Connect with us

NEWS

Phone Data Maps the Ebola Weeks Congo Missed

Phone records from three Ituri towns predicted Bundibugyo Ebola’s June map weeks before Congo declared the outbreak, now past 11,114 cases.

Published

on

Anonymized phone records of people leaving three Ituri towns already predicted which health zones would report Bundibugyo Ebola by 1 June, weeks before Congo declared the outbreak. The reconstruction, published on 12 November in Science, covers travel from Mongbwalu, Bunia and Rwampara between 24 April and 24 May.

The first confirmed patient had symptoms on 25 April, 20 days before the 15 May declaration. The European Centre for Disease Prevention and Control, using Congolese data up to 10 November, now lists 11,114 confirmed cases and 5,433 deaths in the Democratic Republic of the Congo, a crude fatality ratio of 48.9%.

Phone Records Predicted Which Zones Would Light Up

Dav M. Ebengo of the Institut National de Recherche Biomedicale in Kinshasa, with colleagues including epidemiologist Pierre Akilimali of the National Institute of Public Health and public-health professor Samuel Scarpino of Northeastern University, paired case records with mobility data to ask a blunt question. Did the flow of phones out of the first hotspots tell you, in advance, which health zones would have at least one confirmed Bundibugyo virus disease case by 1 June?

The paper says human movement was the main driver of early spatial expansion between 10 May and 1 June, when the number of health zones reporting positive cases rose by 12 and the number of health areas rose by 49. The authors also estimate a 10-day doubling time in that first phase, in line with genomic estimates of how fast the virus was spreading.

WHAT THE MODEL TESTED

  • Straight-line distance: Among places within about 100 kilometers of the first centres, farther zones were less likely to have a confirmed case by 1 June.
  • Travel time: Two towns at the same distance did not carry the same risk if one sat on a used road and the other did not.
  • Actual movement: Counts of people travelling from Mongbwalu, Bunia and Rwampara beat distance and travel time at recreating the early map.

Scarpino said the mobility feed “really tells us how many people are going between the different health zones.” The versions of the simulation that used those phone-derived flows came closest to the case records, the team found, and a shared dashboard let response staff use the results while the analysis was still being written.

The First Confirmed Patient Had Symptoms on 25 April

Congo’s health ministry described the index case as a nurse who died in a health facility in Bunia, the capital of Ituri, after fever, bleeding, vomiting and weakness. Laboratory work at INRB identified Bundibugyo virus. The Science study reports that the first confirmed patient was already ill on 25 April, which is the date that now sits under every later delay.

Africa CDC announced the outbreak on 15 May. Two days later the World Health Organization determined that the epidemic in Congo and Uganda was a public health emergency of international concern, citing infections among health workers, geographic spread and clusters of undocumented deaths. Africa CDC declared a Public Health Emergency of Continental Security on 18 May.

THE MISSED START

  1. 25 April 2026: The first confirmed patient already has symptoms, the Science paper reports.
  2. 24 April to 24 May 2026: Researchers measure anonymized phone flows out of Mongbwalu, Bunia and Rwampara.
  3. 15 May 2026: The DRC declares the outbreak after laboratory confirmation in Ituri.
  4. 17 May 2026: WHO declares a public health emergency of international concern.
  5. 10 May to 1 June 2026: Health zones with positive cases increase by 12 and health areas by 49.
  6. 1 June 2026: The date used to test whether April-May travel predicted confirmed cases.
  7. 12 November 2026: The Science paper is published; Congolese totals through 10 November stand at 11,114 cases.

Researchers have said the virus was circulating for weeks, and likely longer, before that May confirmation. Standard field tests were built around other ebolaviruses, and samples from Ituri had to reach Kinshasa. Scarpino has also pointed to a testing mix-up and to fighting that kept teams off roads. Tambe Akem, an independent public-health consultant in Yaoundé, called the phone findings a reinforcement of what responders already believed at the start: people were moving, and the virus was moving with them.

