In 2019, a team led by Bethany Hedt-Gauthier published a review in BMJ Global Health with a title that says most of it: “Stuck in the middle.” They looked at more than 7,000 health research papers about sub-Saharan Africa published between 2014 and 2016. Overall, just over half of first authors were from the country the paper was about. When the collaboration included a top US university, that figure fell to 23%. And 13.5% of all the papers had no local co-author at all.
Those are not numbers about authorship alone. They are a picture of how careers in global health are built, and of who does the building.
Many collaborations like these describe themselves as including mentorship. Senior researchers from wealthy institutions train junior colleagues in the countries where the work happens. Methods are taught, grant writing is coached, papers are reviewed. By the standards of most mentorship programmes, that is a success.
And yet the junior colleagues stay in the middle of the author list. Which suggests the problem is not a shortage of mentoring. It is that mentoring, on its own, is the wrong tool.
Advice is cheap. Access is not.
Mentorship, as usually practised, is about development: feedback, encouragement, skills, perspective. It is valuable, and most people who have built a career in global health can name someone who gave them good advice at the right time.
Sponsorship is different. A sponsor uses their own position to open doors: naming you as principal investigator on a grant, putting you forward for a panel, handing you first authorship on the paper that will define your field, stepping back so you can lead the call with the funder. Sponsorship costs the sponsor something. Mentorship usually does not.
That difference explains a lot. A senior researcher at a Northern university can mentor dozens of early-career colleagues without giving up anything that matters to their own career. The grants still come through their institution. The high-profile papers still carry their name in the senior position. The collaborators become more skilled, but the structure stays the same.
The incentives run the wrong way
It would be easy to blame individuals, and some deserve it. But the more honest explanation is structural.
Academic promotion in high-income institutions rewards first and last authorships, grant income as principal investigator and international visibility. Every one of those is a finite resource in a collaboration. Handing them to a partner in Kampala or Lusaka is, from the point of view of a promotion committee, a loss.
Funders reinforce this. Many large grants still require a lead institution with the financial systems, audit history and indirect-cost arrangements that Northern universities have and many Southern institutions are still building. The money flows through the North, and control tends to follow the money.
So well-meaning senior researchers are caught. They can mentor generously and still be rewarded for keeping the prizes. Asking them to sponsor instead means asking them to act against the incentives of their own employer.
Follow the overheads
There is one mechanism that deserves more attention than it gets: indirect costs. When a grant is awarded to a university, a slice of it goes to the institution to cover buildings, finance staff, compliance and the rest of the research machinery. For a Northern university, overheads from international grants help pay for its own research infrastructure. When that grant passes through the Northern institution and a sub-award goes to a Southern partner, the partner often receives a smaller overhead rate, or none, on its share.
The consequence is circular. The Southern institution cannot build strong grants offices and finance systems because it does not get the overhead money that pays for them. Because it lacks those systems, funders judge it too risky to hold the prime award. So the grant goes North again, and the overheads with it. No individual has to behave badly for this loop to persist. It runs on administrative defaults. And because those defaults sit in contracts and finance offices rather than in anyone’s research plan, they rarely come up when collaborators sit down to talk about equity. The conversation happens about authorship, which is visible, while the money that decides who can lead next time moves quietly in the background.
That is why the institutional point matters more than it sounds. Sponsorship of individuals will hit a ceiling if their institutions are structurally unable to hold the money.
What handing over actually looks like
Imagine a successful five-year collaboration on maternal health in a single district. The Northern principal investigator held the grant; a Southern co-investigator ran the fieldwork, managed the team and knows the data better than anyone. The renewal is due.
The mentoring version of the renewal keeps the same structure. The Northern investigator stays as principal investigator, the Southern colleague is promoted to co-principal investigator, and the application talks warmly about capacity strengthening. Everyone’s career moves a little.
The sponsoring version is different. The Southern colleague is named principal investigator, with the grant held by her institution if the funder allows it, and the Northern investigator becomes a co-investigator who provides specific methods support. The Northern partner writes in the application that this was the plan from the start. The risk is real: the application may score lower with a less familiar name at the top, and the Northern investigator loses a line on the CV that their promotion committee will count. That cost is the whole point. If there is no cost, it probably is not sponsorship.
