Institutional members access full text with Ovid®

The cost effectiveness of antiretroviral treatment strategies in resource-limited settings

Bishai, Davida; Colchero, Arantxaa; Durack, David Tb

doi: 10.1097/QAD.0b013e328137709e
Epidemiology and Social

Background: Optimal resource allocation for antiretroviral treatment (ART) in developing countries requires assessment of different strategies for drug treatment and laboratory monitoring.

Objectives: To compare costs and outcomes for 10 000 simulated HIV-infected patients followed every 6 months for 10 years in a limited-resource setting.

Method: Five nested strategies, with and without the availability of a second-line treatment regimen, were simulated: (a) no ART (NO ART); (b) with ART but without any laboratory markers of HIV other than positive serology (ART ONLY); (c) ART plus total lymphocyte count (TLC); (d) ART plus CD4 cell counts (CD4); and (e) ART plus CD4 cell count plus viral load measurement (VL). Baseline prices of CD4 cell count and viral load measurements were $5.00 and $25.00 per test, respectively.

Results: With no second-line treatment available, treating 10 000 patients with ART ONLY compared with NO ART would cost $14.49 million [95% confidence interval (CI), 14.45–14.52] and would generate an additional 23 060 quality-adjusted life years (QALYS) (95% CI, 22 770–23 360) for a median incremental cost effectiveness ratio (ICER) of $628/QALY. Median ICER values per QALY for CD4 and VL strategies are $238 and $16 139, respectively, when second-line treatment is unavailable. With second-line ART available, the corresponding median ICER values are $8636, and $14 670.

Conclusions: In the absence of second-line ART, the CD4 strategy is a more cost-effective laboratory testing strategy for managing HIV infection than either TLC or VL. Availability of second-line ART plus CD4 cell count and/or viral load measurement would save additional lives, but at high incremental cost.

From the aJohns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA

bBecton Dickinson, Franklin Lakes New Jersey and Duke University, Durham North Carolina, USA.

Received 18 April, 2006

Revised 8 January, 2007

Accepted 19 January, 2007

Correspondence to Professor D. Bishai, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe St., Baltimore, Maryland 21205, USA. E-mail:

© 2007 Lippincott Williams & Wilkins, Inc.