Ateneo - UPO

Introduction to Evidence-Based Medicine (EBM), with particular focus on the statistical aspects of randomized controlled clinical trials and meta-analyses, and on the evaluation of diagnostic test performance

Scientific articles and teaching materials provided in class and available on the DIR platform

Knowledge of the distinctive features of randomized controlled clinical trials and meta-analyses, of the statistical methods for data analysis, and of the methods for evaluating diagnostic test performance. Understanding of the results reported in scientific articles concerning the topics covered in class.

None

Lectures with the aid of PowerPoint slides and exercises on evaluating the conduct and results of studies drawn from scientific articles on randomized controlled clinical trials and meta-analyses, and on assessing study quality. The section on diagnostic test performance is delivered as an asynchronous lesson, with videos available on the DIR platform.

The material used in class will be made available on the DIR platformStudents with disabilities, DSA, BES, once they have contacted the University Staff, can contact the teacher in charge of the course in relation to the declination of the exam methods, regarding the teaching aspects.

For the EBM and Statistics for EBM modules, an exemption exam is scheduled approximately one week before the Medical Pathology III exam, and its grade will remain valid for all subsequent exam sessions. It will not be possible to access the exam for the modules of Hematology, Infectious Diseases, Clinical Immunology and Rheumatology, and Simulated Cases without having passed the exemption exam. The EBM exam consists of a written test with 2 open-ended questions on the group project. The Statistics for EBM exam consists of 5 multiple-choice questions on the evaluation of diagnostic test performance, and 2 open-ended questions requiring comments on the results reported in a meta-analysis and one of the trials from a pool identified by the instructor and made available on DIR at the end of the course. Specifically, a figure or table from one of the proposed trials and meta-analyses will be provided, and a comment on the reported results will be required. A single grade will be calculated as the average of the two exam parts.

Randomized Controlled Trials (RCTs): study design, randomization methods, analysis approaches (intention-to-treat, per-protocol, as-treated), study objectives (superiority, non-inferiority, and equivalence trials), bias (selection, performance, detection, and attrition bias) and methods to avoid them. Frequency measures for the occurrence of dichotomous outcomes (risk, incidence, and prevalence), measures of association and impact, tests for comparing means, survival analysis (Kaplan-Meier method for risk estimation, survival curves, log-rank test, Cox proportional hazards model).
Meta-analysis: defining the research question, defining the literature search strategy, defining inclusion and exclusion criteria, quality assessment of included studies (RoB2 scale for RCTs), data extraction and analysis (calculation of fixed-effect and random-effects meta-analytic estimates and assessment of heterogeneity between studies, Cochrane's Q test, I² index), assessment of sources of variability and robustness of results (subgroup/stratified analysis to identify sources of heterogeneity, influence analysis, publication bias). Overview of network meta-analysis.
Evaluation of diagnostic test performance: definitions of sensitivity, specificity, positive and negative predictive value, accuracy and confidence intervals, serial testing, parallel testing, ROC curve, criteria for determining the optimal cut-off, area under the curve (AUC).

Knowledge: By the end of the course, students will be familiar with the main characteristics of randomized controlled trials and meta-analyses, the main analysis and sensitivity analysis methods, as well as the main techniques for evaluating the performance of diagnostic tests with dichotomous and continuous outcomes.
Competence: By the end of the course, students will be able to interpret the results of scientific articles related to randomized controlled trials/meta-analyses based on the data reported in tables and figures. They will also be able to assess whether a diagnostic test is adequate for identifying a given condition.
Transversal skills: By the end of the course, students will be able to critically read a scientific article and draw conclusions regarding the potential treatment of future patients based on the evidence of efficacy available in the scientific literature.