Influences of diet, physical activity, body size on breast cancer in South Africa: a study of African women in transition

The findings from this research project will inform whether public health strategies to reduce risk in African women can include advice on body size, diet, energy intake and physical activity levels

  • Topic: Breast cancer
  • Institution: International Agency for Research on Cancer (IARC)
  • Country: France
  • Status: Completed

Researcher: Sabina Rinaldi has replaced Isabelle Romieu who has retired.

Scientific abstract

(View plain language abstract)


Reducing obesity and increasing physical activity (PA) levels are important preventive measures for post-menopausal breast cancer (BC), advice predominantly based on studies in Caucasian women. The effect of obesity and PA on BC risk in African women is less clear. Their breast cancers have a higher proportion of triple-receptor negatives, a tumour subtype for which obesity increases risk at all ages in Caucasian women. There are inconsistent findings from Africa. In 2003, Parkin et al commented: “No studies of diet and breast cancer in African populations appear to have been reported”. The Transition and Health during Urbanisation of South Africa (THUSA) study demonstrated rapid population changes in dietary intake, endocrine hormones, metabolic syndrome and BMI upon urbanisation, and lack of PA leading to obesity that may account for rising breast cancer rates in that country.

Objectives and hypotheses

Our objectives are to assess the association of excess body weight, physical activity and dietary factors with breast cancer in South African women; to investigate whether these associations differ by menopausal status and breast cancer subtype or differ to established associations in Caucasians. We hypothesise that: (i) body fatness and adult weight gain is associated with risk of breast cancer at pre-menopausal ages with differential effects according to receptor status and menopausal status; (ii) waist circumference and waist-hip ratio are stronger than BMI as predictors of breast cancer risk (where one exists); (iii) changes in dietary patterns upon urbanisation, in particular a decreased intake of traditional food and increase in simple carbohydrates and saturated fats, are associated with increased breast cancer risk in black South African women.

Settings and methods

We propose to conduct a case-control study of diet, PA and body size in relation to breast cancer in Johannesburg, South Africa. Breast cancer cases will be recruited from the Chris Hani Barragwanath Hospital in Soweto. The breast clinic in this hospital (led by a co-PI) diagnoses 300 women a year, approximately 90% are black African women. Receptor status is routinely tested and tumour characteristics are recorded. Friends of cases who have never been diagnosed with breast cancer, age-matched and living in the same neighbourhood will be recruited as controls. All women will be interviewed to collect data on breast cancer risk factors, and exposures of interest: diet, PA and body size. PA and dietary assessment methods appropriate for this population have already been developed, validated and successfully applied by co-PIs who will train interviewers. A PA index will be calculated based on a short questionnaire, a modified version of Baecke. DEXA measurements of whole body fat will be performed. A culturally sensitive quantitative food frequency questionnaire using South African food composition tables will be used to assess diet. Detailed anthropometric measures will be taken. We also propose to obtain a blood and spot urine sample from all women (no funds requested for assays), for future research potential.

Potential impact

The findings will inform whether public health strategies to reduce BC risk in African women can include advice on body size, diet, energy intake and physical activity levels. Primary prevention is particularly important in African settings as tumours are diagnosed at late stages.

Plain language abstract


Breast cancer is either the first or second most common cancer in women in Africa, and rates of this disease are increasing in the continent partly due to the decline in family sizes. Survival prospects from breast cancer are worse for most of African women diagnosed with breast cancer than for those in western countries. Very little is known about the risk factors for breast cancer in Africa. Findings from the west suggest that higher levels of physical activity and avoidance of weight gain are protective against breast cancer, but these factors have not yet been conclusively studied in African women in Africa. These precise factors are in transition during urbanisation in Africa and they may have an impact on breast cancer risk.

Aims and goals

We aim to conduct a study of the influence of diet, physical activity and body size on breast cancer risk in black South African women. We will investigate whether associations observed are the same or different to that in Caucasian women, and whether they differ by the type of breast cancer, or by a woman's age at the time of her breast cancer diagnosis.

How it will be done

We will invite all women diagnosed with breast cancer at a single breast cancer clinic to participate in the study. The clinic is situated in Soweto, Johannesburg, South Africa, and the vast majority of its patients are black South African women and are public patients. At the time of diagnosis, women will be asked whether they would be willing to take part in an interview on lifestyle factors, conducted by a nurse within the hospital grounds, provide a blood and urine sample, have body size measurements taken, and give consent to their data and tumour samples to be used for research. Each patient with breast cancer will also be asked to bring a friend of a similar age with her to her next visit. This friend will be asked to undergo the same interviews and measurements. The characteristics of the women with breast cancer will be compared to those of the friends without breast cancer to determine whether dietary factors or body size and physical activity may have played a role in the development of the disease. All participants will be reimbursed for their travel costs to the hospital.

Potential impact

This study will be the first large scale study of dietary and body size influences on breast cancer risk in black South African women. Its findings will help to identify much needed ways to lower breast cancer risk. The findings will be relevant both to black South African and African women in general.