Workshop overview
The primary objective of this workshop is to offer participants an
in-depth understanding of concepts and methodologies for conducting
dose-response meta-analyses, utilizing either individual patient data or
aggregated (summarized) data. The workshop will cover a range of topics,
including statistical models tailored for dose-response meta-analyses
across various endpoints such as continuous, binary, and time-to-event
data. Additionally, participants will explore different dose-response
strategies, including linear trends, polynomials, and splines, along
with techniques for assessing and quantifying heterogeneity,
meta-regression models, and sensitivity analyses. Throughout the
workshop, we will elucidate different research questions and
applications, formalize statistical inference methods for the proposed
models, delineate the types of available data, and demonstrate how to
effectively present results in tabular and graphical formats suitable
for publication.
The workshop will employ a combination of lectures and hands-on
computer sessions. There will be a strong emphasis on practical
implementation, with all analyses and exercises conducted using both
Stata and R software. By the conclusion of the workshop, participants
should feel confident in their ability to conduct a dose-response
meta-analysis using their own data.
Content
Although research questions about dose-response mechanisms based on
multiple studies of either aggregated or individual data are
increasingly popular, formal training on this topic is lacking. Aim of
this workshop is to introduce participants to the type questions, data,
statistical models, and visualizations widely used in a variety of
research fields (i.e. nutrition, cancer, cardiovascular,
environmental).
This workshop will cover methods for estimating dose-response
associations from either individual or aggregated patients data,
flexible modelling using splines and polynomials, meta-analysis of
study-specific dose-response functions, hypothesis testing,
quantification and evaluation of heterogeneity, meta-regression models
and sensitivity analyses. Focus will be on the specification of an
appropriate research question, choice of analysis, interpretation of
results, and presentation of dose-response results either in a tabular
or graphical presentation.
Who is this course for?
This course is aimed at researchers and students in biostatistics,
epidemiology and related fields. Some knowledge of basic statistical
concepts (i.e. central limit theorem, test of hypothesis, confidence
intervals) and multiple linear regression models will be assumed.
Participants should have some prior experience of using either R or
Stata, though not necessarily in the context of meta-analysis.
Evaluation
At the end of the workshop, you will be asked to provide some
feedback and evaluate strengths and limitations of the proposed learning
activities.
Schedule structure
Morning
| 09:30 - 10:15 |
Lecture: |
|
| 10:15 - 10:30 |
Break |
|
| 10:30 - 12:00 |
Practical: |
|
Afternoon
| 13:00 - 14:30 |
Lecture: |
|
| 14:30 - 14:45 |
Break |
|
| 14:45 - 16:00 |
Practical: |
|
Acknowledgements
This workshop is supported by the Strategic Research Program
(SFO-Epi) in Biostatistics and Epidemiology at Karolinska
Institutet.