NAS Featured scientist: Prof Samuel Manda
Extraordinary Professor in the Department of Statistics and Director of the Biostatistics Unit at the South African Medical Research Council

Q: Why did you choose to study statistics/biostatistics?
A: I think you can call it, being at the right place at the right time. Studying statistics offered me a noteworthy academic and scientific career in Biostatistics (a subdiscipline of statistics (the science of obtaining an understanding from data in the presence of variability and uncertainty).

Biostatistics is a branch of statistical science, which uses quantitative methods to develop and apply techniques and methods to provide solutions to problems that affect people's health and quality of life, either individually or as part of a group of people in a population. Biostatisticians collaborate with research scientists such as public health, bioinformatics and computational biology; computer science and mathematics.

Q: Why is science, (including biostatistics) important?
A: Biostatistics plays an important role in ensuring that empirical evidence from research in biomedical sciences is scientifically and ethically relevant. For example, in the assessment of efficacy and safety of therapeutic (treatment) interventions, biostatistics is used to determine whether the therapy is beneficial to a patient regarding the desired result (efficacy) and adverse effects (safety). In the conduct of new clinical research, such as drug and vaccine development (e.g. COVID-19 vaccines), biostatistics plays a key role in all the steps of the drug development process, which ensures that the submission to national medical boards is of high quality, thus shortening on timelines and reducing coast. In health care, biostatistical design and analysis provide results that enhance knowledge for improving the efficiency of health care delivery. Biostatistics research can also contribute to the creation of effective patient care plans as individualised treatment of cancerous tumours, whole-genome sequencing to predict a patient’s likelihood of healthy HIV treatment outcome.

Q: Highlights of your career so far, including at UP?
A: Soon after my postdoctoral fellowship in the Department of Statistics at Auckland University (New Zealand), I took up an academic position in the newly created Biostatistics Research Unit (under the Directorship of Prof Mark Gilthorpe) at the University of Leeds. I spent just over five years here, and it was the most formative and productive period of my career. Recently, in 2019, the Biostatistics Unit (SAMRC) successfully organised and hosted the first Joint Conference of the Sub-Saharan Network of the International Biometrics Society and DELTAS Africa Sub-Saharan Africa Consortium for Advanced Biostatistics.

Q: Give us a glimpse of your most recent research
A: Currently, I have been working with other biostatisticians and data scientists from the Biostatistics in offering biostatistical input for COVID-19 related research projects. We have already supported the South Africa Medical Research Council and Department of Health in the analysis of COVID 19 mortality data in South Africa, which was published in the South African Medical Journal. I am also involved in the spatial analyses of COVID-19 in South Africa, in collaboration with colleagues in the SAMRC, UP and CSIR, and Africa in collaboration with the University of Limpopo.

The emerging era of big data requires Biostatisticians with an advanced comprehension of fundamental statistical concepts and methods, computational tools to handle dimensionality reduction, and machine learning. I am currently supervising six (6) PhD students in different biostatistics research areas, from causal inference for multiple outcomes; multilevel discrete time-to-event models in big data space; to precision medicine in clinical trials
 

Q: How can the field of Statistics/Biostatistics contribute to cancer research?
A: Most of the developments in Biostatistics have been in cancer research. At individual patient-level analysis, biostatistics is used to estimate the relative risk of cancers, say lung cancer, between smokers and non-smokers adjusting for other known prognostic factors in epidemiological studies. In clinical trials comparing the safety and efficacy of cancer treatments, biostatistical methods are incorporated in every step including trial design, protocol development, data management and monitoring, data analysis and clinical trial reporting. Working with cancer radiologists and scientists, biostatistical techniques could be used to compare the diagnostic accuracy of magnetic resonance imaging (MRI), mammography or ultrasound-guided biopsy for prostate or breast cancer detection accounting for rates of access, recommendation, and acceptance of each diagnosis methods.

At the population level, biostatistics methods are used to assess whether morbidity and mortality levels of individual, or groups of, cancers tend to cluster in some areas of region or country; or have discernible temporal trends. These biostatistics analyses are useful in indicating where the policy to improve cancer morbidity and mortality might be best targeted or where further investigation to understand the causes of the poor cancer outcomes might be best carried out. Temporal trends could also be indicative of changes in cancer detection and testing.

Q: Describe a day in the life of Prof Manda?
A: A lot of emails to attend to, every morning. Managerial duties to attend to; these also include mentorship and supporting staff in their research work. The majority of my work revolves around my research, supporting research of the other biomedical scientists and postgraduate supervision, and research meetings with local and international institutions. And there are papers from all the research activities including my students' work to be published in peer-reviewed scientific journals. Journals and funding bodies also require that I review other scientists’ research papers and proposals, and universities would ask me to examine their postgraduate dissertations and theses.

Q: What are the (qualities) of a good scientist?
A: Flexible thinking, with an indirect and creative approach to concluding problems. A scientist should provide room for scientific improvisation and adaption of well-tested methods to new research need, even improving on them.

Q: Advice to young scientists…
A: I think young scientists in today’s world have plenty of opportunities to hold their hands on; so many scientific innovations; the internet, fast-changing technology in commerce, transport, health (e.g genetic profiling and computational biology, health data capturing technology) and education. As a young scientist, you can find your niche area, outside of traditional scientific research and careers, and make it work for you; of course, you need to have good mentors and networks with both young and scientists of stature. Please do not carry and bring an attitude and negativity into your science, it will kill your scientific career.

Q: What words/beliefs do you live by?
A: I take open and constructive criticism of one's research paper or conduct as very enriching and enabling of scientific interaction. I am one of those that believe in doing scientific research honestly and ethically sound.