Malawi Short Course repost
Background
In many public health studies, researchers use randomised controlled trials (RCTs) to estimate the causal effects of exposures on health outcomes. In RCT studies, participants are randomly assigned to an intervention or a control group, thereby balancing measured and unmeasured factors across the groups. By employing randomisation, RCTs have become the gold standard for undertaking causal inference in health research. For example, vaccine trials use RCTs to determine whether the new vaccine affects outcomes. However, conducting RCTs can be expensive, unethical, or infeasible. So, instead, researchers use observational studies to estimate the causal associations. However, causal inference from observational studies is challenging because of confounding (where other factors influence the results) and selection bias, arising from non-random study participants.
This short course introduces the fundamental concepts and practical tools of causal inference, equipping participants with the skills to assess causal association and apply basic causal methods to observational data.
Course Aim
To provide participants with a foundational understanding of causal inference principles and commonly used analytical methods for observational studies.
Learning Objectives
By the end of the course, participants will be able to:
- Distinguish between causal inference and association-based analysis
- Understand key causal concepts such as counterfactuals, confounding, and bias
- Use causal diagrams (DAGs) to clarify assumptions
- Apply basic causal methods for observational data analysis
Target Audience and Requirements
This workshop is designed for public health researchers involved in the design, analysis, and interpretation of results from observational studies. Participants are expected to have basic statistical skills, descriptive statistics, linear and logistic regression, confidence intervals, and hypothesis testing; basic epidemiology, including exposure, outcome, confounding, and bias; and observational study designs such as cohort, case–control, and cross-sectional studies. Some hands-on experience with statistical software (e.g., R, Stata) is required. However, no prior formal training in causal inference, potential outcomes, or directed acyclic graphs (DAGs) is needed, as these will be introduced from first principles.
Logistics
To register, please fill in the Google form
https://docs.google.com/forms/d/e/1FAIpQLSfxshxnt_1cgc8kuosv1SmoUdMyM9WnIQokhbkQ2tknmfcT0Q/viewform?usp=publish-editor.
For enquiries, please contact Dr Halima Twabi at htwabi@unima.ac.mw.
Fee: [Free for participants, limited spaces available]
Workshop Facilitators:
Samuel Manda is a Professor of Statistics at the University of Pretoria. He has conducted extensive and highly impactful research on Bayesian methodology and its applications in modern biostatistics within the health sciences. He holds a PhD in Bayesian Statistics from the University of Waikato, where he investigated advanced methods for analysing correlated survival data. He subsequently undertook postdoctoral research at the University of Auckland, focusing on nonparametric and multilevel Bayesian survival models. His current research includes developing and applying state-of-the-art survival models, causal inference methods and spatial statistics. Professor Manda has authored and contributed to nearly 200 scholarly publications in mainstream statistical, applied and public health journals. He has served as Principal Investigator and statistical consultant for major international projects. He is also involved in mentorship, capacity building, and advancing the field of statistical sciences. Prof. Manda serves as Principal Investigator for the United Nations Global Index of Volunteer Engagement (GIVE). He is also an Associate Editor for the journals Biometrics, Frontiers in Public Health and the South African Statistical Journal.
Halima S. Twabi is an Associate Professor of Statistics at the University of Malawi. She holds a PhD in Biostatistics from the University of Malawi, where she investigated causal inference methods for longitudinal studies with multivariate outcomes. She is a biostatistician with research interests in causal inference for observational studies, complex longitudinal analysis, and multivariate statistics. Her application research applies causal methods to routine health data to support evidence-based public health decision-making in Sub-Saharan Africa. She has published her work in peer-reviewed statistical, epidemiological, and public health journals and has contributed to multidisciplinary research projects at national and international levels. She is actively involved in postgraduate supervision, mentorship, and capacity building in biostatistics and public health research.