatistica. Concetti di base per l'analisi statistica delle scienze dell'area medico-sanitaria. Edises. 2019
M. Pagano K. Gauvreau Biostatistica. Idelson-Gnocchi 2003
D.G. Altman. Practical statistics for medical research. Chapman and Hall, London. 1991.
Casazza, G., Costantino G. Introduzione all'approccio critico alla decisione clinica. Milano: Milano University Press, 2024. ISBN 979-12-55101-02-4 (print). Per scaricare gratuitamente il volume: https://doi.org/10.54103/milanoup.164
Bibliografia consigliata:
M. Bland Statistica medica. Maggioli Editore. 2019
W. Daniel Biostatistica. Concetti di base per l'analisi statistica delle scienze dell'area medico-sanitaria. Edises. 2019
M. Pagano K. Gauvreau Biostatistica. Idelson-Gnocchi 2003
D.G. Altman. Practical statistics for medical research. Chapman and Hall, London. 1991.
Casazza, G., Costantino G. Introduzione all'approccio critico alla decisione clinica. Milano: Milano University Press, 2024. ISBN 979-12-55101-02-4 (print). Per scaricare gratuitamente il volume: https://doi.org/10.54103/milanoup.164
Assessment methods and Criteria
Assessment of student learning will be based on a PC based written test that may consist of questions, exercises for calculating and/or interpreting the results reported in scientific papers (critical reading of tables, graphs and numerical results reported in the "Results" part of a scientific article). Questions and exercises can be formulated in the form of open questions or multiple choice questions. Each question will be assigned a score, which will be attributed to the student in case of a correct answer. The final grade, expressed in /30, will be obtained from the sum of the scores obtained for the individual questions. During the test students will be allowed to use written or printed material (books, slides and notes) and a pocket calculator. The use of smartphones, tablets and laptops is not allowed. The results of the tests will be published on the educational website (Ariel) or will be communicated by email.
MED/01 - MEDICAL STATISTICS - University credits: 6
Lessons: 52 hours
: 20 hours
: 20 hours
Professors:
Ambrogi Federico, Turati Federica
Shifts:
Linea San Giuseppe
Course syllabus
Introduction
Presentation of the course: objectives and methods of development.
The role of statistical science in the biomedical field
Scientific evidence
Characteristics of the various types of observational and experimental scientific studies: cohort, case-control, cross-sectional and experimental studies (randomized clinical studies). The 4 phases of experimental studies. Tools and methods for evaluating the effectiveness of an intervention, or exposure to a risk factor, in scientific studies.
Endpoints (qualitative and quantitative).
Data.
Information, variables and data.
Methods and tools for data collection.
Construction of a questionnaire for data collection.
Descriptive statistics
Descriptive statistics I: construction and reading of frequency tables. Frequencies (absolute and relative; cumulative) and proportions. One-way and two-way tables.
Descriptive statistics II: construction and reading of graphs. The most common graphs and charts.
Descriptive statistics III: measures of central tendency (mode, median and mean) and measures of dispersion (interquartile range and standard deviation).
Introduction to probability and diagnostic test accuracy
Uncertainty and probability. Introduction to probabilistic reasoning and basic probability rules.
Probabilistic models: the normal and the binomial model.
Biological variability
Systematic and random errors: accuracy and precision
Reference individual and reference ranges of biochemical parameters.
Basic measures of clinical epidemiology
Measures of occurrence: incidence and prevalence.
Measures for the assessment of the efficacy of treatments or for the assessment of the effect of potential risk factors: risk differences, relative risks, odds ratios, number needed to treat. Methods of calculation and interpretation.
Confounding factors in clinical and epidemiological studies: definition and statistical tools for control.
Diagnostic test accuracyMeasures for evaluating the accuracy of a diagnostic or screening test: sensitivity, specificity, predictive values and likelihood ratios. The concordance within and between operators.
Diagnostic test based on the measurement of a continuous parameter: the ROC curve.
