Research integrity in low- and middle-income countries: Systematic review of prevalence of poor authorship practice, plagiarism and other misconduct




Poster session 1 Wednesday: Evidence production and synthesis


Wednesday 13 September 2017 - 12:30 to 14:00


All authors in correct order:

Rohwer A1, Young T1, Wager E2, Garner P3
1 Centre for Evidence-based Health Care, Faculty of Medicine and Health Sciences, Stellenbosch University, South Africa
2 Sideview, United Kingdom
3 Liverpool School of Tropical Medicine, United Kingdom
Presenting author and contact person

Presenting author:

Anke Rohwer

Contact person:

Abstract text
Background: Good reporting practices are important to preserve research integrity. There are few published studies on research integrity in low- and middle-income countries (LMICs). Taking stock of these existing studies from LMICs is important to inform future research and promote best practices.

Objectives: To identify and summarise empirical studies about research integrity in LMICs in relation to poor authorship practice, plagiarism and other misconduct.

Methods: We included quantitative and qualitative studies on research-reporting practices (e.g. authorship, plagiarism, conflicts of interest, data fabrication) amongst health researchers in LMICs. We searched electronic databases, conference abstracts and contacted experts in the field to identify relevant studies. Study selection, data extraction and quality assessment was done by one author and checked by another. We contacted authors in cases of missing data. We narratively synthesised results.

Preliminary results: We screened titles and abstracts of 6003 citations. Of 113 full-texts, we included 33 studies comprising 20 cross-sectional surveys of healthcare researchers, 10 cross-sectional surveys of journals, 2 qualitative studies and 1 case study. Included studies were conducted in Asia (n=11), Middle East (n=10), sub-Saharan Africa (n=4), Latin America (n=6), and across more than one region (n=2). Most studies were judged as having moderate to high risk of bias, had a small sample size and were poorly reported. Across studies, self-reported prevalence of guest authorship varied from 8.3-65% (n=9); ghost authorship from 6.4-43% (n=4); plagiarism from 0-73%; data fabrication from 0-14.4% (n=3); and, data falsification from 0-27% (n=3). Factors influencing misconduct included lack of knowledge and experience, institutional shortcomings such as lack of consequences, pressure from funders and need for recognition.

Conclusions: Studies from LMICs reported high rates of research misconduct. Studies were generally small and based on survey participants’ observations. Studies conducted across LMICs were limited to cross-sectional surveys of journals.