Last week, I lined up the men’s World Cup teams by health spending and life expectancy. As discussed, averages mask significant regional variation within each country. For example, rural communities often report higher mortality, and it is not uncommon for data from places outside the big cities to be missing altogether, excluding them from countries’ overall figures.

This week I want to use the last two World Cup finals as waypoints in an appraisal of infant mortality.

Infant mortality is like a barium meal on a country’s health system. Life expectancy compresses what happens across the life course into one average, but infant mortality (obviously) is set in the first twelve months, most of it in the first twenty-eight days. It is heavily dependent on how well the mother was fed and monitored, whether the birth was attended by a skilled health worker and whether any subsequent infection or deterioration was caught in time.

Share

Children of the Qatar final

Argentina beat France in Lusail on 18 December 2022, the World Cup having to be played in the winter.

A child born that week is three now; its first year is well over, and the risk it carried can now be evaluated. If our figurative World Cup final baby was born in Japan, it had a very good chance of still being with us, with approximately one death in every 550 live births.

Were it born in the United States, the figure was around one in 180; if it had been born in Cameroon, roughly one in 24. Basically, one for every player on the football pitch and both managers.

A Cameroonian infant was more than twenty times more likely to die in its first year than a Japanese one.

The US number is the strange one on that list, as a baby born in the United States that week arrived during the first twelve-month period in two decades in which American infant mortality rose: 5.6 deaths per 1,000, or 20,538 babies, up three per cent on the year before. The rate did not continue rising in the following two years, so this was not a sustained reversal, but it was still a warning from the richest health system among the 32 teams that competed in the Qatar men’s World Cup.

The first month

Most infant deaths occur in the first twenty-eight days, and neonatal deaths now account for 47 per cent of all deaths under five, up from 40 per cent in 1990. Diarrhoea, malaria and pneumonia have been beaten back over decades, while newborn deaths have declined more slowly and therefore make up a growing proportion of the overall figure.

Newborn deaths are caused mainly by prematurity, complications during labour and birth, infection and congenital conditions. The risks are shaped by whether the birth was attended by a skilled health worker and whether the woman and baby had access to effective antenatal, obstetric and postnatal care.

Midwife-led continuity of care improves women’s experiences and reduces some interventions during birth, although the latest systematic review is less certain than earlier evidence about its effects on preterm birth, fetal loss and neonatal death. One of the most effective parts of neonatal care remains a midwife who is present before, during and after birth, backed by functioning referral and specialist services when they are needed.

Subscribe now

Between now and the next final

Last week I built a table from 2023 spending on health per capita: the United States at $13,473 a head, the Democratic Republic of Congo at $23 and Morocco at $232. By the time of the next final, in 2030, that table will look different, and probably not for a reason anyone will want to be particularly associated with.

In 2025, development assistance for health was estimated to have fallen by 26.9 per cent against the year before, much of it following the dismantling of USAID in the United States, with Britain, Germany and France cutting alongside it.

Child deaths under five had fallen in most years since 2000. Using a separate modelling framework, the Institute for Health Metrics and Evaluation projected that they would rise from 4.6 million in 2024 to 4.8 million in 2025. If confirmed, this would be the first increase this century.

Modelling in the same report estimates what happens if the money never returns. A permanent 20 per cent cut to foreign aid would mean approximately 12.5 million additional child deaths by 2045. A permanent 30 per cent cut would mean around 16.3 million additional deaths.

Of course, it is the countries clustered just above the DRC’s $23 health spend per person per year that are now seeing sharply reduced aid spending. It is hard to see how they escape worsening outcomes.

I do not know which countries will do better or worse in the next four years, but I do know that a baby born during the week of the 2030 final will spend its first month in a global health system carrying considerably less of that funding than the one that provided the data for this article. The number of deaths among children under five, which had been falling for a quarter of a century, is now flattening and, under some projections, going the other way.

Full time

Spain beat Argentina in New Jersey on 19 July, one goal in extra time. It was a bad game, and worse for the comparison: the worst final of the tournament, coming three and a half years after arguably the best one. We saw a worsening not only of the overall game of football but also of the sportsmanship, notably from Argentina.

Their victory in 2022 showed this great nation can do better than this. Over the same period, progress on child mortality has slowed sharply and, under some projections, may reverse.

Infants born in 2022 entered a world in which progress on child survival was very uneven but still moving in the right direction. For those born last Sunday, that direction is considerably less secure.

Leave a comment

I spent a fair portion of my career treating fevers, so it seems reasonable to spend the next couple of weeks on the febrile illness caused by a four-yearly event: the men’s football World Cup.

Sports medicine is the only healthcare on display at a World Cup, a medical team sitting in the dugout or running onto the pitch when a player stays down, It is the best-resourced care those players will ever receive, and arguably the least important in any of their lives.

Every player at this tournament got there through decades of mundane healthcare: the midwife at the birth, the childhood immunisations, the nurse who cleaned and dressed the ten-year-old’s cut knee, the physiotherapy after the first serious injury at seventeen.

Having made a career looking at health systems and nursing, I wanted to see how those systems would fare in a tournament of their own. Forty-eight countries have just been put on a level(ish) pitch. How do they match up in the healthcare World Cup?

There are a thousand ways to run this exercise. I am going to start with two numbers: health spending per person and life expectancy.

_{World Bank data 2023}

_{Please note Curaco didn’t submit data and England and Scotland Data presented as UK}

Subscribe now

Line up the 48 teams by what their countries spend on health per person and you get a very different set of winners and losers. The United States sits at the top: $13,473 per person in 2023. The Democratic Republic of Congo sits at the bottom: $23 for a year of healthcare per person.

That is a 586-fold difference.

Two caveats: These are current US dollars, unadjusted for purchasing power. A dollar buys more nursing time in Kinshasa than in Boston, so the real gap is smaller but not 586 times smaller.

They are also national averages, which is its own distortion. The Congolese figure averages Kinshasa with the rural provinces, and averages in countries like the DRC conceal enormous variation, almost always in the same direction.

Money doesn’t automatically buy years.

Money is not the only thing that matters, and the teams at this year’s tournament prove it.

Female life expectancy in the United States is 81.4 years. South Korea, spending $3,137 per person, under a quarter of the American figure, reaches 86.6. Switzerland spends $11,784 and reaches 86.2. Korea buys five more years of female life than the United States at a fraction of the cost; Switzerland spends nearly everything the Americans do to arrive at roughly the Korean result. Spending is associated with outcomes, but it is not as simple as more money gets you more years on earth.

Share

Why not? Because what the money buys is a system, and we make choices as to who is covered, where the staff go, what gets done early cheaply in primary care rather than expensively late in tertiary hospitals.

Morocco has proved really effective both on the pitch and off it. Women in Morocco live to 77.8, within four years of the United States, on a fiftieth of the spend. Algeria gets its men to 75.1 on almost the same budget.

Demographers worked out the shape of this fifty years ago: if you plot life expectancy against national income, you get a curve that climbs steeply, then flattens

The first hundred dollars per person buys immunisation, an educated health worker at birth, and antibiotics for a child’s pneumonia. The last five thousand buys weeks, mostly at the end.

This is why the DRC’s $23 is a catastrophe and much of the USA’s $13,473 is inefficient and wasteful.

Congo healthcare means every dollar withheld (either through government spending or aid cuts) costs years. The United States is not just far along the curve, it’s falling below it. Some of the cause is price: an American hospital pays more than a Swiss one for the same scan, the same ICU bed, the same hour of agency nursing.

