SEARCH

SEARCH BY CITATION

Keywords:

  • cost analysis;
  • voluntary counselling and testing;
  • injecting drug user;
  • HIV/AIDS;
  • Indonesia
  • Analyse des coûts;
  • conseil et dépistage volontaire;
  • usager de drogues injectables;
  • VIH/SIDA;
  • Indonésie
  • Análisis de costes;
  • Aconsejamiento y Prueba Voluntario;
  • Usuario de drogas inyectables;
  • VIH/SIDA;
  • Indonesia

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objective  To evaluate costs and outcomes of voluntary counselling and testing (VCT) service delivery models in urban Indonesia.

Methods  We collected primary data on utilization, costs and outcomes of VCT services in a hospital clinic (568 clients), HIV community clinic (28 clients), sexually transmitted infection (STI) community clinic (784 clients) and prison clinic (574 clients) in Bandung, Indonesia, in the period January 2008–April 2009.

Results  The hospital clinic diagnosed the highest proportion and absolute number of HIV infections, but with the lowest average CD4 cell count and with the highest associated travelling and waiting time. The prison clinic detected fewer cases, but at an earlier stage, and all enrolled in HIV care. The community clinics detected the smallest number of cases, and only 0–8% enrolled in HIV care. The unit cost per VCT was highest in the hospital clinic (US$74), followed by the STI community clinic (US$65), the HIV community clinic (US$39) and the prison (US$23).

Conclusion  We propose a reorientation of the delivery models for VCT and related HIV/AIDS treatment in this setting. We call for the scaling up of community clinics for VCT to improve access, promote earlier detection and to perform (early) treatment activities. This would reduce the burden of the hospital clinic to orient itself towards the treatment of AIDS patients. This is one of very few studies addressing this issue in Asia and the first of its kind in Indonesia, which has a rapidly growing HIV epidemic. The conceptual framework and overall conclusions may be relevant to other low-income settings.

Objectif:  Evaluer les coûts et les résultats des modèles de prestation des services de CDV dans les zones urbaines d’Indonésie.

Méthodes:  Nous avons collecté des données primaires sur l’utilisation, les coûts et les résultats des services de CDV dans une clinique d’hôpital (568 patients), une clinique communautaire du VIH (28 patients), une clinique communautaire MST (784 patients) et une clinique de prison (574 patients) à Bandung, en Indonésie, durant la période de janvier 2008 à avril 2009.

Résultats:  La clinique de l’hôpital a diagnostiqué la plus grande proportion et le plus grand nombre absolu d’infections VIH, mais avec le plus bas taux moyen de cellules CD4 et la plus longue durée pour le déplacement et l’attente associées. La clinique de la prison a détecté peu de cas, mais à un stade précoce et tous ont été inscrits aux soins VIH. Les cliniques communautaires ont détecté le plus faible nombre de cas et seulement de 0 à 8% inscrits aux soins VIH. Le coût unitaire par CDV était le plus élevé dans la clinique de l’hôpital (74 $ US), suivi par la clinique communautaire MST (65$ US), la clinique communautaire VIH (39$ US) et la prison (23$ US).

Conclusion:  Nous proposons une réorientation des modèles de prestation du CDV et autres traitements associés du VIH/SIDA dans ce contexte. Nous appelons à l’élargissement des cliniques communautaires de CDV pour améliorer l’accès, promouvoir le dépistage précoce et initier tôt des activités de traitement. Cela permettrait de réduire la charge sur la clinique de l’hôpital afin de l’orienter vers le traitement des patients SIDA. Il s’agit ici d’une des rares études sur cette question en Asie et la première dans son genre en Indonésie, pays qui connaît une croissance rapide de l’épidémie VIH. Le cadre conceptuel et les conclusions générales peuvent être utiles à d’autres régions à faibles revenus.

Objetivo:  Evaluar los costes y resultados de los modelos de entrega de servicios de aconsejamiento y prueba voluntarios (APV) en zonas urbanas de Indonesia.

