Health research publications by South African authors from 1996 to 2015: a bibliometric analysis

Introduction: research publications have a vital role in the scientific process, providing a strategic connection between the generation of new knowledge and its conversion into policy, practice, and positive health outcomes. There was a substantial increase in research funding in South Africa from the dawn of multi-party democracy in the mid-1990s to 2015. However, it is not known whether there was a corresponding increase in research publications from the country. Therefore, the objective of this bibliometric study was to assess trends and factors associated with health research publications from South Africa between 1996 and 2015. Methods: in July 2016, we searched Scopus for health science articles published between 01 January 1996 and 31 December 2015 with at least one author affiliated to an institution based in South Africa. We sought annual data on national-level indicators from Statistics South Africa and World Bank data. We used Poisson regression to examine trends in publication outputs and negative binomial regression to explore national-level factors associated with a change in the number of publications over time. Results: we identified 51,133 publications, with a mean of 2,557 publications per year. Four universities (University of Cape Town, University of the Witwatersrand, Stellenbosch University, and the University of Pretoria) contributed more than half of the publications. The top destination journals were the South African Medical Journal (14.57% of the articles), PLoS ONE (5.77%), South African Family Practice (4.68%), Journal of the South African Veterinary Association (2.48%), and The Lancet (2.37%). The annual number of publications increased five-fold from 1133 in 1996, with an upsurge after 2003, to 5820 in 2015. The average annual percentage growth in the number of publications rose from 3.31% in 1996-2000 to 13.63% in 2011-2015. Year of publication (incidence rate ratio 1.16, 95% confidence interval 1.14 to 1.18) and annual private expenditure on health (incidence rate ratio 1.08, 95% confidence interval 1.05 to 1.10) were independent predictors of publication output. Conclusion: the number of health research publications from South Africa grew substantially between 1996 and 2015, with wide variation in output among universities. Private expenditure on health may be a proxy of health research funding, which probably explains its association with publication output in this study.