The HIV pandemic continues to have the greatest impact in sub-Saharan Africa with an estimated 22.5 million adults and children living with HIV [www.UNAIDS.org. (Accessed 1 August 2008)]. It is a public health crisis for the world's poorest countries causing major social and economic disruption [www.UNAIDS.org. (Accessed 1 August 2008)]. In the developed world, highly active antiretroviral therapy (ART) in the late 1990s brought significant improvement in the quantity and quality of life and is a very cost-effective intervention [1,2]. Provision and scale-up of ART in resource-limited settings is also feasible and cost-effective [3–9]. In the developing world, the initiation of ART is guided by the patient's CD4 cell count with a level of 200/μl or less considered appropriate for therapy and in some settings a CD4 cell count of 350/μl or less. Where reliable CD4 cell count testing is not available, WHO clinical staging of HIV infection is recommended [10–12,13]. Symptomatic stages 3 and 4 indicate advanced immunosuppression or AIDS [10–12]. Studies comparing the reliability of WHO staging to CD4 cell counts for determining advanced immunosuppression indicate that as many as 70% of individuals are not correctly identified at screening [11,14–16]. This may have significant cost and benefit implications for ART programs in resource-limited settings. New low-cost CD4 cell count tests have also been studied in this setting and found to have similar sensitivity to conventional flow cytometry methods [17–21].
Utilizing an economic evaluation approach, we developed a Markov lifetime model of HIV infection and compared the direct healthcare costs and benefits in life years or quality-adjusted life years (QALYs) gained using routine and low-cost CD4 cell count versus WHO clinical staging of HIV to initiate ART. We estimated an incremental cost-effectiveness ratio (ICER) in US dollars per life year or QALY gained.
Markov state transition model
A Markov state transition probability model, following a hypothetical cohort of 10 000 HIV-infected individuals starting with a CD4 cell count more than 350 μ/l, was developed comparing the two approaches to guide initiation of ART (Figs 1 and 2). The model incorporates costs, survival and quality of life and is derived from the published natural history studies of HIV cohorts prior to the advent of ART, indicating means and 95% confidence intervals for incubation time to development of AIDS or death [22–29]. The sensitivity of WHO clinical staging is derived from published studies in this setting using absolute CD4 cell count by flow cytometry as the reference [10,14,15].
Patients in the model are reviewed annually for consideration of ART by either CD4 cell count of 200/μl or less or by WHO stage 3 or 4 criteria. We allowed for reversibility so that individuals may recover on ART at a probability based on previous studies and re-enter the model with a CD4 cell count of more than 350/μl [3,2]. A small proportion will progress despite commencement of ART and die in the first 90 days [4,30,31]. The relative effectiveness of ART is derived from large cohort studies undertaken in resource-limited settings. It closely corresponds to outcomes observed in developed nations [3,30,32–36] and incorporates failures related to adverse effects of therapy and poor adherence with treatment over a prolonged period of time.
To clearly present the importance of CD4 cell count testing in guiding initiation of ART, we did not incorporate the cost and effects of specific opportunistic infections, but instead utilized summary healthcare utilization and costs derived from primary data in sub-Saharan African settings [3,5,7,32]. The cost and effect of HIV viral load testing was not evaluated, given that this is unlikely to be available in most routine resource-limited settings [23,24,37,38]. In the sensitivity analysis, we examined the cost and effect of changing to second-line ART therapy after 10 years.
The transition probabilities for HIV health states, the sensitivity of WHO clinical staging compared to CD4 cell count testing and the relative effectiveness of ART are shown in Table 1.
A cohort of 10 000 HIV-infected individuals are followed for a total of 20 years. The effectiveness of ART has been demonstrated over at least 10 years and it is believed that some additional benefit is conferred beyond this time. We chose 20 years to follow at least half the cohort to death.
We undertook the perspective of the public health services in a sub-Saharan African setting. Wherever possible, health-related costs and health service utilization for HIV were derived from primary sources in sub-Saharan Africa: the Republic of South Africa (RSA) [3,5,7,32] where recent data are available and from Cote D'Ivoire . This is also included in the sensitivity analysis examining variations in parameters and generalizability.
