Alumnus Thrives at South African Medical Research Council

Alumnus Thrives at South African Medical Research Council

Recent Statistics PhD graduate, Dr Yusentha Balakrishna works as a Specialist Statistician within the Biostatistics Research Unit of the South African Medical Research Council (SAMRC).

Balakrishna’s work involves collaborating with researchers on the design and analysis of various health studies. She has contributed to some 40 publications and reports in the areas of non-communicable diseases, environmental health, nutrition, and various other medical disciplines.

‘No day is the same and there is always a new research area to explore,’ said Balakrishna. ‘Being exposed to different study designs, data structures and study objectives requires me to research and learn about different and new statistical methods.

‘I am also passionate about evidence-based interventions and policies. In this regard, I collaborate on systematic reviews by conducting meta-analyses and am currently an Associate Statistical Editor for the journal BMJ Evidence-Based Medicine.’

During the early stages of the COVID-19 pandemic Balakrishna collaborated on rapid review studies to ensure that the best available evidence regarding interventions was used to inform national health guidelines.

‘Statistics is a key aspect of any quantitative study, so I am also involved in teaching biostatistics courses to health researchers from all sectors,’ she added.

Balakrishna received her PhD in Statistics from UKZN in September 2023. Supervised by Professor Samuel Manda, Professor Henry Mwambi and Dr Averalda van Graan, she focused on the South African Food Composition Database. ‘The South African Food Data System (SAFOODS) is a division within the Biostatistics Research Unit at the SAMRC and I learnt about their vital work and data during my employment,’ she explained. ‘Food composition data is generally used to determine nutritional intake in dietary surveys. However, through our interactions, we realised the potential of combining sound statistical methodology with the data.

‘Food composition data also has a unique data structure with several characteristics such as being compositional, high dimensional, and containing missing data, correlation, extreme values, and skewness. This was an exciting and intriguing statistical challenge and led to the publication of three journal articles thus far with many other aspects to further investigate.’

Balakrishna explained that an understanding of the relationships between nutrients, food items and health informs nutritional guidelines and policies. ‘The South African Food Composition Database contains the nutritional content of different food items consumed by the population,’ she said. ‘My research applied statistical clustering (grouping) techniques to this data to objectively determine the different classes or groups of food items.

‘While the study of statistical methods that are applied to food composition data is reported in international literature, minimal literature from South Africa was found with the focus of the available research tending to be on dietary patterns rather than the nutritional content of the food items themselves. My research set out to fill that gap and to determine our country’s unique food patterns.’

The study found that South Africa’s nutritional landscape can be described by eight nutrient patterns which are comparable to the country’s food-based dietary guidelines. She was also able to identify food items that are generally high in sodium, which correlated with the food items that are targeted by the national salt regulations.

The significance of Balakrishna’s research is that it has shown that by analysing South African food composition data using statistical techniques, current food-based dietary guidelines and the national salt regulations can be supported with data-driven results. This encourages the use of evidence-based nutritional guidelines and policies.

‘My research assists the average consumer to make informed food choices while demonstrating how modern statistical methods, such as the machine-learning techniques we used, can be employed to progress research in nutrition science,’ she said.

Balakrishna has nothing but praise for UKZN, particularly her academic mentor over many years, Mwambi. ‘Choosing UKZN for my PhD degree was a clear decision and I’m humbled to be counted amongst UKZN’s accomplished alumni,’ she said. ‘I also always admire the achievements of my colleagues at the SAMRC and I am extremely grateful to be part of an organisation that inspires excellence in the pursuit of knowledge.’

Words: Sally Frost