Some is administration as a system organised around billing, employing a vast army of people to do all that comes with chasing people for payment around an event of illness.

Korea and Switzerland sit above the curve for really important reasons: everyone is covered, primary care is dense, and disease is identified while it is still cheap to treat.

Morocco is on track: 99 per cent of its infants get their third diphtheria-tetanus-pertussis dose, and have done so for more than a decade . Eighty-seven per cent of births are attended by skilled staff, up from around two-thirds in the early 2000s.

Its entire health ministry budget for Morocco 2025 was about $2.5 billion: less than the annual turnover of Guy’s and St Thomas’

That is one London hospital trust.

The work that buys years cheaply is done by nurses, midwives and community health workers: the inexpensive end of the workforce, in every system, on every continent.

World Cups and health system performance may differ in many ways, but there are certainly winners and losers.

Related reading: The world’s next nursing shortage may not be where you think and What eleven specialists can do if they move around.

Leave a comment

We held an event on Monday; it went really well, and we got a great turnout with lots of buzz. Sadly, there was one element that didn’t go to plan. One of our keynote speakers was refused a visa and thus had to present online.

The keynote at this event was to form part of a trip that would have been invaluable for all concerned. Yet despite our best supportive effort, the visa was refused.

Initially I put this down to bad luck, the wrong person considering this on the wrong day. But then, when I mentioned it to people in my orbit, everyone had a story to tell.

So, I started digging, and what I found is worse than I had ever imagined.

Last month the International Confederation of Midwives held its Triennial Congress in Lisbon. The theme was “One Million More Midwives”. More than 3,000 midwives, researchers, educators and policymakers gathered to discuss maternal and newborn health: the 260,000 women who die each year in pregnancy and childbirth , and the 4.2 million babies who are stillborn or die in their first month. Nearly all of those deaths happen in Africa and South Asia , Sub-Saharan Africa alone accounts for around 70 per cent of maternal deaths worldwide.

At least twenty invited speakers from Africa and Asia were refused entry at the last minute. Many more delegates may have been caught in the same net. The refusals fell on midwives from Nigeria, Ghana, Rwanda, Burundi, Uganda, Ethiopia, Sierra Leone, the Democratic Republic of the Congo, Bangladesh and India. These were leaders, researchers and educators, formally invited, several with presentations already accepted into the programme .

Subscribe now

Harriet Akello directs midwifery for Mother Health International, which works in remote communities in Uganda. She was due to run a session on how evidence-based guidelines keep women and babies alive in the hardest conditions. She applied months in advance, but her visa was refused the Thursday before the congress opened.

An Ethiopian assistant professor, selected by ICM for its leadership sponsorship programme, was refused over her bank statement and what the consulate called an unreliable purpose of travel (the same reason we were given for our keynote speaker). Two delegates from the Bangladesh Midwifery Society were also refused. They are the people who persuaded their government to hire five thousand more midwives. In a congress about one million more midwives, they had a right to be in the very front row.

Portugal’s foreign ministry said that visa applications are assessed rigorously, objectively and factually, in line with the Schengen Visa Code

In case you are wondering, yes, it seems every delegate from a wealthy country got in without an issue.

There is a particular kind of global health absurdity in holding a conference about the world’s shortage of midwives, then refusing visas to midwives from the countries where that shortage is most evident.

Four years ago, Canada hosted the International AIDS Conference in Montreal. Canada rejected 1,020 visa applications for the conference (36 per cent of the total) and left another 10 per cent unprocessed by the time it ended, so effectively closer to half. The rejection rate for applicants from Nepal was 83.5 per cent; from Nigeria, 55.8 per cent; from Cameroon, the DRC, Ethiopia and Ghana, over 40 per cent.

Activists took over the opening ceremony and the president of the International AIDS Society called it global inequity and systemic racism from the stage. Canada’s immigration department later reviewed its handling of the conference and found no fault. When the International Council of Nurses brought its own Congress to Montreal the following year, Canada’s official guidance made the same point more honestly: a special event code had been assigned for visa applications, but it did “not offer any advantage to the applicant”.

The asylum fallacy

Aha! But some of these people claim asylum; scream the usual crowd. It’s true, there is no point denying that this happens in a very small number of cases, but without widening this post to the length of a book, I think this points more to the failure of ‘safe and legal routes’ than to the character of the people applying who are dedicated health professionals and experts in their fields.

Migration is very often driven by failures within the health systems people are leaving in the first place: weak health services, poor pay and limited career development are the classic push factors that lead health workers (and everyone else!) to look abroad. Conferences and international collaboration are among the few mechanisms that work in the other direction, strengthening the workforce and the system at home rather than depleting it.

Making it harder for health workers to attend those conferences does not close off migration. It closes off the exchange that might have reduced the need for it. Frankly tarring everyone with the same brush is deeply cynical and counterproductive. The benefits completely outweigh any perceived risk, even if you happen to view a nurse claiming asylum as a risk.

It would be comfortable, from London, to treat this as other countries’ bureaucracy, but this isn’t the case at all.

In 2018, at least seventeen researchers fourteen from sub-Saharan Africa, three from Asia, were blocked from attending the Women Leaders in Global Health conference at the London School of Hygiene and Tropical Medicine. Heidi Larson, director of the school’s Vaccine Confidence Project, called the refusals discrimination and read out letters from the absent on stage .

Share

The following year, twenty-four of the twenty-five researchers invited to the LSE Africa Summit were refused visas. Seventeen delegates were barred from the European Conference of African Studies. At a Wellcome Trust meeting on Ebola preparedness for the Democratic Republic of the Congo, a £1.5 million flagship programme, convened while the outbreak was live only six of the invited participants could attend. Seventy senior leaders wrote to the government in protest. Sierra Leonean academics invited to LSE events waited weeks for refusals that arrived without explanation, or received their visas after the events had ended. LSE began moving conferences to Belgium. A pandemic-preparedness meeting the host country’s own scientists could not fully convene, four months before anyone had heard of COVID.

The international president of Médecins Sans Frontières said her organisation faced the same problem constantly: staff from Africa and the Middle East invited to MSF’s own general assembly simply never make it, to the point where MSF now issues invitations a year in advance.

The refusal letters themselves are something else. One professor was refused because he had not previously been sent on similar training in the UK. A senior Ugandan researcher was told there was no evidence he would benefit from his trip. A consular officer, ruling on whether an epidemiologist would benefit from an epidemiology meeting.

The structure below the surface

In a twist that will surprise none of you, the Royal Society found the highest refusal rates for UK visas similarly skewed towards the global south. The gradient is not subtle, and it maps almost exactly onto the gradient of disease burden. Sub-Saharan Africa alone accounts for roughly two-thirds of all people living with HIV worldwide . The countries that carry the most maternal deaths, the most HIV, the most epidemic risk, are the countries whose health professionals find it hardest to be in the room.

It starts early. (In 2017, every medical student from Kenya, Uganda and Tanzania who applied to attend the International Federation of Medical Students’ Associations meeting in Montenegro was rejected. The exclusion is not just of today’s workforce but of the next one — the students who would have built the networks that careers in global health are made of.

Voice still needs a visa

Global health loves “voice”, “representation” and “lived experience” and it builds whole panels around those words. But voice still needs a visa and representation still needs a passport that works in the manner intended.