Métodos:  Se recolectaron datos primarios sobre utilización, costes y resultados de los servicios de APV en un hospital (568 clientes), un centro sanitario comunitario para VIH (28 clientes), un centro sanitario comunitario para ETS (784 clientes), y la clínica de una prisión (574 clientes) en Bandung, Indonesia, durante el periodo comprendido entre Enero 2008 y Abril 2009.

Resultados:  En el hospital se diagnosticó la proporción más alta y el mayor número absoluto de casos de infección por VIH, pero con el menor promedio de recuento de células CD4 y con el mayor tiempo asociado de viaje y de espera. La clínica de la prisión detectó el menor número de casos, pero en una etapa más temprana, y todos ellos fueron incluidos dentro de un programa de atención del paciente con VIH. Los centros sanitarios comunitarios detectaron el menor número de casos, y solo 0–8% fueron incluidos en programas de atención de pacientes con VIH. El coste unitario del APV era mayor en el hospital (US$74), seguido por el centro sanitario comunitario para ETS (US$65), el centro sanitario comunitario para VIH (US$39) y la prisión (US$23).

Conclusión:  Proponemos un cambio en la orientación de los modelos de entrega del APV y de los tratamientos relacionados el VIH/SIDA en este emplazamiento. Insistimos en la importancia de llevar a mayor escala el APV en los centros sanitarios comunitarios, de mejorar el acceso, promover una detección más temprana y realizar actividades de tratamiento tempranas. Todo ello reduciría la carga sanitaria del hospital, permitiendo así su reorientación hacia el tratamiento de los pacientes con SIDA. Este es uno de los pocos estudios en los que se aborda este problema en Asia, y el primero en que se realiza en Indonesia, la cual tiene una creciente epidemia de VIH. El marco conceptual y las conclusiones generales podrían ser relevantes en otros emplazamientos de países con bajos ingresos.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The HIV epidemic in Indonesia is among the fastest growing in Asia (UNAIDS 2007). It is concentrated among injecting drug users (IDUs) and their sexual partners in most parts of the country, with the exception of parts of Papua, where the epidemic is generalized. As of 2009, the cumulative Indonesian death toll of HIV/AIDS amounted to nearly 17 000, of which 42% were reported to be IDU-related, while hetero and homosexual transmission accounted for 48% and 4%, respectively (MOH-RI 2009). HIV/AIDS control efforts have increased considerably in Indonesia in recent years (Siregar et al. 2009), with voluntary counselling and testing (VCT) being a key component in the national strategic plan (Afriandi et al. 2009). VCT offers an entry point to treatment of HIV-positive individuals and is instrumental to prevent further spread of the disease by reducing risk behaviour of tested individuals (UNAIDS 2001). VCT services are only accessed by approximately 30% of all high-risk groups (UNAIDS 2007), and scaling up of services is warranted.

Various delivery models of VCT exist, ranging from integration in primary health services, specialized STD clinics to tertiary-level hospital services. In Indonesia, VCT is typically delivered by tertiary level hospitals, as this is where most individuals present for testing, while it has been stressed that more efforts are needed to increase early testing at community level (Wisaksana et al. 2009). Integration of VCT in primary health care in Tanzania shows that this may lead to improved VCT uptake, reduced stigma and improvement of general health service delivery (Mmbando et al. 2004). However, it is not clear which delivery models are most appropriate in which context. Different aspects may be important, such as VCT costs, outcomes of services including the number of positive cases identified and follow-up with treatment, sustainability of the services, accessibility for clients, provider capacity to accommodate scaling up of services and the target population of the specific delivery model.

This paper concentrates on the former two aspects, and presents the costs and outcomes of four VCT delivery models – as delivered at the hospital clinic, the HIV community clinic, the STI community clinic, and the prison clinic – in an urban setting in Indonesia. It thereby responds to the almost complete paucity of evidence on the costs of VCT services in Asia (Walker 2003) – so far the evidence-base only comprises two studies on costs of VCT in Thailand (Teerawattananon et al. 2005; HITAP 2008), one study at the south-east Asian sub-regional level (Hogan et al. 2005), and one in India (Dandona et al. 2008). The present costing study is unique for three reasons. First, it provides the first costing study on VCT in Indonesia, which is an important input for rational decision-making on the scaling up of service delivery (Siregar et al. 2009). Second, the study does not only estimate costs of VCT service delivery, but also costs of implementing VCT as incurred at the more administrative level (so called programme costs) – these costs are typically ignored in economic analysis (Johns et al. 2003). Third, the study also estimates costs of clients of seeking and undergoing VCT. Both programme costs and patient costs are also important criteria when making decision on scaling up VCT services.

