NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on the Economics of Antimalarial Drugs; Arrow KJ, Panosian C, Gelband H, editors. Saving Lives, Buying Time: Economics of Malaria Drugs in an Age of Resistance. Washington (DC): National Academies Press (US); 2004.

Cover of Saving Lives, Buying Time

Saving Lives, Buying Time: Economics of Malaria Drugs in an Age of Resistance.

Show details

2The Cost and Cost-Effectiveness of Antimalarial Drugs

In one sense, the economics of antimalarial drugs are simple: currently effective drugs—those recommended by the global community for most of Africa and Asia—are not affordable by the endemic countries, so few people are getting them. This is in contrast to the previous generation of effective drugs, which were exceedingly cheap by any standards, but which no longer work because drug resistance is so widespread. As a result, malaria mortality rates are rising in Africa, and more cases are inadequately treated.

The Immediate Problem of ACT Affordability

Chloroquine—the standard, effective drug for decades—costs about 10 U.S. cents per course of treatment for an adult. The new effective drugs— artemisinin combination therapy (ACT)1—today cost US$2.40 per course wholesale, and can be marked up to five times that amount in pharmacies in Africa. There are limited alternatives to ACTs for effectiveness and safety for widespread use, and their adoption is recommended by WHO and virtually all other expert bodies. Even 10 cents is too much for the poorest of the poor in every endemic country (which equals substantial numbers of people), but it is affordable for many, and governments generally have been able to afford that price (or find external financing) for their public-sector needs. Multiply the per-dose difference in price by the millions or tens of million courses used per year in each country in Africa, and the money adds up. Continent-wide, an estimated 200-400 million courses of treatment are used per year, and an additional 100 million courses in the rest of the world. Realistically, the endemic countries are able to contribute very little of the incremental cost, and that is not likely to change for many years, given the pace of economic development in Africa and in poor nations elsewhere. One positive change that is almost certain, however, is that several different ACTs will become available,2 and their price will come down by more than half, to about 10 times the current price of chloroquine, or about US$1 per adult course and US$0.60 for an average child's course.

The Cost of Producing ACTs and Future Prices

Why are ACTs so much more expensive than chloroquine? Even at prices anticipated after the massive scale-up that would be necessary to supply the African market—should funding become available for large-scale purchase—the price as sold by the producer will likely remain at about 10 times the price of chloroquine (Table 2-1 lists prices offered by major producers to Médecins Sans Frontières [MSF] in 2003). These prices reflect the production costs of artemisinins without a premium for any exclusivity related to patents.3 They are high because the process involves growing the source plant, Artemisia annua, extracting the active moiety, and creating the desired artemisinin derivative (artesunate, artemether, etc.). Coformulation with the companion drug follows. The price of the finished product is driven mainly by the cost of the artemisinin derivative, but also is affected by the companion drug (e.g., lumefantrine, the companion drug in Coartem, is a relatively expensive drug on its own, which contributes to the high price of the coformulation).

TABLE 2-1. Wholesale Prices (Offers) for Artesunate and ACTs in 2003 (US$).


Wholesale Prices (Offers) for Artesunate and ACTs in 2003 (US$).

There is something of a chicken-and-egg quality about the current price of artemisinins, and the prospect for significantly lower prices. Lower prices can be expected in response to large-scale demand, which, in turn, will induce competition among producers. However, without assurances from the global community that there will be a market for large quantities of ACTs, manufacturers will not have the incentive to scale up production, and prices will not drop. In addition to competition driving prices down (i.e., companies being willing to accept lower profits per dose with higher volumes), technological improvements in the process could bring down the actual production costs, which would be passed along to purchasers in a competitive environment.

The real price breakthrough will likely occur only when a fully synthetic artemisinin is developed, eliminating the growing and extraction process. The Medicines for Malaria Venture (MMV) has such a compound under development, which they predict could be available in 5-6 years. The ultimate price is not known, but it should be significantly less expensive than current artemisinins (assuming no premium for exclusivity). If the synthetic product is better than, or at least as effective as the extracted ones, the market would change dramatically. A global subsidy might still be needed, but it could be less than what is needed now.