Lived experience still needs an embassy official to believe you.

The visa system is global health’s invisible gatekeeper as it decides who gets to be “global”, who presents evidence, who networks, who gets remembered by donors and policymakers, and whose experience remains conveniently local.

Nurses and midwives make up the largest share of the world’s health workforce, and the shortage of both, an estimated 900,000 midwives alone is concentrated overwhelmingly in Africa and low-resource settings. The people working inside that shortage are precisely the people the system finds easiest to exclude. Global health does not only exclude people by ignoring them. There is a particular cruelty in inviting people to share expertise only for your government to refuse them later.

Often there is no right to appeal.

What gets lost

A conference can survive missing speakers. The programme runs, the coffee gets drunk, the social media posts get written.

But what gets lost when the people working closest to the world’s heaviest burdens of illness and death are not there to shape the conversation, the interventions that get funded, whose evidence counts, which model of care gets exported back to the very countries whose experts were kept at home.

A global health workforce movement cannot be global if the nurses and midwives most affected by the crisis are the ones most easily excluded. Until the machinery of travel matches the language of inclusion, global health will keep losing the insight it cannot replace.

Leave a comment

Twenty years ago this week I sat in my parents’ house in the leafy suburbs, two-thirds excited and one-third terrified, and started a blog. The next morning I flew to a country I wasn’t allowed to name (Jonglei state, Southern Sudan; the war made discretion sensible) and spent a year there as an outreach nurse, writing it all down by head torch. I was 25 and couldn’t really have told you what the job was had you asked me. Neither, it turned out, could the job description.

I found the blog recently, sitting inside an abandoned hollow of the internet.

First off, the title is really problematic; I broad-brushed ‘Africa‘ when I could have been more imaginative and respectful of a continent I am so privileged to have visited many times.

It’s the sort of thing I tell my students off for doing now. It was naivety at its grandest form, and I wince a bit now when I think of it.

Reading it two decades on is odd, and what has spooked me a bit is that nearly everything I now say for a living, in journals and on conference stages, I had already written by Christmas 2006.

At age 25 I described the clinics I worked in as “islands of treatment in an ocean of nothing”. Within weeks I had worked out that the foreign aid worker is “much overrated in many ways we are a nice optional extra”, and that the national staff who kept looted clinics open through a war were doing the real work: “sadly you don’t see movies about them”. I watched an American doctor land, assess one child for a disease she turned out not to have, and fly off again past a crowd of unimmunised children, and I did the sums on what her plane had cost.

I wrote about a village three hours’ walk from clean water and the swamp its people drank from, the same swamp they used as a toilet. I noticed that our clinic was treating diarrhoea with medicine the patients washed down with the water that had given them the diarrhoea in the first place. When the village’s hand-operated water pump broke, I could date it from the clinic records. Gastroenteritis was up 67 per cent within three days.

Subscribe now

I have spent much of the past two decades saying slightly more sophisticated versions of the same things I wrote in the blog: history doesn’t repeat, but it often rhymes, as Twain said.

The evidence base I lean on now has improved enormously, but many of the observations haven’t changed at all.

It would also be lazy and overly negative to suggest that progress has not been made in the time since I left Sudan.

The world I was writing about killed a child with measles every few hours, and that has genuinely changed. Global measles deaths fell from roughly 780,000 in 2000 to around 107,500 by 2023 . Under-five mortality has roughly halved over the same period. The meningitis A vaccine I once nursed through a failing cold chain (cool boxes cosseted, as I put it then, “like vintage bottles of wine or champion race horses”) was replaced a few years later by MenAfriVac, which has all but ended epidemic meningitis A across the belt where I worked. When we vaccinated 7,758 people in a week and lost none of the nearly fifty meningitis cases our clinic saw, I was, without knowing it, working at the tail end of a problem that was about to be solved for good. So the machinery of international aid can work. Money and attention, held on a single problem long enough, will fix it.

There is almost no media attention on this astonishing progress; you would find it a challenge to find anyone who doesn’t work in our industry who knows about it.

However, progress in other areas is far less impressive and in places it’s hard to see any improvement at all.

When I was in South Sudan, there was a Comprehensive Peace Agreement in place, which was notable because the peace was limited, the agreement often far from certain, and it could never be described as comprehensive.

The death of the key political leader John Garang in a mysterious helicopter crash in 2005 had made South Sudan yet more unstable, and I still remember even now the various incidents when the war came very close indeed.

Southern Sudan was five years from independence when I arrived, and, despite the challenges I wrote at the end, hopefully about “a country supposed to be grabbing the opportunity of peace with both hands”. Sadly, that didn’t go as I wished, and plenty of those villages still have no pump. The islands of treatment are still islands. We have simply spent twenty years mapping the ocean of limited healthcare in higher resolution.

Share Globally Nursing

This February a plane dropped a bomb on Lankien. The hospital in Jonglei where I had worked had run continuously since 1995 and was the only advanced healthcare facility for a quarter of a million people; the compound of huts and the airstrip I flew in and out of are recognisable in the blog. Staff discharged their patients in the hours before the strike, having had warning; some of them were women in labour. The bomb hit the warehouse, the town was taken and looted, and in April MSF announced that after thirty-one years it was closing Lankien for good.

It is the fourth of its hospitals in the country shut by attacks since the start of 2025. MSF will only say that government forces are the sole party with the means to bomb from the air; local investigators have reached less guarded conclusions.

Twenty years ago I helped rebuild clinics that had been stripped bare in exactly this way, and I wrote then that the work was the work of a generation. How tragic that I have turned out to be right.

What would really bewilder the 2006 version of me is who I’ve become. I was scathing then, in an amateur way, about some of the machinery of aid. My entry on data collection calls it “spirit crushingly dull” and admits that six hours of work had achieved nothing except that “the Excel columns are now in different colours”. I wrote whole paragraphs about reports nobody would read, while a clinic full of patients waited outside.

My life now is often spent trying to find and support data collection across the various programmes I support. I didn’t realise then that the healthcare deserts mapped closely to the data ones.

There were some other parts on data that I had completely forgotten. I reread a blog post I had written about an afternoon spent completing government tax forms for staff who did not know their dates of birth, and yet the form required one, so we found ourselves in a quandary.

The only way out? Fraud.

So we built them together: what year were you married? Let’s say you were twenty then (year of birth) . Pick a number between one and twelve (month of birth) and (my favourite) how many cooking pots do you have at home? (Day of birth). Thus quite a few people were born on the fourth of July!

I committed that fraud upwards of thirty times, and it remains the most complete description of what happens when bureaucracy designed in one world lands on another that I have ever produced.

Nothing I have written since, with a reference list attached, has said it better. The same goes for the shop I passed one Sunday selling pillowcases, pocket mirrors and intravenous gentamicin. South Sudan’s entire health system summarised in a sentence.

I ended the final entry, somewhere over the Mediterranean, with the only line in ten months I didn’t undercut with a joke: but for an accident of birth, it could be you or me there. Twenty years on, the accident of birth still decides who nurses and who is nursed, and whose clinic is safe.

The blog’s most obvious finding was the one about us: the foreign aid worker as a nice optional extra. I have spent twenty years testing that hypothesis with my own career, and I would say that, broadly, it holds. Certainly we should focus on the nurses who are working to develop services in their own countries and give far less press attention to the foreign aid worker. I often hear it said that UK audiences won’t focus on a story unless there is a UK angle. I think this grossly underestimates the audience.

