Ebola Is Spreading in Eastern Congo, and Nobody Is Watching

By mid-June, the outbreak in eastern Congo had passed 200 confirmed deaths and was nearing 900 confirmed cases — up from eight laboratory-confirmed cases a month earlier. The true toll is higher still: testing lags, tracers are reaching only about half of known contacts, and people die untested in villages and displacement camps that no surveillance team gets to.

This is the 17th Ebola outbreak the Democratic Republic of Congo has lived through, and already the largest ever caused by the Bundibugyo strain. Declared on 15 May, it has spread across more than thirty health zones in Ituri, North Kivu and South Kivu, with confirmed cases over the border in Uganda. There is no licensed vaccine for this strain and no approved treatment. The World Health Organization has declared the outbreak a public health emergency of international concern.

eGov Africa builds digital government systems in this region, and the honest answer on whether technology can help is: a little, in one specific place, and only if we are clear about everything it cannot do. The dangerous version of that answer — the one that gets funded and then collapses — is that an app will stop the outbreak. It will not.

A catastrophe almost nobody is watching

Notice how little of this reached you before today. A disease with no vaccine and no treatment is moving through a region of millions, and it loses the competition for attention to a World Cup and to whichever ceasefire is collapsing this week in the Middle East. Congo's own team is preparing to play in that tournament; one of its warm-up matches was quietly moved and played to an empty stadium. The cameras will follow the players, not the virus they left at home. The honest question is why we look away. Because it is Africa? Because Congo has been at war so long that another emergency reads as background noise? Because there is always a crisis closer to home with better pictures?

Pick whichever answer sits easiest. None of them survives a look at the map of where the modern economy gets its raw materials.

Eastern Congo is one of the most important mineral regions on earth. The DRC produces around seventy percent of the world's cobalt and holds most of its coltan — the ore that becomes tantalum, the metal inside the capacitors of nearly every phone, laptop, and electric vehicle. The provinces now burning with Ebola, Ituri and the two Kivus, are the centre of the country's tin, tungsten, tantalum, and gold trade. Mongbwalu, the town where this outbreak began, is a gold-mining town. Thousands of artisanal miners move between informal sites scattered across the region, and the roads that carry the ore carry the virus with them: mining traffic and cross-border trade are among the reasons contact tracing has fallen apart.

The minerals of the green transition and the path of an untreatable virus move along the same roads.

So the supply chain the green transition is built on runs directly through an outbreak in a war zone. If this gets worse — wider spread, harder lockdowns, deeper conflict — the disruption would not politely stop at Congo's border. Cobalt and tantalum are not commodities the rich world can wait out; they are inputs for the batteries and electronics the whole decarbonisation effort depends on. And yet the money is not following the risk: of a 319-million-dollar response plan, close to 290 million remains unfunded. The reason to care is that people are dying, and that should be enough. If it is not enough, here is the colder version: indifference to eastern Congo is a wager against the supply chain you are counting on.

The virus is not the hardest problem

Ebola is, in cold mechanical terms, a containable disease. You find the sick, you isolate them, you trace everyone they touched, you bury the dead safely, and you win the race against the chain of transmission. That playbook has worked in past outbreaks. It is failing here, and the reasons have almost nothing to do with the virus.

The epicentre is Mongbwalu, a mining town. Ituri is a commercial and migratory hub where people, goods and traders cross borders constantly. The WHO director-general put the consequence plainly: insecurity, attacks on health facilities and population movement have made it nearly impossible to trace contacts and isolate cases. Armed groups — the ADF, CODECO, the Rwanda-backed M23 — control or contest much of the terrain. Health workers have borne the brunt: the outbreak began as a cluster among hospital staff in Bunia, and by mid-June around seventy-five had been infected and seventeen had died, in a system that had too few of them to begin with.

So the binding constraint is not that the virus is mysterious or untreatable, though the absence of a vaccine makes everything worse. The constraint is that the people who carry it move, and the system meant to follow them cannot. By mid-June, tracers were reaching only about half of known contacts, against a target of ninety percent. Tedros put it without euphemism: “We cannot build community trust or isolate the sick while bombs are falling.”

The virus is the easy part to understand. The hard part is a population that has to keep moving to survive.

Quarantine assumes a savings account

There is also a quieter reason containment fails. Isolation assumes a person can stop moving for two or three weeks — stop trading, stop walking to the field, stop crossing the checkpoint to sell — and still eat. With close to ten million people facing acute hunger across the eastern provinces, a quarantine order with no food attached is not a health measure. It is a choice between two ways to die, and most people take the slower one.

