GCSRT Program | Foundation Courses

More... Share to Twitter Share to Facebook
Foundation Courses

PROGRAM CURRICULUM: JUNE 21, 2014 TO JUNE 6, 2015

Foundation Courses - "F101 to F107"
 

F101 - Introduction to Biostatistics

This course provides a thorough introduction to the most commonly used biostatistics techniques for clinical research. Specific topics include tools for describing central tendency and variability in data; methods for performing inference on population means and proportions via sample data; statistical hypothesis testing and its application to group comparisons; and issues of power and sample size in study designs. There is an introduction to simple linear regression and survival analysis. While there are some formulae and computational elements using Stata, the emphasis is on interpretation and concepts.

F102 - Introduction to Epidemiology

This introductory course in epidemiology presents an overview but not a detailed discussion of the basic methods of epidemiology and their applications to clinical research. Lectures explore such basic principles of epidemiology as the importance of measurement, including types of outcome measures and measures of association; diverse array of study designs available in clinical research, including cross-sectional studies, cohort studies, case-control studies and experimental designs; types of potential biases, including selection bias and measurement bias; confounding and methods for its avoidance and control; and effect modification.

F103 - Applied Regression

The course is designed for students who have completed A101, i.e., students with a good working knowledge of elementary descriptive statistics; sampling distributions; one and two sample tests for means and proportions; correlation and basic linear and multiple regression model building. initially, lectures will explore general concepts in linear regression and consider residual analysis and data transformations. lectures will address multiple linear regression, including consideration of confounding and effect modification. model building will be emphasized. Lastly, several lectures will explore topics in logistical regression including, 2x2 Tables and stratification, model building and assessment of goodness of fit, and smoothing and generalized additive models.

F104 - Survival Analysis

This course is designed for students who have completed A101 and A103. It builds on the basic concepts of survival analysis discussed in A101, including hazard functions, survival functions, types of censoring and truncation, Kaplan-meier estimates, log-rank tests and their generalization. The course introduces statistical models and methods useful for analyzing univariate and multivariate failure time data. After completing this course, students will be able to describe time-to-event data and compare groups with a time-to-event outcome; interpret the coefficients and control for confounding using a Cox proportional hazards model; interpret interaction terms and incorporate time varying covariates in a Cox model as well as assess the proportional hazards assumption. Lastly, students will learn how to complete a sample size calculation for a survival study.

F105 - Causal Design

Causal inference is an overarching objective of most forms of medical and epidemiological investigation. Key questions usually consist of whether an intervention works and the extent of the benefit and whether it causes harm. While a randomized controlled trial design is considered the most powerful way to infer causality, such studies may not be possible or feasible and an observational approach may be necessary to attain causal inference. This course builds on A102. at the end of the course, students will have a deeper understanding of observational approaches, especially from the perspective of overcoming the problem of confounding. Students will be able to define confounding and develop approaches toward identifying confounders. DAGs, as a structural approach to identifying confounders, will be highlighted. Other topics will include the rules of D-separation and conditioning on common effects. Propensity scores will be introduced. The differences between randomized trials and observational studies are considered and quasi-experimental designs introduced.

F106 - Longitudinal and Correlated Data

A longitudinal study refers to an investigation where outcomes and possibly treatments or exposures are collected at multiple follow-up times. a longitudinal study generally yields multiple or “repeated” measurements on each subject, which may correlate over time. With correlated outcomes, it is useful to understand the strength and pattern of correlations. Characterizing correlation can be approached using mixed-effects models or generalized estimating equations (GEE). This course covers methods to analyze longitudinal data, including the use of linear regression models. Topics will include polynomial trends for time (e.g., linear or quadratic) and linear mixed-effects models. at the end of the course, students will be able to interpret the results from a multilevel model and understand how to incorporate multiple random effects into the model. Students will be able to understand the types of missing data that occur in longitudinal and crosssectional analysis as well as understand the assumptions associated with each analysis approach. This course requires completion of a101, a103, and a104.

F107 - Special Topics in Biostatistics

This course seeks to synthesize the importance of key topics that have already been covered (e.g., power analysis and sample size) and discuss other important statistical topics such as agreement studies and factorial design. The lectures will play an integrative role with lectures in F101, F103, and F104. For example, the power analysis lecture identifies and distinguishes the design factors affecting the precision and power of a planned investigation and addresses the trade-offs among design parameters within constraints of time and resources. At the end of the course, students will be able to prepare a thorough and pertinent statistical justification for sample size, precision, power, and detectable effects in a planned investigation as well as interpret and evaluate the power-analysis aspects of a research proposal or published report.


"The GCSRT program has achieved a perfect blend of international exposure and flexibility. It has been able to fulfill my desire to tap into a global wealth of knowledge without compromising on professional growth. I am confident its impact on my career goals will be phenomenal."
 

Ewemade Igbinedion, MBBS, MPH, FRSPH, ACIEH
Igbinedion University Teaching Hospital
Benin, Nigeria