The people who were essential in 2006 were the national staff who reopened looted clinics, and the people who are essential now are those who continue to care for the people of Jonglei.

Leave a comment

Last week I spoke at the International Home Care Nurses Organisation’s conference in London.

So did Howard Catton, Chief Executive of the International Council of Nurses, who said something that I’ve been thinking about all week.

The global nursing crisis is not only a question of how many nurses the world has, but also of where they are.

The 5.8 million shortfall is real – but this sits on top of a second problem, distribution which is the bit that most workforce narratives miss. ‘We don’t have enough staff’ is easier to understand than ‘our staff are in the wrong place and don’t have the right skills’ which is a bit more tricky.

Both are then hiding a third problem the averages. A country’s nurse numbers can climb on paper while the gap on the ground between urban and rural only widens.

In many countries, 60 to 65 per cent of the nursing workforce sits in hospitals.

In England, where he happened to be standing, just 12 per cent of NHS nurses work in dedicated community services, while district nurses, health visitors and school nurses go on quietly declining.

Subscribe now

He set this against the image we commonly carry of community nursing: the district nurse on a bicycle, somewhere pastoral, rural, lemon drizzle cake, all very twee, arguing quite rightly that community nursing is incredibly complex. Today’s community nurse administers chemotherapy in a front room and manages drug regimens that would once have required admission, complex care delivered at home, metrics collected on a tablet.

As with most things the issues extend across our world. The maldistribution described does not stop at the edge of the high-income world. In a wealthy system, the staff exist and are in the wrong place. In the poorest regions, they are in the wrong place and there are too few of them to begin with.

So let’s look at where people actually live in relation to where care is . I have spent most of my career focused on the more purple bits of this map,

Travel time to the nearest health facility. White and pale green indicate under an hour; deep purple indicates more than a day. Source: Malaria Atlas Project.

So at first glance, I will admit I was pleased. It really doesn’t look too bad! A lattice of bright corridors threads along the roads, and most of the inhabited world looks within an hour or two of help.

But there is one very important factor that I didn’t mention. The first map is our patients journey to healthcare if they have access to a car.

Now take the car away and model the journey as most rural patients in the poorest regions actually make it — on foot.

The same map, walking only. The bright corridors collapse into the surrounding purple. Source: Malaria Atlas Project.

The corridors collapse. Across the interior of the Sahel, the Andean highlands, the forest blocks of Central Africa, the nearest facility is now more than a day’s walk away.

The road network was doing the work the first map appeared to credit to the health system, and the people left in the deepest purple are left stranded, adrift on an endless ocean. They suffer just the same health challenges and yet the response and options are far worse. They are the woman in labour who will not reach the clinic, the child whose fever will not be seen.

This is the geography the workforce numbers describe; the bright spots are where almost all of the doctors and many of the nurses already are — cities, the coast, the places a vehicle can reach. The purple is everyone for whom care is a journey measured in days, not minutes. You cannot fix a purple region just by building another hospital but by supporting higher quality community care close to where people live.

The State of the World’s Nursing report, published last May, shows the global workforce has grown to nearly 30 million, and the headline shortage has come down from 6.4 million to 5.8 million. Good news, until you read the next line: 78 per cent of the world’s nurses serve 49 per cent of the world’s population, and the shortfall that remains sits overwhelmingly in low- and lower-middle-income countries. The shortage is shrinking on average and concentrating at the same time.

And a rising density tells you almost nothing about access, because the number that improves is a national average, and a national average hides horrendous disparity. A country can add nurses, watch its nurses-per-thousand climb, and report progress to the WHO, all while the new nurses settle exactly where the old ones already were.

The subnational picture, when anyone bothers to assemble it, is brutal.

Across 58 countries, the region with the most nurses had on average eleven times the density of the region with the fewest, and in the worst cases 13 per cent of a country’s population had access to 45 per cent of its nurses.

India is the cleanest illustration. The headline ratio is about two nurses per thousand — low, but improving, and the kind of number a health ministry can build a narrative around. Underneath it, the density of registered nurses runs from 0.8 per ten thousand in Bihar to nearly 79 in Kerala.

These are two different countries at each end of the same flag.

Nearly two-thirds of Indians are rural, yet the nurses — and most of the colleges that educate them — cluster in a handful of southern and western states. The story repeats across sub-Saharan Africa: in Botswana, one of the better-staffed countries in the region, rural districts had 26 nurses per ten thousand against 77 in urban ones.

Where the nursing education pipeline is strong but the jobs and the pay are not, nurses are trained and then leave — for the capital, or for a high-income system that will pay roughly three times what a low-income one can; in much of Africa, a substantial share of nurses say they intend to emigrate . The destination countries then report a falling shortage and congratulate themselves on a pipeline they did not build. The arithmetic that improves the global figure is, at the source, a story of educating a workforce and watching it walk to where the map is already bright.

In a high-income system, concentrating nurses in hospitals is a misallocation: the staff, largely exist, they are simply in the wrong building, and in principle, with additional education packages and support you could move them.

In LMIC there was never a version of the budget where the rural clinic was staffed. The purple on the map is not an underfunded service, it is a service that was never there at all.

The reflex, when faced with the purple, is to reach for the community health worker, lightly trained, cheap to deploy, close to home. A cadre that is cheap enough to put everywhere is, by construction, a cadre you can underpay and under-equip indefinitely as their career prospects are so much lower. Compared to degree educated nurses they are a far less attractive poach.

The next nursing shortage, then, is not coming, for it is already here: evenly reported and unevenly felt, buried under countrywide averages that obscure profound shortages in our community provision. Follow the map past the merely underserved, all the way into the purple, and ask whether our headline figures are obscuring a distributive disaster.

Leave a comment

There is a number I keep returning to from a paper we recently published in Frontiers in Health Services.

At the time the programme I am about to describe began, Myanmar had eleven paediatric neurologists. Eleven, for a paediatric population of around fourteen million children, the vast majority living outside the countries largest city, Yangon. This, in a country where 70% of the population is rural and the roads between them and any kind of specialist care are long.

Eleven.

A single large children’s hospital in London employs more than that.

The conventional response to a workforce gap of that scale is a training plan. Produce more specialists and build more capacity at the centre with efforts to extend the referral network. These approaches are not wrong, but they take decades and the children in Kale and Kengtung and Myitkyina do not have decades.

So what do you actually do with eleven specialists and a country the size of France?

The answer, in this case, was deceptively simple. A hub-and-spoke model linking the paediatric neurology team at Yangon Children’s Hospital with general paediatricians in seven regional public hospitals.

Quarterly in-person clinics, where specialists travel to the patients’ local hospital. Monthly virtual consultations in between and a mobile messaging group for quick questions.

The neurologist goes to the child. Not the other way around.

This sounds really obvious, and it really isn’t.

The default assumption of specialist medical care, in low-resource settings as much as anywhere else — is that specialist expertise lives at the centre and patients travel to reach it. That assumption is so embedded in how all health systems are funded and organised that it rarely gets questioned.

The problem is that in Myanmar, for parents of children with a neurological diagnosis, the journey to Yangon is not so much an inconvenience as a financial event. Our cost analysis compared the cost per patient of attending a specialist appointment at the tertiary centre in Yangon with that of running an outreach clinic at their local hospital (Linn et al., 2026).

Under the standard referral model, the mean cost per patient consultation was $193 US dollars. Under the new outreach model, it was a more manageable $ 7 per patient.