The analysis of costs and outcomes of VCT delivery models serves as a basis for a broader elaboration of scaling up the respective models – taking into account all other aspects as well. The objective of the paper is to inform policy makers on the options and limitations of scaling up alternative VCT delivery models in Indonesia.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study setting

The study was conducted at four sites in Bandung, the capital of West Java. The first site is a hospital clinic located at Hasan Sadikin hospital, a teaching and provincial top-referral public hospital for West Java (with a population of approximately 40 million people). The clinic delivers HIV-related services such as VCT, anti retroviral treatment (ART), and STI services, and general medical services, and is frequented by both the general population and high risk groups, mostly IDUs. As this clinic resides at a top-referral hospital and it is one of the few clinics that deliver ART within the city, the number of HIV service–related visits is very large and the clinic operates at full capacity. The clinic generates its own revenues through user-fees and otherwise relies on government, hospital and donor funding.

The second site is a newly established HIV community clinic that is integrated with a public community health centre, targeting both general and high-risk populations, but currently is mainly frequented by IDUs. The third site is an STI community clinic owned by an NGO. The clinic delivers mostly STI and VCT services and is visited by high-risk groups other than IDUs (e.g. female sex workers, transgenders and men having sex with men), as well as smaller numbers of the general population. The STI case load in this clinic is high (around 2000 cases within a year), and it is the only clinic with its own outreach team which also performs pretest counselling for VCT in Bandung. The clinic charges a highly discounted price for its services to improve access for high-risk groups from every income level. The funding for the clinic is mostly covered by donors with limited support from the government. The fourth site is an HIV clinic in a prison for drug users including IDUs. The VCT in this clinic is also performed as part of the screening test for new prisoners. At the moment, this clinic mostly relies on donor funding.

All clinics have their own complete VCT package, including pre- and post-test counselling and HIV testing, except for the prison clinic that sends all blood samples to the Hasan Sadikin Hospital laboratory. The HIV-community-based clinic refers samples for confirmatory tests at the hospital-clinic if the screening test is positive result. Each site has its own distinct characteristics of clients, VCT procedures, HIV-positive cases and settings (Table 1). HIV antibodies are measured by different commercially available rapid tests, the hospital laboratory also uses immunoassays, with external quality control showing 100% accuracy (National Serology Reference Laboratory, Australia). CD4 cell measurements are performed at the hospital laboratory using Facscount flowcytometry (BD Biosciences, Jakarta, Indonesia).

Table 1.   Overview of study sites
VariableHospital clinicHIV community clinicSTI community clinicPrison clinic
  1. VCT, voluntary counselling and testing; IDUs, injecting drug users.

  2. †Only includes pre- and post-test counselling; HIV blood-testing is conducted at Hasan Sadikin hospital.

  3. = 9, HIV (+) clients with traceable CD4 cell count.

Number of VCT56828784574
Type of high-risk groupsAll high-risk groupsMostly IDUsMostly sex workers and transgendersPrisoners
Service charge (US$)2.130.323.72n/a
Type of VCT offeredCompleteCompleteCompleteIncomplete†
Period of studyJanuary–December 2008December 2008–September 2009October 2007–September 2008May 2008–April 2009
HIV (+) detected, no. (% of VCT conducted)214 (38)4 (14)26 (3)39 (7)
CD4 cell count on HIV (+), median (IQR)129 (26–338)n/a292 (189–397)‡290 (165–382)
Enrolled in HIV care (%)8808100

