Epidemiology and Social
Clinical impact and cost-effectiveness of co-trimoxazole prophylaxis in patients with HIV/AIDS in Côte d’Ivoire: a trial-based analysis
Yazdanpanah, Yazdana,b; Losina, Elenac; Anglaret, Xaviere,f; Goldie, Sue Jg; Walensky, Rochelle Pd; Weinstein, Milton Cg; Toure, Siakaf; Smith, Heather Ed; Kaplan, Jonathan Eh; Freedberg, Kenneth Ac,d,g; for the Global AIDS Policy Model Investigators
From the aService Universitaire des Maladies Infectieuses et du Voyageur, Centre Hospitalier de Tourcoing, Faculté de Médecine de Lille
bLaboratoire de Recherches Économiques et Sociales, CNRS URA 362, Lille, France
cBoston University School of Public Health, Boston, Massachusetts
dMassachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
eINSERM U593, Bordeaux, France
fProgramme PAC-CI, Abidjan, Côte d’Ivoire
gHarvard School of Public Health, Boston, Massachusetts
hCenters for Disease Control and Prevention, Atlanta, GA, USA.
Received 17 August, 2004
Revised 22 March, 2005
Accepted 17 April, 2005
Correspondence to Yazdan Yazdanpanah, MD, PhD, Service Universitaire des Maladies Infectieuses et du Voyageur, Centre Hospitalier de Tourcoing (Faculté de Médecine de Lille), 135, rue du Président Coty – B.P.619, F 59208 Tourcoing, France. E-mail: firstname.lastname@example.org
Background: In 2000, WHO/UNAIDS recommended co-trimoxazole prophylaxis for persons at early stages of HIV infection (WHO stage ≥ 2) in sub-Saharan Africa.
Objective: To assess the cost-effectiveness of alternative strategies for initiation of co-trimoxazole in Côte d’Ivoire.
Design: Cost-effectiveness analysis with an HIV simulation model using clinical and cost data from a randomized trial of co-trimoxazole in HIV-infected adults.
Methods: The study included HIV-infected patients in Côte d’Ivoire, with median age 33 years. Thirty-four percent were classified as WHO stage 2, 59% as stage 3, and 7% as stage 4. The mean CD4 cell count was 331 × 106 cells/l. The interventions were no prophylaxis, clinical criteria-based co-trimoxazole initiation (early: WHO stage ≥ 2; late: WHO stage ≥ 3), CD4-based co-trimoxazole initiation (< 500, < 200, < 50 × 106 CD4 cells/l). The outcome measures were life expectancy, lifetime costs, and incremental cost-effectiveness.
Results: The most effective strategy, initiation of co-trimoxazole prophylaxis at WHO stage ≥ 2, increased undiscounted life expectancy by 5.2 months, discounted life expectancy by 4.4 months, and lifetime costs by US$ 60, compared with no prophylaxis. Delaying prophylaxis initiation until WHO stage ≥ 3 was less costly and less effective. All CD4-based strategies were dominated. The incremental cost-effectiveness of early versus late co-trimoxazole prophylaxis initiation was US$ 200/year of life gained. Results were stable despite wide variations in plausible assumptions about bacterial resistance and the prophylaxis efficacy on co-trimoxazole-resistant strains.
Conclusions: For HIV-infected adults in Côte d’Ivoire, co-trimoxazole prophylaxis is reasonably cost-effective and most effective if initiated when WHO stage ≥ 2. Early co-trimoxazole prophylaxis will prevent complications prior to antiretroviral therapy initiation and should be considered an essential component of care for early HIV in sub-Saharan Africa.
By the end of 2003, the joint United Nations program on HIV/AIDS estimated that 25 to 28 million people were infected with HIV in sub-Saharan Africa . Although antiretroviral treatment is desperately needed, availability and coverage remain inadequate. As efforts are made to remove barriers for provision of antiretroviral treatment, strategies to decrease morbidity and mortality due to HIV-associated opportunistic diseases (e.g., tuberculosis, bacterial infections, and parasitic diseases) are available [2–4]. In 1999, Anglaret et al. and Wiktor et al. reported results from two placebo-controlled trials conducted in Abidjan, Côte d’Ivoire, showing that co-trimoxazole prophylaxis decreases the incidence of bacterial diseases, malaria, cerebral toxoplasmosis, and isosporiasis in HIV-infected adults [5,6]. In addition, Wiktor et al. reported a significant reduction in mortality rates with co-trimoxazole prophylaxis .