Infection Risk Rose Again 150 Kilometres Out

Close to the first centres, risk fell as distance rose, which is the pattern a local outbreak is supposed to show. Then the curve bent. Akilimali said infection risk rose again in places 150 to 200 kilometers away, because people used established routes into larger towns and cities rather than walking in a ring around the first graves.

Ituri’s gold-mining belt and its trade roads are not a side detail in that finding. Phones leaving Bunia, Mongbwalu and Rwampara were not wandering at random; they were following markets, pits and bus lines that already tied the northeast to North Kivu and, farther on, to Kisangani. A town with a bad road 30 miles away could stay quiet while a hub twice as far lit up, because more people actually went there.

Rachel Howard, senior technical emergency health adviser at the International Rescue Committee, put the limit of a dataset in one line: Congolese public-health staff “know things a dataset can’t show, for example which roads and markets move people between communities.” The phone model got the early map right by counting those trips. It still needed people who already knew which junction mattered on a given Tuesday.

Vodacom’s Masts Were Already Counting the Journeys

The Science paper did not invent the method in November. Flowminder, a Swedish nonprofit, had been turning Vodacom Congo call-detail records into anonymized Vodacom Congo movement maps for the response since the first weeks of the outbreak, building on a partnership the operator and the foundation have run since 2018 for other crises, including COVID-19, the 2021 Nyiragongo eruption and routine immunisation planning.

No individual-level file leaves Vodacom’s premises, Flowminder says. The product is an estimate of how groups of subscribers move between health zones, which is then treated as a proxy for how infectious people move. A 28 May rapid note tracked people who had been in Bunia, Mongbwalu and Rwampara. The 30 June update added Nyankunde and a North Kivu cohort, and ranked destinations by average subscriber-days.

FLOWMINDER’S JUNE CHECK

  • Top ten: All 10 highest-ranked destinations from the Ituri cohort had at least one confirmed case by 25 June.
  • Top thirty: 21 of those 30 zones (70%) had confirmed cases.
  • Ranks 31 to 60: Only 6 of those 30 zones (20%) were affected.
  • First hop outside Ituri: Katwa in North Kivu, then Butembo, Beni, Watsa in Haut-Uele and Makiso Kisangani in Tshopo.

The June check is the operational twin of the Science result. Antenna records were already sorting high-risk zones from the rest while official detection was still catching up to a nurse in Bunia. Raoul Kamadjeu, a UNICEF public-health specialist in Nairobi whose team has used the same Flowminder feeds, said preparedness work did happen in places flagged as high risk, and that limited resources still forced teams to spend most of their effort on zones that were already burning.

Bundibugyo Still Has No Licensed Vaccine

This is Congo’s 17th recorded Ebola outbreak and the largest Bundibugyo outbreak on record, now the second-largest Ebola outbreak anywhere. The 2018-2020 epidemic in North Kivu, Ituri and South Kivu, caused by Zaire ebolavirus, ran from 1 August 2018 to 25 June 2020 and reached 3,470 cases, of which 3,317 were confirmed and 153 probable. Responders then had a licensed ring-vaccination tool, Ervebo, built on the Zaire glycoprotein.

Bundibugyo virus is a different species. ECDC notes that there are still no licensed vaccines or specific treatments for Bundibugyo virus disease, which is why a 20-day detection gap did not meet a stockpile that could have been thrown around the first contacts. Previous Bundibugyo outbreaks, in Uganda in 2007 and in Congo in 2012, were small enough that a commercial vaccine never became a priority. Ervebo, Inmazeb and Ebanga all target Zaire ebolavirus. Some Ervebo doses have gone to Congolese health workers and into a ring-vaccination trial; they are not a licensed Bundibugyo shot.

Two Bundibugyo-specific candidates have now entered trials in the DRC. Médecins Sans Frontières has argued that wider use of Ervebo should not wait on those results.

