logo
ADVERTISEMENT

NICHOLAS OKUMU: Science should put impact on lives at the forefront, not citations

History repeats this truth. The world remembers not who was most cited, but who was most consequential.

image
by NICHOLAS OKUMU

Columnists14 October 2025 - 09:30
ADVERTISEMENT

In Summary


  • The purpose of science was never to be cited; it was to be useful. To light up dark rooms, to heal, to feed, to build.
  • What endures are the systems built, the minds opened, the lives touched. What endures is science that bends toward service, not self-congratulation.
Vocalize Pre-Player Loader

Audio By Vocalize






The question came quietly, like a scalpel sliding between ribs.

“What is your H-index?”

I had been speaking about systems re-engineered, teams rebuilt and patients who could now walk again. Three companies were founded. Forty livelihoods sustained. A public hospital department was reshaped into a modern, self-running service. The only orthopaedic oncology unit in East Africa was built from the ground up.

Then, in one sentence, it all collapsed into a number.

How many times had other scholars mentioned my name in their footnotes? That, apparently, was the currency of my worth. Not the policies influenced or the mothers and fathers who could now access care, but citations.

In that moment, I realised something unsettling: science has created a false economy.

The H-index was meant as a blunt but helpful tool, a way to gauge productivity and influence. The metric has since become a kind of passport in science—used from Boston to Beijing, Nairobi to New Delhi. Yet its universality is mostly procedural, not philosophical. It works best in citation-heavy fields like physics or biomedicine, and falters in disciplines where knowledge is built through teaching, systems or art rather than journal papers.

In the humanities, where books shape generations, the H-index barely registers. In clinical practice, where a single innovation can save hundreds of lives but yield few citations, it misses the point entirely.

It is everywhere, but it does not mean the same thing everywhere. Somewhere along the way, it became a universal badge of legitimacy. Committees cling to it. Donors nod to it. Universities trade on it.

But like any inflated currency, it distorts the marketplace. It rewards visibility within the ivory tower, not impact beyond it. It tells young scientists that prestige comes not from discovery but from being referenced by the right people in the right journals.

And so the incentives twist: papers over patients, citations over cures, bibliographies over breakthroughs. Ask a mother in a rural clinic how many times her doctor’s name appears in PubMed, and she will look at you blankly. Her currency is survival. Her only metric is whether she and her child make it through the night.

Science does not need to abandon its metrics; it needs to rebalance its portfolio. The H-index measures intellectual echo, but we need a parallel measure for human impact.

An Impact Index, perhaps, one that asks: What systems have you built that endure? How many people have you trained, mentored or inspired? What knowledge have you translated into policy, practice or product? How much access have you expanded for those left behind? What real value has your science created for society?

These are not soft questions. They are the hard accounting of progress, the kind that decides whether knowledge lives in journals or in people’s lives.

Rwanda offers a masterclass in this recalibration. Dr Sabin Nsanzimana, its Minister of Health, says if you want to judge a health system, look at two numbers: maternal and child mortality. Over 25 years, both have plummeted.

In 2024, Rwanda stopped a Marburg outbreak before it could spread. In 2025, TIME named Nsanzimana among its 100 Next leaders, not for his H-index but for outcomes: women who lived through childbirth, children who lived through childhood, futures secured.

History repeats this truth. The world remembers not who was most cited, but who was most consequential.

Albert Einstein received his Nobel not for relativity but for the photoelectric effect, a discovery that powered modern electronics. Katalin Karikó and Drew Weissman were once dismissed for pursuing mRNA.

Their reward came not in early citations but in vaccines that saved millions. Norman Borlaug, no bibliometric star, won the Peace Prize for sparking the Green Revolution, a quiet revolution that ended famines.

Citations follow consequence, but consequence begins with courage.

From my vantage point, half in the clinic and half in the boardroom, I see the same disease in both science and business: valuation error. When markets price visibility over value, bubbles form and burst. When academia prizes citations over solutions, the same happens: reputations inflate, relevance deflates.

The purpose of science was never to be cited; it was to be useful. To light up dark rooms, to heal, to feed, to build. If innovation is the research dividend, then human welfare is the return on investment.

 No one erects statues for citation counts. No one gathers their grandchildren to tell stories about impact factors. What endures are the systems built, the minds opened, the lives touched. What endures is science that bends toward service, not self-congratulation.

If we continue rewarding the wrong currency, science may enrich its elite while bankrupting its purpose. But if we recalibrate, if we begin to prize the builders, the doers, the transformers, then science can reclaim its rightful role: not as a marketplace of citations but as a covenant with humanity.

Because the true index of impact will never be written in footnotes. It will be written in lives.

Surgeon, writer and advocate of healthcare reform and leadership in Africa

Related Articles