Following the publication of these findings, a WHO/UNAIDS meeting in Zimbabwe in March 2000 recommended that ‘co-trimoxazole should be used for prophylaxis in adults and children living with HIV/AIDS in Africa as a part of a minimum package of care’ . However, experts expressed concerns related to the cost-effectiveness of prophylaxis, optimal timing of co-trimoxazole initiation, risk of drug resistance, and generalizability of the Abidjan-based spectrum of HIV-related disease [8–10]. In response to these concerns, we evaluated the cost-effectiveness of co-trimoxazole prophylaxis in adults living with HIV/AIDS in Côte d’Ivoire, and assessed alternative strategies for initiation of prophylaxis. We also explored the implications of uncertain parameters and alternative assumptions about disease spectrum, risk of resistance, and reduced efficacy in the presence of resistance.
Based on the structure of our previously developed model used to address HIV treatment issues in both the US and in France [11–14], we developed a simulation model of HIV disease using country-specific data from Côte d’Ivoire. Using a first-order Monte Carlo simulation, the model generated number of opportunistic diseases, life expectancy, and lifetime costs for 1 million hypothetical HIV-infected individuals. The performance of co-trimoxazole prophylaxis was evaluated using incremental cost-effectiveness analysis . We adopted a societal perspective and discounted both costs and clinical benefits at 3% per year . All costs and cost-effectiveness ratios were expressed as dollars (2000 US) per year of life gained.
Health states were defined according to both CD4 cell count and history of clinical events. Progression of HIV disease, linked to CD4 cell count, was modelled as a series of monthly transitions between acute opportunistic disease and chronic health states [11–13]. Effective opportunistic disease prophylaxis with co-trimoxazole resulted in reduction in the incidence of the specific infection against which the prophylaxis was instituted . Death from any cause occurred from either a chronic state or an acute state.
Opportunistic diseases were divided into 11 groups which are common in Côte d’Ivoire: severe bacterial infections, mild bacterial infections, severe fungal infections, mild fungal infections, malaria, tuberculosis, isosporiasis, cerebral toxoplasmosis, Mycobacterium avium complex bacteremia, other severe illnesses, and other mild illnesses.
Estimates for the monthly incidence of primary and recurrent opportunistic diseases (i.e., the first episode of a specific opportunistic disease and recurrences of the same disease) and death were derived using data obtained from the placebo arm of a randomized controlled trial conducted in Abidjan, Côte d’Ivoire (the ANRS 059 trial) . Details of the trial, designed to assess the efficacy of co-trimoxazole prophylaxis at early stages of HIV-infection, have been published elsewhere [5,16].
We estimated the incidence of primary opportunistic diseases as a function of CD4 cell count (Table 1) . Due to the lack of more than three CD4 cell count measurements for all patients, each patient's CD4 decline was modelled as a linear function of time using a mixed-effect model . The probabilities of each clinical illness and death were estimated as functions of the current CD4 cell stratum . Four CD4 cell count strata were considered: ≤ 50, 51–200, 201–500, and > 500 × 106 cells/l. Risks of recurrent opportunistic diseases were estimated as follows . For severe and mild bacterial infections, mild fungal infections, and ‘other mild illnesses’, all patients from the ANRS 059 trial were grouped regardless of CD4 cell count. For oesophageal candidiasis, tuberculosis, malaria, and ‘other severe illnesses’, we assumed the incidence of recurrent opportunistic diseases is equal to the incidence of primary clinical illness in the CD4 stratum ≤ 50 × 106 cells/l, due to insufficient number of recurrent events in the ANRS 059 trial. We assumed that patients with cerebral toxoplasmosis, Mycobacterium avium complex bacteremia, and isosporiasis were not cured but required lifelong maintenance therapy in the absence of antiretroviral therapy.
We estimated the incidence of death stratified by both date and cause of death and CD4 cell count when possible . We assumed that no acute mortality was associated with mild opportunistic diseases. Death rates within 30 days of other clinical illnesses were estimated using ANRS 059 data with the exception of cerebral toxoplasmosis and cryptococcal infection. For those two infections, death rates within 30 days were derived from other published studies in sub-Saharan Africa [20,21]. Death rates in patients with no history of opportunistic disease or with a history of mild or severe opportunistic disease, were estimated as a function of CD4 cell count. Finally, mortality from causes other than HIV were from country-specific lifetables for Côte d’Ivoire .
