Secondary Logo

Journal Logo

Original articles

The economics of effective AIDS treatment in Thailand

Over, Meada; Revenga, Anab; Masaki, Emikoc; Peerapatanapokin, Wiwatd; Gold, Juliane; Tangcharoensathien, Virojf; Thanprasertsuk, Sombatg

Author Information
doi: 10.1097/01.aids.0000279713.39675.1c
  • Free



Thailand is in the vanguard of developing countries that are seeking to provide antiretroviral therapy (ART) as the standard of care to large numbers of individuals with symptomatic HIV disease. As of December 2006, 88 457 people living with HIV/AIDS (PHA) in Thailand had received ART through the National Access to Antiretroviral Program for People Living with HIV/AIDS (NAPHA), and approximately 23 000 are estimated to receive ART through the Social Security Scheme, resulting in a total of more than 111 000 individuals receiving government-financed ART. In 2006, the Royal Thai Government announced a policy of providing universal access for ART through its health and social security funds [1,2].

Provision of antiretroviral therapy in Thailand

The first case of AIDS in Thailand was reported in September 1984. Since then, more than one million Thais have been infected with HIV, and of those more than 400 000 have died. In 2004, an estimated 572 500 Thais were living with HIV/AIDS. Among those individuals, some 49 500 will develop serious AIDS-related illnesses during the year, and approximately the same number will die of AIDS-related complications. It is also estimated that 15 174 new infections will occur in 2006 (continuing the downward trend from 143 000 new infections in 1990 and 23 676 new infections in 2002).

The Royal Thai Government and Thai society have demonstrated a strong commitment to providing comprehensive care and support to PHA, but it is only recently that they have been able to provide ART to large numbers of individuals with symptomatic HIV (Fig. 1). The availability of a domestically produced triple-drug combination, GPO-vir, at affordable prices (approximately B1200, or US$30 per month) has enabled many PHA to access ART who could previously not afford it, and has allowed the Ministry of Public Health (MOPH) to roll out a large-scale campaign to provide triple-drug ART as standard care. GPO-vir is a single tablet that has a three-drug combination regimen (stavudine plus lamivudine plus nevirapine) and is produced by the Government Pharmaceutical Organization.

Fig. 1
Fig. 1:
Timeline for access to HIV/AIDS medical care in Thailand. ART, Antiretroviral therapy; ddC, zalcitabine; ddI, didanosine; ZDV, zidovudine. Source: Ministry of Public Health [3].

As a result, the number of PHA on ART has increased sharply from approximately 3000 at the start of 2002 to 27 000 by the end of 2003, rising to 52 593 by February 2005 and to 111 457 by December 2006. The programme is partly financed by the Global Fund to Fight AIDS, Tuberculosis, and Malaria and partly by the government budget. The number of treatment sites increased from 112 in 2001 to 462 in early 2003, rising to 841 by February 2005.

Table 1[3–6] shows the evolution of the estimated cases of PHA, estimates of AIDS cases, and the numbers of individuals on public ART in Thailand. As of the end of 2004, more than 80% of those living with AIDS were on public ART, a remarkable achievement. Since that year, the ART programme has been able to raise peoples’ CD4 cell levels until they are no longer suffering from AIDS, so that now the number of people on ART is more than double the number living with AIDS.

Table 1
Table 1:
Estimated cases of people living with HIV/AIDS, AIDS cases, reported AIDS cases in public hospitals, and access to public antiretroviral therapy.

Benefits of previous prevention efforts

The successful implementation of NAPHA poses a significant challenge to the Royal Thai Government and to Thai society now and in the future. In a few years, as AIDS patients live longer with ART, the health system will need to provide care not only for 10 000–20 000 new cases per year, but also for those who started treatment in previous years and are still alive. Given the commitment across all segments of Thai society, the country's significant health system capacity, and the availability of internal and external financing, Thailand has good prospects for meeting this challenge. The picture would be very different if Thailand had not succeeded in its past HIV prevention efforts.

In the absence of past successful efforts at national prevention, HIV infections, and thus AIDS cases would now be much more prevalent than they actually are (Fig. 2). We have estimated that without such efforts, Thailand would have 7.7 million HIV cases and 850 000 AIDS cases in 2005, approximately 14 times more of each than exist in reality. See Brown and Peerapatanapokin [7] and Revenga et al. [2] for descriptions of the projection model. The ‘no prevention’ scenario was constructed under the assumption that risk behavior and the prevalence rates of sexually transmitted infections remained at 1990 levels to 2010. If we suppose that Thailand would try to offer ART to all of this much larger number of AIDS patients, its budget requirements would also be 14 times greater and would continue to grow over the next decade. Thanks to its extensive and effective prevention efforts between 1991 and 2001, Thailand has avoided the need to spend an additional US$18.6 billion (B745 billion) on treatment over the decade between 2002 and 2012.

