An outbreak is beaten one house visit at a time

Congo’s health ministry said on Sunday that the Ebola outbreak had reached Nord-Ubangi, an eighth province, with two confirmed cases, one of them fatal, both men in their twenties¹. Authorities have recorded at least 9,400 cases and more than 4,500 deaths since the outbreak was declared in mid-May, the deadliest in the country’s history. It is caused by the Bundibugyo virus, for which there is no approved vaccine or treatment¹. Reuters, citing the latest situation report, put the total at 9,415 cases and 4,536 deaths in the 24 hours to 16 October, with 67 new confirmed cases that day, 44 of them in Ituri³.

That is the headline, and it is grim. It is also the least useful number in the story. A province is a line on a map. An outbreak is beaten, or not, in smaller units: a sample in a laboratory, a contact visited at home, a burial done safely, a health zone that goes six weeks without a case.

Read the situation report, not the map. The ministry’s daily report for 14 October, published the next day, is a dull document in the best sense². Laboratories analysed 506 samples, against 394 the day before, and the share testing positive fell to 12.8% from 17.3%². More tests and a lower share of positives is the combination you want: it suggests the net is being cast wider, not that the virus is being found less because fewer people look. Of 4,048 alerts received, 98.5% had been verified, and every one of the 612 validated suspect cases had been investigated².

Then the uncomfortable lines, printed just as plainly. Of 33,264 contacts to be followed, 25,956 were seen in the previous 24 hours: a follow-up rate of 78.0%, down from 81.8% the day before and below the required threshold of 95%². Only 242 of the 612 suspects, 39.5%, had been transferred to transit or treatment centres². Twenty-two of the day’s 37 confirmed deaths happened in the community rather than in a treatment centre². And 269 front-line workers have been infected since the outbreak began². Those are the gaps an outbreak travels through, and the report names them itself. A system that publishes its own shortfall every day is a system that can be held to fixing it.

Shrinking pockets, a wider map. The World Health Organization’s regional director for Africa, Mohamed Yakub Janabi, put it in one sentence on Friday: “We see pockets of Ebola generally going down, but unfortunately geographically it’s spreading out”¹. Both halves are in the data. Twelve of the 65 health zones touched so far had gone at least 42 days, two full incubation periods, without a new confirmed case, and three more were between 22 and 42 days². Yet the newest provinces sit along the border with the Central African Republic, where armed violence has forced thousands to flee and roads and phone networks are poor¹. Distance is the virus’s ally. Logistics is the response’s.

The authorities in North Kivu, where the WHO says substantial transmission continues¹, are answering with logistics. From today, people coming from Ituri may not enter the province for a week. Motorcycles may carry only the driver and one passenger, taxis four. Burials are limited to 20 family members, and checkpoints must screen temperatures and provide hand-washing. Today is declared a “Stop Ebola Day”, with only authorised services at work⁴. None of this is glamorous. Most of it is enforceable. That is the point.

What was caught, and where. In Kenya an eighth case was confirmed on Sunday: a 49-year-old patient attendant, likely exposed to the country’s first case, who was tested after developing symptoms in quarantine. He was one of 38 people tested that day, and the six other living patients were stable⁵. A case found inside quarantine is the system working: the chain is visible and the contacts are already known. A case found in a market is the system being surprised. Kenya’s cases, so far, are the first kind.

Why this leads on a Monday in Berlin. Because the same lesson keeps arriving from different desks this morning. Oil is moving through a contested strait on escorts and detours rather than a treaty. A protest movement in France is being answered, or not, by a list of meetings and a payroll of substitute teachers. What decides the outcome is rarely the announcement. It is the follow-up.

Seventy-eight per cent is not good enough, by the response’s own standard. It is also a number, published daily, against a threshold, published daily. That is what accountability looks like before anyone holds a press conference about it. Watch it climb, or not. It will tell you more about this outbreak than the next province on the map.