The Global Cost of ACTs and Future Antimalarials

A per-course price of US$1 can be taken as a reasonable upper limit for the wholesale cost of ACTs purchased in large quantities (e.g., 1 million or more courses) within 1 to 2 years. Prices of US$1 and less have already been offered to MSF by several companies, assuming that no major problems are encountered in scaling up, and producers can meet quality standards (no major problems are foreseen for either issue).4 Using our estimate of 300-500 million episodes treated each year, the math is simple: US$300-500 million per year. This quantum jump in the global cost of antimalarials is, in all likelihood, a one-time phenomenon, for the following two key reasons.

First, the volume of antimalarials needed is not expected to increase significantly—assuming that treatment and other control measures remain at least at current levels. If treatment coverage could be expanded, more treatments might be required, but such changes are likely to be incremental and small in relation to the current treatment volume. Currently, treatment failures account for some number of treatments, and these should be reduced with effective drugs, balancing out some of the incremental increases. In the lower transmission areas at least, if used optimally, ACTs have the potential to decrease malaria transmission, and therefore the volume of drugs required, as has occurred in KwaZulu Natal, South Africa, as part of an integrated control program.

The second reason is, as already stated, future drugs are likely to be cheaper to produce than artemisinins (although probably not as cheap as chloroquine). And because, realistically, any new antimalarial with global promise is likely to emerge from MMV or another publicly funded source (e.g., the U.S. military), the need to recoup R&D costs through sales will be minimal. The day of the ten cent course of antimalarials may be gone, but newer, effective combination treatments still will be a relative bargain.

Another thing that is not likely to change quickly, however, is the inability of the endemic countries to absorb the higher costs. This means that without a global subsidy that is assured at least in the medium term, governments (through the public sector), and families (through their largely private-sector purchases) could be left without financial access to ACTs, or newer antimalarials, as they are introduced.

Inability to Rely on Tiered Pricing

Malaria affects only poor nations, those with highly restricted purchasing power. As is well known, the R&D costs of new drugs, together with the profits, are recovered in a markup of price over the costs of producing the drug. For those medicines with worldwide demand, it is possible to recover the fixed costs and make the bulk of the profits from sales in the richer countries, while selling at much lower prices in poorer countries. Price discrimination—tiered pricing—is clearly emerging as the pattern for antiretroviral drugs for people with HIV, for the panoply of drugs needed for tuberculosis, for childhood vaccines, and others.

In the case of malaria, the market is very small outside of endemic nations, nearly all of which are in the low-income category: within these countries, there is no room to charge anyone substantially above the marginal cost of production in order to create incentives for R&D. The exception to this generalization is that antimalarials are needed by travelers, and by the military of high-income countries. These groups, however, need antimalarial prophylactic drugs, and to a much lesser extent, treatments for clinical disease. In the past, chloroquine filled both roles, but just as with treatment, because of widespread resistance, chloroquine can no longer be used for prophylaxis. Some other drugs still can be used for both purposes. Mefloquine and Malarone fall into this category; both are quite expensive when bought in the United States or Europe. However, artemisinins are not good candidates for preventing malaria because they have extremely short half-lives, a characteristic that may prevent the easy emergence of resistant parasites. A good prophylactic drug needs to remain in the body between doses, and the longer the interval between dosing, the better people's adherence to the regim.


Antimalarial drugs are not the only pressing health care need in sub-Saharan Africa, where AIDS, tuberculosis, and a host of other infectious and other diseases also claim lives prematurely. And resources allocated to malaria can be spent on bednets, indoor spraying, or other measures, and not necessarily on antimalarial drugs. Thus drugs must compete for both public and private funds on grounds of their value for the dollars spent: their cost-effectiveness.

The question most relevant to this study is the cost-effectiveness of treating versus not treating acute episodes of malaria (the latter includes using an ineffective drug as well as no treatment). In the short term (possibly the next 5 years), it also is relevant to ask whether it would be beneficial to switch from failing but affordable chloroquine, to an intermediate drug that is as inexpensive as chloroquine, but which can only last a few years because of rapidly evolving resistance. In this scenario, a switch to more expensive ACTs would be expected within a few years, but in the meantime, considerable funds might be saved. The only drug that currently meets the conditions for such an intermediate is SP, but the analysis might apply in the future to other drugs. We look at both the cost-effectiveness of treatment versus no treatment, and of switching to a stopgap drug and then to ACTs.

Benchmarks for Cost-Effectiveness

There are no firm benchmarks or cutoff points to separate interventions that are deemed worth the cost (“cost-effective”) and those that are not, but there are some guidelines.5 At one end, a 1996 WHO Ad Hoc Committee recommended that, for low-income countries with a GDP per capita of less than $765, interventions that cost $25 or less per disability-adjusted life-year (DALY)6 were “highly attractive,” and those costing $150 or less per DALY were “attractive” (all figures in 1996 dollars) (World Health Organization, 1996).

