Published by Maney Publishing (c) Liverpool School of Tropical Medicine Risk factors for TB infection and disease in young childhood contacts in Malawi R. SINFIELD* ,{ , M. NYIRENDA*, S. HAVES*, E. M. MOLYNEUX* & S. M. GRAHAM* ,{,1 *Department of Paediatrics and 1 Malawi-Liverpool-Wellcome Trust Clinical Research Programme, College of Medicine University of Malawi, Malawi and { Liverpool School of Tropical Medicine, University of Liverpool, UK (Accepted March 2006) Abstract Background: Screening of children in household contact with smear-positive tuberculosis (TB) is universally recommended but seldom practiced in resource-poor settings. It has huge potential to reduce the burden of TB disease in children, particularly if streamlined to focus on those at greatest risk. Aims: To assess the prevalence of infection and disease amongst children aged (5 yrs in household contact with smear-positive TB. To identify which source case characteristics are risk factors for infection. Methods: A prospective, hospital-based audit was conducted over a 17-mth period in Southern Malawi. Smear- positive adults were identified and encouraged to bring their children to the outpatient clinic, in accordance with the national TB programme guidelines. Full assessment was performed, including tuberculin skin test. Results: 195 children aged (5 yrs who were contacts of 161 source cases were assessed. Prevalences of TB infection and disease were high (45% and 23%, respectively). The likelihood of a child being infected was significantly greater with increasing smear-positivity of the source case, and also if the source case were female (OR 2.25, 95% CI 1.19–4.27, p50.01). Conclusions: The high prevalence of TB infection and disease in child contacts attending this hospital-based clinic supports the current policy of contact-screening in Malawi. However, community-based studies are needed to provide a more accurate assessment of prevalence and risks for child contacts. Introduction Childhood tuberculosis (TB) is a common cause of morbidity and mortality in African children. 1–3 Contact screening and manage- ment has great potential to reduce the burden of TB disease in children. 4–7 Although recommended by the National Tuberculosis Control Programme (NTP) of Malawi 8 and the World Health Organization, 9 it rarely happens. 10 In a resource-poor setting, it might be useful if contact management were streamlined to focus on children at greatest risk of infec- tion. We aimed to assess the prevalence of infection and disease amongst children aged (5 yrs attending a hospital-based clinic and living in the same house as adults with smear-positive pulmonary tuberculosis (PTB), and to identify which source-case characteristics are risk factors for infection. Methods Over a 17-mth period, a prospective, hospital- based audit was conducted of children who were in household contact with adults with a diagnosis of smear-positive PTB. The Reprint requests to: Dr R. Sinfield, Department of International Child Health, Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, UK. Fax: z44 (0)151 705 3370; e-mail: rebecca.sinfield@liverpool.ac.uk Annals of Tropical Paediatrics (2006) 26, 205–213 # 2006 The Liverpool School of Tropical Medicine DOI: 10.1179/146532806X120291