Back

Actions for fostering cross-disciplinary global health research

Ding, Y.; Tomeny, E.; Bates, I.; the IMPALA Consortium,

2021-10-01 public and global health
10.1101/2021.09.30.21264294 medRxiv
Show abstract

IntroductionThe ability to conduct cross-disciplinary research in international collaborations is critical for improving global health. Published evidence on actions to foster cross-disciplinary research comes mainly from high-income countries with its applicability to global health unclear. Our study investigated the actions taken to foster cross-disciplinary research across a North-South global health programme to recommend actions to improve the effectiveness of future cross-disciplinary research programmes in global health. MethodsWe used an adapted three-component framework to frame data collection and analyses. Building on a recent literature review, we compared actions used by the programme for planning, implementing and managing cross-disciplinary research to those described in the literature. Data sources included interviews with 31 participants (including researchers, administrators, and collaborators), a review of programme documents, a baseline survey, and observations of interactions at meetings and events. Interview data were coded and analysed using the framework. Narrative summaries were created using thematic synthesis and triangulated by the document review and observations. ResultsFor cross-disciplinary research to be successful in global health, a shared vision with explicit goals is essential. These goals concern knowledge integration supported by regular communication, orientation about involved disciplines, and equitable partnerships across institutions and individuals. Fostering cross-disciplinary global health research needs a significant time investment from researchers, support staff, and programme leaders. Indicators for tracking cross-disciplinary working should be agreed at the outset and monitored throughout. ConclusionCross-disciplinary activities should be managed separately from primary research activities. The three-component framework would be helpful in guiding designing cross-disciplinary programmes. Summary BoxO_ST_ABSWhat is already known?C_ST_ABSO_LIGlobal health can be advanced by cross-disciplinary collaboration within and beyond the health sciences. C_LIO_LIIndividual researchers, research team leaders, academic institutions and research funders all have roles in making cross-disciplinary research more effective. C_LIO_LIPublished evidence on actions to foster cross-disciplinary research comes mainly from high-income countries and its applicability to global health is unclear. C_LI What are the new findings?O_LIThe cross-disciplinary aspects of research programmes need to be actively managed. C_LIO_LIPre-agreed indicators should be used to plan and track cross-disciplinary research. C_LIO_LIFostering cross-disciplinary research and managing tensions takes time and explicit continuous discussions. C_LIO_LIOur adapted three-component framework (i.e., planning, implementation and management) is useful for collecting and analysing multi-source and multi-perspective data on a cross-disciplinary programme in real-time. C_LI What do the new findings imply?O_LICross-disciplinary activities should be managed and tracked separately from the primary research activities. C_LIO_LIProgress in research planning, implementation, and management should be reviewed against pre-agreed indicators, with troubleshooting conducted accordingly. C_LIO_LICross-disciplinary research requires the allocation of more time and funds for active management of research planning than would be required for mono-disciplinary projects. C_LIO_LIThe three-component framework would be helpful in guiding the design of cross-disciplinary programmes. C_LI

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.