More recently, the Commission on Macroeconomics and Health suggested that interventions costing less than three times GDP per capita for each DALY averted represent good value for money (Commission on Macroeconomics and Health, 2001). WHO's World Health Report for 2002 built on this principle, defining “very cost-effective” interventions as those that cost less than GDP per capita to avert each additional DALY, and “cost-effective” interventions as those where each DALY averted costs between one and three times GDP per capita (World Health Organization, 2002).

It bears noting that cost-effectiveness analysis is an inexact science, more so in cases, such as this one, where many assumptions must be made on scant data. Uncertainty bounds are very wide. At its best, cost-effectiveness can help to define the relative positions of interventions by grouping them according to reasonably similar cost-effectiveness. Then, broad policy questions related to health care spending can be addressed, e.g., why are some interventions supported that cost more than three times GDP per capita, while others that appear to be better buys are not? One important aspect, relevant to ACTs, is that cost-effectiveness analysis does not address the total cost of an intervention for a country. Something that is very cost-effective—because it costs little and works well—could be very expensive per capita if a large proportion of the population needs it, which is the case for malaria treatment. It may not be affordable without additional resources, even though it is desirable and relatively cost-effective.

Two Models

We present two simplified models to examine the consequences of treating versus not treating (i.e., treating with a wholly ineffective antimalarial): switching drugs once—directly to ACTs—versus twice—to a less expensive but less durable alternative first, thence to ACTs. In the analysis of treating versus not treating, the evolution of resistance is not introduced (although the model from which this was adapted does include resistance), the aim being to define a basic level of benefit from effective treatment. The second analysis (one change versus two) is based on a different model, which does, of necessity, take into account the evolution of resistance. It also incorporates the effects of partial coverage within a population. Neither analysis attempts to model large-scale spatial effects if certain countries adopted effective antimalarials but others did not.

Treatment versus No Treatment

The model used to generate cost-effectiveness estimates for malaria treatment with ACTs was adapted from an earlier model looking at other aspects of malaria treatment (Box 2-1). Here, we compare ACT treatment with no treatment under two sets of treatment conditions. Treatment is either presumptive (i.e., everyone with fever suspected to be malaria is treated) or only after diagnosis of symptomatic individuals with a “hypothetical rapid diagnostic test” (RDT).7 Under the assumptions of this model, treatment with ACTs saves lives, and at a cost that is well under even the low cutoff for a “highly attractive” or “very cost-effective” intervention for children under 5, at less than US$8 per DALY with presumptive treatment, and US$6.23 with an RDT. A life saved for US$209 or US$171 (without and with RDT) is also considered a bargain. For the over-5s, the answer from this model is not as clear-cut, though the point estimate of the cost-effectiveness ratio still falls within the “attractive” or “very cost-effective” range, at US$112 per DALY with presumptive treatment, and US$81.61 with RDT. Saving the life of someone over 5 (the model uses “adult” assumptions for this whole group, so this really applies to the adult population) costs about US$2,800 with presumptive treatment, and US$2,027 with RDT. It is important to note that these benefits apply only to reduced mortality, except for a very small portion for permanent disability caused by severe malaria. The model does not give any value to reduced morbidity; while the benefit per case may be small, for adults, the loss of a few days to 2 weeks of work are not trivial financially, and in the aggregate, can be very large.

Box Icon

BOX 2-1

ACTs versus No Treatment: Key Points in the Cost-Effectiveness Analysis. This cost-effectiveness analysis is adapted by Shillcutt and Mills from a model (Coleman et al., forthcoming) that describes clinic-based treatment for malaria, specifically where (more...)

Most of the variables in the model are subject to greater and lesser degrees of uncertainty, to which the authors have assigned ranges of values. Here we present single values based on the best (or central) estimate for each variable.

The analyses based on presumptive treatment should be closer to the current and near-term reality in Africa, where RDTs are seldom used. RDTs are used routinely in at least some parts of Asia, but the variables in this model (e.g., the proportion with fever who have malaria and the probabilities of death at different ages) are not applicable to most of Asia without significant modification.