That is also why the talk of ignorance gets the problem backwards. Some people do not understand the threat. Many understand it perfectly and break the rules anyway, because the rules were written by people who never had to choose between isolation and dinner.

The paper register already exists, and it already fails

This is not a hypothetical argument against a blank slate. At checkpoints today, staff already try to do exactly this on paper. Everyone who passes is meant to have their name written down, sometimes a temperature reading, sometimes a phone number. The intent is correct. The method cannot carry it.

The data falls apart at the moment of collection. Names are scrawled at the speed of a moving queue, and a large share come out illegible or incomplete. Temperatures get skipped when the line is long. Phone numbers are missing, wrong, or never asked for. So when a case is later confirmed, the register that should let responders warn everyone who stood at that checkpoint on that day is a stack of paper nobody can search, cross-reference, or call. Tracing back through it is not slow. It is impossible.

And that is on a quiet day. When thousands of people cross a single border post, a clerk with a notebook is not running a contact-tracing system — they are performing the ritual of one while the actual information evaporates as fast as it is written. Who was here, when, reachable how: all of it gone by the next shift.

A paper register at the volume of a border crossing is not a slow database. It is no database at all.

What a digital check-in layer actually does

So where does technology fit? Narrowly, and usefully, in one spot. The single function that is breaking down fastest is contact tracing — knowing who passed through where, and who was near a confirmed case. In a region of constant movement, that is exactly the kind of problem a digital register handles better than a clipboard.

The proposal is simple. People check in at checkpoints, markets and transit points using a phone-based register. If someone they were near later tests positive, the system can flag a possible exposure and tell them to seek testing and isolate. It does not diagnose, it does not treat, and it does not enforce. It does one thing: it makes an invisible contact chain visible faster than a team of tracers walking village to village in a war zone can.

Building this is not the hard part. We learned that running provincial tax systems in these same provinces — Ituri, the Kivus, Kongo Central. The software worked within months. The politics took seven years. The transferable lessons from that work are not technical, and they are the only reason a contact-tracing layer would have a chance.

The other lesson is colder. Protect the field agents, because their authorization papers do not protect them. The people who would run registration points in Ituri are working in the same terrain where health workers are being attacked: in Mongbwalu a safe-burial team was assaulted, and at points of entry five workers were taken hostage after being accused of spreading the very disease they were trying to track. Any honest plan budgets for their security first, not as an afterthought.

The reasons it could fail are already visible

Most e-government projects fail, and they fail in predictable ways — a gap between the world the designers imagined and the world the system lands in. The gaps here are not hard to spot, because we already hit them building tax systems in these provinces.

Connectivity, first. Our offline collection app in Ituri struggled with the basics: long initial connection times, dead zones across whole collection areas, broken data synchronization. A contact-tracing register that depends on a live connection will go dark exactly where the outbreak is hottest. It has to work offline and sync later, or it does not work at all.

Trust, second, and this one cuts the other way. In a region where armed groups contest control, a registry of who moved where is not a neutral object. It is a target. People have rational reasons to fear that a record of their movements could be used against them by whoever holds the checkpoint next month. That fear is not ignorance. It is correct risk assessment, and the design has to answer it — minimal data, local control, clear limits — or people will stay away.

And the elites. Outbreak rules, like tax rules, get evaded most reliably by the people with the power to evade them. A check-in system that the connected and the armed simply ignore becomes one more burden on the people who already comply. That is a political problem, not a technical one, and pretending otherwise is how these projects lie to their funders.

The easy twenty percent

None of this is an argument against building the tool. It is an argument against selling it as the answer. A contact-tracing layer, designed to protect the user rather than watch them, built to work offline, run by field staff whose security is funded — that is worth doing, and it could shorten the chain of transmission in exactly the place the current response is failing.

But it sits on top of a stack it cannot replace. Underneath it is community trust, which the WHO is right to put at the centre. Under that, food and cash, without which isolation is a fiction. Under that, treatment and safe burials. And at the bottom, the thing no app touches: humanitarian access, and the ceasefire that would allow it. Tedros said stopping transmission depends entirely on access. He is right, and no software changes that sentence.

The technology is the easy twenty percent. It was true for tax, and it is true for Ebola. The other eighty percent is whether anyone with power is willing to feed the people they are asking to stay home, and stop shooting long enough to let the tracers in. Build the app. Then go fight for the part that actually ends the outbreak.