Data collection and cost estimation

We distinguished health-care costs (costs related to the consumption of resources in the health-care system), and non-health-care costs (costs falling on the patient for seeking and undergoing care). Health-care costs were estimated on the basis of service utilization and service cost data for each clinic (Table 1). We retrieved data on service utilization from each clinic’s records including information on number of VCT and resources used. Service costs were estimated using a micro-costing approach (Drummond et al. 2005). All resources consumed and salaries were listed and estimated on the basis of clinical records, interviews with medical staff, government standard, or market price. Costs to the clinic consist of recurrent and capital costs. Personnel recurrent costs were estimated on the basis of their actual wage whenever possible, or estimated by using government salary scales (issued by the Ministry of Finance in 2007). Other recurrent costs such as the cost of medical and administration goods consumed during the 1-year period of observation were estimated using both actual and market prices. Laboratory cost was estimated using the current service charge. Capital costs included trainings and workshops attended by the clinic staff, and unit costs on organizing these activities were obtained from budget or government records. Market prices were used to estimate other capital costs, which comprise the cost of equipment, furniture and start-up costs (such as renovation cost, if applicable). Capital costs were subsequently annualized on the basis of the life time of the capital items, using a 3% discount rate (Drummond et al. 2005). We omitted the cost of utilities (i.e. water, electricity). Although HIV tests administered in the clinic were sent to a laboratory outside the prison for diagnosis, associated costs were included in the prison clinic costing analysis.

Non-health-care costs (i.e. patient costs) were estimated by conducting a survey among a sample of high-risk group clients utilizing the VCT service at the hospital and STI clinic between September and October 2009. For the HIV community clinic patient costs were obtained from another survey conducted between June and August 2009. We collected information on, among others, travel costs, travelling time, the monthly income of the client, the average number of daily working hours and monthly expenditures. Such study was not conducted at the prison. All costs were measured in Rupiah and converted to US$ using the 2008 exchange rate (World Bank 2009). Both the utilization and cost data were entered into and analysed with Microsoft Excel 2007.

In addition, we collected the data on the number of HIV-positive clients and their CD4 cell count from each site. We also interviewed the clinic managers to put the observed data in context.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Number of patients and visits

The number of VCTs conducted varied considerably between settings. Most of the VCTs were conducted at the STI community, but most positive cases were found at the hospital clinic (Table 1). The median CD4 cell count at the time of HIV diagnosis was relatively low in the hospital clinic, higher in the STI community clinic and prison and unavailable for the HIV community clinic. In the hospital clinic and prison, most patients diagnosed through VCT enrolled in HIV care after VCT, whereas very few did so in the community clinics (Table 1). Table 2 provides the characteristics of the sampled clients for each of the clinic. Most clients were young and employed. Travelling and waiting time were the highest for the hospital clinic, lower for the community clinics and not applicable for the prison clinic.

Table 2.   Characteristics of sampled high-risk group clients getting voluntary counselling and testing (excluding inmates)
Variable/clinicHospital clinic (= 66)HIV community clinic (= 57)STI community clinic (= 61)
Male, no. (%)37 (56)52 (91)32 (53)
Age, mean282825
Occupation, no. (%)
 Employed58 (93)46 (81)57 (93)
 Unemployed and/or not in labour force4 (7)11 (19) (n = 22)4 (7)
Mode of transportation to clinic, no. (%)
 Public transport33 (50)4 (18)23 (38)
 Motorcycle29 (44)17 (77)24 (39)
 Others4 (6)1 (5)14 (23)
Travelling time to the clinic, minutes, mean4539 (n = 22)20 (n = 60)
Spent time at the clinic, minutes, mean94n/a38
Client’s monthly estimated income, USD (median)48 (36–72)77 (36–128)124 (52–246)

Cost of VCT services from different clinic sites

Table 3 summarizes the results of VCT costing analysis. The unit cost per VCT was highest in the hospital clinic (US$74), followed by the STI community clinic (US$65), the HIV community clinic (US$39) and the prison (US$23). Personnel recurrent cost was the largest cost component for both hospital and STI community clinic (70% and 41%). At the HIV community clinic, equipment costs constituted the largest portion of costs (30%), while the estimated laboratory cost was the largest cost item for the prison clinic (34%). Non-health-care costs (i.e. patient costs) per VCT were lowest at the HIV community clinic, and approximately five times higher at the STI community clinic. Interviews with managers of the clinics reveal that the hospital clinic is operating at full capacity (i.e. cannot always treat clients promptly), the HIV community clinic is operating at low capacity and the STI community clinic is operating at a capacity between the previous two clinics.

