Skip to main content

Statistical Methods in Controlled Trials and Survival Analysis

Credits
6
Types
Optional
Requirements
This subject has not requirements , but it has got previous capacities
Department
CS;EIO
Statistical methods such as randomized trials, sequential designs, and survival analysis, originally developed in biomedicine, are now widely used across diverse fields including economics, digital marketing, engineering, insurance, sociology, and criminology. This course introduces these techniques, combining causal inference, randomized controlled trial design, and survival analysis, with hands-on implementation in R.

Teachers

Person in charge

  • Marta Bofill Roig (marta.bofill.roig@upc.edu)

Weekly hours

Theory
2
Problems
1
Laboratory
1
Guided learning
0
Autonomous learning
7.54

Objectives

  1. Objective 1 (Methodological Evaluation & Problem Solving): To critically analyze, model, and solve complex technical problems in medical and clinical research by applying fundamental statistical principles, causal inference, and advanced experimental or observational designs.
    Related competences:
  2. Objective 2 (Data Management & Survival Analysis): To efficiently manage, structure, analyze, and visualize complex biomedical data, specifically implementing advanced survival analysis techniques and modeling time-to-event outcomes using statistical software.
    Related competences:
  3. Objective 3 (Clinical Decision Making & Application): To design and implement advanced statistical (such as group sequential methods and adaptive designs) to optimize decision-making processes in clinical trials, including early termination for efficacy or futility.
    Related competences:
  4. Objective 4 (Ethical Judgements & Uncertainty): To integrate incomplete or limited clinical data to make reasoned, critical judgments, taking into full account the social, ethical, and safety responsibilities associated with human clinical trials and medical research.
    Related competences:
  5. Objective 5 (Scientific Communication & Argumentation): To reason, argue, and clearly communicate scientific-technical conclusions, study rationales, and statistical reports to both specialized (medical/statistical) and non-specialized audiences without ambiguity.
    Related competences:
  6. Objective 6 (Autonomous Learning & Professional Drive): To develop self-directed learning skills and a continuous improvement mindset, enabling the student to autonomously master new advanced methodologies and face future professional challenges in the fast-evolving field of medical statistics.
    Related competences:

Contents

  1. Introduction and Statistical Foundations
    Description of the different types of medical studies and conceptualization of the challenges in the field and the scope of the subject. The concepts of bias, confounding, and causal inference are introduced, along with a review of the statistical tools necessary for evidence-based inference. In particular, data types and descriptive statistics, probability and distributions, hypothesis testing, p-values and confidence intervals, and sample size and statistical power.

    1. Observational study designs (cohort, case-control, cross-sectional)
    2. Experimental study designs (non-randomized trial, randomized controlled trial)
  2. Clinical Trial Design
    This part focuses on clinical trials as a central tool in experimental medical research. The phases and objectives of trials, the main randomization methods, and the statistical techniques that allow for the decision of early trial termination due to signs of efficacy or futility are examined.

    1. Phases and objectives of clinical trials
    2. Randomization methods
    3. Group sequential methods for early termination
  3. Survival Analysis
    Many medical studies evaluate the time until an event (death, relapse, recovery) instead of simple presence or absence. Survival analysis is the branch of statistics designed for this context, with applications that go beyond medicine: engineering (time-to-failure) or economics (unemployment duration).

    1. Observation schemes and censoring
    2. Basic concepts in survival analysis
    3. Estimation of the survival function and cumulative hazard
    4. Comparison of survival curves between treatment groups
    5. Design and sample size in survival clinical trials
    6. Regression models in survival analysis

Activities

Activity Evaluation act


A. Introduction and Statistical Foundations

Desenvolupament del tema Bloc A. Introduction and Statistical Foundations
Objectives: 1 5
Theory
4h
Problems
2h
Laboratory
1h
Guided learning
0h
Autonomous learning
6h

B. Clinical Trial Design

Progress in B. Clinical Trial Design
Objectives: 1 3 4 5
Contents:
Theory
10h
Problems
5.5h
Laboratory
5h
Guided learning
0h
Autonomous learning
30h

C. Survival Analysis

Progress in C. Survival Analysis
Objectives: 2 5 6
Contents:
Theory
10h
Problems
5.5h
Laboratory
5h
Guided learning
0h
Autonomous learning
30h

Exam 1

Exam 1
Objectives: 1 3 4 5 6
Week: 9
Theory
0h
Problems
0h
Laboratory
0h
Guided learning
0h
Autonomous learning
0h

Exam 2

Exam 2
Objectives: 1 2 3 5 6
Week: 18
Theory
0h
Problems
0h
Laboratory
0h
Guided learning
0h
Autonomous learning
0h

Test 1

Test 1
Objectives: 1 3 4 6
Week: 4
Theory
0h
Problems
0h
Laboratory
0h
Guided learning
0h
Autonomous learning
0h

Test 2

Test 2
Objectives: 2 4 5 6
Week: 16
Theory
0h
Problems
0h
Laboratory
0h
Guided learning
0h
Autonomous learning
0h

Teaching methodology

Lectures: Magisterial sessions where the faculty will introduce the theoretical concepts, methodological foundations, and the statistical and design tools of the subject. Interactivity will be encouraged through the resolution of doubts and brief discussions on key concepts.

Application sessions: Practical classroom sessions aimed at problem-solving, case study discussion, or the critical analysis of medical articles. The objective is to apply the learned theory to practical situations and debate the results in groups.

Laboratory sessions: Practical sessions where students will learn to use the statistical software (R). Work will focus on data manipulation, analysis programming (such as survival analysis or sample size calculation), and the interpretation of computer output.

Evaluation methodology

The assessment of the subject is continuous and consists of the following elements:

2 Midterm exams, EP1 and EP2 (Theoretical-practical): Written tests that will assess both the understanding of theoretical concepts and the ability to apply them in solving problems and practical cases.

2 Quizzes, T1 and T2: Short tests aimed at verifying regular progress in the subject and the achievement of key concepts.

The final mark will be computed using the following formula: 0.35(EP1+EP2)+0.15(T1+T2)

Bibliography

Basic

Previous capacities

Recommended prior knowledge:

Fundamental statistics: Basic concepts of parameter estimation, hypothesis testing, and regression models.

Data management and analysis: Ability to perform basic data manipulation (cleaning, filtering, transforming variables) and apply descriptive statistics techniques (calculation of frequencies, means, deviations, and graphical representations).