Treatment costs, CD4 lymphocyte count test correlation and discounting
The costs of treating patients with HIV include the costs of providing ART and other direct healthcare treatment costs such as treating opportunistic infections, monitoring and treating adverse effects of therapy. The costs of first-line and second-line ART are derived from referenced sources for developing countries [5,6,39]. The cost of standard CD4 cell count testing is based on several sources [5,6,17,39]. Low-cost CD4 cell test prices are based on studies that have demonstrated acceptable correlation to reference methods tested in field conditions in resource-limited settings [17,18]. In the base case analysis, future costs and outcomes were discounted at 3% per annum as recommended in a developing country setting .
Healthcare utilization, related costs and CD4 cell count test correlation are shown in Table 2.
Health-related quality of life
Health-related quality of life analysis has expressed outcomes in terms of life years and QALYs gained. QALYs are calculated by weighting life expectancy by the health-related quality of life (HRQoL) spent in any given health state. The utility values used to calculate QALYs were taken from a survey of South African patients at different stages of HIV disease who completed a SF-36 questionnaire [41,42] or the EuroQol (EQ-5D) [43,44]. Both are widely validated instruments that describe health states in terms of multiple dimensions [3,5,43,45]. Wherever local valuations for different health states were not available, utility scores were also derived from a US population survey of a cohort of HIV-infected individuals using the time trade-off method .
The utility values for the stages of HIV infection are shown in Table 2.
The cost-effectiveness threshold used in this analysis is based on an ICER per QALY gained below the per capita gross domestic product (GDP) for the RSA and Cote d'Ivoire, as recommended by the Commission on Macroeconomics and Health [47,48].
Sensitivity analysis for uncertainty
Sensitivity analysis is subject to uncertainty in terms of the choice of analytic methods, the model structure, parameter variability and generalizability across different settings. Using the responsiveness to change of the ICER for the base case scenario, across a range of variables, the uncertainty of the cost-effectiveness estimate can be assessed.
Uncertainty in the analytic methods has been assessed through a deterministic one-way sensitivity analysis comparing the base case with varying time horizons from 20 to 10 years, given the known effectiveness of ART to at least 10 years. In the base case analysis, future costs and outcomes were discounted at 3% per annum  and then compared with 6% often used in a developed country setting. The cost-effectiveness threshold used in this analysis was GDP per capita for two sub-Saharan African nations for which reliable primary cost and effects data were available.
The assumptions in the model such as the transition probabilities between health states were drawn extensively from large cohort studies in the HIV literature. The sensitivity of WHO clinical staging of HIV compared to CD4 cell count testing was varied to a possible maximum estimate to examine the effect on the ICER. The effectiveness of ART was also varied across a plausible range.
Parameter variability has been assessed by deterministic one and multiway sensitivity analysis comparing the base case with plausible changes to the sensitivity of low-cost CD4 cell testing compared to routine flow cytometry as the reference method. The cost of the routine CD4 cell test was also varied to a maximum in the reported range from several sources to assess the effect on the ICER.
In order to assess the possible effect of the emergence of widespread ART resistance after 10 years of first-line therapy, we examined the effect on the ICER of implementing second-line ART after 10 years in the model.
The generalizability of the results to varying settings has been assessed by varying the healthcare costs through a feasible range derived from the two countries studied. We also examined the effect on the ICER, given a hypothetical ‘worst case scenario’ with a maximum sensitivity of WHO staging, a minimum sensitivity for the low-cost CD4 cell testing, a maximum cost for CD4 cell testing and also switching to second-line ART after 10 years.
Base case analysis
In the base case analysis, the total costs and effects, in terms of life years and QALYs for individuals assessed by WHO clinical staging only for eligibility for ART over 20 years and applying a discount factor of 3%, were $8351, 10.32 and 9.08, respectively (Table 3).
For those individuals assessed by annual routine CD4 cell count testing, the costs, life years and QALYs were $8543, 10.54 and 9.29, respectively. This produced an ICER of $884 per life year gained and $939 per QALY gained. When the low-cost CD4 cell test was used, the cost reduced to $8368 and produced an ICER of $80 per life year gained and $85 per QALY gained.
For sensitivity analysis, see Table 4.
Methods: time horizon and discounting
Reducing the time horizon to 10 years resulted in a small increase in the ICER for QALYs gained of 2.6% using routine CD4 cell testing. When low-cost CD4 cell tests were used, the ICER showed a substantial cost-saving compared to WHO staging with –$311 per QALY gained.