Table 3.   Annual cost of delivering VCT service for different settings (US$, 2008 exchange rate)
Type of cost/clinicHospital clinicHIV community clinicSTI community clinicPrison clinic
  1. VCT, voluntary counselling and testing.

  2. †Productivity losses signify income losses because of seeking and undergoing VCT.

  3. ‡Not costed separately, but included in the capital and recurrent cost items .

Health care costs
 Capital costs (annualized)
  Personnel (trainings and workshops)14520742204194
  Building/space6571812661624
  Equipments65729835130
 Subtotal145968548375848
 Capital costs per VCT324610
 Recurrent costs
  Personnel21 6335721 0512116
  Supplies259126310 007539
  Outreach team activities6186
  Laboratory3255314439
 Subtotal27 47935137 2447093
 Recurrent cost per VCT62114812
Non-health care costs
 Patient costs2157468862n/a
 Access costs per VCT10.33n/a
 Estimated productivity loss per VCT†41.38n/a
 Patient costs per VCT5211n/a
 Total annual cost31 095108350 94212 941
 Number of VCT42128784574
 Unit costs per VCT74396523
 Unit costs per HIV positive case1782711959332

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

There is a need to scale up VCT services in Indonesia, and a variety of service delivery models has been developed, each with its own characteristics in terms of its clients and services. The findings of this paper are summarized in Table 4 in dimensions of economic aspects (costs and accessibility), clinical aspects (VCT performed, % HIV positive, CD4 cell count, enrolment in HIV care) and other aspects (VCT target group), and lead to the following observations.

Table 4.   Economic, clinical and other aspects of VCT delivery models†
Service delivery modelEconomic aspectsClinical aspectsOther aspects
Unit cost per VCT (US$)Physical AccessibilityVCT performed% HIV positiveCD4 countEnrollment in HIV careVCT target group
At presentIf scaled upAt presentIf scaled up
  1. †Perceptions of these aspects as relative to other clinics. Perceptions relate to present situation, unless indicated otherwise.

  2. VCT, voluntary counselling and testing; IDU, injecting drug users.

Hospital clinicHighHighLowManyManyLowHighSymptomaticSymptomatic
HIV community clinicLowLowHighFewFewNALowMostly IDUAll
STI community clinicHighHighHighManyFewHighLowMostly sex workers and transgendersAll
Prison clinicLowNANAManyFewHighHighPrisonersPrisoners

The hospital clinic does not seem the most appropriate venue for scaling up the delivery of its VCT services to the broader population. The costs per VCT performed at the hospital-based clinic (US$74) are higher than those provided at the HIV and STI community clinics. Also, the clinic is operating at full capacity and thus further economies of scale from scaling up services along with cost reductions may not be possible. The clinic is also less physically accessible than the other clinics as indicated by the relatively long travelling and waiting time. The role of the clinic merely seems to be that of a referral centre for treatment of late and/or complicated cases and for treatment of opportunistic infections, given that it is frequented by individuals presenting for testing with symptomatic and late-stage disease (as indicated by the relatively low CD4 count among HIV-positive-tested patients) (Wisaksana et al. 2009).

The HIV community clinic holds potential to function as an entry point to VCT for the broader population, and the following observations in the present study qualify this perspective. The clinic has the lowest costs per VCT performed (US$39) compared with the hospital clinic and STI community clinic. Considering its recent establishment and currently low capacity utilization, cost reductions can be expected when these services are scaled up. Also, the HIV community clinic is relatively physically accessible as indicated by the short travelling time, and this holds potential for a high uptake of VCT when scaled up. However, as the present clinic is now mainly frequented by IDUs, more efforts are needed to target and also attract the (symptomatic and asymptomatic) population as a whole. Concerns are the (present) absence of HIV care or a good referral system to a clinic with HIV care, and its current inability to cover all costs by collecting fees (nor does it receive financial support from the government).