Another way the model has been simplified is to assume that ACT resistance does not exist, thus the costs and benefits would remain perpetually the same. The model from which this was adapted does build in the spread of drug resistant malaria. As time goes on, costs increase (because of the cost of re-treating initial failures) and benefits decrease. It is reasonable to assume that, even if ACTs are used well, malaria strains resistant to ACTs will eventually develop and spread (as discussed elsewhere in this report), but it also is reasonable to anticipate that ACTs could remain effective for a relatively long period. In fact, they could well be superseded by newer, less expensive drugs before they lost effectiveness. Leaving resistance out of the model means possibly overestimating the cost-effectiveness of ACTs in the long term, but it may still be realistic in the short and medium term.

One Change versus Two: Chloroquine → ACTs versus Chloroquine → “Intermediate” → ACTs

ACTs are recommended by WHO as the drug of choice for policy change in countries with significant chloroquine resistance. Within a few years, this will include every country in sub-Saharan Africa. But if these countries were to switch to an inexpensive intermediate drug or an inexpensive combination that does not include an artemisinin (e.g., chloroquine + SP) as an interim measure, this would delay the higher treatment cost of ACTs. As recently as a year ago, changing to SP was a potentially viable option, with the understanding that resistance was likely to develop within a few years. A model was developed by a member of the IOM committee (Laxminarayan, forthcoming) to compare the costs and effects of a direct change to ACTs versus changing first to SP and then to ACTs. The model incorporates aspects of malaria transmission, levels of immunity, and drug resistance, and the costs of a malaria episode (including the cost of the drug, but no dispensing costs; and the costs of morbidity and mortality, set at US$0.50 per patient per day with malaria). Although based on SP as the intermediate, the model is indicative of a change to any relatively cheap alternative antimalarial with the potential for rapid development of resistance.

Characteristics that affect both the costs and effects of the two strategies are the level of coverage (i.e., the number and proportion of the population who get treatment for malaria when needed); the level of immunity to malaria in the population, which is related directly to the intensity of malaria transmission; and how quickly patients are treated after becoming symptomatic (this is dependent on access).

The level of coverage has an obvious influence on both costs and effects, which would be proportional to the number treated, but it has a second, less obvious, but extremely important, effect. The level of coverage is the most important factor in determining the level of “drug pressure,” which determines the rate at which drug resistance spreads (see Chapter 9 for more detail on drug pressure and drug resistance). In the medium to long term, this effect dominates the results.

For the individual, immunity to malaria reflects exposure through the bites of infective mosquitoes. At the community level, immunity increases with the intensity of transmission and decreases with the extent of treatment coverage with an effective drug (Pringle and Avery-Jones 1966; Cornille-Brogger et al., 1978).

Where transmission is most intense, the level of immunity is highest and becomes evident at the earliest ages. Where infective bites are rare (e.g., in much of Asia), very little immunity develops. Immunity to malaria is a temporary phenomenon, waxing and waning with exposure.

For the purposes of this model, an initial frequency of drug-resistant parasites of 10–12 (a number very close to 0) was arbitrarily chosen for the artemisinin-based combinations, and for SP, 10–3 (one in one thousand) (Laxminarayan, forthcoming). Over time, treatment selection pressure leads to a greater prevalence of infected individuals who carry the drug-resistant strain relative to those carrying a drug-sensitive strain. The model was used to compare the economic consequences of two strategies: 1) replacing chloroquine (CQ) with ACTs and 2) replacing CQ with SP and waiting for resistance (to a level of 20 percent) to develop before introducing ACTs. A 3 percent discount rate is used for all relevant aspects of the model.

Using either strategy, the total costs of infection decrease with increasing levels of coverage with either strategy. This is attributable to faster cure rates, lower morbidity, and less need for re-treatment because of initial failures. At very low levels of treatment coverage (and low drug pressure), resistance to the intermediate drug is not a problem, so the least expensive drug gives good results for less money. At high levels of treatment coverage (and high drug pressure), resistance evolves so rapidly regardless of which strategy is followed that the faster acquisition of immunity with a less effective drug plays a critical role in determining the superior strategy.

The bottom line is that if one were interested in only the short term, using the less expensive drug makes better economic sense since the costs of resistance-related morbidity do not enter into the considerations. However, for longer planning horizons, a direct switch to ACTs is advantageous, given the costs of higher morbidity associated with increasing resistance to the intermediate drug. In theory, with higher intensity of disease transmission, the benefit of switching to ACTs directly may be diminished because of greater immunity associated with higher transmission, and hence a lower risk of resistance developing to the intermediate drug, whatever it is. Resistance to the intermediate would be expected to take longer to develop and, therefore, the benefits of switching to using it first, then switching to ACTs, increase.

