Sputum smear-positive TB patients, diagnosed in the laboratory, who never start anti-TB treatment are classified as 'initial defaulters'. In Ridge Hospital, Accra, Ghana, there were 84 laboratory confirmed TB cases in 2009, of whom 32 (38%) were initial defaulters. Cure and default rates based on this cohort were 54% and 43% respectively, compared with rates of 87% and 8% when using the cohort based on 52 patients registered for treatment. This study highlights the problem of initial defaulters, and shows that programme performance may be poor when patients in laboratory registers are used as the cohort to evaluate treatment outcomes.
Copyright © 2012 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.