Fig. 2
Fig. 2:
Importance of previous prevention efforts. Source: Brown and Peerapatanapokin [7].

During the 1990s, Thailand spent more than most countries on its HIV/AIDS programme. Over this period (1991–2001) its total budget expenditure on HIV prevention and on very limited treatment (primarily of opportunistic illnesses) was a substantial US$434 million in 2002 dollars (B17.3 billion) [8,9]. By averting the need to spend US$18.6 billion (B745 billion) over the subsequent decade, each dollar (or baht) invested in the 1990s saved US$43 (or B1720) in needed treatment expenditure in the subsequent decade. It is doubtful that any other Thai government investment achieved such a high benefit–cost ratio. The finance ministries of countries such as China and India [10], where the HIV/AIDS epidemic is at an earlier stage, should be aware of the high financial return to HIV/AIDS prevention campaigns when they make decisions on allocations between HIV/AIDS prevention and other programmes.

Even in Thailand, however, there is no room for complacency. Although policies have been successful in lowering the rate of new infections, the prevalence of HIV in high-risk groups is still high, especially among those groups that past prevention efforts did not explicitly target, such as intravenous drug users or male sex workers. There is also evidence that the effect of past prevention campaigns is waning [11]. Recent rounds of the Behavioral Surveillance Survey show that the percentage of male conscripts reporting sexual relationships with commercial sex workers, after declining for several years, started to increase again in 2002 and 2003 [12]. The same pattern is visible with other female sex partners and among married conscripts who have extramarital sex. Condom use among those male conscripts is not high: only 59% report consistent use of condoms with sex workers, and only 25% do so with non-regular female sex partners [12]. New risk behaviors by other groups, such as youth, also need to be addressed.

Methods: measurement of the effect of antiretroviral drug policy

The objective of ART policy is to lengthen and improve the lives of the recipients, and therefore a natural measure of the effectiveness of a policy is the number of life-years that ART adds to the population [13,14]. Public policy makers need to consider not only the direct effects of ART policy on the patients receiving ART, but also the indirect effects of ART on the creation of new HIV infections. For example, increasing evidence suggests that ART patients are significantly less infectious than they would be in the absence of ART or if taking only one or two drugs. Growing evidence also indicates that greater access to ART may lead to complacency and increased risk behavior by individuals on ART and by people (whether HIV negative or positive) in surrounding communities [15–18].

Assessing the spillover effects of ART on individuals other than the patient is an indispensable part of designing a policy. Table 2[19] provides a classification of the indirect (or external) effects into biological and behavioral effects on transmission. Within each of the categories, the effects could be beneficial by slowing transmission or could be adverse by accelerating it. The measure of the importance of any of those effects is the rate of new HIV infections, or the HIV incidence rate.

Table 2
Table 2:
Possible effects of antiretroviral therapy on HIV transmission.

To project the costs of ART policy, we have adopted the simplifying assumption that most programme costs will be related to the individual patient in the form of provider time, pharmaceutical products, diagnostic tests, and disposable paper and rubber products. Therefore, the costs will not vary much with scale or scope. The exception we are making to this rule is the cost of equipping a healthcare facility that is not a district hospital with the capacity to administer ART. We assume that in order to qualify to manage one or more ART patients, any facility must train a minimum number of providers in ART protocols, and that to keep providers abreast of the rapidly changing technology of ART, the facility must retrain those staff members every year. This category of costs might be called recurrent fixed costs. The costs recur every year as a function of the number of facilities but can be spread over all the patients at the facility.

To estimate the effect of NAPHA and then of the cost effectiveness of various modifications of NAPHA, we must model how the specific changes introduced by NAPHA influence the behavior of patients and providers. Our approach is to construct a model of the links between government instruments and policy outcomes. We do this in five steps summarized in Figure 3.

Fig. 3
Fig. 3:
Policy instruments to epidemiology to performance. AEM, Asian Epidemiological Model; ART, antiretroviral therapy; OI, opportunistic infection; PHA, people living with HIV/AIDS; VCT, voluntary counselling and testing. Source: Revenga et al. [2].

First, we model the link from the two primary policy instruments, price and availability (or supply), to the distribution of patient demand for care. Second, we project the evolution of prices and availability into the future, and then we use conditional logit models of demand for testing and for care to compute the projected distribution of demand across treatment options under alternative policy options [2,20]. Third, we apply the expected demand to the Thai population of eligible infected individuals, as projected by an updated version of the Asian Epidemiological Model (AEM), in order to deduce the future of ART use. See Brown and Peerapatanapokin [7] for a description of the AEM without ART and Revenga et al. [2] for a description of the modifications introduced to incorporate ART. Fourth, we estimate the direct and spillover effects of ART use on health and HIV incidence. We draw on the literature on the effectiveness of ART in order to make detailed assumptions regarding patient survival under early or late recruitment and for each policy scenario [21,22]. Also see Chapter 3 and Appendix A of Revenga et al. [2]. Fifth, we apply unit costs to estimate the financial burden of the NAPHA policy.

