Coronavirus has begun , but in one of the world’s largest refugee camp complexes a worst-case scenario may have been avoided. A mathematical model run earlier this year by researchers at Johns Hopkins University – and now published following peer review – may have encouraged authorities to amplify measures that slowed the virus’s spread.
The  in Bangladesh, containing 23 separate camps, is home to about 600,000 Rohingya people who fled ongoing violence in neighbouring Myanmar. The population density surpasses that of many cities, with 46,000 people per square kilometre. The first case of coronavirus .
Months before the first cases appeared, Paul Spiegel at the Johns Hopkins Center for Humanitarian Health and his colleagues used what is called a “Susceptible Exposed Infectious Recovered” () model to simulate how the virus might spread in the camp under different levels of transmission, interventions and precautions.
Advertisement
Their simulation suggested that under a high transmission scenario with no interventions to slow the virus’s spread essentially all of the camps’ residents – 590,000 people – might have been infected within 12 months, with 2800 fatalities.
Spiegel says the authorities were already planning interventions, but were given fresh impetus to do so when they were presented with his team’s results. “We have been told that the scenarios were a wake-up call and had a positive effect,” he says. To date, just 66 people in the camps have been confirmed to have coronavirus – although Spiegel says it is likely that there are more cases that haven’t been confirmed through testing.
Measures to prevent and limit the spread of the coronavirus in the camps were implemented as early as March, says Catalin Bercaru, communications officer at the World Health Organization office in Dhaka, Bangladesh. Up to mid-July, the WHO trained 280 health facility staff and more than 1500 refugee community health work volunteers on how to deal with cooronavirus in the crowded conditions.
Bercaru says the WHO also supported the establishment of a testing laboratory in Bangladesh’s Cox’s Bazar district which now has the PCR machine capacity to run between 750 and 1000 tests per day. Additionally, the WHO has worked within the camp to ensure key health messages are shared regularly with the refugees, she says.
However, Mohammad Kamrujjaman at the University of Dhaka says that Bangladesh lacks the number and speed of testing needed to know how many cases there really are, both in its refugee camp population and nationally. Kamrujjaman says Spiegel’s model is sound, but he points out that statistics available from refugee camps are often poor, making them challenging to fit with SEIR or any other epidemic models.
“Since there’s no combination of real data and model prediction, it is very hard to see the real scenario,” he says.
Spiegel says there aren’t solid global covid-19 statistics for refugee populations, given that more than .
Nicole Bohme Carnegie, a biostatistician at Montana State University, says models are useful because they make it relatively easy to simulate the potential effects of implementing a variety of interventions. This can provide policymakers with critical information on the likely impacts of different approaches.
“I think this is under-utilised as an approach to prioritise interventions, especially when we are trying to find a delicate balance between economic, social and health impacts of covid-19 and quarantine and shutdown efforts,” she says.
“We are already doing similar studies with Johns Hopkins University in other refugee camps globally, including in Africa, Asia and the Middle East,” says Andrej Mahecic, spokesperson for the UN High Commissioner for Refugees (UNHCR). But he stresses that the studies are most useful for planning purposes. “It is important to note that these studies are not predictions, they are estimates of how the situation might unfold,” he says.
PLoS Medicine