The efficacy of co-trimoxazole prophylaxis in preventing opportunistic diseases and rates of toxic events were derived from the ANRS 059 trial (Table 1) . Based on the results of the trial, we considered that co-trimoxazole reduces the occurrence of mild and severe bacterial infections, malaria, isosporiasis, and acute unexplained fever (categorized as other severe clinical illness). Because a trend towards reduction in incidence for cerebral toxoplasmosis was observed in the co-trimoxazole arm of the trial and has been reported elsewhere , we also considered that co-trimoxazole reduces the occurrence of this infection. In addition, we incorporated an increased risk of mild fungal infections with use of co-trimoxazole because episodes of oral candidiasis were more frequently observed in the co-trimoxazole arm than the placebo arm of the trial. We also considered minor and major toxicity related to co-trimoxazole use. Patients in the trial were not receiving antiretroviral therapy.
Direct medical costs of HIV-related care with and without co-trimoxazole prophylaxis were included in the analysis. The following components of direct medical costs were considered: (1) the number of outpatient visits and inpatient admissions; (2) length of stay for inpatient admissions; and (3) for each admission, laboratory tests, clinical procedures, and the doses and quantity of drugs dispensed. Direct non-medical costs, patient time costs associated with clinic visits, and the monetized values of lost work time associated with illness were not included.
We used data on resource use collected in the placebo arm of the ANRS 059 trial to estimate components of the direct medical costs. We estimated resources used in three clinically important and economically relevant stages of HIV infection that correspond to the structure of our model of disease: (1) no acute clinical illnesses (stratified by CD4 cell count); (2) acute clinical illness; and (3) final month of life. We applied a unit cost to each resource identified and measured in each stage of disease; these unit costs were summed over all resource types to develop stage-specific average cost estimates (Table 1).
Total lifetime costs with no prophylaxis and co-trimoxazole prophylaxis were calculated by multiplying the estimated average monthly cost of care in each stage by the corresponding model-projected occupancy time in that stage in the absence and presence of co-trimoxazole, respectively. Patients receiving prophylaxis also incurred the monthly cost of co-trimoxazole, and the cost of minor or major toxicity attributable to co-trimoxazole.
Data on average costs per day for inpatient bed-day (US$ 13.90), length of stay ≥ 1 day, and day-care admission (US$ 6.90), length of stay < 1 day, including both room/overhead costs and average physician and nurse fees per day, were from Youpougnon University Hospital in Abidjan. Average costs per outpatient medical consultation were from urban community clinics in Abidjan (US$ 0.69). Unit costs for laboratory tests and procedures performed were from the CeDReS laboratory of the Treichville University Hospital cost database. For tests or procedures performed in private practice, we used unit costs based on their cost databases. Unit costs of medications were from the pharmacy records of Médecins sans Frontières-Logistique (Bordeaux, France), Pharmacie de Santé Publique de Côte d’Ivoire (the national public drug supplier), or private drug suppliers in Côte d’Ivoire. Major toxicity attributable to co-trimoxazole was assumed to incur a 7-day inpatient bed-day stay, whereas minor toxicity incurred an outpatient visit plus two additional blood counts.
To assess the internal validity of the model, we projected the number of clinical illnesses and deaths at 9.6 months, the median duration of follow-up of patients in the ANRS 059 trial. Outcomes projected by the model were within 10% of reported results in the ANRS 059 trial for each opportunistic disease and for death.
Base case analysis
The characteristics of simulated patients were drawn from a distribution of patients similar to those enrolled in the placebo arm of the ANRS 059 trial : median age 33 years, 40% male, baseline mean CD4 cell count 331 × 106 cells/l, 34% WHO clinical stage 2, 59% stage 3, and 7% stage 4. We compared the provision of prophylaxis to adults with symptomatic HIV disease who were at early stages of HIV infection (WHO stage ≥ 2) to provision in later stages (WHO stage ≥ 3 or lower CD4 cell count thresholds). To perform the analysis by WHO stage we assumed that diseases categorized as severe bacterial infections, severe fungal infections, tuberculosis, isosporiasis, cerebral toxoplasmosis, Mycobacterium avium complex bacteremia, and other severe illnesses match the stage 3 and 4 WHO classification diseases.