For hospitals the very furthest from Yangon, Kengtung in the east, Myitkyina in the north the standard referral cost reached $441 per consultation. The outreach visit cost $29 and $8 respectively. The cost saving ranged from 81% to 98% depending on location.

In a country where many rural families live close to subsistence, $441 is months of household income. And the comparison assumes families could make the journey at all. Before the programme, most of them couldn’t and so the reality was not expensive referrals to Yangon, it was no referral.

Rather crucially, the cost of the specialist visits to district hosptials was covered by the Ministry of Health and Sports, reducing the costs to families to nearly zero.

Between 2017 and 2020, the programme facilitated 2,603 patient consultations across the seven regional hospitals. Epilepsy was the most common diagnosis at 54%, followed by cerebral palsy at 12%. Attendance rose steadily and consistently across the period. This was not the ebb and flow of a new initiative, but genuine building demand as families learned that a specialist would actually come.

The diagnostic profile is important as Epilepsy and cerebral palsy are not acute presentations. They are conditions requiring careful management across childhood: repeated follow-up, medication titration, monitoring of development over time. They are also, in low-resource settings, among the most undertreated conditions precisely because of that burden of access. Evidence from population-based studies in Myanmar showed a wide treatment gap for childhood epilepsy, with children remaining undiagnosed, untreated, or managed on inappropriate regimens. The outreach model did not just provide access, it provided the kind of access these conditions actually require.

The telemedicine element is worth examining separately, because it was not telemedicine in the conventional sense.

Rather than the standard model of doctor-to-patient interaction over a video platform, the approach incorporated a non-specialist doctor at the patient’s end who would conduct the physical examination under real-time instruction from the specialist in Yangon. The regional paediatrician became the hands, eyes and ears. The neurologist became the clinical reasoning. Between visits, cases were triaged through the messaging group; regional hospitals eventually began pre-screening patients so that specialist time during outreach visits focused on the children with greatest needs.

The standard critique of telemedicine in specialist care is that it cannot replace physical examination, which is true. The response this model developed was not to abandon physical examination but to restructure who conducts it.

It is not a perfect solution. Access to diagnostics and key medications varied across hospitals. Reliable internet was not guaranteed. The neurology team in Yangon was doing this alongside their full clinical duties. There is a ceiling at which this model becomes non-viable. We could not identify exactly where that ceiling was during the evaluation period, which tells you something, but it exists.

Then 2021 happened.

The coup dismantled the health system the programme had spent four years embedding itself within. The hospitals remained, but the Ministry of Health and Sports they reported to was controlled by a military junta rather than a democratically elected government.

I say this not to end on defeat, unlike last weeks post, this programme worked. The numbers are real: 2,603 consultations, $7 per patient, cost savings of up to 98%. Those children were seen who would not otherwise have been seen. The evidence is published and available to health systems facing similar constraints, which is most of them.

The broader argument the model makes is one that global health absorbs slowly: specialist scarcity is as much a distribution problem as a supply problem.

More than 70% of countries classified as low-income lack consultant-led paediatric neurology services across all regions. The response is almost always framed as a training question, we must produce more specialists goes the cry. But it does almost nothing in the short or medium term for the child with epilepsy in Kale whose family cannot afford the journey to Yangon.

What this programme demonstrated is that a small number of specialists, a quarterly schedule, a mobile phone, and a genuine willingness to redesign who travels where can reach a significant number of those children.

No new technology, no large external budget, no international volunteers.

Existing capacity, organised differently.

This was not a resource problem solved by more resources. It was a distribution problem solved by a quarterly visit, a messaging group, and a willingness to question which direction the specialist should move.

I wanted to write this week about failure, and about why a programme I led and developed did not work. One of my major criticisms of the sector I work in is its ability to adopt the brace position every time even mild criticism appears. That habit limits our ability to learn from things that did not go well.

So, here goes.

I led the development of the Ukraine Clinical Guidance website, and it failed.

Take yourself back to 2022: the tail of the pandemic and the unfolding crisis in eastern Europe. Across our screens came images of tanks, shelling and fighting across Ukraine. Within weeks the camera angles widened, and we saw attacks on hospitals, including the maternity hospital in Mariupol. Health workers were trying to keep services going while facilities, ambulances, supply chains and power systems were under attack or under strain.

A clinical guidance website felt practical. It could be built quickly, bring together trusted resources, translate useful material into Ukrainian and make it available to doctors, nurses and others working in disrupted conditions. It was not going to repair a damaged hospital, replace a missing colleague or reopen a supply route, but it seemed reasonable to think that clear guidance in the right language might help people practising in circumstances that were anything but normal.

The model was Myanmar. We had seen a similar approach work there after the 2021 coup, where health workers were also trying to sustain care under political violence, insecurity and institutional collapse. The Myanmar Clinical Guidance website I developed was never glamorous — a practical repository of clinical and nursing resources, forming part of wider virtual and blended support for people still trying to educate, supervise and practise in a fractured system.

In Myanmar, the website worked because it was not really just a website. It sat inside a network. There were relationships that pre-dated the crisis. Clinicians and educators knew each other, trusted each other and could say bluntly what was useful and what was not. There were people who could tell us whether a resource fitted practice, whether it needed translating, whether the format was wrong, or whether the best thing we could do was stop producing content and listen. The website had a route into practice because the relationships were already there.

That lesson should have been obvious: digital support works best when it is relational, specific and connected to people who can use it. I understood that in principle. I did not apply it well enough in Ukraine.

Ukraine looked similar only at the surface level — a country at war, health workers under pressure, clinical systems disrupted, an urgent need for practical support. That was enough to make the Myanmar model feel transferable. Emergencies create pressure to act quickly, and the digital answer has an obvious appeal: no travel, no visas, no security plan, no waiting for buildings to reopen. You can translate, upload and share.

But Ukraine was not Myanmar, which now sounds obvious. It had its own health system, professional networks, institutions, reform agenda, clinical culture, but my relationship with them was non-existent. We were left with having to try and launch from a standing start.

The Ukraine Clinical Guidance website was described publicly as a collaborative partnership containing clinical guidelines and resources for doctors, nurses and others in Ukraine . It included Ukrainian-translated clinical procedure videos and covered paediatrics, neonatal care, emergency care, obstetrics, nursing, mental health, surgery, general practice and first aid.

Over the three years it ran, barely anyone used it.

The resources existed, had been translated, and could be shared — but availability is a weak substitute for implementation. A clinical resource does not become part of practice because it has been uploaded; it becomes useful when people trust it, use it, adapt it and have some route to make it matter. A website can be live and still not be alive in the system. It can sit online, looking professional, without becoming part of anyone’s working day — without reaching the nurse in a cold ward, the doctor in another mass casualty episode, or the educator supporting students whose education has been disrupted by war.

What we learnt in Myanmar was that virtual resources travel through trust, repetition, ownership, adaptation, teaching, feedback and use. That was where we were weak in Ukraine. We had too many people involved, insufficient coordination, and no equivalent of the Myanmar network. There were good people around the work, but goodwill is not ownership, and it certainly missed the crucial convening layer between ourselves and clinicians. The website became one more well-intentioned offer in a crowded emergency information space.

That is not a satisfying failure, because nothing dramatic happened. The website did not collapse. It did not cause obvious harm. It did not become a scandal. It failed in the quieter way that global health programmes often fail: it existed, it looked plausible, and it did not become important enough to matter.