Projection scenarios

The effect of a policy choice can be defined only in comparison with what would have happened in the absence of this choice. This alternative scenario, called a baseline or counterfactual, is a projection of the future course of AIDS treatment if the Royal Thai Government had continued HIV prevention but refrained from introducing its expanded NAPHA programme. Several alternative baseline scenarios could have been chosen, see cells (A), (B) and (C) in Table 3. Each of these scenarios corresponds to a different combination of public financing of ART and government subsidization of the production and sale of low-cost ART. The chosen baseline corresponds to cell (A), which represents what would have happened if the government had kept only its HIV prevention programmes and its previously existing (pre-2001) voluntary treatment programme, with only branded drugs available. The effect of NAPHA is obtained by comparing outcomes from cell (A) with those from cell (D). In terms of Figure 2, the baseline constitutes the lower of the two trend lines up to 2003. The total effect could be separated into one part because of the availability of low-cost generic ART and another part as a result of the public finance of ART provision. Such a deconstruction would enable us to attribute some portions of the benefits of NAPHA to each of its two components and another portion to the synergy between them.

Table 3
Table 3:
Potential baseline scenarios or counterfactuals to National Access to Antiretroviral Program for People Living with HIV/AIDS.

In addition to the NAPHA policy scenario described above, the report considers two enhancements to NAPHA and a third policy that would combine those two enhancements (Table 4). The enhancements are chosen to address what are perceived by knowledgeable Thai and international observers to be potential weak points in the NAPHA programme and, indeed, in all publicly financed and publicly provided ART programmes worldwide.

Table 4
Table 4:
Policy scenarios for the National Access to Antiretroviral Program for People Living with HIV/AIDS programme.

Early analyses of the general effectiveness and cost effectiveness of publicly provided ART assumed that many HIV-infected patients would be recruited to treatment when their immune systems first dropped below an eligibility threshold. As a result, the benefits of ART would be maximized. Experience in Thailand, however, as well as in several other countries (for example, Botswana, Brazil, Malawi, and member states of the Organization for Economic Co-operation and Development), shows that most patients are identified as being ART eligible only when their opportunistic illnesses lead them to the hospital, which is usually when their CD4 cell counts are already well below the threshold at which they would most benefit from care. It is thus useful to analyse an alternative version of NAPHA that would include a much more vigorous promotion of voluntary counselling and testing (VCT) in an effort to attract patients into treatment when they first become eligible for it. As an alternative to the promotion of VCT, the Thai government could encourage earlier initiation of ART by instituting a policy of routine or ‘opt-out’ testing in government health centers [23–25]. We model this as ‘VCT scenario (D2)’.

A major challenge for ART programmes will be to attain and sustain high levels of adherence among their patients. Ministry-sponsored training programmes for public sector ART providers are currently teaching the importance of adherence. Experience around the world, however, suggests that as ART treatment is scaled up it will be increasingly difficult to attain high levels of adherence among new patients and to sustain those levels among all patients [26]. One promising approach with which Thailand has already experimented is both to subsidize and to facilitate the establishment of non-governmental organizations that provide emotional, physical, and sometimes even financial support to patients [27]. In this report, we refer to public sector delivery that has been strengthened by adding those demand-enhancing programmes as augmented public delivery of ART. Our ‘Augmented (D3)’ scenario is intended to capture the incremental benefits and costs of such a programme.

We also model a ‘Both (D4)’ programme, which includes the costs both of expanded VCT and of augmented adherence, and models a synergistic benefit between them. Because the AEM epidemiological model represents the entire adult Thai population, not just the patients, it can capture all of the effects of treatment on transmission that are listed in Table 2. In scenario D4, when both augmented interventions are implemented, lower viral loads and longer durations of infectiousness as a result of one intervention have interaction effects, both positive and negative on the effects of the other intervention.

Costs of antiretroviral therapy

Costs of ART can be defined in many ways, such as costs to the public sector, to individual patients, and to society. To evaluate the various policy options for expanding the public provision of ART in Thailand, we adopt the perspective of, and estimate the costs to, the public sector. Average costs of ART per patient are estimated according to the types of treatment regimens (first-line therapy and second-line therapy), the modes of service delivery (public, augmented public, and private service delivery), and stages of the disease (asymptomatic and symptomatic HIV). Specific cost components included in estimating average costs of ART per patient are the costs of antiretroviral drugs, laboratory tests, and monitoring, costs of treating opportunistic infections, and PHA support expenses.

Cost data were obtained from existing studies in Thailand, both published and unpublished, and from informal consultations with local and international experts.