When the discount rate was increased to 6% per annum, the ICER did not change significantly for routine CD4 cell testing. For low-cost testing, there was a very significant 97% reduction from $84.5 to $2.5 per QALY gained.
Model: sensitivity of WHO clinical staging and effectiveness of antiretroviral therapy
When the sensitivity of WHO clinical staging was increased from 50 to 75%, the ICER per QALY gained reduced slightly, 9–10% for routine CD4 cell testing. For the low-cost CD4 cell count testing, there was a very significant increase in the ICER per QALY gained of over 600%, or from base case of $85 to $621 per QALY gained.
When the effectiveness of ART was minimized to an estimate of 44%, the ICER per QALY gained decreased slightly by 4% for routine CD4 cell testing and by 25% for low-cost CD4 cell testing.
Parameters: sensitivity of low-cost CD4 cell count test, cost of CD4 cell count test and second-line antiretroviral therapy for 10 years
When the sensitivity of the low-cost CD4 cell testing compared to routine flow cytometry was minimized to 80% agreement, the ICER per QALY gained increased significantly by 500%, to $509 from the base case of $85.
Comparing the cost of routine flow cytometry to that of the low-cost CD4 cell count test, the ICER per QALY gained was a 10-fold, or 90% decrease from $939 to $85.
When first-line ART was further discounted to 2009 estimates, the ICER per QALY gained was further reduced by 67% or a decrease from $939 to $304.
When second-line ART was included for the cohort after 10 years, the ICER per QALY gained decreased by 16% for the routine CD4 cell test, but substantially increased by 700% for the low-cost CD4 cell count testing group from $85 to $709.
Generalizability: healthcare costs and ‘worse case scenario’
When the healthcare costs of inpatient and outpatient care were varied from those of the RSA to the Cote D'Ivoire, the ICER per QALY gained increased significantly. The increase for routine and low-cost CD4 cell testing was 131 and 1441%, respectively compared to the WHO staging approach.
When a multiway ‘worst case scenario’ was developed combining increased sensitivity of WHO staging, reduced sensitivity of the low-cost CD4 cell testing, increased CD4 cell cost and switching to second-line ART after 10 years, the ICER per QALY for the CD4 cell test approach compared to the WHO staging approach decreased from the base case by 50%.
The objective of an economic evaluation in healthcare is to better inform decision-making about the cost and benefits of a new intervention compared to an existing level of care. ART in the developed world has brought significant improvement in the quantity and quality of life for people infected with HIV. Increased global funding and reduced drug costs has meant that the provision and scale-up of ART in resource-limited settings is both feasible and cost-effective [www.UNAIDS.org. (Accessed 1 August 2008)] [3–9,39]. WHO guidelines have recommended that the initiation of ART is guided by the patient's CD4 cell count of 200/μl or less, consideration at CD4 cell count of 350/μl or less, when this test is available, or alternatively clinical staging of HIV infection symptomatic stages 3 and 4 (AIDS). The poor sensitivity of clinical staging, however, means that as many as 70% of eligible individuals are not correctly identified at screening [11,14–16].
Using a Markov lifetime model, we have shown that routine CD4 cell count compared to WHO clinical staging to guide initiation of ART has clear benefits in terms of improved quantity and quality of life and appears to be a very cost-effective intervention. The base case estimated an ICER of $939 per QALY gained. Low-cost CD4 cell tests have also been studied in this setting and found to have similar sensitivity to conventional flow cytometry methods [17–20]. Compared to WHO staging, this approach appears to be extremely cost-effective with an ICER of $85 per QALY gained. Both approaches are well below the cost-effectiveness thresholds of GDP per capita for the two sub-Saharan African countries studied [49,50]. Further development of this model at higher starting CD4 cell count may also be valuable.
Using one and multiway sensitivity analyses across a range of plausible values, the ICER per QALY gained remained well below the cost-effectiveness threshold. The ICER for routine CD4 cell testing was most sensitive to changes in the sensitivity of WHO staging, the cost of CD4 cell testing, the cost of first-line ART and the valuation of healthcare utilization. The last of these was related to the proportionately higher cost for outpatient compared to inpatient visits between the countries studied.