The STI community clinic could play an important role in scaling up VCT services, specifically by continuing its focus on high-risk groups as it does now. The clinic appears to be able to detect HIV-positive cases in an early stage, as indicated by the relatively high CD4 cell count of HIV positive clients tested. The clinic’s outreach activities and the resulting network with high-risk groups probably contribute to its success. The present clinic is located near the area of where the sex workers live and work, which improves physical accessibility, as indicated by relatively low travel and waiting time. Moreover, the clinic’s active campaign on delivering STI services gives it good access to other high-risk groups such as transgenders and male having sex with male. This holds potential for a high uptake of VCT when this decentralised service model is scaled up. In relation, the focus of the clinic on the provision of STI services may also reduce HIV transmission (Kreiss et al. 1989; Schwandt et al. 2006). The absence of a good referral system to a clinic with HIV care is a matter of concern, and currently the discounted price charged cannot cover the full cost of delivering the services (the clinic relies heavily on donor funds).

The VCT services as delivered in the prison clinic hold potential to be scaled up to other prisons. During the period of study, both the number of VCT performed and the CD4 cell counts of the HIV-positive clients were relatively high, as was the enrolment in HIV care. This demonstrates the ability of the clinic to identify and treat HIV infection in prison at an early stage, although one should realize that HIV care for prisoners after release might pose a bigger challenge. Furthermore, the delivery of VCT in the prison clinic is relatively cheap (US$23). Personnel costs are relatively low and patient cost of seeking and undergoing the service are absent. Currently, the prison clinic is relying on donor funding for its VCT services. Whether VCT services should also be implemented in other prisons targeting other groups is difficult to answer in the present study and warrants further investigation.

Based on our study findings we can make several recommendations (Table 5). VCT services may best be scaled up in community-based clinics to the broader population, to improve their uptake and hence early detection of HIV/AIDS. These clinics may also initiate treatment of early/uncomplicated cases. A system should be developed for referral of late cases and those with opportunistic infections to the hospital. The HIV clinic in the hospital should be less oriented towards conducting VCT. We also emphasize the need for measures that address the general barriers to VCT such as (travel) costs, lack of confidentiality of test results, and perceived lack of benefit of conducting HIV test (Suherman et al. 2009).

Table 5.   Recommendation on role of clinics in VCT and treatment†
Service delivery modelVCTUncomplicated ARTComplicated ARTTreatment of opportunistic infections
  1. VCT, voluntary counselling and testing; ART, anti retroviral treatment.

  2. †‘+’ and ‘−’ denote respectively a role of high and low importance for the clinic in the specified activities in HIV/AIDS control.

Hospital clinic+/−+/−++
HIV community clinic++
Community clinic++
Prison clinic++

Our results should be interpreted with some caution. First, this study evaluated highly contextualised VCT service delivery models, which hampers their generalizibility. Yet, the conceptual decision framework can be applied to other settings. Second, we estimated patient costs of seeking and undergoing care on the basis of (assumed) productivity losses – these may not be realized in practice and we may then have overestimated the total patient costs. Differences in these patient costs are related to income levels of the clinic’s clients and do therefore not necessarily reflect differences in access to the different clinics. Third, personnel costs in clinics may include inefficiencies in the delivery of VCT services, and we may have overestimated total costs – a time motion study is warranted to assess the required resources to scale up VCT services in further detail.