One factor that is not taken account of in this model is the likelihood that, during a period of use of an intermediate drug, artemisinin monotherapy might gain in popularity, and during that time, some of the ACT companion drugs would continue to be used as monotherapy. This could result in higher levels of resistance to ACTs before they are introduced formally as first-line treatment. In general, the greater the coverage with ACTs, and the sooner their use begins, the lower the likelihood of widespread monotherapy use, which could lower the likelihood of resistance to ACTs.

If it were easy for countries to switch drugs, it might make sense to introduce the cheaper drug first, and then move to ACTs before resistance has had much effect on malaria morbidity. However, this is not likely to happen for two reasons. First, malaria-endemic countries have found it difficult to modify their malaria treatment policies proactively in response to impending resistance-related morbidity. The fact that CQ is being used even with high treatment failure rates when an alternative drug is available is emblematic of these difficulties. Second, each change in treatment policy incurs costs associated with retraining health workers, printing material that explains new dosing regimes, restocking new drugs, and so forth, which can be significant in the short term. In the case of a switch to an intermediate that lasts only a few years, these policy change costs would have to be amortized over a much shorter life of the drug than in the case of a switch to ACTs. At a human level, knowledge of which antimalarial is recommended, and how it should be used must penetrate deep into society not only through the formal sector but through the many-layered informal sector. People must gain collective experience with a new drug, and observe its effects consistently to make the change complete, which can take years. This model does not incorporate policy change costs or other difficulties, so it may mask a portion of the true costs of switching first to an intermediate antimalarial and then to ACTs, versus a single switch to ACTs.

Antimalarial R&D: Effects of Lack of Global Purchasing Power

Malaria has never been an attractive target for the research-based pharmaceutical industry for the very reason the current crisis exists: lack of funds to pay for even modestly priced drugs. Most of the antimalarial drugs developed in the past half century are the products of publicly financed R&D. Up to now, the largest number of new products has come from R&D conducted by the Chinese government and the U.S. military. Most recently, the early development of the artemisinins was launched by Chinese government scientists, in response to concerns of the North Vietnamese government over the toll of chloroquine-resistant malaria on its soldiers during the conflict with the United States. The U.S. military continues its research program (though at a considerably lower level than its peak, during the Vietnam War), but in the past 5 years, has been up-staged—in a positive way—by the Medicines for Malaria Venture (MMV), an international “public-private partnership” with the sole mission of developing new antimalarial drugs. MMV today has the most active pipeline of research and products in various stages of development that has ever existed (see chapter 10 for a more detailed discussion of antimalarial drug R&D).

The current inaction over funding ACTs could call into question the sense of continuing to allocate public funding and solicit private contributions toward R&D for new antimalarials. The unstated implication of the failure to purchase ACTs is that only drugs that are about as inexpensive to produce as chloroquine are worth developing for first-line treatment. There may well be new drugs, cheaper than ACTs, in another 10 years, but they may still be unaffordable to most users. Rationalizing current R&D expenditures is not a reason to establish a global subsidy for ACTs, but the failure to do so may well have consequences for the R&D process that are detrimental and difficult to reverse.


Families bear much of the financial burden of malaria treatment in sub-Saharan African and elsewhere. Funds deployed by governments come from internal country resources as well as from the major donors of development assistance, including support from individual countries (bilateral aid), and from international organizations, most importantly, the International Development Association of the World Bank, and other United Nations agencies. Nongovernmental organizations (particularly church-run medical services and groups like MSF) have for decades directly provided significant amounts of medical care, including drugs, in some cases (e.g., Zambia) acting as part of the public medical care system. Newer foundations—in particular, the Bill and Melinda Gates Foundation—and the beginning of operations by the Global Fund have provided significant increments of new funding, but also have highlighted the very low levels available in comparison to stated needs.

