biostatistics

The main objective of this subject is for students to become familiar with the basic concepts and techniques of Descriptive Statistics, Probability Theory and Statistical Inference. It is intended that students understand the need and usefulness of statistical methodology in research in Health Sciences, particularly in the field of Medicine.

The specific objectives of the subject are detailed below:

- To know how to discriminate between the objectives of a statistical analysis: descriptive or inferential.
- To know how to distinguish between a statistical population and a sample of it.
- To synthesize and describe a large amount of data, selecting the appropriate statistics for the type of variables and analyzing the relationships between them.
- To know the probabilistic basis of Statistical Inference, as well as the general principles of the most common probabilistic models.
- To know how to estimate unknown parameters of a population from a sample.
- To know the principles and applications of hypothesis testing.
- To know how to compare two populations based on their characteristic and unknown parameters.
- To know how to formulate real problems in statistical terms and apply Statistical Inference to their resolution.
- To be able to handle statistical software packages.
- To assume the need and usefulness of Statistics as a tool in their professional practice, being aware of the degree of subjectivity and the risk of decisions based on statistical results.

Topic 1.- Descriptive statistics.
Definition and objectives of Statistics. Statistics in medical research. Design of a study, population and sample. Types of statistical variables. Summary of the information contained in a sample: frequency tables and graphical representations. Measures of centralization, position, dispersion and shape.

Topic 2.- Probability calculus.
Random experiment. Event and sample space. Operations with events. Axiomatic definition of probability. Conditional probability. Independence of events. Product rule. Theorem of total probabilities. Bayes rule. Prevalence and incidence of a disease. Diagnostic tests: sensitivity, specificity and predictive values.

Topic 3.- Discrete random variables.
One-dimensional random variable concept. Discrete random variable. Probability mass, distribution function and survival function. Characteristic measures: expected value and variance. Binomial distribution. Poisson distribution.

Topic 4.- Continuous random variables.
Continuous random variable. Density function, distribution function and survival function. Characteristic measures: expected value and variance. The normal distribution. Cut-off points for binormal diagnostic tests. Central Limit Theorem. Approximation of the binomial distribution by the normal. Distributions associated with the normal: Chi-Square, T-Student.

Topic 5.- Pointwise and interval estimation.
Objectives of Statistical Inference. Parameter and statistical concepts. Samplig distributions of statistics. Pointwise estimation of the mean, variance, and proportion. Bias and variance. Confidence intervals for the mean (in normal populations) and for the proportion. Determination of sample size.

Topic 6.- Introduction to hypothesis testing.
Basic concepts: Null and alternative hypotheses; unilateral contrast and bilateral contrast; acceptance and rejection zones; type I error and significance level; type II error and power; p-value. Test on the mean (in normal populations) and on the proportion. Test of means comparison in normal populations (for two independent or paired samples) and of proportions.

Topic 7.- Tests for categorical variables.
Contingency tables. Observed frequencies and expected frequencies. Chi-square test of independence. Yates Correction. 2x2-contingency tables in the field of Medicine. Association measures: Relative risk and odds ratio.

Topic 8.- Simple linear regression model.
Scatterplot. Covariance and linear correlation coefficient. Least squares method. Inference on model parameters. Variability decomposition. The F-Test. Determination coefficient. Diagnosis of the model. Prediction.

- Alonso Pena, M., Bolón Rodríguez, D., Ameijeiras Alonso, J., Saavedra Nieves, A. and Saavedra Nieves, P. (2024). Manual de R para prácticas de Bioestadística. Servizo de Publicacións da Universidade de Santiago de Compostela. DOI: https://dx.doi.org/10.15304/9788419679536.
- Álvarez Cáceres, R. (2007) “Estadística Aplicada a las Ciencias de la Salud”. Editorial Diaz de Santos.
- Daniel, W.W. (2006) “Bioestadística. Base para el análisis de las ciencias de la salud”. (2ª ed). Editorial LIMUSA. Wiley.
- Douglas G. A. (1997) “Practical Statistics for Medical Research”. Ed. Chapman & Hall.
- Martín Andrés, A. and Luna del Castillo, J. (1994) “Bioestadística para las ciencias de la salud”. (4ª ed). Ediciones Norma.
- Martín Andrés, A. and Luna del Castillo, J. (1995) “50 +/- 10 horas de Bioestadística”. Ediciones Norma.
- Martínez González, M.A; Sánchez, A. and Faulin, J. (2006). “Bioestadística amigable”. 2ª ed. Editorial Diaz de Santos.
- Milton, J.S. (1994) “Estadística para biología y ciencias de la salud”. (2ª ed). Ed. Interamericana, McGraw-Hill.
- Paradis, E. (2003). R para principiantes. R Cran. Disponible en https:/ cran.r- project.org/doc/contrib/rdebuts_es.pdf
- Quesada, V. and others (1982) “Curso de ejercicios de estadística”. (2ª ed). Editorial Alambra.
- Rosner, B. (2000) “Fundamentals of Biostatistics”. (5ª ed). Wadsworth Publishing Company. Duxbury Press.
- Venables, W.N., Smith, D.M. and the R Core Team (2020). An Introduction to R. Notes on R: A Programming Environment for Data Analysis and Graphics (Version 3.6.3). Disponible en https:/ cran.r-project.org/doc/manuals/r-release/R-intro.pdf.
- Verzani, J. (2005). Using R for Introductory Statistics. Chapman and Hall.