In conclusion, we propose a reorientation of the delivery models for VCT in Indonesia, which would improve access and promote earlier detection, and reduce the burden of the hospital clinic to orient itself towards the treatment of complicated HIV/AIDS cases. Whereas this study is specific to Bandung, Indonesia, the conceptual framework and overall conclusions may also hold relevance to other low-income settings.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank the staff of the clinics involved for making the needed data available to us for analysis. We acknowledge the financial support of the European Commission (Sixth Framework Programme; Specific Targeted Research Project no: 318/10150A – EC SANTE/2005/105-033).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Afriandi A, Aditama T, Mustikawati D et al. (2009) HIV and injecting drug use in Indonesia: epidemiology and national response. Acta Medica Indonesiana 41(Suppl. 1), 7578.
  • Dandona L, Kumar SGP & Ramesh YK et al. (2008) Changing cost of HIV interventions in the context of scaling-up in India. AIDS 22: S43S49, doi:10.1097/01.aids.0000327622.24232.aa.
  • Drummond MF, Sculpher MJ, Torrance GW et al. (2005) Methods for the Economic Evaluation of Health Care Programmes, 3rd edn. Oxford University Press, New York.
  • General Directorate of Communicable Disease Control and Environmental Health, Ministry of Health, Republic of Indonesia (MOH-RI) (2009) Case Statistics of HIV/AIDS in Indonesia – Report of March 2009. MOH-RI, Jakarta.
  • Health Intervention and Technology Assessment Program (HITAP) (2008) The Potential of Provider-Initiated Voluntary HIV Counselling and Testing at Health Care Settings in Thailand, Final Report. 2008. HITAP, Thailand.
  • Hogan DR, Baltussen R & Hayashi C et al. (2005) Cost effectiveness analysis of strategies to combat HIV/AIDS in developing countries, achieving the millennium development goals for health. British Medical Journal 331, 14311437.
  • Johns B, Baltussen R & Hutubessy R (2003) Programme costs in the economic evaluation of health interventions. Cost effectiveness and resource allocation 1, 1, http://www.resource-allocation.com/content/1/1/1 .
  • Kreiss K, Coombs R, Plummer F et al. (1989) Isolation of human immunodeficiency virus from genital ulcers in Nairobi Prostitues. Journal of Infectious Diseases 160, 380384.
  • Mmbando D, Kironde S, Mbekem G, Richardson J, Reeler A & Bhatt P (2004) Scaling-up VCT Service Delivery: Results from an ‘Integrated’ Approach Using Rapid Tests in Primary Health Care Facilities in 3 Districts of Tanzania. Presented on International Conference on AIDS (15th: 2004: Bangkok, Thailand)
  • Schwandt M, Morris C, Ferguson A, Ngugi E & Moses S (2006) Anal and dry sex in commercial sex work, and relation to risk for sexually transmitted infections and HIV in Meru, Kenya. Sexually Transmitted Infections 82, 392396, doi:10.1136/sti.2006.019794.
  • Siregar AYM, Komarudin D, Leuwol B, Afriandi I, Djuhaeni H & Baltussen R (2009) Economic aspect of HIV/AIDS control and injecting drug use in Indonesia. Acta Medica Indonesiana 41 (Suppl. 1), 7074.
  • Suherman H, Sodjakusumah TI & Fitriana E et al. (2009) Perceived barriers of taking HIV test (VCT) among IDUs in Bandung. IMPACT; unpublished.
  • Teerawattananon Y, Vos T, Tangcharoensathien V & Mugford M (2005) Cost-effectiveness of models for prevention of vertical HIV transmission – voluntary counseling and testing and choices of drug regimen. Cost Effectiveness and Resource Allocation 3: 7; doi:10.1186/1478-7547-3-7.
  • The World Bank (2009) WDI Online (World Development Indicators) 2008. Available from: http://ddp-ext.worldbank.org/ext/DDPQQ/member.do?method=getMembers &userid=1&queryId=6. (accessed 20 October 2009).
  • UNAIDS (2001) The Impact of Voluntary Counselling and Testing: A Global Review and Challenges: June 2001. UNAIDS, Geneva.
  • UNAIDS, WHO (2007) AIDS epidemic update: December 2007. Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Organzation (WHO), Geneva.
  • Walker D (2003) Cost and cost-effectiveness of HIV/AIDS prevention strategies in developing countries: is there an evidence base? Health Policy and Planning 18, 417.
  • Wisaksana R, Alijsahbana B, Crevel R et al. (2009) Challenges in delivering HIV-care in Indonesia: experience from a referral hospital. Acta Medica Indonesiana 41(Suppl. 1), 4551.