Table 5[28] summarizes the costs of various regimens available in 2004–2006 and recommended by the Thai MOPH and World Health Organization in their treatment guidelines. Antiretroviral drug costs in Thailand are somewhat higher than the average costs of similar drugs for middle-income countries reported in World Health Organization, UNAIDS, UNICEF [29]. In May 2007, the Thai MOPH was still paying the price for lopinavir/ritonavir negotiated with Abbott in 2004. Subsequent lower offers by Abbott have come with conditions, such as no further price negotiations until 2011, which the Thai MOPH had not accepted at the time of writing. Negotiations with generic companies by the Thai MOPH were dropped following support for a pooled procurement procedure facilitated by the Clinton AIDS Foundation [30].

Table 5
Table 5:
Costs of antiretroviral drugs per patient by types of regimens in Thailand.

The annual costs of antiretroviral drugs vary significantly between first-line and second-line regimens, ranging from B14 400 (using GPO-vir) to B273 864 (using expensive protease inhibitors) per patient per year. The average cost of first-line ART regimens is estimated at B19 271 (US$481.80) per patient per year, using a weighted average of three categories of ART drug regimens under the MOPH treatment guideline. Weights are distributed by 80, 15, and 5%, respectively, for the antiretroviral drug regimes (1), (2), and (3) as shown in Table 6[31].

Table 6
Table 6:
Annual cost per patient by types of drug regimens.

The average cost of second-line regimens is estimated at B270 000 (US$6740) per patient, costing 14 times more than the average cost of first-line regimens.

In addition to the cost of antiretroviral drugs, significant costs are associated with providing and monitoring ART. The costs of outpatient and inpatient services are not negligible because providers must be trained and repeatedly retrained, and because the individual patient's use of medical services increases at the time of initiating ART. On the basis of the above estimates, the annual average cost of ART using first-line therapy is estimated at B33 700 (US$840) per patient (Table 6). The costs of antiretroviral drugs and laboratory monitoring represent nearly 60% of the total ART expenses when first-line therapy is used, and the costs increase to 95% of the ART expenses when patients are on second-line therapy.

We expect that continuous care and support from PHA groups will become increasingly important in expanding the public provision of ART, especially by supporting patients in their adherence to ART. Although there are many publicly funded treatment sites with associated PHA groups, little is known about their resource requirements and financial sustainability. With the help of Medecins Sans Frontieres Thailand, we obtained preliminary estimates of the costs of PHA support to improve adherence. These estimates suggest that such support costs approximately B3100 (US$78) per patient per year; or in other words, adds some 8–9% to the total cost of ART per patient per year [32]. The cost of PHA support is used to estimate the annual cost of augmented ART per patient.

Cost effectiveness of first-line therapy

NAPHA with first-line therapy only, at a cost of US$736 or B29 440 per discounted life-year saved, is the most affordable and cost-effective policy option modelled in this report. Figure 4 presents the cost per life-year saved of NAPHA without second-line drugs when compared with the cost per life-year saved of NAPHA with second-line therapy. The central finding is that the NAPHA programme with second-line drugs can save years of life for US$2145, (B85 800), whereas NAPHA with only first-line therapy is far more cost effective at US$736 (B29 440) per discounted life-year saved.

Fig. 4
Fig. 4:
Cost-effectiveness of National Access to Antiretroviral Program for People Living with HIV/AIDS with first-line antiretroviral therapy compared with National Access to Antiretroviral Program for People Living with HIV/AIDS with both first and second-line antiretroviral therapy. ▪ National Access to Antiretroviral Program for People Living with HIV/AIDS (NAPHA), first-line antiretroviral therapy (ART) only; □ NAPHA, first and second-line ART. Source: Authors.

In the absence of a significant change in the price of second-line drugs, the cost of saving an additional life-year through second-line therapy is high when compared with the substantial health benefits of first-line policy only. Using a pure cost-effectiveness criterion, a policy that subsidizes only first-line therapy would be superior. Other considerations may, however, weigh heavily on the government's final decision as to what policy to adopt. From the perspective of this study, affordability and equity are also relevant criteria.

An argument for horizontal equity (i.e. fairness across patients with different life-threatening conditions) would compare NAPHA with and without second-line therapy with the cost of saving life-years through the subsidized treatment of other adult illnesses, such as cancer, heart disease, or end-stage renal disease. This comparison would show that NAPHA with second-line therapy is cost effective compared with those other interventions. For example, a recent study of the cost effectiveness of hemodialysis in Thailand found the cost per quality-adjusted life year to be approximately US$63 000 per year [33]. Advocates of vertical equity would argue that the government should ensure that the bottom of the income distribution has access to care that the top one-fifth would purchase for themselves (including second-line therapy).

By 2015, the current NAPHA policy will have added approximately 220 000 individuals per year to the living population. Even at the end of the projection horizon, when the Thai AIDS epidemic is predicted to slow, the NAPHA policy will be saving approximately 190 000 life-years each year (Fig. 5). Therefore, 10% more life-years would be saved under the current NAPHA policy than under an equivalent NAPHA without second-line therapy. By keeping people alive longer, NAPHA will cause an increase in the number of HIV-infected individuals in Thailand, as well as a significant increase in the number of PHA who are on treatment. As treatment success by definition increases HIV prevalence rates, the government will need to measure the success of its HIV prevention programme by reductions in HIV incidence rather than by reductions in prevalence.