Uganda contained a travel-linked cluster and declared its outbreak over after 42 days without a new confirmed case. Kenya has reported a cluster of nine cases, including two deaths, with seven recoveries and all remaining contacts off 21-day watch as of 11 November. Nigeria confirmed one case in a traveller from Uganda, later evacuated to Switzerland, and listed 62 contacts. Those imported clusters were finite. The Congolese map is not.

Eight Provinces Are Still Adding Cases

As of 11 November, with data through 10 November, eight provinces have reported confirmed Bundibugyo virus disease. Ituri still holds 69.5% of national cases. The latest daily increment in the ECDC wrap-up was 72 confirmed cases and 46 deaths, 44 of those cases in Ituri, 24 in Nord-Kivu, three in Haut-Uele and one in Nord Ubangi. Some 981 patients were in isolation and 3,031 people had recovered. Contact follow-up sat at 85.5% of identified contacts.

CONFIRMED CASES BY PROVINCE

Province Confirmed cases Deaths Health zones affected
Ituri 7,728 3,655 28 of 36
Nord-Kivu 2,761 1,523 17 of 34
Haut-Uele 499 216 7 of 13
Tshopo 98 32 7 of 23
Bas-Uele 13 4 4 of 11
Nord Ubangi 10 1 2 of 11
Sud-Kivu 3 1 1 of 34
Sud Ubangi 2 1 1 of 16
Total 11,114 5,433 67 of 178

ECDC says the Congolese figures remain under review and harmonisation. Neighbouring countries that share a land border are still classed as high risk because of trade and travel, and because many transmission chains were never documented. Several medically evacuated patients have been treated in Europe, including a British health worker moved to the United Kingdom on 5 November.

In Ituri, Only Nine Percent Were Known Contacts

A mobility model can tell a district officer which town is next. It cannot put a name on a contact list. On 12 November, Médecins Sans Frontières said the outbreak remained out of control and that surveillance, which is supposed to be the backbone of any Ebola response, was still failing in large parts of the map. In Ituri, only nine percent of Ituri patients treated by MSF teams had previously been identified as contacts. Over the past month, close to two-thirds of reported Ebola-related deaths happened in communities rather than in treatment centres.

Despite the vast amounts of money provided for the response, many frontline responders are left trying to contain the outbreak with empty hands and empty pockets.

Dr Javid Abdelmoneim, International President, Médecins Sans Frontières

MSF described unpaid and untrained staff, ambulances without fuel, surveillance teams that exist on paper, and community workers without phones or mobile data to send alerts. Insecurity in the east has cut roads and emptied clinics at the same time the phone maps have been pointing to the next junction. The Science study shows that those junctions were already visible in April. The 11,114 confirmed infections are what grew in the months when looking still waited on a laboratory, and when looking still cannot name most of the people who will sit in the next ward.

Disclaimer: This article is news reporting and analysis of a published epidemiology study and of official outbreak figures. It is for information only and is not medical advice, not a diagnosis, and not a guide to travel, vaccination or treatment. Anyone who may have been exposed to Ebola, or who has fever or other symptoms after time in an affected area, should seek care from a qualified clinician or local public-health authority and follow the health ministry and WHO guidance in force where they are. Case counts, death totals, contact-tracing rates and trial status are those published by the cited agencies as of the dates given above and will change as the outbreak continues.

Harry is the editor of THE iBULLETIN, an independent publication he owns and runs. He has been in journalism for ten years, first reporting and later editing, and much of what the site covers now begins in its inbox. Reader mail is read in full, every message of it. A tip is treated as a lead to be verified, not a story to be printed, and a challenge to a published fact is checked against the original filing, statement or transcript within the day, with the article corrected under a public policy if the reader is right. Questions that several readers ask become articles. That exchange feeds coverage of news, business and technology, of science and sports, and of entertainment, lifestyle, travel, auto and gaming, written for readers spread across many countries rather than one. Harry works from primary sources and checks each number himself before publication, and he would rather run a shorter story than an unconfirmed one. The address for all of it, tips, corrections and questions alike, is support@theibulletin.com.

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Trending