What the data hints at
The same BMJ Global Health review contained a quieter finding that deserves more attention. Local representation was highest when the collaborators came from another African country, not from Europe or North America.
That does not mean Northern partners have nothing to offer. It suggests that collaborations between peers with less power difference produce more shared credit. It is an argument for investing in South-South mentoring networks, regional research consortia and senior researchers in low- and middle-income countries who can sponsor the next generation from a position of equal standing.
The data has limits worth stating. Authorship is a proxy for leadership, not a direct measure of it, and the review covers papers from a decade ago. Some things have shifted since, with more funders asking questions about equitable partnership. But anyone who sits on grant panels or journal boards can check for themselves whether the pattern has broken. Most will find it has bent, not broken.
The honest counterargument
People who resist fast handovers are not always protecting turf. Funders are accountable for money, and some institutions really do lack the systems to manage large grants, report on time or pass an audit. A failed grant, especially a first one, can do more damage to a young researcher’s career than a slower route. Pushing someone into a principal investigator role before their institution can support it is not generosity; it can be a setup for failure.
That argument partly holds, and it points to sequencing rather than refusal. Grants can be designed to transfer leadership in stages, with the prime award moving to the Southern institution partway through once agreed financial milestones are met. Northern partners can second finance staff rather than hold the money. The test is whether the plan has a date on it. Capacity-building with no end point is often simply a way of never handing over.
There are second-order effects to think about too. If funders suddenly require Southern leadership, the few researchers already established will be overwhelmed with requests to be named as lead. Institutions may compete for the same small pool of senior people, pulling them out of teaching and mentoring the next cohort. Building the pipeline takes longer than changing the rules, and rules that outrun it will produce figurehead leadership instead of the real thing.
What better models look like
If the goal is careers rather than capacity-building reports, a few changes matter more than any mentoring curriculum:
- Make sponsorship explicit. Collaboration agreements should state, at the start, who will be principal investigator on follow-on grants and who will hold first and last authorship on the main outputs.
- Fund the institution, not only the individual. A brilliant researcher cannot lead grants if their university cannot manage them. Money for finance offices and research management is less photogenic than fellowships, and often more important.
- Change what gets counted. Northern institutions could credit senior researchers for the independent grants their former trainees go on to win, not just for their own outputs.
- Build peer networks. Regional cohorts of early-career researchers, mentoring one another across countries, create relationships that outlast any single project.
Who holds the levers
Funders hold the biggest one. They decide who can be the prime recipient, what overhead rate applies, and what gets scored. A funder that asks every applicant to state who will lead the follow-on grant, and then checks five years later, changes behaviour faster than any guidance document. Funders can also pay directly for grants management capacity in Southern institutions, which is unfashionable spending that unlocks everything else.
Journals hold a smaller but sharper lever. Several already ask about the contribution of local researchers to work done in their countries. Asking is cheap. Declining to publish papers with no local authors, where local researchers clearly could have been involved, is the version that bites.
Heads of Northern departments control promotion criteria. If a senior researcher’s file can show that three former collaborators now hold their own major grants, that should count for as much as another last-author paper. Until it does, the incentive to keep the prize will beat the intention to share it.
And senior researchers in low- and middle-income countries are sponsors too. The pattern of hoarding opportunity is not unique to the North; it appears wherever seniority and scarcity meet. The South-South finding in the BMJ Global Health review is encouraging precisely because it suggests that regional leaders can open doors for the next generation from a position of shared context. That work deserves recognition and funding in its own right.
What to watch
The practical signals are measurable. Is the share of grants awarded directly to institutions in low- and middle-income countries rising, and do those grants carry full overhead rates? Are journals asking about local authorship and contribution, and do they act on the answers? Do Northern universities’ promotion criteria start to credit sponsorship? These are unglamorous indicators, but they will say more about whether global health leadership is shifting than any number of partnership statements.
The uncomfortable question for senior people
None of this requires anyone to stop mentoring. It requires senior researchers to ask a harder question of themselves: what have I actually given up so that someone else could lead?
If the answer is nothing, it is worth being honest about what the mentoring relationship is. It may be generous, and useful, and sincerely meant. It is not yet changing who holds the pen.
The next decade of global health leadership will be decided less by who received the best advice than by who was handed the first real opportunity to lead, and by whom.
Who sponsored your career, not just advised it, and what did it cost them? We would like to hear your story in the comments.