Fig. 5
Fig. 5:
Benefits (life-years saved) and costs of National Access to Antiretroviral Program for People Living with HIV/AIDS relative to baseline.
Fig. 1
Life-years saved;
Fig. 1
net cost of antiretroviral therapy (ART). Source: Authors.

The total cost of NAPHA with second-line therapy reaches a ceiling at US$500 million or B20 billion per year in 2020. Beginning in 2008, expenditure on second-line therapy accounts for more than one-half of total ART spending. By the end of the projection, a quarter of the patients receiving second-line therapy will absorb three-quarters of the treatment budget (Fig. 6). The projected cost of NAPHA will increase Thailand's AIDS spending from its current level of approximately US$100 million or B4 billion per year to more than five times that amount in 2020. Even at its peak, however, total spending on AIDS treatment will require increasing total healthcare spending by less than 25%. We judge this level of expenditure to be affordable to the Thai government.

Fig. 6
Fig. 6:
After 2010, most costs are for second-line therapy. ‘Symptomatic’ refers to patients whose CD4 cell counts are below 50 cells/μl and are presumed to manifest symptoms of HIV infection. ‘Asymptomatic’ refers to patients whose CD4 cell counts are between 50 and 200 cells/μl and are assumed not to manifest such symptoms.
Fig. 1
Cost of public antiretroviral therapy (ART), first-line asymptomatic;
Fig. 1
cost of public ART, first-line symptomatic; □ cost of public ART, second-line asymptomatic; ▪ cost of public ART, second-line symptomatic. Source: Authors.

Policy options to enhance adherence and to recruit patients earlier

Timely patient recruitment and enhanced adherence buy additional life-years. See Revenga et al. [2] for the detailed assumptions and their basis in international and Thai data. If initiated now, the expanded VCT, augmented adherence, and ‘both’ policies would benefit more patients each year until the approaches save, respectively, 18 000, 50 000, and 60 000 additional life-years in 2020, on top of the 210 000 life-years saved that were generated by NAPHA alone in that year (Fig. 7). Therefore, for 2020, the alternative policies offer the possibility of improving NAPHA benefits by almost 30%.

Fig. 7
Fig. 7:
Projected annual life-years saved under alternative scenarios relative to baseline.
Fig. 1
Scenario D1;
Fig. 1
scenario D2;
Fig. 1
scenario D3;
Fig. 1
scenario D4. Source: Authors.

The expanded policies, however, also involve additional costs [2]. Of the four policies considered, the current NAPHA policy is the most cost effective (Fig. 8). The second-most cost effective is the augmented treatment policy, which enhances patient adherence. We estimate that systematically adding patient support groups to all treatment sites in Thailand would increase the cost per life-year saved by less than US$40 (B1600), thereby making this approach a good investment. As stated above, the annual cost of PHA groups is approximately US$78 per year. For this expenditure, we posit that the country achieves somewhat longer life expectancies, and thus postpones the treatment costs that occur when treatment fails, and also somewhat reduced annual HIV transmission during the patient's lifetime. These offsetting benefits bring the net incremental cost per life-year down from US$78 to less than US$40. Under the central assumptions of the model, spending the resources on expanded HIV testing in order to recruit patients in a more timely manner would increase the cost per life-year saved by only another US$60 (B2400), which also seems like a good buy. We thus recommend that Thailand undertake both of the two policies analysed to strengthen treatment, bringing the estimated cost per life-year saved to US$2243 or B89 720. These policies to strengthen ART are independent of, and much less costly than, the decision to finance second-line therapy. They would be even more affordable and advisable if public finance paid only for first-line therapy.

Fig. 8
Fig. 8:
Cost-effectiveness of National Access to Antiretroviral Program for People Living with HIV/AIDS and alternative scenarios relative to baseline.
Fig. 1
National Access to Antiretroviral Program for People Living with HIV/AIDS; ▪ voluntary counselling and testing; □ adherence;
Fig. 1
VCT plus adherence. Source: Authors.

Public financing will help ensure equitable access

Suppose that Thailand had reduced the price of first-line ART by authorizing the production of GPO-vir, but had refrained from subsidizing treatment. For the top two quintiles in the income distribution, first-line ART costs less per year than or equal to average household medical expenditures, and therefore could be considered to be affordable through user fees (Fig. 9). Even for the bottom two quintiles, the US$842 or B33 680 cost of first-line therapy compares with the medical expenses of the sickest households for a single year. The problems for the poorest households are likely to be caused by two unusual features of the cost of treatment. First, the treatment must continue for the rest of the patient's life. For those households in the lowest 80% of the income distribution that are able to raise the resources to pay the US$842 for one year, the second and third year will become increasingly onerous. Second, poverty-induced laxity in treatment will lead to treatment failure, to the development of resistant strains of the virus, to the spread of those resistant strains to others, and to the requirement that the patient move to second-line therapy.