Using the results of a diagnostic accuracy study in clinical practice: pre- and post-test probability of disease.
Sampling, random variability and statistical inference
Concept of population and sample. Sampling and sample studies.
Population parameters and sample estimates; sampling distributions and statistical inference. Probabilistic models: Gaussian, Student's t and chi-square distributions. Central limit theorem.
Sampling and uncertainty: the standard error of the estimate of a parameter. Interval estimate of parameters: confidence intervals.
The logic of statistical inference in observational and experimental scientific studies: the hypothesis test. Null hypothesis and alternative hypothesis. Type I and II error. Power of a study. The concept of p value.
Qualitative endpoints. Comparison of two proportions: confidence intervals and appropriate statistical tests (z-test, chi-square test, Fisher's exact test, and McNemar's test)
Quantitative endpoints. Comparison of means of two or more groups: confidence intervals and appropriate statistical tests (z-test, t-test, analysis of variance, non-parametric tests).
Difference between statistical significance and clinical relevance.
Methods for defining the number of patients needed in a clinical study (sample size).
Time-to-events. The Kaplan-Meier method and the log-rank test
Modelling approach to the analysis of clinical data.
Introduction to regression models. Independent (causes) and dependent (outcomes) variables.
When the outcome is a quantitative variable: linear regression. Interpretation of results.
When the outcome is a qualitative variable: analysis with logistic regression models. Interpretation of results.
When the outcome is time-to-event: survival analysis: the Cox model. Interpretation of results.
The concept of multiple regression analysis: strategies and meaning.
Presentation of the course: objectives and methods of development.
The role of statistical science in the biomedical field
Scientific evidence
Characteristics of the various types of observational and experimental scientific studies: cohort, case-control, cross-sectional and experimental studies (randomized clinical studies). The 4 phases of experimental studies. Tools and methods for evaluating the effectiveness of an intervention, or exposure to a risk factor, in scientific studies.
Endpoints (qualitative and quantitative).
Data.
Information, variables and data.
Methods and tools for data collection.
Construction of a questionnaire for data collection.
Descriptive statistics
Descriptive statistics I: construction and reading of frequency tables. Frequencies (absolute and relative; cumulative) and proportions. One-way and two-way tables.
Descriptive statistics II: construction and reading of graphs. The most common graphs and charts.
Descriptive statistics III: measures of central tendency (mode, median and mean) and measures of dispersion (interquartile range and standard deviation).
Introduction to probability and diagnostic test accuracy
Uncertainty and probability. Introduction to probabilistic reasoning and basic probability rules.
Probabilistic models: the normal and the binomial model.
Biological variability
Systematic and random errors: accuracy and precision
Reference individual and reference ranges of biochemical parameters.
Basic measures of clinical epidemiology
Measures of occurrence: incidence and prevalence.
Measures for the assessment of the efficacy of treatments or for the assessment of the effect of potential risk factors: risk differences, relative risks, odds ratios, number needed to treat. Methods of calculation and interpretation.
Confounding factors in clinical and epidemiological studies: definition and statistical tools for control.
Diagnostic test accuracyMeasures for evaluating the accuracy of a diagnostic or screening test: sensitivity, specificity, predictive values and likelihood ratios. The concordance within and between operators.
Diagnostic test based on the measurement of a continuous parameter: the ROC curve.
Using the results of a diagnostic accuracy study in clinical practice: pre- and post-test probability of disease.
Sampling, random variability and statistical inference
Concept of population and sample. Sampling and sample studies.
Population parameters and sample estimates; sampling distributions and statistical inference. Probabilistic models: Gaussian, Student's t and chi-square distributions. Central limit theorem.
Sampling and uncertainty: the standard error of the estimate of a parameter. Interval estimate of parameters: confidence intervals.
The logic of statistical inference in observational and experimental scientific studies: the hypothesis test. Null hypothesis and alternative hypothesis. Type I and II error. Power of a study. The concept of p value.
