Chair and organiser: Prof Ngianga-Bakwin Kandala, PhD
Discussant: To be appointed
Sub-Saharan Africa faces a high disease burden in communicable diseases and an increasing burden in non-communicable diseases with a strong spatial and temporal structure. More recently, increased funding for research from donor initiatives has generated high-quality household data volume, but there is a lack of capacity for advanced data analysis.
Globally, the fields of geographical epidemiology and public health surveillance have benefited from combined advances in hierarchical model building and in geographical information systems. Exploring and characterising a variety of spatial patterns of diseases at the disaggregated fine geographical resolution has become possible (Banerjee et al, 2004).
Insight into the sensitivity of the resulting inference to the choice of the structure of the different components of the hierarchical model has been gained through the use of simulation studies (Best et al, 2005) and numerous case studies worldwide with few in SSA. Baseline results on how to use the posterior distribution of relative risk estimates to detect areas of increased risks have been implemented and tested mostly in western countries.
Extension of hierarchical disease mapping models to models that simultaneously consider space and time leads to a number of benefits in terms of interpretation and potential for detection of localised excesses to inform local policies. Such extension is accompanied by an increase of the complexity of the model structures that might be specified and the rationale for the many choices that have to be made is less clearly documented and tested in complex sampling in household surveys in SSA.
In this session, we present and discuss classes of Bayesian hierarchical space time models that can be used to characterise the patterns of communicable and non-communicable disease burden in SSA (Kandala, 2014). Particular attention will be paid to the influence of the geographic location and time discretisation on the resulting inference and how the space-time consideration of patterns can strengthen the inference. Secondly, we show case applications of Global health issues in SSA using different household data sources from complex survey data such the Demographic and Health Surveys (DHS) and the Multiple Indicators Cluster Surveys (MICs) and use the above models for detecting space-time clusters in a number of scenarios. Finally, we seek papers that use these models to analyse the spatio-temporal variations of HIV, TB, HSV-2, malaria, hypertension, diabetes, malnutrition, and obesity) in selected SSA countries.
Chair and organiser: Prof Bernard Omolo
Chair and organiser: Prof Din Chen
Sponsored by the South Africa DST-NRF-SAMRC SARChI Chair in Biostatistics, this special session is organized to discuss the current development in biostatistical methods and the applications to public health research and evidence-based decision-making. Five experts are invited to present their findings. Specifically, Dr. Arashi will discuss the development of a penalized multivariate t mixed model for multiple longitudinal data analysis and show the application to the Mayo Clinic Primary Biliary Cirrhosis sequential data. And Dr. Burger will discuss the development of a generalized Bayesian nonlinear mixed-effects regression model to analyse the zero-inflated longitudinal count data in tuberculosis clinical trials. Continuing with biostatistical clinical trials, Dr. Ring will discuss the development of the expected power (statistical assurance) for bioequivalence trials and show examples on the use of the assurance concept in comparisons to power calculations. Further to biostatistical big data analytics, Dr. Mwambi will discuss the development of joint spatial disease modelling for the prevalence of HIV and HSV-2 in Kenya and Dr. Ferreira will discuss an investigation of hepatitis B vaccination behaviour using unsupervised- and supervised statistical methods.
Chair and organiser: Prof Bernard Rachet
Time-to-event (‘survival’) analysis provides valuable information on the disease of interest, at population or individual level, depending on the metric chosen. The estimation method varies according to that metric. Furthermore, analyses using population-based data require specific methods to account for the competing risks of death because of the lack of reliable information on the cause of death. Methods developed for relative survival data setting are employed, instead of the more conventional approaches within the cause-specific setting.
Concepts and estimation methods used in relative survival setting will be described and their use and challenges will be illustrated through different applications.
Chair and organiser: Dr. Misrak Gezmu
Abstract: In the last two decades, South-North collaborative biomedical research studies have increased and are being conducted in many sub Saharan African countries. The collaborative research projects require the participation of interdisciplinary research teams both from the South and North. Biostatisticians are critical members of the research teams and play important roles in conducting the research projects. Because of the shortage of biostatisticians from the South, the biostatisticians from the North become project biostatistical leaders. In-country biostatistical leaders are essential part of the local research team. As a project biostatistical leader, they will participate in the thinking process of developing the study design, choosing the analysis methods, assuring that the research questions addressed are those that offer the greatest relevance for the local people and interpreting the study results to help policy makers revise or change health policies. Many collaborative researchers have recognized the shortage of local biostatisticians, but training local biostatisticians is a challenge.
Efforts made towards biostatistics capacity building in the region include conducting workshops, building biostatistics programs, training the trainers and biostatistics courses conducted with South-South and South-North collaborations. In this session, we will hear from those involved in this effort of strengthening the local biostatistics resources. Speakers will discuss success and challenges encountered and funding opportunities used in building biostatistics capacity in the region.
Organisers: Ziv Shkedy and Khangelani Zuma.
One of the main problems in high education at both under graduate and master levels in developing countries is the lack of high quality , R based, materials for courses in education programs. The >eR-Biostat initiative (https://er-biostat.github.io/Courses) is focused on education programs (both undergraduate and master) in Biostatistics/Statistics and for non-statisticians and aim to develop new E-learning system for courses at different education level.
We believe that accessibility to free high quality education materials is crucial to ensure a high standard in education. We offer an “open sources”, R based, education materials in statistics. Everybody (teachers and students) can download and use the courses from free.
The >eR-Biostat initiative introduces a new, R based, learning system, the multi-module learning system, in which the students in the local universities in developing countries are be able to follow courses in different learning format, including e-courses taken online and a combination between e-courses and local lectures given by local staff members. R software and packages are used in all courses as data analysis tool for all examples and illustrations. The >eR-Biostat initiative provides a free, accessible and ready to use tool for capacity building in biostatistics/statistics for local universities in developing countries with current low or near zero capacity in these topics.
The session consists of four presentations:
Please send proposals to email@example.com