Production and use of post–graduate students’ research: a multiple methods study of universities in Uganda and other low- and middle-income countries

Abstract

Background: Sir Ian Chalmers, a leading health scientist noted, “…For ethical, economic and scientific reasons, health-relevant degree theses must be made publicly accessible, […] otherwise it is a waste…”. Post–graduate students’ research in institutions of higher learning involves considerable investment in time, effort, and money. However, the developments from these post–graduate students’ research projects, for example, the volume of publications and whether or not they are used to inform decisions of practice or policy in the field of health was unclear. The thesis attempted to address this gap in four papers or sub-studies focusing on Uganda and other low- and middle-income countries. Specific objectives: 1. Sub-study 1: To review and summarize existing evidence on the effectiveness of approaches that increase research productivity (proportion published) or the use (proportion cited) of post– graduate students’ research; and assess the determinants of post–graduate students’ research productivity and use. 2. Sub-study 2: To quantify and characterize Masters students’ research and their outputs over a 15-year period (from 1996 to 2010) at Makerere University College of Health Sciences, Uganda. 3. Sub-study 3: To determine the extent to which research generated by Masters students at Makerere University contributes to decision–making or policy formulation in health in Uganda and internationally. 4. Sub-study 4: To identify and evaluate current support mechanisms for research generation and application for Masters students in four universities in Uganda. Page 22 of 164 Methods: A multiple methods approach was used. Sub–study 1 was a systematic review of studies about how post–graduate students furthered their research project work or theses in low– and middle– income countries. These studies investigated whether the post–graduate students published their work or presented in conferences or were used to inform technical reports or policy related documents. This review was guided by a protocol registered in PROSPERO (CRD42016042819). Briefly, published articles in PubMed/MEDLINE and the ERIC databases were searched for through to July 2017. Duplicate assessments for included primary studies was performed and discrepancies resolved by consensus. Outcomes of interest were research productivity (proportion published) or the use (proportion cited). A meta-analysis of the findings was done using both fixed and random effects approaches. Sub–study 2 was a retrospective review of post–graduate students’ research records enrolled for masters’ degree courses from 1996 to 2010 at Makerere University College of Health Sciences, and followed to 2016. The outcome measures were publications (primary) and citations, electronic dissertations found online or conference abstracts (secondary). Descriptive and multivariable logistic regression analyses were performed to identify determinants of the primary outcome. Sub–study 3 was a multiple case study of ‘research use’ from masters’ students’ dissertations at Makerere University College of Health Sciences completed between 1996 and 2010. A case of ‘research use’ was defined as citation of research products from post–graduate students’ dissertations in publicly available health policy–related documents. Sub–study 4 was a self–assessment cross–sectional survey by masters students in four schools of public health and medicine in Uganda, namely: Makerere, Mbarara, Mukono and Nkozi Universities. The four domains assessed for were: research question identification (priority Page 23 of 164 setting), proposal development for the post-graduate research project, knowledge transfer of postgraduate research products and promoting use of post-graduate research products Results: In sub–study 1, the proportion of manuscripts published from post–graduate research projects in the 12 studies included from 8 low and middle income countries was 7% (95% CI: 7% – 8%; Higgins I–squared and Cochran’s Q, p value of < 0.01) and 23% (95% CI: 17% – 29%; Higgins I–squared of 98.4% and Cochran’s Q, p value of < 0.01) using fixed effects and random effects models respectively. Hardly any studies reported on the citation of post-graduate students’ studies, or the determinants of productivity or use of post-graduate students’ research. No study was found that reported effectiveness of interventions to increase publication or use of post–graduate students research. In sub–study 2, there were 1,172 dissertations were found with Masters of Public Health being the single most popular course, enrolling 363 students (31%). Manuscripts from 209 masters’ dissertations at Makerere University were published (18%, 95% CI: 16%–20%) and 196 were cited (17%, 95% CI: 15%–19%). Few (4%), policy–related documents (technical reports and guidelines) cited these dissertations. In terms of research priorities, the bulk of the research was about MDG 6, predominantly on infectious diseases (42%), and the health services delivery pillar (66%) was the most common World Health Organization health system pillar researched. Three quarters (75%) of the post-graduate research projects were cross-sectional studies whilst a fifth (21%) used qualitative case-study methodology. Variables that significantly affected publication proportion in the multivariable model were students’ age at enrolment into the masters programme, with adjusted coefficient -0.12 (95% CI: -0.18 – -0.06),