The ICER for low-cost CD4 cell testing was most sensitive to changes in sensitivity of WHO staging, sensitivity of CD4 cell testing, cost of the CD4 cell test, change to second-line ART after 10 years and the valuation of healthcare utilization. When we developed a ‘worst case scenario’ combining several of the following variations – reduced sensitivity of CD4 cell testing, increased sensitivity of WHO staging, a maximum cost of CD4 cell testing and switching to second-line ART after 10 years – this still produced a net decrease of the ICER by 44% or $527 per QALY gained.
This analysis has many limitations. The methods and perspective undertaken could have been varied more extensively. The choice of the time horizon is based on the known effectiveness of ART followed to 10 years. We chose to set the base case at 20 years and follow at least half the cohort through to death, assuming that ART remained effective. Drug resistance may emerge after 10 years and longer term studies are required. The discounting rate of 3% was varied to 6% in the sensitivity analysis and this produced an even more attractive ICER/QALY gained for both CD4 cell testing approaches.
We did not include indirect costs outside healthcare costs in this analysis, but we would anticipate additional cost-effectiveness if HIV-infected individuals regain productivity on treatment.
Our HIV transition model structure was simplified to assess the importance of CD4 cell testing in guiding initiation of ART. The transition probabilities across the health states and all-cause mortality used in the model are derived from long-term HIV cohort studies, but the estimates are mean values from a range. Re-running the model across further plausible ranges of values may further refine the model.
We assessed the cost and effects of annual CD4 cell testing only, often utilized in many ART programs , but more frequent testing would require further evaluation. The relative effectiveness of ART is derived from large cohort studies undertaken in resource-limited settings and closely corresponds to outcomes observed in developed nations [3,30,32–36]. These studies incorporate treatment failures related to adverse effects and poor adherence over a prolonged period of time. Our model did not specifically accommodate the additional complexity of varying patient adherence over prolonged periods of time, which may result in reduced effectiveness and drug resistance. In the sensitivity analysis, we examined the cost and effect of changing to second-line therapy in the entire cohort after 10 years. Surprisingly, this did not appear to significantly increase the ICER for routine CD4 cell testing, but did increase that for the low-cost CD4 cell approach. This overestimates the actual switch rate to second-line therapy, often reported in the range of 25% in the first 2 years .
We did not incorporate the cost and effects of specific opportunistic infections, but instead utilized summary healthcare utilization and costs derived from primary data in sub-Saharan African settings [3,5,7,32]. The cost and effect of HIV viral load testing was not evaluated, given that this is unlikely to be available in most resource-limited settings [24,25,37,38].
In terms of parameter uncertainty, we used one and multiway sensitivity analysis to vary both the sensitivity of the low-cost CD4 cell test compared to that of routine flow cytometry to a possible minimum value in the reported range. We also assessed the effect of varying costs of CD4 cell testing and this did not significantly change the cost-effectiveness estimate. Ideally, a probabilistic approach to sensitivity analysis could provide a more robust measure of uncertainty across a range of parameter values. Generalizability was assessed by varying the healthcare costs from two different sub-Saharan African countries where reliable costs were available. We developed a ‘worst case scenario’, which still produced a net decrease of the ICER/QALY by 44% compared to the base case or $527 per QALY gained.
Given the current healthcare cost valuation, utilizing CD4 cell testing to guide initiation of ART appears to be very cost-effective even when using ‘worst case scenario’ estimates. With further efforts to source discounted ART  and develop low-cost reliable CD4 cell testing systems for resource-limited settings, this cost may reduce even further, making this approach a more attractive intervention to nations with even lower GDP per capita. There is emerging evidence from developing nations that applying the criteria and guidelines for treatment of HIV infection from the developed world is highly cost-effective [3,32,37,52–54]. Identifying individuals at earlier stages of HIV infection necessitates the availability and routine use of CD4 cell testing. One of the major obstacles to implementing widespread CD4 cell count testing for ART programs in sub-Saharan Africa remains the lack of technical expertise  and some of the newer low-cost systems supported by comprehensive quality assurance may go some way to solving this problem. For many countries, affordability with competing healthcare needs and an estimate of the total annual cost for this approach for a high HIV prevalence country as part of total health budget would be very useful.