Fig. 9
Fig. 9:
Affordability of antiretroviral therapy by income level. ▪ Mean household income;
Fig. 1
non-food expenditure per household;
Fig. 1
medical expenditure per household; –– –– cost of second-line antiretroviral therapy (ART); – – – cost of first-line ART. Source: Authors.

Although it is conceivable that the cost of first-line therapy could be partly financed with user fees, second-line therapy is much more expensive, exceeding the total household income of 40% of the population. Most individuals on first-line therapy will eventually need second-line therapy, and they will not be able to afford it without public support.

Public financing can strengthen positive spillovers and limit negative spillovers of antiretroviral therapy

ART may be used to increase the effectiveness of prevention activities, especially VCT, but this beneficial effect of ART requires a greater integration of treatment and prevention efforts than currently exists in Thailand.

Poor adherence to first-line therapy will speed the development of viral resistance to those drugs and will hasten the day when the patient must move to second-line therapy. Public intervention to support adherence can limit the spread of resistant virus. From a social as well as an individual perspective, adherence support mechanisms such as the augmented public care that we model in this report are likely to be cost effective as well as therapeutically beneficial.

Complacency about prevention could cause future costs to rise substantially

If the availability of ART is accompanied by a sustained government prevention programme and if it leads individuals to reduce risk behaviors such as injecting drugs and having unprotected sex, then the cost effectiveness of ART is improved by approximately 9%, and future government expenditures on ART will reduce by US$926 million (B37 billion) or 14%.

Conversely, if the availability of ART crowds out government expenditure on prevention and leads individuals to increase their risk behavior back to its levels of 1992, government treatment expenditure will increase more than threefold [2] [Table 6 (1)]. In such a case, the cost per life-year saved would increase nearly threefold, from US$2145 to US$6243 (from B85 800 to B249 720). In a similar modelling exercise in India, disinhibition was estimated to have an even more costly effect, potentially outweighing all the benefits of the ART programme [10]. It will clearly be important for the government to monitor closely the effect of its treatment programme on risk behavior.

Negotiated agreements on intellectual property rights for pharmaceuticals

The drugs used in second-line therapy are patented, produced, and sold by multinational pharmaceutical corporations, and therefore Thailand must either pay the high prices demanded by those monopolies or exercise its rights under World Trade Organization treaties to grant a compulsory licence for the manufacture of the drug, subject to negotiated royalties.

Thailand stands to gain a great deal from bilateral agreements to reduce trade barriers with trading partners such as the United States, and therefore the Royal Thai Government may be tempted to relinquish its rights to grant compulsory licences for AIDS drugs in exchange for proffered trade advantages. The report finds that the cost of such concessions would be large. For example, by exercising compulsory licensing to reduce the cost of second-line therapy by 90%, the government would reduce its future budgetary obligations for NAPHA by US$3.2 billion discounted (B127 billion discounted) to 2025, and would cut by more than half the cost per life-year saved of the NAPHA programme, from US$2145 to US$940 (or B85 800 to B37 600).

The size of royalty payments that the World Trade Organization mandates to accompany compulsory licensing is indeterminate and is subject to negotiation. Thailand could enhance its bargaining power vis-à-vis the multinational pharmaceutical industry by coordinating its negotiations with other middle and low-income countries.

In conclusion, in its current form, Thailand's NAPHA programme is affordable. Under the model's assumptions, it is also cost effective relative to the baseline scenario. Furthermore, although enhancement of ART with the two policies we analyse (early recruitment through expanded VCT and improved adherence through PHA groups) renders NAPHA slightly less cost effective, these enhancements are still a good bargain, particularly if both are enacted.

Much of the cost of ART over the long term is associated with the provision of second-line treatment. One way to limit the potential financial burden is for the Thai government to make explicit the scope of its commitment to providing public ART: is it a limited commitment to provide only first-line treatment, or is it a more open commitment to provide whatever level of treatment is required by the patient? Estimates of cost effectiveness show that a version of NAPHA that includes only first-line drugs is much more cost effective, at only US$736 or B29 440 per life-year saved, than the policy with second-line therapy. NAPHA with second-line therapy, however, saves a quarter of a million more life-years. Sensitivity analysis shows that, if the price of second-line drugs were to be reduced by 50%, either by concessions from multinational patent holders or through compulsory licensing, the cost per life year of NAPHA first and second-line therapy would be reduced by 31%, from US$2145 to US$1476 [2]. A second way for the government to limit its expenditures on second-line therapy is to grant compulsory licences for the manufacture of patented second-line pharmaceutical products. Doing so will require high-level political resolve that is based on an accurate understanding of the costs to Thailand, the health benefits, the budgetary savings, and the trade repercussions of such action.

