Tuberculosis treatment delay in adults and household transmission to children: a community-based study in a setting with high burden of tuberculosis and HIV

Rose, Penelope Cathryn

Tuberculosis treatment delay in adults and household transmission to children: a community-based study in a setting with high burden of tuberculosis and HIV

Master Thesis

2015

Publisher

University of Cape Town

Department

Department of Public Health and Family Medicine

Faculty

Faculty of Health Sciences

Abstract

Background: Tuberculosis (TB) control depends on interrupting transmission through rapid diagnosis and treatment initiation of infectious TB cases. With increasing delay in the diagnosis and treatment of pulmonary TB, disease is likely to progress, leading to progressive lung cavitation and increased sputum bacillary load, likely increasing TB transmission. This study investigated the effect of treatment delay in adult TB patients on the risk of TB infection and disease in child household contacts. Methodology: Secondary analysis was performed using data from a community-based household contact investigation study. Cross-sectional analysis was conducted of baseline data collected at enrolment. Children aged three months to fifteen years with documented household exposure to an adult with TB were enrolled between December 2007 and June 2012. These children were screened for TB infection (Mantoux tuberculin skin test [TST] and two interferon-gamma release assays [IGRA]) and disease. Total treatment delay was measured in adult TB source cases as the time from cough onset until treatment initiation, with those reporting no cough serving as the reference category. Logistic regression models were used to evaluate the effect of total treatment delay in adults on the risk of TB infection in child household contacts, with TB disease evaluated as a secondary endpoint. Results In total 671 children were enrolled as household contacts of 290 adult TB source cases. In multivariate analysis, the odds of TST positivity increased with cough duration ≥4 weeks prior to TB treatment initiation (odds ratio (OR) = 1.77 [95% CI 1.02-3.09] for cough <4 weeks; OR = 2.74 [95% confidence interval ( CI ) = 1.39-5.40] for cough 4-12 weeks; OR = 2.39 [95% CI = 1.19-4.82] for cough >12 weeks, compared to non-coughing adult TB patients), child's age ≥5 years (OR = 4.51, [95% CI = 2.60-7.83]), sharing the same bedroom (OR = 2.17, [95% CI = 1.43-3.31]), more than one household TB contact (OR = 2.70, [95% CI = 1.35- 2 5.42]) and with household tobacco smoke exposure (OR = 2.10, [95% CI = 1.22-3.61]). Adult TB source case HIV status did not modify the association between cough duration and risk of infection in children. Results of analyses of TB infection indicated by IGRA positivity were consistent with TST results. Prevalent TB disease in child contacts was associated with source case sputum smear and culture positivity, additional household TB contacts and decreasing age of the child. Conclusions: Delays of longer than four weeks from cough onset until TB treatment initiation were associated with increased risk of TB infection in child household contacts. These findings confirm the importance of reducing delays in TB diagnosis and treatment in adults to reduce transmission, ideally to less than four weeks. Although HIV co -infected TB patients are often considered less infectious, delayed treatment initiation remained associated with TB transmission, even amongst HIV co-infected adults with TB. In addition to the traditional risk factors for developing TB disease after infection, source case exposure factors also increased the risk of exposed children developing TB disease.