We evaluated the impact of varying co-trimoxazole efficacy and toxicity, incidence of recurrent clinical illnesses, death rates, and clinical illness-related costs on the clinical impact and cost-effectiveness of co-trimoxazole. To explore the generalizability of our results we also varied region-specific parameters, including the incidence of clinical illnesses and the prevalence of co-trimoxazole resistance among bacterial pathogens. To evaluate the impact of resistance, we used a two-way sensitivity analysis to vary the prevalence of co-trimoxazole resistance and the responsiveness of co-trimoxazole-resistant strains across a broad range. Assumptions about the prevalence of co-trimoxazole resistance among bacterial pathogens were based on available data on resistance among non-typhi Salmonella isolates in sub-Saharan Africa [24–27].
Base case analysis
The number of lifetime cases of primary mild and severe bacterial infections decreased from 612 and 514 per 1000 patients without prophylaxis to 480 and 367 per 1000 patients with co-trimoxazole given to all patients with WHO stage ≥ 2. The number of cases of other co-trimoxazole-sensitive opportunistic diseases – malaria, isosporiasis, and cerebral toxoplasmosis – also decreased with co-trimoxazole. The number of other opportunistic diseases increased, because patients protected from co-trimoxazole-sensitive opportunistic diseases remained susceptible to these diseases [12,13].
Table 2 shows the results of the cost-effectiveness analysis. With no prophylaxis, mean undiscounted life expectancy was 41.4 months, discounted life expectancy was 38.0 months, and total projected lifetime costs were US$ 1260. Initiation of prophylaxis in patients with WHO stage ≥ 3 provided an undiscounted life expectancy gain of 2.4 months, a discounted life expectancy gain of 2.4 months, and an incremental cost-effectiveness ratio of US$ 110/year of life saved compared to no prophylaxis. In comparison, earlier initiation of prophylaxis (WHO stage ≥ 2), as recommended by WHO/UNAIDS, nearly doubled this benefit and had an incremental cost-effectiveness ratio of US$ 200/year of life saved compared to late initiation. Compared to no prophylaxis, this early-initiation strategy had an incremental cost-effectiveness ratio of US$ 150/year of life saved. Starting prophylaxis based on a variety of CD4 cell count thresholds was always more costly and less effective than strategies relying on clinical staging.
Results were most sensitive to the efficacy of co-trimoxazole and the cost of care for clinical illnesses (Table 3). If the efficacy of co-trimoxazole was reduced by 50%, the cost-effectiveness ratio for early prophylaxis compared to no prophylaxis increased from US$ 150/year of life saved to US$ 350/year of life saved. When the cost of care for acute clinical illnesses was 25% of baseline, the co-trimoxazole prophylaxis cost-effectiveness ratio increased to US$ 250/year of life saved.
Figure 1 illustrates the impact of varying the incidence of specific preventable opportunistic diseases such as bacterial infections and malaria, demonstrating that prophylaxis is less effective and cost-effective when the incidence of these infections decreases. However, even with a 75% decrease in the incidence of bacterial infections or malaria, prophylaxis still resulted in 3.7 and 4.3 month gains in discounted life expectancy, respectively. Under these scenarios, the cost-effectiveness ratios for early versus no co-trimoxazole prophylaxis were US$ 210 and US$ 180/year of life saved.
Figure 2 depicts a two-way sensitivity analysis in which we simultaneously varied the prevalence of co-trimoxazole resistance and the efficacy of prophylaxis on co-trimoxazole-resistant strains. Under the pessimistic assumption that prophylaxis for co-trimoxazole resistant strains is not effective at all, the gain in discounted life expectancy was estimated to be 3.9 months in an area with 43% co-trimoxazole resistance among bacterial pathogens (Senegal), 3.6 months with 57% resistance (Kenya), 3.0 months with 83% resistance (Malawi), and 2.9 months with 90% resistance (Ethiopia) [24–27]. Under these scenarios, co-trimoxazole cost-effectiveness ratios were US$ 180, US$ 200, US$ 250, and US$ 260/year of life saved, respectively.