Utilizing routine or low-cost CD4 cell counts compared to WHO clinical staging in order to guide initiation of ART for patients infected with HIV appears to be a very cost-effective intervention for sub-Saharan Africa. The ICER per QALY gained is well below the threshold of the per capita GDP. We recommend the implementation of routine CD4 cell testing as an integral part of the scale-up of ART programs in the sub-Saharan African public health services, as an important healthcare priority and that ongoing iterative evaluation is needed to monitor this policy.
The present study did not receive financial support. We are grateful to Professor John Cairns at the Department of Health Economics, the London School of Hygiene and Tropical Medicine for support in developing this analysis.
1. Freedberg KA, Losina E, Weinstein MC, Paltiel AD, Cohen CJ, Seage GR, et al. The cost effectiveness of combination antiretroviral therapy for HIV disease. New Engl J Med 2001; 344:824–831.
2. Beck EJ, Mandalia S, Gaudreault M, Brewer C, Zowall H, Gilmore N, et al. The cost effectiveness of highly active antiretroviral therapy, Canada 1991–2001. AIDS 2004; 18:2411–2418.
3. Vijayaraghavan A, Efrusy MB, Mazonson PD, Ebrahim O, Sanne IM, Santas CS. Cost effectiveness of alternative strategies for initiating and monitoring highly active antiretroviral therapy in the developing world. AIDS 2007; 46:91–100.
4. Stringer JSA, Zulu I, Levy J, Stringer EM, Mwango A, Chi BH, et al. Rapid scale up of antiretroviral therapy at primary care sites in Zambia. JAMA 2006; 296:782–793.
5. Cleary SM, McIntyre D, Boulle AM. The cost effectiveness of antiretroviral treatment in Khayelitsha, South Africa: a primary data analysis. Cost Eff Resour Alloc 2006; 4:20.
6. Goldie SJ, Yazdanpanah Y, Losina E, Weinstein MC, Anglaret X, Walensky RP, et al. Cost effectiveness of HIV treatment in resource poor settings: the case of Cote d'Ivoire. New Engl J Med 2006; 355:1141–1153.
7. Badri M, Maartens G, Mandalia S, Bekker LG, Penrod JR, Platt RW, et al. Cost effectiveness of highly active antiretroviral therapy in South Africa. PLoS Med 2006; 3:e4.
8. Ivers LC, Kendrick D, Doucette K. Efficacy of antiretroviral therapy programs in resource-poor settings: a meta-analysis of the published literature. Clin Infect Dis 2005; 41:217–224.
9. Akileswaran C, Lurie MN, Flanigan TP, Mayer KH. Lessons learned from use of highly active antiretroviral therapy in Africa. Clin Infect Dis 2005; 41:376–385.
10. Proposed WHO staging system for HIV infection and disease: preliminary testing by an international collaborative cross sectional study. AIDS 1993; 7:711–718.
11. Gallant JE, Somani J, Chaisson RE, Stanton D, Smith M, Quinn TC. Diagnostic accuracy of three clinical case definitions for advanced HIV disease. AIDS 1992; 6:295–299.
12. WHO. Case definitions of HIV for surveillance and revised clinical staging and immunological classification of HIV related disease in adults and children. Geneva: WHO; 2006.
14. McGrath N, Kranzer K, Saul J, Crampin AC, Malema S, Kachiwanda L, et al. Estimating the need for antiretroviral treatment and an assessment of a simplified HIV/AIDS case definition in rural Malawi. AIDS 2007; 21(Suppl 6):S105–S113.
15. Kagaayi J, Makumbi F, Nakigozi G, Wawer MJ, Gray RH, Serwadda D, Reynolds SJ. WHO HIV clinical staging or CD4 cell counts for antiretroviral therapy eligibility assessment? An evaluation in rural Rakai district, Uganda. AIDS 2007; 21:1208–1210.
16. Callens SFJ, Kitetele F, Lusiama J, Shabani N, Edidi S, Colebunders R, et al. Computed CD4 percentage as a low cost method for determining pediatric antiretroviral treatment eligibility. BMC Infect Dis 2008; 8:31.
17. MacLennan CA, Liu MKP, White SA, Van Oosterhout JJG, Simukonda F, Bwanali J, et al. Diagnostic accuracy and clinical utility of a simplified low cost method of counting CD4 cells with flow cytometry in Malawi: diagnostic accuracy study. BMJ 2007; 335:190–194.