One reason this is worth writing is that global health is still not very good at talking honestly about failure. We are excellent at describing lessons learned once they have been made safe enough for a final report. We can say a programme was “challenging”, implementation was “complex”, engagement was “variable”, or the context was “fluid”. We are much less willing to say that something did not work — or that we copied the visible part of a model without copying the relationships that made it useful.

Polite language protects the wrong thing. It protects institutions, reputations and future funding bids, but it does not protect the next nurse, doctor, midwife or health worker asked to use a tool that has never really been tested in their context.

Ukraine is a good place to learn this, because the pressure on the health system leaves little room for romantic thinking about digital solutions. WHO has verified more than 3,000 attacks on health care in Ukraine since the full-scale invasion.

This matters for nursing. Nurses do not need another distant repository of things they might read if the shift ever calms down — they need usable support that connects with the decisions they are already making. A website can help, but only as part of a delivery route: with Ukrainian clinical ownership, a professional body or ministry partner, and educators and clinicians who request content, test it, and reject what does not work. It might also mean not building the website at all.

That is difficult for organisations to hear, because building something is satisfying. It gives everyone a task, and it can be announced and shared with language about solidarity. It lets institutions feel they have responded — and I understand the appeal, because I led exactly this kind of response. The problem is that emergencies reward visible action, while implementation looks slow, relational and dull. The slow work determines whether the visible action has any effect.

The network is often the crucial part of the model. Digital resources can be useful in war because they move faster than people can. They can cross borders, survive travel restrictions, be translated, updated and shared, and they can offer some support to clinicians cut off from ordinary teaching, supervision and professional contact. But they do not become useful simply because they exist.

If I were doing it again, I would start more slowly, even in an emergency. I would spend less time asking what we could build and more time asking who in the country wanted it, who would own it, who would reject content that did not fit, who would update it, who already had the trust of clinicians — in short, building a network. I would build less and listen more, and I would be much more suspicious of any solution that can be completed before the relationships exist.

That is not a comfortable lesson. It is not the sort of thing that looks good in a funding report. But it is probably the right one.

London has already had its first warning shot of the summer.

In late May, the UK Health Security Agency issued its first amber heat-health alert of 2026, covering London and other parts of England, warning that high temperatures could put vulnerable people at increased risk and that health and social care services needed to prepare. The heat then broke records. Temperatures reached 35.1°C at Heathrow and Kew Gardens, the highest May temperature recorded in the UK.

The media, as ever, reported the whole thing with pictures of young people enjoying the sun and people eating ice creams.

Heat waves such as this are lethal.

The emerging El Nino patterns make an urgent issue much worse

As older people suffer in flats that do not cool down, children’s asthma worsens in hot and polluted air, care homes try to keep residents hydrated, and nurses toil in hospital buildings designed for another climate.

The UK does have a plan for this. The UKHSA Adverse Weather and Health Plan sets out how health and social care, local government, voluntary organisations and communities should prepare for weather-related harm. It is sensible enough as a document, but talking is one thing; walking is quite another.

Much of that ‘walking’ workload will land on nurses.

Climate change is now routinely described as a health emergency, and that is right, although the phrase can still sit rather comfortably above the more awkward practical question of who is expected to notice, absorb and respond to the consequences. In ordinary clinical life, the response begins with someone recognising the heat-exhausted patient before they deteriorate, someone keeping an eye on the older person who lives alone through three hot nights in a flat that does not cool down, and someone noticing that a child’s breathing has worsened because the air outside has worsened too.

In most health systems, that someone is often a nurse.

WHO describes climate change as a threat to the basic conditions for health: clean air, safe drinking water, food and shelter. That is useful because it keeps the issue out of the abstract. Climate change does not only create new problems. It makes old problems harder to manage, and it presses on the bits of the system that were already weak.

Heat makes this obvious. WHO’s heat and health guidance says heatwaves and prolonged excess heat are increasing because of climate change, and that even low and moderate heatwaves can affect vulnerable people. That sounds simple until you imagine the actual shift required.

The UK has already had a small taste of what this means in practice, even if we are still inclined to treat it as an unusual inconvenience rather than a predictable pressure on health services. Older buildings overheat, hospital wards become uncomfortable and sometimes unsafe, and staff respond in the way health workers usually respond when the system has not caught up: they improvise with fans, fluids, ice, opened windows, closed blinds and a fair amount of common sense.

That is something, and it may prevent harm in the moment, although it should not be mistaken for adaptation. It is people making do inside infrastructure built for a climate that no longer quite exists.

In lower-resource settings, the margin is thinner because heat may arrive where clinics already lack reliable electricity, flooding may arrive where roads are already fragile, and drought may arrive where malnutrition is already part of the clinical picture. Climate risk rarely arrives as a neat, single hazard that can be managed in isolation; it tends to turn up alongside the existing failures in housing, transport, staffing, water, food security and access to care, which is why the burden falls so quickly on the people trying to keep services working.

This is why climate health planning has to move beyond risk registers and tidy lists of hazards. A plan that identifies heat, flooding and disease risks is only half useful unless it is equally clear about the workforce expected to respond when those risks become patients, families and disrupted services. Someone has to visit the isolated older person, triage the breathless child, explain to families what deterioration looks like, maintain routine care while emergency work expands, and keep enough of the ordinary health system functioning so that the climate response does not simply consume everything else.

If the answer is “the existing workforce”, then the plan needs to admit what that means.

Nurses already hold together parts of health systems that do not quite work. In many countries they are the most accessible professional group, especially in rural and community settings. They are also the workforce most likely to see the early signs that climate pressure is becoming clinical risk. Heat stress, dehydration, diarrhoeal illness after flooding, missed medicines after displacement, worsening asthma, maternal risk, mental distress after repeated shocks.

The health of Children should make this harder to ignore. UNICEF describes climate change as a major threat to children’s health, nutrition, education, development and survival. Children are less able than adults to withstand extreme weather and are more vulnerable to temperature changes and disease.

That language is grant, but the consequences are ordinary enough to recognise: a child missing school after displacement, a baby with diarrhoea after contaminated water, a teenager anxious after another flood, a child with asthma breathing hotter and dirtier air.

These are not only paediatric problems. They become family problems, school problems, community problems and nursing problems.

Global nursing needs to say this more clearly because climate adaptation is not only about greener hospitals, although health care emissions matter. Adaptation is also about whether nurses are educated, staffed and supported to deliver care in a changing climate. A carbon plan without a workforce plan is useless.

This is where global health language becomes slippery. We talk about resilient systems as if resilience is something that can be written into a strategy. In practice, resilience often means people absorbing more than they should. Nurses are told to adapt. Communities are told to cope. Facilities are told to prepare. Then the heatwave, flood or outbreak arrives, and the same thin workforce is expected to make the plan real, often from a standing start.

I am not against plans. I have seen enough chaos to appreciate a decent one. The problem is the gap between a plan and the conditions needed to deliver it. A climate health plan that does not reach the rota, the curriculum, the placement, the medicine supply chain, then it is worthless.

Nursing education has to catch up with this as it is clear climate change cannot sit as a decorative lecture somewhere near the end of a public health module. It needs to be woven into the practical formation of nurses. What does heat illness look like in an older adult? How does flooding change infection risk? What happens to long-term medication when families are displaced? What does safe care look like when a clinic loses power?