We utilized a comprehensive simulation model of HIV disease, incorporating country-specific natural history, efficacy, and cost data, to assess the benefits and cost-effectiveness of various strategies for co-trimoxazole prophylaxis in HIV-infected adults in Côte d’Ivoire . Compared with no prophylaxis, co-trimoxazole prophylaxis in symptomatic adults at early stages of HIV infection reduced the incidence of severe bacterial infections by more than 30% and increased per person projected discounted life expectancy by 4.4 months. The incremental cost-effectiveness of starting co-trimoxazole prophylaxis in symptomatic adults at early stages of HIV-infection was US$ 150/year of life gained when compared to no prophylaxis, and US$ 200/year of life gained when compared to late initiation. In contrast, co-trimoxazole prophylaxis initiation based on CD4 cell counts was more costly and less effective.
Because some of the beneficial effect of co-trimoxazole prophylaxis in Côte d’Ivoire may relate to the high prevalence of malaria and the relatively low prevalence of bacterial pathogen resistance in that country [8,9], we explored the impact of these two parameters in sensitivity analyses. Results were robust to decreases in the incidence of malaria. Co-trimoxazole prophylaxis remained effective, and its cost-effectiveness ratio did not exceed US$ 250/year of life gained even under conservative assumptions regarding both the differences in bacterial resistance patterns between Côte d’Ivoire and other sub-Saharan African countries and the efficacy of prophylaxis on co-trimoxazole-resistant strains. The stability of these conclusions is due to the fact that co-trimoxazole prevents other opportunistic diseases in addition to bacterial infections. These results are consistent with those of a recent observational cohort study from rural Uganda that reported a significant reduction in morbidity and mortality rates with co-trimoxazole prophylaxis in an area with high rates of bacterial resistance to co-trimoxazole .
The use of co-trimoxazole prophylaxis effectively prevents the occurrence of opportunistic infections. However, the present study did not find co-trimoxazole prophylaxis to be cost-saving. The long-term increase in total costs for patients on prophylaxis results from increases in life expectancy due to successful treatment. Although no clear cost-effectiveness threshold exists to guide policymaking decisions, the results of this analysis may be contextualized by comparing them to cost-effectiveness ratios calculated for other clinical interventions adopted in sub-Saharan Africa. Studies in this setting have found the cost-effectiveness of prevention of mother-to-child HIV transmission to range from US$ 1–731/disability-adjusted life year gained (DALY) [29–32], preventive therapy for tuberculosis from US$ 169–288/DALY , malaria control from US$ 1–121/DALY , and onchocerciasis vector control from US$ 171–327/DALY (2000 US$) . Although we recognize the need for caution in directly comparing these results because of differences in study design, the cost-effectiveness of co-trimoxazole prophylaxis in adults at early stages of HIV-infection in Côte d’Ivoire compares favorably with these interventions. The Commission on Macroeconomics and Health has recently defined interventions with a cost-effectiveness ratio of less than the gross domestic product per capita of a country as very cost-effective . Our findings of US$ 200 per life year gained are substantially less than the gross domestic product per capita of Côte d’Ivoire, which is US$ 754 .
There are several limitations to this analysis. First, the generalizability of these results requires caution since a clinical trial population was used to estimate the natural history of HIV disease and the cost of care . In a rural population that is not as closely monitored, mortality rates are likely to be higher, and cost of care lower, than in a randomized trial population. Second, we were not able to estimate the incidence of all recurrent events and death rates for each specific clinical event, or to stratify these estimates by CD4 cell count due to the relatively short duration of follow-up in the placebo arm of the ANRS 059 trial and the consequent occurrence of a relatively small number of opportunistic diseases. Third, disability weighting associated with years of life lived within categories defined in our model was not available for either the ANRS 059 trial or for Côte d’Ivoire. As such, we expressed our results in cost per year of life gained. Fourth, we did not consider the impact of co-trimoxazole prophylaxis on the evolution of drug resistance. Widespread use of co-trimoxazole for prophylaxis may lead to an increase in the prevalence of co-trimoxazole resistance among bacterial pathogens [38–40]. Moreover, it is possible that co-trimoxazole use may have a similar effect on levels of sulfadoxine/pyramethamine-resistant malaria . A fully specified dynamic model would be needed to take into account emerging resistance in both HIV-infected patients and in the general population .
