Climatic variations influence the emergence of cholera in Africa

Global climate change has for several years been contributing greatly to the spread of cholera through associated increase in the frequency of torrential rain, floods and periods of drought. It is now established that the spread of zooplankton which harbours the Vibrio cholerae bacterium follows that of phytoplankton, whose growth is directly related to climate variations.

However, a host of factors act on the climatic conditions and they are difficult to study. Certain parameters vary depending on the regions of the world whereas others act on the global scale. The interactions between the climate and emergence of cholera must therefore be studied region by region. Research has been ongoing in Bangladesh and also in South America for many years, but up to now few studies have been conducted in Africa. Yet it is on that continent that the public health situation is giving the most cause for concern.

A study published by research scientists of the Laboratoire de Génétique et Evolution des Maladies Infectieuses (GEMI), mixed research unit IRD/CNRS (2), is the first to yield evidence of the correlations between the outbreak of cholera epidemics and climatic data in 5 West African countries (Togo, Ivory Coast, Ghana, Benin and Nigeria). The research team set up an epidemiological database founded on cases recorded by the WHO over a 20-year period, between 1975 and 1995, in each of these countries. Comparison of these figures with parameters of local and global climate variations showed the factors particularly involved to be the volume of rainfall and the Indian Oscillation Index (IOI), an indicator of the global climate variability constructed from variations in atmospheric pressure in the Indian Ocean. Values of this index lower than –1 showed an association with a warm event such as an increase in sea surface temperature. In contrast, values above +1 coincided with cold events.

The annual rainfall regime and the IOI act on the aquatic environment in which Vibrio cholerae develops (estuaries, sea shores, river beds and so on). In the wild, the cholera bacillus lives in contact with small aquatic crustaceans, copepods, are a component of the zooplankton. These microscopic animals, which constitute the principal reservoir of the bacterium, feed on phytoplankton. They therefore have a tendency to congregate in the zones where the density of microscopic algae is highest. This relationship is fundamentally important. It provides a means to monitor the areas rich in plankton by remote sensing, and therefore to detect from space the potential reservoirs for Vibrio around the coasts.

Outbreaks of cholera appear irregular and, in order better to understand the epidemic pattern dynamics, the GEMI researchers used an adapted statistical tool that favours a wavelet analysis method. This novel process allows comparison of the frequencies of epidemic outbreaks with a range of climatic or environmental parameters (climate variability index, volume of precipitation, phytoplankton concentration near the coasts). This approach also takes into account the random variation of frequencies of emergence of epidemic foci.

The research team thus successfully linked the number of new cases of cholera first to the global climate variability index and then to monthly rainfall readings between 1989 and 1994. For that period, a frequency of epidemic occurrence of 2 to 3 years was indicated for the countries studied, except for the Ivory Coast. A significant correlation was also observed between the IOI and the annual rainfall regime for these same four countries. Furthermore, analysis of the interannual variability of rainfall between 1975 and 1996 indicated the existence of a 3 to 5–year long cycle in the appearance of the disease for the whole of the areas covered by the study. The IOI and volume of precipitation are therefore two climatic variables that appear to be strongly correlated with the appearance of epidemic foci of cholera. The latter usually develop during seasonal periods but their rhythm period can also exceed the annual cycle (between 2 to 5 years). In other words, indirect relationships between climatic variations or variations in rainfall volume and emergence of foci of infection can persist for several years. These results agree with those obtained previously in Bangladesh and South America.

Over the coming years, the results of this research work should contribute to the creation of an early warning system that takes climatic parameters into account for prediction of cholera epidemic dynamics. That should facilitate the organization of prevention actions, such as drinking water filtration schemes, and the planning of care provision for people by supplying medical kits and rehydration kits. This kind of approach could also be applied to the understanding and prevention of other climate-sensitive illnesses such as malaria, dengue and other vector-borne diseases.

Grégory Fléchet -IRD