Article 09 · Health
Alert: The plague ship
What the hantavirus outbreak on the MV Hondius tells us about cruise ships spreading new diseases, and whether the WHO is right to call the global risk low.
A cruise ship with 150 people from 23 countries has been floating off the coast of West Africa for days, because nobody will let it come ashore. Three passengers are dead. The World Health Organization (WHO) says the risk to the public is "low". Cape Verde refused permission. Morocco refused to refuel a plane sent to evacuate patients. Spain's central government said the ship could dock in the Canary Islands, but the islands' president told Madrid he would still not accept it.
The MV Hondius is a disease story. It also tests every system that is supposed to work when a dangerous germ turns up on a floating international city. Those systems are breaking exactly where you would expect them to.
Why the ship's design matters
A cruise ship is a closed space where people mix, and the business model then sends those people all over the world.
The Hondius left Ushuaia, Argentina, on April 1 with 88 passengers and 59 crew. It stopped in Antarctica, South Georgia, Tristan da Cunha, St. Helena and Ascension Island. These are remote places with many kinds of wildlife, and passengers walked among the animals and among each other. They ate together and breathed air from the same systems. They also shared the small boats used for trips ashore.
Then the Andes strain of hantavirus began to kill people. The first to fall ill, a 70-year-old Dutch man, got a fever and headache five days into the trip. He was dead by April 11. His wife collapsed at Johannesburg airport on April 25 and died in the emergency department. A German passenger died of pneumonia on board on May 2. A British expedition guide is in intensive care in Johannesburg, and a former passenger tested positive in Zurich.
There are eight cases in at least four countries, and the ship has not yet finished its voyage.
No other civilian setting mixes people this closely and then scatters them this fast. An infected passenger can be sick in Cape Verde on Monday and in Zurich with no symptoms by Friday. After infection, hantavirus takes one to six weeks to cause symptoms. Every passenger who left the ship at any stop on the route could carry it onward, and none of them would be flagged for weeks.
Who gets hurt
The passengers who died and their families are hurt first, but the harm reaches further.
Health workers carry more than their share of the risk. The medical team on the Hondius had to care for very sick patients with only what the ship carried. They had no quick test for hantavirus and no set way to treat it, because no treatment or cure exists. When patients were flown to South Africa, hospital teams there were exposed to a virus that no doctor had on their list of possible causes until lab results came back from Johannesburg.
Port communities face a choice nobody prepared them for. Cape Verde has under 600,000 people. Its health system is not built to cope with a new disease arriving on an international cruise ship. Refusing to let passengers off made sense: a small island nation worked out what would happen if even one case spread into the community.
Crew members are the most trapped group in all this. Passengers had embassies to call on and ways to be evacuated. The crew are still on board. On cruise ships they usually live and work in tighter spaces than passengers, with shared cabins and common rooms, so keeping a sick person apart is close to impossible.
When ports say no
The Hondius is showing, live, a case our model marked as one of the hardest ways for things to go wrong: ports refusing entry to a ship with a confirmed outbreak.
Cape Verde said no. The ship sat at anchor off Praia while three passengers were flown by air ambulance to the Netherlands and Germany. Morocco refused to refuel one of those flights. The Canary Islands president publicly opposed Madrid's decision to let the ship dock in Tenerife. He said his government did not have enough information to guarantee public safety.
Spain's health minister said the ship would dock in Tenerife within days and that passengers who are not Spanish citizens would be sent home. The Canary Islands president said his region could not allow it. Two levels of the same government disagreed in public, on the news, about a quarantine ship.
That political split is predictable, and it will happen again. Regions that live on tourism have built-in reasons to turn away disease ships, whatever national health officials advise. The Canary Islands receive over 16 million tourists a year. Even a brief link to hantavirus is an economic risk, and a regional politician weighs it differently from a national health minister in Madrid.
A ship reports an outbreak and the nearest port refuses. The ship sails on, and the next port refuses or sets conditions. Every day at sea gives the virus more time to incubate. Each refusal makes the politics harder, and every delay shortens the time doctors have to respond. The Hondius spent days between countries while passengers stayed in their cabins and crew brought meals in protective gear.
Cooperation that held, barely
South Africa deserves specific credit. When the Dutch man's wife collapsed at Johannesburg airport, South African health officials started tracing the people on her flight. They confirmed hantavirus with a PCR lab test and shared the result with other countries within days. That result turned a "mysterious respiratory illness on a cruise ship" into a named virus whose way of spreading is known.
The WHO coordinated at least eight countries for this one outbreak. The UK was notified as the flag state, the country where the ship is registered. Argentina was the port of departure. South Africa confirmed the virus in its lab and cared for patients. Cape Verde was the port of refuge, and Spain is where the ship will finally dock. Switzerland had a case among passengers who had already left. The Netherlands and Germany were home to passengers and took in the evacuated patients.