WHO/UNAIDS recommends that patients receive antiretroviral therapy, when available, at WHO stage 4 CD4 cell count ≤ 200 × 106 cells/l or WHO stage 3 and CD4 cell count < 350 × 106 cells/l . In Côte d’Ivoire, co-trimoxazole prophylaxis is now recommended at WHO stage 2 – prior to the point at which patients become eligible for antiretroviral therapy. In this study, we found that co-trimoxazole prophylaxis is both effective and cost effective. These findings should encourage clinicians and policymakers to adopt such standards of care for early HIV disease. The implementation of programs for early HIV care, including prophylactic treatment with co-trimoxazole, does not preclude the development of initiatives that will increase access to antiretroviral therapy. Rather, prophylaxis and antiretroviral therapy are complementary treatment strategies; early initiation into care will ensure that patients develop fewer opportunistic diseases, and, by fostering an early connection to care, may enhance patients’ eventual access to antiretroviral therapy. In addition, early prophylaxis allows patients to become accustomed to daily medication intake, providing an important step towards preparation for antiretroviral therapy. With antiretroviral therapy becoming increasingly available, further studies should explore the cost-effectiveness of co-trimoxazole prophylaxis in the presence of antiretroviral therapy.
The results of this study support the WHO/UNAIDS provisional recommendations that co-trimoxazole prophylaxis in sub-Saharan Africa should be provided to all patients with symptomatic HIV disease, and will be most effective if initiated when patients meet criteria for WHO stage ≥ 2. For HIV-infected adults in Côte d’Ivoire, and potentially in other settings with similar epidemiological and socio-demographic profiles, co-trimoxazole prophylaxis can be expected to substantially reduce morbidity and mortality at reasonable cost, thus allowing patients to eventually start antiretroviral therapy without an extensive prior history of opportunistic diseases or the attributable mortality associated with these diseases.
We acknowledge the assistance of Sylvie Deuffic-Burban, PhD for suggestions on data management and analysis. We are indebted to A. David Paltiel, PhD for his valuable suggestions during the development of the model and for helpful comments on the analysis, to Hong Zhang, SM for his programming expertise, and to Tammy M. Muccio and Lindsey L. Wolf for administrative assistance.
Sponsorship: This study was supported by grants from the Agence National de Recherches sur le SIDA (ANRS 1286), the Centers for Disease Control and Prevention (Cooperative Agreements U64/CCU 114927 and U64/CCU 119525), and the National Institute of Allergy and Infectious Diseases (NIAID AI058736, K23 AI0794, K24 AI062476, K25 AI50436 and CFAR P30 AI42851).
Conflict of interest disclosure: none of the authors has any financial interests in this manuscript. Y.Y. had full access to all the data in the study and had final responsibility for the decision to submit for publication.
2. Holmes CB, Losina E, Walensky RP, Yazdanpanah Y, Freedberg KA. Review of human immunodeficiency virus type 1-related opportunistic infections in sub-Saharan Africa. Clin Infect Dis 2003; 36:652–662.
3. Attia A, Huet C, Anglaret X, Toure S, Ouassa T, Gourvellec G, et al
. HIV-1-related morbidity in adults, Abidjan, Cote d’Ivoire: a nidus for bacterial diseases. J Acquir Immune Defic Syndr 2001; 28:478–486.
4. Corbett EL, Churchyard GJ, Charalambos S, Samb B, Moloi V, Clayton TC, et al
. Morbidity and mortality in South African gold miners: impact of untreated disease due to human immunodeficiency virus. Clin Infect Dis 2002; 34:1251–1258.
5. Anglaret X, Chêne G, Attia A, Toure S, Lafont S, Combe P, et al
. Early chemoprophylaxis with trimethoprim-sulphamethoxazole for HIV-1-infected adults in Abidjan, Côte d’Ivoire: a randomised trial. Cotrimo- CI Study Group. Lancet 1999; 353:1463–1468.
6. Wiktor SZ, Sassan-Morokro M, Grant AD, Abouya L, Karon JM, Maurice C, et al
. Efficacy of trimethoprim-sulphamethoxazole prophylaxis to decrease morbidity and mortality in HIV-1-infected patients with tuberculosis in Abidjan, Cote d’Ivoire: a randomised controlled trial. Lancet 1999; 353:1469–1475.
8. Grant AD, Kaplan JE, De Cock KM. Preventing opportunistic infections among human immunodeficiency virus-infected adults in African countries. Am J Trop Med Hyg 2001; 65:810–821.
9. Grimwade K, Gilks C. Cotrimoxazole prophylaxis in adults infected with HIV in low-income countries. Curr Opin Infect Dis 2001; 14:507–512.
10. Badri M, Ehrlich R, Wood R, Maartens G. Initiating co-trimoxazole prophylaxis in HIV-infected patients in Africa: an evaluation of the provisional WHO/UNAIDS recommendations. AIDS 2001; 15:1143–1148.