Many years ago I worked in South Sudan and watched as carefully applied wound dressings washed away as the population walked home from the clinic through a series of knee-deep rivers, without the tools to adapt; I fear our efforts were often undone by the water that surrounded us.

There is also a leadership issue. Nurses are often close enough to see what is changing, but not always powerful enough to shape the response, hardly unique to this topic. The people who understand the practical system are consulted late, asked to implement quickly, and then blamed if reality refuses to match the plan.

Climate adaptation will fail if nurses are treated only as delivery staff. They need to be involved in risk mapping, service design, education planning, emergency preparedness and community communication. That is not because nurses are morally better than anyone else. It is because they know where the plan on paper will fall apart in reality.

In conflict-affected and fragile settings, climate risk does not wait politely behind political crisis. Floods hit displaced populations. Heat affects people living in temporary shelters. Drought worsens malnutrition. Disease patterns shift while surveillance systems are already weak. Nursing staff are then expected to continue care with fewer supplies, less security and less support.

It is also the places facing the greatest climate health risks that are often not the places most responsible for creating them. They are also often the places with the least spare workforce capacity to respond. When global climate finance talks about health adaptation, nursing should be in the middle of the conversation.

That means investing in community nursing, public health nursing, emergency preparedness, infection prevention, mental health support, maternal and child health, and the education systems that produce nurses capable of working in more unstable conditions. It means protecting nurses during disasters and conflict.

It means designing facilities where staff and patients can survive heat.

The phrase “emerging infection” can make an outbreak sound like a biological event first and a systems problem second. The two outbreaks that raced across our newsfeeds this week show something more.

In one setting, there is testing, quarantine, repatriation, monitoring, and a public health machinery that moves quickly. In another, health workers lack masks, transport, basic medicines and enough staff to keep services open, they are dying as a result. The virus matters, but the response is more clearly shaped by the system it lands in.

This week gives us two very different examples.

In the Democratic Republic of the Congo and Uganda, the WHO has declared an Ebola disease outbreak caused by Bundibugyo virus (Also known as Ebola) a Public Health Emergency of International Concern. WHO was alerted on 5 May 2026 to a high-mortality illness in Mongbwalu Health Zone in Ituri Province, DRC. The outbreak includes deaths among health workers, and the WHO has specifically noted that the infection and death of four healthcare workers within four days at Mongbwalu General Referral Hospital point to serious infection prevention and control breaches.[

At the same time, a hantavirus cluster linked to the MV Hondius cruise ship has produced a very different public health response: laboratory investigation, contact monitoring, quarantine, repatriation arrangements and international coordination across several countries. WHO reported severe respiratory illness cases aboard the ship in early May, with testing confirming Andes virus infection. The CDC later confirmed that 18 recently repatriated US passengers had been asked to remain at the Nebraska Quarantine Facility until 31 May 2026, the 21-day mark of their monitoring period.

The temptation is to write about these as two scary viruses, but Ebola and hantavirus are not the same problem. Ebola spreads through direct contact with the blood or body fluids of symptomatic patients and has a long history of transmission in health facilities. The Andes virus (hantavirus) is unusual because person-to-person transmission is rare. However, the wider public health risk in the current cruise ship outbreak has been described by WHO as low.

One outbreak is unfolding in conflict-affected central Africa, where responders are dealing with deaths among health workers, uncertain spread and shortages of basic supplies. The other is linked to a cruise ship, where passengers can be traced, repatriated and monitored through formal systems with a level of logistical capacity that most outbreak responses can only envy.

Reuters reported that first responders in the DRC were short of basic supplies, including pain medicine, masks, and motorbikes for contact tracing. Without wishing to state the obvious, a contact tracer without transport is not a contact tracer in any useful sense.

A nurse without protective equipment is at an exceptionally high risk. A facility without basic infection prevention is not simply under-resourced; it can become part of the chain of transmission. The damage beyond those suspected of having the virus and the delays in care can be colossal.

In outbreaks, nurses are not a decorative part of the response. They are usually the people closest to the risk - Triaging the fever, liaising with close contacts, and cleaning up infectious body fluids.

When health workers die early in an outbreak, it is not only a personal tragedy; it is a screaming warning about the system. It tells us something about training, supervision, protective equipment, water and sanitation, triage, isolation, staffing and trust.

The hantavirus outbreak on the cruise ship shows another side of outbreak management. Here, the issue is less about the collapse of basic infection prevention and more about public health logistics. Who was exposed? Who needs monitoring? Who can travel? Where can people be safely quarantined? Who explains the risk without turning uncertainty into a macabre theatre?

To move away from the more sensationalist to, for me, the central question - why some outbreak responses have transport, testing, quarantine capacity and international coordination, while others are still asking for masks and motorbikes?

We are very good at saying that pathogens do not respect borders. We are less good at admitting that outbreak response capacity is profoundly affected by the places immediately at risk, the countries where it is detected, and often, the nationalities of the people it kills.

A cruise ship outbreak involving international passengers triggers a certain machinery. An outbreak in a conflict-affected health zone in the DRC triggers another, slower and far less effective one.

That does not mean wealthy countries should ignore imported risk and wait until an outbreak becomes their problem. It should mean the opposite, if the world wants early containment, it has to fund the basic infrastructure that makes it possible. Surveillance does not begin with a dashboard in Geneva.

It begins with a nurse who spots a symptom or a pattern among patients, a laboratory that can process a sample, a vehicle that can reach a village, a facility that can safely isolate, and a workforce that trusts it will not be left to cope with the pathogen unprotected.

The Global Preparedness Monitoring Board has again warned that preparedness reforms have not kept up with pandemic risk.

Nursing has a dog in this race because outbreak response relies on forms of work that are often poorly valued until they fail. Infection prevention is not glamorous. Triage is not glamorous. Cleaning, waste disposal, donning and doffing, explaining isolation, monitoring contacts, and maintaining records do not gain world attention in the way a scary, uncontrollable virus does.

Ebola has taught this repeatedly. Health facilities can save lives, but they can also amplify transmission if the basics are missing. That is not a criticism of individual nurses. It is the predictable consequence of asking staff to manage high-risk care without the conditions required to do it safely.

Hantavirus teaches a different lesson: when we really want to do outbreak control properly, we can. Still, good quality nursing, from the first nurse on the cruise ship all the way along the chain, is essential and demands effective funding.

The danger is that we talk about nurses in outbreaks only when they become symbols: brave nurse, fallen nurse, heroic nurse in PPE. It’s not difficult to see why, but it would seem preferable to focus on what would have kept the nurse safer before bravery became necessary.

International Nurses Day is meant to make nursing visible. It often does this through familiar language about gratitude, contribution, and skill; a word salad of positive vibes.

Look at almost any social media post about nursing, and you will see pictures of diverse nursing teams, often with specific reference to the global nature of the workforce. This did not happen overnight and it did not happen by accident.

Wealthy health systems have relied on internationally educated and experienced nurses for years, often while speaking warmly about partnership with the countries those nurses leave behind. Nurse migration can be good for individual nurses. It can also expose the utter failure of destination countries to educate and retain enough staff of their own.

In the UK, Health and Care Worker visas issued to nursing professionals fell from around 26,100 in 2022 to 1,777 in 2025, a fall of about 93%. The route that once brought more than 26,000 nursing professionals in a year is now producing fewer than 2,000.