It worked, barely. The response needed close to perfect work in the first week after the lab result, before passengers spread out faster than tracing could follow them. One Swiss case slipped through. That passenger went to University Hospital Zurich on their own, after answering an email from the cruise operator. The system worked with almost no room to spare.
What the model shows
Our model weighs the possible outcomes. In roughly 2 in 5 of them, the virus gets out of early control in a meaningful way. That means spread in communities beyond the people on the ship, started by passengers who passed symptom checks and flew home. It need not be a pandemic.
About 1 in 10 outcomes end with the global cruise industry facing a question about whether its business model can survive. Hantavirus itself is no special threat to the industry. The problem is that any germ serious enough to make ports refuse entry exposes a weakness in how the business is built, and no screening method can fully close it. You cannot run ships that mix people from around the world and also guarantee that a germ stays on board.
In the model, containment works best when two things happen at once. Countries must coordinate fast, which happened here. The hospitals receiving patients must also have enough spare capacity, which held, barely. Our model suggests both lined up well for the Hondius. Whether they would line up for a ship carrying 5,000 passengers instead of 150 is a very different calculation.
Our base case
We are moderately confident on this one. The Hondius outbreak stays contained. The biggest factor in favor of containment is how little the Andes strain spreads between people: only through close contact, and not through the air. The WHO's "low risk" call will probably prove correct for this event.
We are watching for the precedent this sets, more than for a hantavirus pandemic. The port refusals, and the public fights between regional and national governments while the ship drifted with nowhere to go, set a pattern for how the next outbreak plays out. The next ship might carry a respiratory germ that does not need shared cabins to spread.
The cruise industry carries roughly 30 million passengers a year across its global fleet. The Hondius carried 150 people, and it stretched the coordination of eight countries past what they could handle. Make that even ten times bigger, and the system the WHO only just held together would collapse under its own weight.
Cruise ships make an outbreak move faster than current health monitoring at sea can follow. On the Hondius, it took weeks, not days, to spot the outbreak. By the time anyone knew what they were dealing with, passengers were spread across continents.
Where we could be wrong
Our argument rests on two assumptions that might not hold.
First, we treat the Hondius as if it stands for the whole cruise industry, and it does not. It was a 150-person expedition ship visiting penguin colonies in Antarctica, with hikes through remote wildlife areas where rodent diseases live. That is a very specific kind of trip. A Royal Caribbean megaship with 5,000 passengers sailing loops around the Caribbean is very different. Those passengers are not walking through rodent territory, and those ports have real hospitals. The jump from "expedition cruise in the sub-Antarctic" to "all cruise ships make pandemics spread faster" may be bigger than we admit.
Second, we treat the port refusals as a preview of how the world will handle the next outbreak. The refusals came from particular places. Cape Verde has fewer than 600,000 people and limited health care. The Canary Islands fight is as much about Spanish regional politics as it is about disease. Clavijo, the Canary Islands president, has argued with Madrid for a long time, and this is the newest subject. Rotterdam, Miami or Sydney would probably not turn away a cruise ship in the same position. We may be drawing global conclusions from places whose geography made the problem look worse than it usually would.
The biggest weakness in our analysis concerns why this outbreak stays contained. The Andes strain spreads only through very close contact, such as sharing a bed or a cabin. It does not drift through the air the way flu does. If that is why everything worked out, our whole argument that cruise ships will spread the next pandemic depends on a different, worse virus that has not appeared yet. We are raising the alarm about something the Hondius does not show. The ship showed that international coordination is fragile and that politicians fight during health emergencies. Those are real findings. Whether it proved anything about cruise ships spreading pandemics is a bigger claim than the evidence supports.
What to track
- 01How fast gene sequencing comes back for the remaining suspected cases on board.Five cases are still unconfirmed. If any of them turns out to be a different strain, or shows mutations compared with the confirmed Andes samples, the picture of how the virus spreads changes completely.
- 02Whether the Canary Islands accept the ship when it reaches Tenerife.Clavijo's public opposition to Madrid's decision is the kind of political friction that grows once cameras arrive at the port. If Tenerife refuses at the last moment, the Hondius becomes a ship without a country again.
- 03How far contact tracing gets for passengers who have already left.The Swiss case came forward on their own, and we do not know how many others did not. With 23 nationalities and an incubation period of one to six weeks, tracing has to hold across countries that have no duty to cooperate and differ in their ability to do so.
- 04How the cruise industry responds on design standards for air systems (HVAC) and isolation.If the answer is "enhanced screening protocols" and crew training, nothing about how ships are built changes. If anyone starts talking about air systems split into separate sections and testing labs on board, the industry is taking the risk seriously.
We will also watch whether any of this is still in the news in two weeks.