According to the subject file that appears in the Report for the Verification of the Official Title of Graduate in Medicine by the USC, the general competences related to the subject of Biostatistics are:

CG28.- Obtain and use epidemiological data and assess trends and risks for decision-making on health.
GC31.- Know, critically assess and know how to use clinical and biomedical information sources to obtain, organize, interpret and communicate scientific and health information.
GC33.- Maintain and use the records with patient information for subsequent analysis, preserving the confidentiality of the data.
GC34.- Have, in professional activity, a critical, creative point of view, with constructive skepticism and research-oriented.
GC35.- Understand the importance and limitations of scientific thought in the study, prevention and management of diseases.
GC36.- Being able to formulate hypotheses, collect and critically assess information for problem solving, following the scientific method.
GG37.- Acquire basic training for research activity.

In the field of Biostatistics, the CG32 competence (related to the use of information and communication technologies in clinical, therapeutic, preventive and research activities) will also be worked on, despite not appearing on the subject sheet.

The specific competencies that students must acquire through the Biostatistics subject are listed below:

CEMII.32.- Know the basic concepts of Biostatistics and its application to medical sciences.
CEMII.33.- Be able to design and carry out simple statistical studies using computer programs and interpret the results.
CEMII.34.- Understand and interpret statistical data in the medical literature.

Competences CEMII.31 and CEMII.378 (related to the critical use of technologies and the management of computer applications, respectively) will also be worked on in the matter, even though they do not appear on its file.

Lecture and interactive sessions will be face-to-face. In the Virtual Campus of the subject, students will find notes and problem sets. In addition, through the Virtual Campus, students will be able to take tests and deliver continuous assessment tasks, as described in the corresponding section.

Lecture sessions (36 hours): in the lecture sessions, the faculty will explain the theoretical and practical concepts of the contents, supported by multimedia presentations. Some standard problems will also be solved, so that the students can work on the exercise bulletins that will be provided. Regarding the material for monitoring the subject, apart from the recommended bibliography, students will have additional teaching material on the Virtual Campus.

Interactive sessions (20 hours): the interactive teaching is distributed in seminars to solve exercises and computer practices. In these sessions, the students will be introduced to the use of the R package for the statistical analysis of data, working on practical cases.

Tutorials (4 hours): the tutorials are designed to monitor student learning. These sessions will fundamentally work on those skills related to critical reasoning and communication skills.

Continuous evaluation (30%): two tests will be carried out to solve questions and problems, similar to those solved in the seminar sessions, with a weight of 5% each. With a weight of 20%, a third test will be proposed to solve practical cases with the R statistical software. These activities could be carried out in person or online.

Final exam (70%): the final exam will consist of several theoretical-practical questions and problems on the contents of the subject.
The weight of the continuous evaluation in the recovery opportunity will be the same as in the ordinary call of the semester.

Note that, in cases of fraudulent completion of exercises or tests, the provisions of the "Regulations for evaluating the academic performance of students and reviewing grades" will apply.

Students have 60 hours of face-to-face teaching (36 hours of lecture sessions, 20 hours of interactive teaching and 4 hours of tutorials).

For each hour of lectures, it is considered necessary to dedicate around 1.5 hours of individual work by the students. In relation to interactive teaching, for each hour approximately one additional hour is considered necessary for the preparation and subsequent review of the exercises and tasks carried out in class. In a complementary way, students must practice solving problems from the bulletins or from the recommended bibliography.

It is recommended to follow the expository and interactive sessions, as well as the activities proposed as fundamental means for the use of the subject.

To successfully pass the subject, it is also advisable to follow the proposed work plans. It is also recommended that students practice using the R statistical package to explore the possibilities of the various techniques explained throughout the course.

The course material will be available to students through the USC Virtual Campus. We intend that this platform will be the main means of communication with students, reinforced with MS Teams and email.