The UK is one example of a wider pattern. High-income countries have treated international recruitment as a pressure valve for domestic workforce failure. When services were short, recruitment expanded. When migration became politically difficult, the route narrowed. The result is nurses left in limbo, source countries unable to employ all the staff they have trained, and destination countries behaving as though workforce supply can be switched on and off.

Ghana shows why rich nations switching the vacuum off and on again is so bad for everyone. In 2025, Ghana’s Ministry of Health confirmed that more than 100,000 trained health professionals were unemployed, including nurses and other licensed staff, largely due to insufficient financial resources for employment. At the same time, Ghana remains a country with serious health workforce pressures, and its government has explored managed migration agreements with countries including the UK, the US, Jamaica, and Barbados, but that clearly isn’t going to work overnight. In the meantime, a globe with a nursing shortage has thousands of degree-qualified nurses making cappuccinos.

That is the damage caused by the current model. A country may decide to expand nurse training because it has real health workforce shortages, because students see nursing as a route into secure work, and because overseas recruitment has made nursing education appear to be a viable investment for the country. The country may educate nurses to meet long-term workforce needs, replace staff lost to migration, or expand access to care.

Why would they do this if they weren’t sure they could employ them?

Bluntly, because we live in an age of short-term governmental thinking and because education programs are often easier to fund and politically easier to announce than recurrent salaries, funded posts, and functioning long-term workforce planning systems.

This represents the arrival of a problem that seemed very far away, when, in many cases, a previous administration decided to open the tap on nursing education without a plan for where all the water would go, but with clear assurances that richer nations would employ a sizable proportion.

But when the destination-country route narrows, often through horrible immigration ‘island of strangers’ type narratives, the individual nurse is left between two failures: no funded job at home and no stable route abroad.

The Philippines has lived with a version of this problem for decades. It has trained nurses for a global market, and many Filipino nurses have built careers in destination countries. Their movement has brought income, professional opportunities, and remittances. It has also made the Philippines central to rich-country workforce planning in a way that should make those countries stop and think, what happens when we treat three and four-year nursing education programmes as a vacuum we can turn off and on.

The Nursing and Midwifery Council has reported this vacuum-on, vacuum-off effect in its numbers. In December 2025, it said growth of the register had slowed significantly because of a sharp fall in international recruitment. During the April to September periods from 2021 to 2024, international joiners made up around half of new recruitment to the register. From April to September 2025, they made up 25.8%.

Canada is working through a version of the same problem. Its federal government says the health workforce has faced shortages for some time and has invested in faster integration of internationally educated health professionals. That may be sensible policy if you are Canada’s Ministry of Health but also shows how quickly wealthy systems reach for overseas-trained staff when domestic supply falls short. The vacuum that was off is now on again.

Australia tells a similar story. The government has been developing a National Nursing Workforce Strategy, using modelling to understand workforce supply and demand over the next 12 years. At the same time, registration reform has created a faster route for internationally qualified nurses from comparable countries.7 Workforce strategy sits beside accelerated international recruitment because services need staff now.

Nursing is a global profession. The point is not that the UK, Canada or Australia should stop recruiting internationally. The point is that this model becomes fragile when wealthy systems assume the global workforce will keep absorbing their planning failures. Destination countries should also pay properly towards the education systems that produce the nurses they recruit (more on that in another post)

Put at its most basic, why should the people of, to pluck an example, Uganda pay to educate nurses whose clinical careers may then benefit the UK/Canada/Australia, when we screwed up our workforce planning?

WHO’s State of the World’s Nursing 2025 report says the global nursing workforce grew from 27.9 million in 2018 to 29.8 million in 2023, while distribution remains deeply unequal. The same report projects a global shortage of 4.1 million nurses by 2030 if current trends continue.

Those numbers should sit uneasily beside any simple claim that high-income countries are merely filling their own gaps. The world does not have a spare nursing workforce waiting at the airport when rich countries get a bit short. It has nurses making choices inside unequal systems. These are choices that the vast majority of us would make if we were in similar circumstances.

The ethical argument is often handled as if a code can tidy it away - ‘we don’t actively recruit from Red List countries’ is, frankly, doublespeak.

Responsible recruitment codes matter; do not remove the central problem: a high-income country that fails to plan and turns its global recruitment vacuum on and off haphazardly causes damage far beyond its borders.

That is hard to square with the language of gratitude when it looks a lot more like asset stripping. No amount of International Nurses Day messaging fixes that.

To be clear, there is no fully correct version of the world map. I am willing to concede that any two-dimensional representation of a three-dimensional globe will have to make some compromises.

The Mercator version, which adorns our schools, offices and homes, was designed largely for navigation and does an exceptionally bad job of showing the countries of the world as they really are. Mercator’s projection champions direction, but it severely distorts land area, especially the further you go from the equator.

Greenland can appear similar in size to Africa. In reality, Africa is around fourteen times larger.

If we can misread something as basic as the size of countries in our world, we should be careful about the other assumptions we carry into global health and nursing.

Global nursing has its own maps. They are less visible, but they shape how we think. They tell us where knowledge sits, where expertise comes from, where problems are located and where solutions are expected to begin.

These maps are often wrong too.

Much of global nursing still works from assumptions that are rarely examined. Some countries are treated as natural sources of expertise. Others are treated as sites of need. Some nurses are invited to teach, but others are invited to describe hardship.

Distorted assumptions sit inside curricula, conferences, funding models, authorship patterns and the language of international work. They sit inside words that sound reasonable: support, capacity building, technical assistance, partnership.

Capacity building is a useful example. The phrase often starts from a deficit assumption. It implies that capacity exists in one place and needs to be transferred somewhere else. Sometimes that may be true, but it is rarely the whole story. Sometimes it is simply wrong.

Nurses working in low-resource settings, crisis-affected systems, rural services, overcrowded hospitals or politically unstable environments often develop highly sophisticated forms of judgement, improvisation, leadership and professional discipline.

Their expertise may not always be written up in the journals we read. It may not sit inside the regulatory frameworks we recognise. It may not be described using the language of quality improvement, implementation science or systems strengthening.

That does not make it less important.

A distorted map can be very misleading in ways that go beyond the relative size of each country. It makes some places look larger than they are, and in doing so, it makes others look smaller.

The same thing happens in nursing.

If UK nurses are educated mainly through UK examples, UK regulatory assumptions, UK workforce debates and UK models of practice, familiarity can start to look like universality. We may know that nursing is practiced differently across the world. That is not the same as understanding it. It is certainly not the same as allowing it to change how we think.

Global nursing should therefore begin with more humility.

It requires us to notice when one part of the world is treated as the default and another as the variation. It requires us to ask who sets the agenda before we praise the partnership. It requires us to be more careful about what we think we know before we arrive.

Weak global understanding leads to weak global engagement.

For those of us in Europe we can easily overestimate what we can offer and then underestimate what we need to learn. Too many times, we design interventions and programmes around our own interpretation of what we think we see and then wonder why they do not sustain after we have gone.

We use the language of partnership while retaining control of the money, the agenda, the authorship and the microphone. Calling it a partnership does not make it so.

For global nursing, humility should be a professional discipline.

Before we teach, we should ask what we have failed to understand.

Before we build capacity, we should ask what capacity is already there.

Before we describe a problem, we should ask who has already defined it differently.

Before we claim a partnership, we should ask who holds power when decisions become difficult.

The world map most of us learned has shaped what we see.

Global nursing should care because the same is true of the professional maps of the assumptions we still use