11. Freedberg KA, Losina E, Weinstein MC, Paltiel AD, Cohen CJ, Seage GR, et al
. The cost effectiveness of combination antiretroviral therapy for HIV disease. N Engl J Med 2001; 344:824–831.
12. Goldie SJ, Kaplan JE, Losina E, Weinstein MC, Paltiel AD, Seage GR 3rd, et al
. Prophylaxis for human immunodeficiency virus-related Pneumocystis carinii pneumonia: using simulation modeling to inform clinical guidelines. Arch Intern Med 2002; 162:921–928.
13. Yazdanpanah Y, Goldie SJ, Paltiel AD, Losina E, Coudeville L, Weinstein MC, et al
. Prevention of human immunodeficiency virus-related opportunistic infections in France: a cost-effectiveness analysis. Clin Infect Dis 2003; 36:86–96.
14. Yazdanpanah Y, Goldie SJ, Losina E, Weinstein MC, Lebrun T, Paltiel AD, et al
. Lifetime cost of HIV care in France during the era of highly active antiretroviral therapy. Antivir Ther 2002; 7:257–266.
15. Gold MR, Siegel JE, Russel LB, Weinstein MC (editors). Cost-Effectiveness in Health and Medicine
. New York, NY: Oxford University Press; 1996.
16. Anglaret X, Messou E, Ouassa T, Toure S, Dakoury-Dogbo N, Combe P, et al
. Pattern of bacterial diseases in a cohort of HIV-1 infected adults receiving cotrimoxazole prophylaxis in Abidjan, Cote d’Ivoire. AIDS 2003; 17:575–584.
17. Losina E, Anglaret X, Yazdanpanah Y, Toure S, Pollock A, N‘Dri-Yoman T, et al
. Incidence of opportunistic infections (OIs) and mortality within specific CD4 strata in HIV-infected patients in Côte d’Ivoire. XIV International AIDS Conference
. Barcelona, Spain, 2002 [abstract TuPeC4685].
18. Diggle P, Liang K, Zeger S. Analysis of Longitudinal Data
. Oxford: Oxford University Press; 1994.
19. Miller DK, Homan SM. Determining transition probabilities: confusion and suggestions. Med Decis Making 1994; 14:52–58.
20. Grant AD, Djomand G, Smets P, Kadio A, Coulibaly M, Kakou A, et al
. Profound immunosuppression across the spectrum of opportunistic disease among hospitalized HIV-infected adults in Abidjan, Cote d’Ivoire. AIDS 1997; 11:1357–1364.
21. Hakim JG, Gangaidzo IT, Heyderman RS, Mielke J, Mushangi E, Taziwa A, et al
. Impact of HIV infection on meningitis in Harare, Zimbabwe: a prospective study of 406 predominantly adult patients. AIDS 2000; 14:1401–1407.
22. Lopez A, Ahmad O, Guillot M, Ferguson B, Salomon J, Murray CJL, et al
. World Mortality in 2000: Life Tables for 191 Countries
. Geneva: World Health Organization; 2002.
23. Ioannidis JP, Cappelleri JC, Skolnik PR, Lau J, Sacks HS. A meta-analysis of the relative efficacy and toxicity of Pneumocystis carinii prophylactic regimens. Arch Intern Med 1996; 156:177–188.
24. Sow AI, Faye Niang MA, Dieng M, Toure K, Fall D, Soumare M, et al
. Sensitivity to cotrimoxazole of bacteria isolated at the Central University Hospital of Fann, Dakar. Dakar Med 1999; 44:20–24.
25. Kariuki S, Gilks C, Corkill J, Kimari J, Benea A, Waiyaki P, et al
. Multi-drug resistant non-typhi salmonellae in Kenya. J Antimicrob Chemother 1996; 38:425–434.
26. Boeree MJ, Harries AD, Zijlstra EE, Taylor TE, Molyneux ME. Co-trimoxazole in HIV-1 infection. Lancet
:334; author reply 335.
27. Aseffa A, Gedlu E, Asmelash T. Antibiotic resistance of prevalent Salmonella and Shigella strains in northwest Ethiopia. East Afr Med J 1997; 74:708–713.