Another option would be for the government to explore other financing mechanisms for ART, including a greater use of user fees and health insurance schemes. In view of the government's commitment to provide free and universal access to ART through NAPHA, any such plan would have to be carefully designed to avoid excluding individuals from treatment or discouraging adherence. Inclusion of AIDS treatment within the ‘30-Baht’ national healthcare plan, a policy currently under discussion in Thailand, must take into account both the large cost of NAPHA and its uneven geographical distribution across the country. Space constraints prevent an analysis of alternative financing mechanisms for ART in Thailand. Although affordable by the government, expanding ART represents a long-term financial commitment that must be integrated into the budget process. Once the Thai government begins to finance a patient's AIDS treatment, that access becomes an entitlement that cannot be sacrificed to budget cycles without incurring large political costs. Continuing to support existing ART patients for the rest of their lives and absorbing new patients while maintaining other health programmes will require a 24% increase in the total health budget by 2013. As no cure for AIDS is in sight, NAPHA represents a long-term government commitment.

The biggest challenge for Thai health policy makers will be to resist complacency about HIV infection and instead to build a synergistic relationship between treatment and prevention. This approach may require the devolution of responsibility for both treatment and prevention to the province level or below so that government units that succeed with prevention will benefit from the saved treatment costs.

Success in rolling out treatment will make achieving the national AIDS strategy objective of less than 1% prevalence difficult to attain, because individuals with HIV will live longer. The first objective of the national AIDS strategy should thus be redefined in terms of HIV incidence, and it should be accompanied by measures to strengthen prevention in the light of expected (and already documented) changes in the risk behavior of both vulnerable groups and the broader population.

The cost of US$2145 or B85 800 per life-year saved through ART may be much more than Thailand would have to spend to save life-years with other interventions. The study recommends that Thailand accompany its expansion of the ART programme with a vigorous investigation of other promising opportunities to improve health costs effectively [34]. Prime candidates among those alternatives would be inexpensive HIV prevention programmes, including condom distribution and peer education. The expansion of immunization programmes, of traffic safety and trauma management, of nutrition programmes, and of water supply are all candidates for cost-effective interventions that would save life-years at probably much less than US$2000 or B80 000 per life-year.


This paper summarizes the results of a study requested of the World Bank by the Thai Ministry of Public Health and previously published in book form [2]. The book acknowledges others who participated in or supported the study and it is dedicated to Nicholas Prescott, a friend and colleague of many of the study participants.

Disclaimer: The production of this special Supplement was supported by the World Bank, the Joint United Nations Programme on HIV/AIDS and the World Health Organization. The findings, interpretations and conclusions presented in this paper do not necessarily reflect the views of these institutions or their constituent agencies or governments.


