London School of Hygiene & Tropical Medicine

The 2018 rainy season in Yemen – which began in mid-April and typically runs until the end of August – could trigger another outbreak of cholera, with conservative estimates suggesting that millions of people may be at risk of infection if the disease continues to spread uninterrupted, according to new research published in The Lancet Global Health.

The observational study of 1.1 million suspected cases was conducted by the London School of Hygiene & Tropical Medicine, Epicentre, Paris, France, the World Health Organization and Médecins sans Frontières.

It helps identify the key factors driving the epidemic in Yemen for the first time. The results of the study have been used by Yemen’s health authorities and the World Health Organization (WHO) to help mitigate the risk of a new outbreak by increasing control efforts and prioritising which districts to distribute cholera vaccinations in before the rainy season begins.

While the rain began in mid-April 2018, there is a lag between the start of the season and the increase in cholera risk, and whether another outbreak begins will become clearer in the next few weeks.

Dr Anton Camacho from the London School of Hygiene & Tropical Medicine and Epicentre, Paris, said: “Due to the explosiveness of cholera outbreaks, reactive control strategies are much less likely to succeed due to limited resources, especially in the midst of a war. Our findings demonstrate the importance of planning during the lull period between cholera outbreaks.

“As a result of our findings, mitigation efforts are already in place to help reduce the impact of another outbreak. If the relationship between rainfall and cholera incidence is causal, it will be important to build on these efforts by continuing the vaccination campaign, and increasing water, sanitation, and hygiene (WASH) interventions during the rainy season.”

The current outbreak of cholera in Yemen began with a small outbreak in September 2016, followed by a larger outbreak in 2017. It is the largest documented cholera epidemic of modern times, with 1.1 million suspected cases reported so far.

The ongoing conflict in Yemen (which began in March 2015), means the population is extremely vulnerable. Three million people have been displaced, and 55% of health facilities are no longer fully functional. The war has also damaged the water supply infrastructure, meaning water is scarce, and prices have surged. The United Nations Office for the Coordination of Humanitarian Affairs estimates that half of the Yemeni population are in need of water and sanitation.

In the study, the authors used data from a national cholera surveillance system, which collected data from more than 2000 health structures across Yemen. This included suspected cases’ age, gender, location, date of illness, disease severity, whether their cholera was laboratory confirmed, and whether they died. This was combined with rainfall data, which was estimated using satellite imagery as many local weather stations have ceased to function since the beginning of the conflict in Yemen.

The authors analysed the association between cholera incidence and levels of rainfall over three periods – the first outbreak (from 28 September 2016-23 April 2017), the increasing phase of the second outbreak (from 24 April 2017-2 July 2017), and the decreasing phase of the second outbreak (from 3 July 2017-12 March 2018).

Overall, more than 1.1 million suspected cholera cases were reported between 28 September 2016-12 March 2018, including 2,385 deaths. Almost a third of cases (29%) were in children under the age of five years. Testing of the bacterial strains suggested that the same Vibrio cholerae O1 Ogawa strain circulated in both waves, but this needs further confirmation.

Using a modelling approach, the authors investigated the increasing phase of the second epidemic wave. They estimated that a weekly rainfall of 25mm was associated with a 42% increase in the chance of a person developing suspected cholera in the following 10 days, compared to a week with no rain.

The authors conclude that the first smaller outbreak in 2016 seeded cholera across Yemen during the dry season, and totalled 25,839 suspected cases. The outbreak was later exacerbated by the return of the rainy season in late April 2017 and peaked two months later when more than 50,000 suspected cases were reported in the first week of July. During the second outbreak, there were a total of 1.08 million suspected cases of cholera.

Over the first four weeks of the spring rainy season in 2017 (15 April-15 May), the daily national cholera incidence increased by 100 times (from 29 to 2900 daily cases), which led to the disease spreading across the whole country. The summer rainy season (July-August 2017) had a smaller impact on transmission, potentially due to increased levels of water, sanitation, and hygiene (WASH) interventions, which tripled between spring and summer.

Using the most recent figures (20 February-12 March 2018), the authors estimate that cholera transmission was still active in at least 11 highly populated, mostly urban, districts. They also modelled future cholera risk in 2018, and, based on their most conservative estimates, suggest that 54% of districts in Yemen are at risk of an epidemic, totalling a vulnerable population of more than 13.8 million people.

While the authors cannot confirm that the link between rain and cholera incidence is causal, they suggest that the use of unsafe water sources during the drought season, contamination of water sources during the rainy season, and changing levels of zooplankton and iron in water which help cholera bacteria survive, may have contributed to the increasing levels of cholera during the rainy season.

With the 2018 rainy season having started, the authors will continue to monitor suspected cholera cases over the coming weeks to determine whether another outbreak is likely to occur.

The authors note some limitations, including that their projections for an outbreak this year should be interpreted with caution, and that the association they have found should be reassessed to include data from this year if another outbreak happens. The use of aerial data to estimate rainfall is likely to have underestimated the amount of rain in Yemen, however this is not likely to affect the association between rain and cholera transmission.

This study was funded by Health Authorities of Yemen, WHO, and Médecins Sans Frontières

Publication

Anton Camacho, Malika Bouhenia, Reema Alyusfi, Abdulhakeem Alkohlani, Munna Abdulla Mohammed Naji, Xavier de Radiguès, Abdinasir M Abubakar, Abdulkareem Almoalmi, Caroline Seguin, Maria Jose Sagrado, Marc Poncin, Melissa McRae, Mohammed Musoke, Ankur Rakesh, Klaudia Porten, Christopher Haskew, Katherine E Atkins, Rosalind M Eggo, Andrew S Azman, Marije Broekhuijsen, Mehmet Akif Saatcioglu, Lorenzo Pezzoli, Marie-Laure Quilici, Abdul Rahman Al-Mesbahy, Nevio Zagaria, Francisco J Luquero. Cholera epidemic in Yemen, 2016–18: an analysis of surveillance data. The Lancet Global Health. DOI:10.1016/S2214-109X(18)30230-4