Research article

Analysis of risk factors for T. brucei rhodesiense sleeping sickness within villages in south-east Uganda

Thomas Zoller¹ , Eric M Fèvre² , Susan C Welburn³ , Martin Odiit⁴ and Paul G Coleman⁵

¹  Medizinische Klinik mit Schwerpunkt Infektiologie und Pneumologie, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany

²  Centre for Infectious Diseases, University of Edinburgh, Ashworth Labs, Kings Buildings, West Mains Road, Edinburgh, EH9 3JT, UK

³  Centre for Infectious Diseases, University of Edinburgh, Easter Bush, Roslin, Midlothian, Edinburgh EH25 9RG, UK

⁴  UACP, PO Box 25589, Kampala, Uganda. Formerly, Sleeping Sickness Programme, LIRI Hospital, PO Box 96, Tororo, Uganda

⁵  London School of Hygiene and Tropical Medicine, Disease Control and Vector Biology Unit, Keppel Street, London WC1E 7HT, UK

author email corresponding author email

BMC Infectious Diseases 2008, 8:88doi:10.1186/1471-2334-8-88

Published:	30 June 2008

Abstract

Background

Sleeping sickness (HAT) caused by T.b. rhodesiense is a major veterinary and human public health problem in Uganda. Previous studies have investigated spatial risk factors for T.b. rhodesiense at large geographic scales, but none have properly investigated such risk factors at small scales, i.e. within affected villages. In the present work, we use a case-control methodology to analyse both behavioural and spatial risk factors for HAT in an endemic area.

Methods

The present study investigates behavioural and occupational risk factors for infection with HAT within villages using a questionnaire-based case-control study conducted in 17 villages endemic for HAT in SE Uganda, and spatial risk factors in 4 high risk villages. For the spatial analysis, the location of homesteads with one or more cases of HAT up to three years prior to the beginning of the study was compared to all non-case homesteads. Analysing spatial associations with respect to irregularly shaped geographical objects required the development of a new approach to geographical analysis in combination with a logistic regression model.

Results

The study was able to identify, among other behavioural risk factors, having a family member with a history of HAT (p = 0.001) as well as proximity of a homestead to a nearby wetland area (p < 0.001) as strong risk factors for infection. The novel method of analysing complex spatial interactions used in the study can be applied to a range of other diseases.

Conclusion

Spatial risk factors for HAT are maintained across geographical scales; this consistency is useful in the design of decision support tools for intervention and prevention of the disease. Familial aggregation of cases was confirmed for T. b. rhodesiense HAT in the study and probably results from shared behavioural and spatial risk factors amongmembers of a household.