Evaluating temporal patterns of snakebite in Sri Lanka: the potential for higher snakebite burdens with climate change.

Dileepa Senajith Ediriweera,Shaluka Francis Jayamanne,Hithanadura Janaka De Silva,Geoffrey Kennedy Isbister,Nipul Kithsiri Gunawardena,David Griffith Lalloo,Peter John Diggle,Anuradhani Kasturiratne,Andrew Dawson,Arunasalam Pathmeswaran

doi:10.1093/ije/dyy188

Abstract

BackgroundSnakebite is a neglected tropical disease that has been overlooked by healthcare decision makers in many countries. Previous studies have reported seasonal variation in hospital admission rates due to snakebites in endemic countries including Sri Lanka, but seasonal patterns have not been investigated in detail.MethodsA national community-based survey was conducted during the period of August 2012 to June 2013. The survey used a multistage cluster design, sampled 165 665 individuals living in 44 136 households and recorded all recalled snakebite events that had occurred during the preceding year. Log-linear models were fitted to describe the expected number of snakebites occurring in each month, taking into account seasonal trends and weather conditions, and addressing the effects of variation in survey effort during the study and of recall bias amongst survey respondents.ResultsSnakebite events showed a clear seasonal variation. Typically, snakebite incidence is highest during November–December followed by March–May and August, but this can vary between years due to variations in relative humidity, which is also a risk factor. Low relative-humidity levels are associated with high snakebite incidence. If current climate-change projections are correct, this could lead to an increase in the annual snakebite burden of 31.3% (95% confidence interval: 10.7–55.7) during the next 25–50 years.ConclusionsSnakebite in Sri Lanka shows seasonal variation. Additionally, more snakebites can be expected during periods of lower-than-expected humidity. Global climate change is likely to increase the incidence of snakebite in Sri Lanka.

Highlights

Sri Lanka—a tropical island nation in the Indian ocean—is host to more than 100 terrestrial snake species
This study demonstrates a general approach to addressing the effects of variable survey effort and recall bias associated with epidemiological surveys
Previous local studies have reported seasonal variation in hospital admission rates due to snakebites, e.g. the peak incidence in admissions due to bites by Bungarus caeruleus in September–October[6] and by Daboia russelii in March–April and October–November.[7]

Summary

Introduction

Sri Lanka—a tropical island nation in the Indian ocean—is host to more than 100 terrestrial snake species. Sri Lanka is an agricultural country whose agricultural activities are mainly carried out in the ‘Yala’ and ‘Maha’ seasons.[14,15] Snakebite is considered as an occupational hazard; high incidences of snakebites have been reported during the periods of high rains and agricultural activity.[3,6,7,16]. Previous studies have reported seasonal variation in hospital admission rates due to snakebites in endemic countries including Sri Lanka, but seasonal patterns have not been investigated in detail. The survey used a multistage cluster design, sampled 165 665 individuals living in 44 136 households and recorded all recalled snakebite events that had occurred during the preceding year. Snakebite incidence is highest during November–December followed by March–May and August, but this can vary between years due to variations in relative humidity, which is a risk factor. Low relative-humidity levels are associated with high snakebite incidence

Methods

Results

Discussion

Conclusion