Although difficult to track accurately, the level of external support for malaria control from the key bilateral and multilateral agencies appears to have remained relatively static in recent years. The Commission on Macroeconomics and Health (Commission on Macroeconomics and Health, 2001) reported on contributions to health development assistance by bilateral and multilateral donors in the late 1990s, which totaled about US$1.7 billion for all diseases and programs. The biggest share—about US$287 million per year—went to HIV/AIDS. Malaria-specific funding was US$87 million, and for tuberculosis, US$81 million. These are the funds earmarked for specific diseases, but it is likely that other general funds were also used for them. A more recent estimate for international funding of malaria control, including bilateral aid and low-cost loan assistance from multilateral organizations for the years 2000 and 2002, totals about US$100 million per year (Narasimhan and Attaran, 2003). This figure is for all aspects of control (insecticide-treated nets, household spraying, and other preventive measures), and not just treatment. The amount that would have gone toward the cost of antimalarial drugs would be some fraction of the total.

Foundations have, for many years, contributed to global public health initiatives. In recent years, the Bill and Melinda Gates Foundation has entered at a level surpassing any other such efforts, spending close to US$1 billion per year on a variety of programs, some funded independently, and some through existing programs (e.g., the Gates Foundation is a major donor to the Global Fund, MMV, GAVI, and a number of other initiatives). This single entity is spending well over half of the amount spent by all governments and multilateral institutions together.

The advent of the Global Fund to Fight AIDS, Tuberculosis and Malaria represents the other big change in the funding landscape for global health. In their first three rounds of grants (from mid-2002 through the end of 2003), the Global Fund approved US$2 billion in 2-year grants to about 100 of the world's poorest countries. About one-quarter—US$475 million—was allocated to malaria projects, although very little so far involves funding the purchase of ACTs.


In April 2000, more than 20 African heads of state met in Abuja, Nigeria for the first-of-a-kind political summit on malaria. It was one of the seminal events of the Roll Back Malaria (RBM) partnership. The African leaders called on the world community to allocate substantial new resources—at least US$1 billion per year—toward reaching the RBM goal of halving malaria deaths by the year 2010. Development partners also were called upon to cancel the debt of poor and heavily indebted nations so that more resources could be released to address malaria and otherwise alleviate poverty. In addition, the Abuja summit sought resources to support R&D for the whole range of malaria control measures.

The derivation of the US$1 billion was not stated explicitly, but this amount does not seem to overstate how much is needed. In the following year, 2001, the Commission on Macroeconomics and Health published a comprehensive and systematic analysis of how much it would cost to “scale up” a set of “priority” interventions for the world's poor (“priority” being defined by acceptable effectiveness, cost-effectiveness, and overall cost, with costs representing incremental costs over existing expenditures).

Malaria was singled out for individual analysis as one of the most important illnesses (in terms of both the burden of disease and its costs to society) to be addressed by increased resources for worthwhile interventions (Table 2-2). What do these figures say about the costs of scaling up malaria treatment interventions? First, US$0.5 billion for malaria treatment in 2007 (US$1 billion in 2015) does not include treating children under 3 years old, because treatment of childhood-related illness (a few lines down) is linked to a specific mechanism for delivering care (Integrated Management of Childhood Illness, IMCI). Likewise, intermittent preventive therapy (IPT) for pregnant women is included in the estimate for “maternity-related illnesses.” Malaria features large in the childhood burden of illness, in particular, so some substantial amount of the IMCI figure could be directly attributable to malaria (but the figures cannot be broken out in that way). The costs included in these estimates are basically the local cost of the interventions, their delivery, and local management support, but not of broader health sector improvements that would be needed to make them possible. Taking those higher-level costs into account roughly doubles the disease-specific estimates (Commission on Macroeconomics and Health, 2001).

TABLE 2-2. Annual Incremental Cost of Scaling Up Priority Interventions.


Annual Incremental Cost of Scaling Up Priority Interventions.

If we take current external annual funding to be about US$100 million from bilateral and multilateral sources, plus another US$250 million from the Global Fund, and generously, $100 million from other sources, the total is still less than half of the US$1 billion suggested by the African leaders, and an even smaller fraction of the amount the Commission on Macroeconomics and Health projected. Both estimates of global need include some provision for incremental spending on ACTs, although not at the level of subsidy suggested in this report, which means that some portion of the estimated US$300-500 million for a global antimalarial drug subsidy should be added to the total.


  • Coleman PG, Morel CM, Shillcutt SD, Goodman CA, Mills AJ. Forthcoming. A threshold analysis of the cost-effectiveness of artemisinin-based combination therapies in sub-Saharan Africa. American Journal of Tropical Medicine and Hygiene . [PubMed: 15331838]
  • Commission on Macroeconomics and Health. 2001. Macroeconomics and Health: Investing in Health for Economic Development . Geneva: World Health Organization.
  • