SAMJ: South African Medical Journal
On-line version ISSN 2078-5135
STEFAN, D C et al. Childhood cancer incidence in South Africa, 1987 - 2007. SAMJ, S. Afr. med. j. [online]. 2015, vol.105, n.11, pp. 939-947. ISSN 2078-5135. http://dx.doi.org/10.7196/SAMJ.2015.V105I11.9780.
BACKGROUND: Childhood cancer is an emerging problem in Africa. Its extent is hazy because data are scarce, but it should be addressed. This is the first report from the South African Children's Tumour Registry (SACTR), which covers the whole of South Africa (SA). It provides minimal estimates of cancer incidence and discusses the challenges of cancer surveillance and control in a child population in a middle-income country. Only about 2% of the African population is covered by cancer registries producing comparable incidence data. OBJECTIVE: To present and interpret incidence patterns and trends of childhood cancer over a 21-year period. The results should raise awareness of the problem of childhood cancer in an African population and provide sensible data for taking this problem in hand. METHODS: All eligible and validated cancer cases registered in the SACTR over the period 1987 - 2007 and classified according to the International Classification of Childhood Cancer were included. Population data were retrieved from official sources and estimated for the population subcategories. Incidence rates were standardised to the world standard and time trends were evaluated using joinpoint models, adjusting for sex and age. RESULTS: Based on the 11 699 cases, the overall age-standardised average annual incidence rate was 45 per million. Threefold differences in the overall incidence rates were observed between the ethnic groups, ranging from 116 for whites to 37 for black Africans, and they differed by diagnostic group. Differences between the nine provinces of SA relate to the ethnic composition and prevailing socioeconomic status. The overall incidence rate declined by 1.2% per year for the whole country (p<0.01). However, the decline was mainly observed during the first few years of the study period, after which rates stabilised or increased. CONCLUSIONS: Diagnosis and notification of childhood cancer should improve. The differences in incidence between ethnic groups suggest the priorities for cancer control.