18. Pattanapanyasat K, Phuang-Ngern Y, Lerdwana S, Wasinrapee P, Sakulploy N, Noulsri E, et al. Evaluation of a single platform microcapillary flow cytometer for enumeration of absolute CD4+ T lymphocyte counts in HIV-1 infected Thai patients. Cytometry B Clin Cytom 2007; 72B:387–396.
19. Janossy G, Jani IV, Bradley NJ, Bikoue A, Pitfield T, Glencross DK. Affordable CD4+ T cell counting by flow cytometry: CD45 gating for volumetric analysis. Clin and Diag Lab Immun 2002; 9:1085–1094.
20. Dieye TN, Vereecken C, Diallo AA, Ondoa P, Diaw PA, Camara M, et al. Absolute CD4 T cell counting in resource poor settings. Direct volumetric measurements versus bead based clinical flow cytometry instruments. AIDS 2005; 39:32–37.
21. Balakrishnan P, Dunne M, Kumarasamy N, Crowe S, Subbulakshmi G, Ganesh AK, et al. An inexpensive, simple, and manual method of CD4 T cell quantitation in HIV infected individuals for use in developing countries. AIDS 2004; 36:1006–1010.
22. Bacchetti P, Moss AR. Incubation period of AIDS in San Francisco. Nature 1989; 338:251–253.
23. Rutherford GW, Lifson AR, Hessol NA, Darrow WW, O'Malley PM, Buchbinder SP, et al. Course of HIV-1 infection in a cohort of homosexual and bisexual men: an 11 year follow up study. BMJ 1990; 301:1183–1188.
24. Mellors JW, Munoz A, Giorgi JV, Margolick JB, Tassoni CJ, Gupta P, et al. Plasma viral load and CD4 lymphocytes as prognostic markers of HIV-1 infection. Ann Intern Med 1997; 126:946–954.
25. Phillips A, Pezzotti P, CASCADE Collaboration. Short-term risk of AIDS according to current CD4 cell count and viral load in antiretroviral drug-naive individuals and those treated in the monotherapy era. AIDS 2004; 18:51–58.
26. Bachetti P, Osmand D, Chaisson RE, Dritz S, Rutherford GW, Swig L, Moss AR. Survival patterns of the first 500 patients with AIDS in San Francisco. J Infect Dis 1988; 157:1044–1047.
27. Rothenberg R, Woelfel M, Stoneburner R, Milberg J, Parker R, Truman B. Survival with the acquired immunodeficiency syndrome: experience with 5833 cases in New York City. New Engl J Med 1987; 317:1297–1302.
28. Osmond D, Harlebois E, Lang W, Shiboski S, Moss A. Changes in AIDS survival time in two San Francisco cohorts of homosexual men, 1983 to 1993. JAMA 1994; 271:1083–1087.
29. Saravolatz L, Neaton J, Sacks L, Deyton L, Rhame F, Sherer R. CD4+ lymphocyte counts and patterns of mortality among patients infected with human immunodeficiency virus who were enrolled in community programs for clinical research on AIDS. Clin Infect Dis 1996; 11:513–520.
30. Badri M, Lawn SD, Wood R. Short term risk of AIDS or death in people infected with HIV-1 before antiretroviral therapy in South Africa: a longitudinal study. Lancet 2006; 368:1254–1259.
31. Lawn SD, Myer L, Harling G, Orrell C, Bekker LG, Wood R. Determinants of mortality and nondeath losses from an antiretroviral treatment service in South Africa: implications for program evaluation. Clin Infect Dis 2006; 43:770–776.
32. Badri M, Cleary S, Maartens G, Pitt J, Bekker LG, Orrell C, Wood R. When to initiate highly active antiretroviral therapy in sub Saharan Africa? A South African cost effectiveness study. Antivir Ther 2006; 11:63–72.
33. Hammer SM, Squires KE, Hughes MD, Grimes JM, Demeter LM, Currier JS, et al. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimetre or less. New Engl J Med 1997; 337:725–733.
34. Montaner JS, Reiss P, Cooper D, Vella S, Harris M, Conway B, et al. A randomised, double blind trial comparing combinations of nevirapine, didanosine, and zidovudine for HIV-infected patients: the INCAS trial: Italy, the Netherlands, Canada and Australia study. JAMA 1998; 279:930–937.
