Share this article on:

Impact of HIV on novel therapies for tuberculosis control

Sánchez, María Sa; Lloyd-Smith, James Ob; Porco, Travis Cc; Williams, Brian Gd; Borgdorff, Martien We,f; Mansoer, Johng; Salomon, Joshua Ah,i; Getz, Wayne Ma

doi: 10.1097/QAD.0b013e3282f7cb4b
Epidemiology and Social

Objective and design: The increased risk for tuberculosis in HIV-infected people has fueled a worldwide resurgence of tuberculosis. A major hindrance to controlling tuberculosis is the long treatment duration, leading to default, jeopardizing cure, and generating drug resistance. We investigated how tuberculosis is impacted by reducing treatment duration alone or combined with enhanced case detection and/or cure under different HIV prevalence levels.

Methods: Our model includes HIV stages I–IV and was calibrated to long-term tuberculosis and HIV data from Kenya. Benefits were assessed in terms of absolute and relative reductions in new tuberculosis cases and deaths.

Results: Compared with present-day strategies, at 3–20% HIV prevalence we attain a 6–20% decrease in incidence and mortality in 25 years when reducing treatment duration alone; benefits exceed 300% when combined with increased detection and cure. Benefits vary substantially according to HIV status and prevalence. Challenges arise because in absolute terms the number of infected people and deaths increases dramatically with increasing HIV prevalence, and because the relative efficacy of tuberculosis control policies displays a nonlinear pattern whereby they become less effective on a per capita basis at HIV prevalence levels greater than 15%. Benefits of reducing treatment duration may even be reversed at extreme HIV prevalence levels. Benefits of increasing cure versus detection increase as HIV prevalence increases.

Conclusion: Reducing tuberculosis treatment duration, alone or in combination with other control strategies, can provide enormous benefits at high HIV prevalence. Tuberculosis control policies need to account for HIV levels because the efficacy of different interventions varies substantially with HIV prevalence.

From the aDepartment of Environmental Science, policy and Management, University of California, Berkeley, California, USA

bCenter for Infectious Disease Dynamics, Pennsylvania State University, University Park, Pennsylvania, USA

cDepartment of Epidemiology and Biostatistics, University of California, San Francisco, California, USA

dWorld Health Organization, Stop TB Department, Geneva, Switzerland

eKNCV Tuberculosis Foundation, The Hague, The Netherlands

fDepartment of Infectious Diseases, Tropical Medicine and AIDS, University of Amsterdam, Amsterdam, The Netherlands

gNational TB Programme, Nairobi, Kenya

hDepartment of International Health, Harvard School of Public Health, Boston, USA

iHarvard University Initiative for Global Health, Cambridge, Massachusetts, USA.

Received 5 June, 2007

Revised 7 December, 2007

Accepted 9 January, 2008

Correspondence to María S. Sánchez, MS, PhD or Wayne M. Getz, PhD, DSc, Department of Environmental Science, Policy and Management, 137 Mulford Hall, University of California, Berkeley, CA 94720-3114, USA. Fax: +1 510 666-2352; e-mail:,

© 2008 Lippincott Williams & Wilkins, Inc.