28. Mermin J, Lule J, Ekwaru J, Malamba S, Downing R, Ransom R, et al
. Effect of co-trimoxazole prophylaxis on morbidity, mortality, and CD4-cell count, and viral load in HIV infection in rural Uganda. Lancet 2004; 364:1428–1434.
29. Marseille E, Kahn JG, Mmiro F, Guay L, Musoke P, Fowler MG, et al
. Cost-effectiveness of single-dose nevirapine regimen for mothers and babies to decrease vertical HIV-1 transmission in sub-Saharan Africa. Lancet 1999; 354:803–809.
30. Marseille E, Kahn JG, Saba J. Cost-effectiveness of antiviral drug therapy to reduce mother-to-child HIV transmission in sub-Saharan Africa. AIDS 1998; 12:939–948.
31. Wilkinson D, Floyd K, Gilks CF. Antiretroviral drugs as a public health intervention for pregnant HIV-infected women in rural South Africa: an issue of cost-effectiveness and capacity. AIDS 1998; 12:1675–1682.
32. Soderlund N, Zwi K, Kinghorn A, Gray G. Prevention of vertical transmission of HIV: analysis of cost-effectiveness of options available in South Africa. BMJ 1999; 318:1650–1656.
33. Bell JC, Rose DN, Sacks HS. Tuberculosis preventive therapy for HIV-infected people in sub-Saharan Africa is cost-effective. AIDS 1999; 13:1549–1556.
34. Goodman CA, Coleman PG, Mills AJ. Cost-effectiveness of malaria control in sub-Saharan Africa. Lancet 1999; 354:378–385.
35. Jamison D, Mosley W, Measham A, Bobadilla J. Disease Control Priorities in Developing Countries
. New York: Oxford University Press; 1993.
36. WHO Commission on Macroeconomics and Health. Macroeconomics and Health: Investing in Health for Economic Development. Report of the Commission on Macroeconomics and Health
. Geneva: World Health Organization; 2001.
38. Abdel-Haq N, Abuhammour W, Asmar B, Thomas R, Dabbagh S, Gonzalez R. Nasopharyngeal colonization with Streptococcus pneumoniae in children receiving trimethoprim-sulfamethoxazole prophylaxis. Pediatr Infect Dis J 1999; 18:647–649.
39. Meynard JL, Barbut F, Blum L, Guiguet M, Chouaid C, Meyohas MC, et al
. Risk factors for isolation of Streptococcus pneumoniae
with decreased susceptibility to penicillin G from patients infected with human immunodeficiency virus. Clin Infect Dis 1996; 22:437–440.
40. Martin JN, Rose DA, Hadley WK, Perdreau-Remington F, Lam PK, Gerberding JL. Emergence of trimethoprim-sulfamethoxazole resistance in the AIDS era. J Infect Dis 1999; 180:1809–1818.
41. Iyer JK, Milhous WK, Cortese JF, Kublin JG, Plowe CV. Plasmodium falciparum
cross-resistance between trimethoprim and pyrimethamine. Lancet 2001; 358:1066–1067.
42. Anderson R, May R. Infectious Disease of Humans: Dynamics and Control
. Oxford: Oxford University Press; 1991.
Global AIDS Policy Model Investigators:
Service Universitaire des Maladies Infectieuses et du Voyageur, Centre Hospitalier de Tourcoing, Faculté de Médecine de Lille, France: Yazdan Yazdanpanah.
Laboratoire de Recherches Économiques et Sociales, CNRS URA 362, Lille, France: Yazdan Yazdanpanah.
Boston University School of Public Health, Boston, Massachusetts, USA: Kenneth A. Freedberg, Elena Losina.
Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA: Nomita Divi, Kenneth A. Freedberg, Charles B. Holmes, April D. Kimmel, Heather E. Smith, Rochelle P. Walensky.
INSERM U593, Bordeaux, France: Xavier Anglaret, Roger Salamon.
Programme PAC-CI, Abidjan, Côte d’Ivoire: Xavier Anglaret, Thérèse N'Dri-Yoman, Roger Salamon, Siaka Toure.
Harvard School of Public Health, Boston, Massachusetts, United States: Kenneth A. Freedberg, Sue J. Goldie, George R. Seage III, Milton C. Weinstein.
Centers for Disease Control and Prevention, Atlanta, Georgia, USA: Jonathan E. Kaplan.
HIV/AIDS; co-trimoxazole prophylaxis; cost-effectiveness; sub-Saharan Africa
© 2005 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.