35. Lucas GM, Chaisson RE, Moore RD. Highly active antiretroviral therapy in a large urban clinic: risk factors for virologic failure and adverse drug reactions. Ann Intern Med 1999; 131:81–87.
36. Staszewski S, Morales-Ramirez J, Tashima KT, Rachlis A, Skiest D, Stanford J, et al. Efavirenz plus zidovudine and lamuvidine, efavirenz plus indinavir, and indinavir plus zidovudine and lamuvidine in the treatment of HIV-1 infection in adults. New Engl J Med 1999; 341:1865–1873.
37. Diomande FVK, Bissagnene E, Nkengasongn JN, Maurice C, Monga B, Laga M, Nolan ML. The most efficient use of resources to identify those in need of antiretroviral treatment in Africa: empirical data from Cote d'Ivoire's drug access initiative. AIDS 2003; 17:S87–S93.
38. Collazos J, Knobel H, Casado JL, On behalf of the Grupo Espanol para el Estudio Multifactorial de la Adherencia (GEEMA). CD4 count and viral load time courses in patients treated with highly active antiretroviral therapy and association with the CDC staging system. HIV Med 2006; 7:504–513.
39. Untangling the web of price reductions: a pricing guide for ARVs in developing countries. 12th ed. Geneva: Medicins sans Frontieres; 2009. [Accessed 10 June 2009].
40. Gold MR, Siegel JE, Russell LB, Weinstein MC, editors. Cost effectiveness in health and medicine. New York: Oxford University Press; 1996.
41. Brazier J, Roberts J, Deverill M. The estimation of a preference based measure of health from the SF-36. J Health Econ 2002; 21:271–292.
42. Brazier J, Usherwood T, Harper R, Thomas K. Deriving a preference based single index from the UK SF-36 health survey. J Clin Epidemiol 1998; 51:1115–1128.
43. Jelsma J, Maclean E, Hughes J, Tinise X, Darder M. An investigation into the health related quality of life of individuals living with HIV who are receiving HAART. AIDS Care 2005; 17:579–588.
44. Dolan P, Gudex C, Kind P, Williams A. A social tariff for EuroQol: results from a UK general population survey. York: Centre for Health Economics, York Health Economics Consortium, NHS Centre for Reviews & Dissemination; 1995.
45. O'Keefe EA, Wood R. The impact of human immunodeficiency virus (HIV) infection on quality of life in a multiracial South African population. Qual Life Res 1996; 5:275–280.
46. Schackman BR, Goldie SJ, Freedberg KA, Losina E, Brazier J, Weinstein MC. Comparison of health state utilities using community and patient preference weights derived from a survey of patients with HIV/AIDS. Med Decis Making 2002; 22:27–38.
47. WHO. Report of the Commission on Macroeconomics and Health. Geneva: World Health Organisation; 2002.
48. Garber AM. Advances in cost effectiveness analysis of health interventions. In: Culer AJ, Newhouse JP, editors. Handbook of health economics. Vol 1A. Amsterdam: Elsevier; 2000. pp. 181–221.
51. Pujades-Rodriguez M, O'Brien D, Humblet P, Calmy A. Second-line antiretroviral therapy in resource-limited settings: the experience of Medecins sans Frontieres. AIDS 2008; 22:1305–1312.
52. Mauskopf J, Kitahata M, Kauf T, Richter A, Tolson J. HIV antiretroviral treatment: early versus later. AIDS 2005; 39:562–569.
53. Moh R, Danel C, Messou E, Ouassa T, Gabillard D, Anzian A, et al. Incidence and determinants of mortality and morbidity following early antiretroviral therapy initiation in HIV infected adults in west Africa. AIDS 2007; 21:2483–2491.
54. Braitstein P, Brinkhof MW, Dabis F, Schechter M, Boulle A, Miotti P, et al, the Antiretroviral Therapy in Lower Income Countries Collaboration; ART Cohort Collaboration groups. Mortality of HIV-1 infected patients in the first year of antiretroviral therapy: comparison between low income and high income countries. Lancet 2006; 367:817–824.
55. Madhivanan P, Krupp K. Technological challenges in diagnosis and management of HIV infection in resource limited settings. BMJ 2007; 335:165–166.
© 2010 Lippincott Williams & Wilkins, Inc.