1. Thailand National HIV and AIDS Program. Towards universal access by 2010. Bangkok: Government of Thailand, Ministry of Public Health, Department of Disease Control; 2006.
2. Revenga A, Over AM, Masaki E, Peerapatanapokin W, Gold J, Tangcharoensathien V, et al. The economics of effective AIDS treatment: evaluating policy options for Thailand. Washington, DC: World Bank; 2006.
3. Ministry of Public Health. NAPHA report for March 2004. Bangkok, Thailand: Ministry of Public Health; 2004.
4. Thai Working Group. Projections for HIV/AIDS in Thailand: 2000–2020. Bangkok, Thailand: Karnsana Printing Press; 2001.
5. Gold J, Duncombe C, Ratanasuwan W, Wilson D, Ford N. Clinical background paper for expanding access to ART in Thailand report. Bangkok, Thailand and Washington, DC: Ministry of Public Health/The World Bank; 2005.
6. UNAIDS, UNFPA. 2002 Thailand country profile. HIV/AIDS situation in Thailand and national response to the epidemic. Bangkok, Thailand: Joint United Nations Programme on HIV/AIDS in Thailand; 2002.
7. Brown T, Peerapatanapokin W. The Asia epidemic model: a process model for exploring HIV policy and program alternatives in Asia. Sex Transm Inf 2004; 80:19–24.
8. Bureau of AIDS TaSTI. Data on costs. Bangkok: Government of Thailand, Ministry of Public Health; 2004.
9. World Bank. Confronting AIDS: public priorities in a global epidemic. New York: Oxford University Press; 1999.
10. Over M, Marseille E, Sudhakar K, Gold J, Gupta I, Indrayan A, et al. Antiretroviral therapy and HIV prevention in India: modeling costs and consequences of policy options. Sex Transm Dis 2006; 33(10 Suppl.):S145–S152.
11. World Bank. Social Monitor Thailand: Thailand's response to AIDS: building on success, confronting the future. Washington, DC: World Bank; 2000.
12. Division of Epidemiology. Monthly epidemiological surveillance report. Bangkok: Government of Thailand, Ministry of Public Health; 2006.
13. Over M, Piot P. Human immunodeficiency virus infection and other sexually transmitted diseases in developing countries: public health importance and priorities for resource allocation. J Infect Dis 1996; 174(Suppl. 2):S162–S175.
14. Teutsch SM. A framework for assessing the effectiveness of disease and injury prevention. MMWR Recomm Rep 1992; 41:1–12.
15. Dukers NH, Goudsmit J, de Wit JB, Prins M, Weverling GJ, Coutinho RA. Sexual risk behaviour relates to the virological and immunological improvements during highly active antiretroviral therapy in HIV-1 infection. AIDS 2001; 15:369–378.
16. Stolte IG, Dukers NHTM, de Wit JBF, Fennema JSA, Coutinho RA. Increase in sexually transmitted infections among homosexual men in Amsterdam in relation to HAART. Sex Transm Infect 2001; 77:184–186.
17. Van de Ven P, Rawstorne P, Nakamura T, Crawford J, Kippax S. HIV treatments optimism is associated with unprotected anal intercourse with regular and with casual partners among Australian gay and homosexually active men. Int J STD AIDS 2002; 13:181–183.
18. Van de Ven P, Mao L, Fogarty A, Rawstorne P, Crawford J, Prestage G, et al. Undetectable viral load is associated with sexual risk taking in HIV serodiscordant gay couples in Sydney. AIDS 2005; 19:179–184.
19. Over M, Heywood P, Gold J, Gupta I, Hira S, Marseille E. HIV/AIDS prevention and treatment in India: modeling the costs and consequences. Washington, DC: World Bank; 2004.
20. McFadden D. Conditional logit analysis of qualitative choice behavior. In: Zarembka P, editor. Frontiers in econometrics. New York: Academic Press; 1974.
21. 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. N Engl J Med 2006; 355:1141–1153.
22. Walensky RP, Paltiel AD, Losina E, Mercincavage LM, Schackman BR, Sax PE, et al. The survival benefits of AIDS treatment in the United States. J Infect Dis 2006; 194:11–19.
23. Branson BM, Handsfield HH, Lampe MA, Janssen RS, Taylor AW, Lyss SB, et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep 2006; 55:1–17.
24. Lifson AR, Rybicki SL. Routine opt-out HIV testing. Lancet 2007; 369:539–540.
25. Weiser SD, Heisler M, Leiter K, Percy-de KF, Tlou S, DeMonner S, et al. Routine HIV testing in Botswana: a population-based study on attitudes, practices, and human rights concerns. PLoS Med 2006; 3:e261.
26. Freedberg KA, Hirschhorn LR, Schackman BR, Wolf LL, Martin LA, Weinstein MC, et al. Cost-effectiveness of an intervention to improve adherence to antiretroviral therapy in HIV-infected patients. J Acquir Immune Defic Syndr 2006; 43(Suppl. 1):S113–S118.
27. Medecins Sans Frontieres. 2005 Operational overview. Medecins Sans Frontieres; web site accessed 6/7/07:
28. Ministry of Public Health. Data on costs from the Bureau of AIDS, Tuberculosis, and Sexually Transmitted Infections. Bangkok: Ministry of Public Health; 2004.
29. World Health Organization, UNAIDS, UNICEF. Towards universal access: scaling up priority HIV/AIDS interventions in the health sector. Progress report. Geneva: WHO; April 2007.
30. Ford N, Wilson D, Lotrowska M, Costa Chaves G, Kijtiwatchakul K. Sustaining access to antiretroviral therapy in developing countries: lessons from Brazil and Thailand.AIDS 2007; 21: in press.
31. Supakakunti S, Phetnoi W, Tsunekawa K. An evaluation of economic impacts of improving access to care to highly active therapy (HAART) program for HIV/AIDS patients in Thailand. Bangkok, Thailand: World Health Organization, Thailand and WHO Collaborating Centre for Centre for Health Economics, Faculty of Economics, Chulalongkorn University; 2004.
32. Masaki E. Background paper on evaluating financial implication of antiretroviral treatment (ART) policy in Thailand. Washington, DC: World Bank; 2004.
33. Teerawattananon Y, Mugford M, Tangcharoensathien V. Economic evaluation of palliative management versus peritoneal dialysis and hemodialysis for end-stage renal disease: evidence for coverage decisions in Thailand. Value Health 2007; 10:61–72.
34. Laxminarayan R, Mills AJ, Breman JG, Measham AR, Alleyne G, Claeson M, et al. Advancement of global health: key messages from the Disease Control Priorities Project. Lancet 2006; 367:1193–1208.

ART; HAART; HIV; antiretroviral; cost; cost-effectiveness; economic; multinomial logit; demand; elasticity; voluntary counseling and testing; VCT; projection; fiscal; survival; NAPHA

© 2007 Lippincott Williams & Wilkins, Inc.