Leptospirosis epidemics in tropical countries are often related to heavy rainfall and flooding.^1–4 The Indian state of Kerala has witnessed post-monsoon epidemics of leptospirosis in recent years. We investigated the relation between the pattern of daily rainfall and the incidence of disease in Calicut, North Kerala by plotting the number of confirmed cases having onset of illness on each day with the daily rainfall recorded for the district by the state meteorology department between July and October, 2002 (fig 1). The day of onset was calculated by subtracting the duration of fever from the day of admission. Disease confirmation was by positive microscopic agglutination test (titres above 1/100; 282 of the 340 suspected cases). IgM enzyme linked immunosorbent assay was also positive in 255 of the 282 cases.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

 Daily rainfall and leptospirosis incidence (disease onset) for the months July–October 2002.

The three peaks of disease onset were in August, September, and October. Each of these was preceded by heavy rainfall peaking about 7–10 days previously. Rainfall peaks were followed by rain cessation resulting in troughs in the graph. The baseline of the troughs preceded the disease onset peak by five to six days. The patients did not have direct occupational exposure to animals. Some 62.9% of the patients had either fissures or wounds on the feet.

Thus, periods of heavy rain followed by days of little or no rain seemed to be the setting for leptospirosis epidemics in this part of the world. Most cases seemed to occur by cutaneous exposure of the legs while walking in stagnant water or moist soil. This implied that leptospira multiplied in the walking paths where water remained undrained for a period of two to three days after the rains was responsible for most cases. Most people in the state wore rubber chappals during the rainy season, which offered little protection against possible infection.

Climatic and environmental factors were probably responsible for epidemic leptospirosis seen in Kerala in the recent years. The pattern of rainfall has changed in the western ghats region of India—which includes Kerala—in the past century, with more rainless days during the monsoon months.⁵ There had also been rapid urbanisation and construction activities in the past two decades, resulting in blockage of natural drainage of rainwater and consequent water logging near human habitats.

If our hypothesis is correct, future epidemics of leptospirosis can be anticipated by studying daily rainfall patterns. The thrust of community action can then be oriented towards improved water drainage and if necessary by disinfection or salination of water logged walking paths and wearing of effective protective footwear.