Qualitative endpoints. Comparison of two proportions: confidence intervals and appropriate statistical tests (z-test, chi-square test, Fisher's exact test, and McNemar's test)
Quantitative endpoints. Comparison of means of two or more groups: confidence intervals and appropriate statistical tests (z-test, t-test, analysis of variance, non-parametric tests).
Difference between statistical significance and clinical relevance.
Methods for defining the number of patients needed in a clinical study (sample size).
Time-to-events. The Kaplan-Meier method and the log-rank test
Modelling approach to the analysis of clinical data.
Introduction to regression models. Independent (causes) and dependent (outcomes) variables.
When the outcome is a quantitative variable: linear regression. Interpretation of results.
When the outcome is a qualitative variable: analysis with logistic regression models. Interpretation of results.
When the outcome is time-to-event: survival analysis: the Cox model. Interpretation of results.
The concept of multiple regression analysis: strategies and meaning.
Prerequisites for admission
The course has no specific prerequisites
Teaching methods
The course consists of 6 ECTS. In addition, one extra ECTS is dedicated to practical skills activity.
Formal teaching: lectures with the use of slides and with the active participation of students.
Blended learning will be used with students working, individually or in groups, on some specific topics and a feedback will be given in the form of classroom discussion.
Formal teaching: lectures with the use of slides and with the active participation of students.
Blended learning will be used with students working, individually or in groups, on some specific topics and a feedback will be given in the form of classroom discussion.
Teaching Resources
Diapositive PowerPoint utilizzate a lezione ed eventuale materiale didattico aggiuntivo fornito dal docente resi disponibili sul sito Ariel del corso.
Bibliografia consigliata:
M. Bland Statistica medica. Maggioli Editore. 2019
W. Daniel Biostatistica. Concetti di base per l'analisi statistica delle scienze dell'area medico-sanitaria. Edises. 2019
M. Pagano K. Gauvreau Biostatistica. Idelson-Gnocchi 2003
D.G. Altman. Practical statistics for medical research. Chapman and Hall, London. 1991.
Casazza, G., Costantino G. Introduzione all'approccio critico alla decisione clinica. Milano: Milano University Press, 2024. ISBN 979-12-55101-02-4 (print). Per scaricare gratuitamente il volume: https://doi.org/10.54103/milanoup.164
Bibliografia consigliata:
M. Bland Statistica medica. Maggioli Editore. 2019
W. Daniel Biostatistica. Concetti di base per l'analisi statistica delle scienze dell'area medico-sanitaria. Edises. 2019
M. Pagano K. Gauvreau Biostatistica. Idelson-Gnocchi 2003
D.G. Altman. Practical statistics for medical research. Chapman and Hall, London. 1991.
Casazza, G., Costantino G. Introduzione all'approccio critico alla decisione clinica. Milano: Milano University Press, 2024. ISBN 979-12-55101-02-4 (print). Per scaricare gratuitamente il volume: https://doi.org/10.54103/milanoup.164
Assessment methods and Criteria
Assessment of student learning will be based on a PC based written test that may consist of questions, exercises for calculating and/or interpreting the results reported in scientific papers (critical reading of tables, graphs and numerical results reported in the "Results" part of a scientific article). Questions and exercises can be formulated in the form of open questions or multiple choice questions. Each question will be assigned a score, which will be attributed to the student in case of a correct answer. The final grade, expressed in /30, will be obtained from the sum of the scores obtained for the individual questions. During the test students will be allowed to use written or printed material (books, slides and notes) and a pocket calculator. The use of smartphones, tablets and laptops is not allowed. The results of the tests will be published on the educational website (Ariel) or will be communicated by email.
MED/01 - MEDICAL STATISTICS - University credits: 6
Lessons: 52 hours
: 20 hours
: 20 hours
Educational website(s)
Professor(s)
Reception:
On appointment (email)
Laboratorio di Statistica Medica, Biometria ed Epidemiologia "G.A. Maccacaro", Via Celoria 22, Milano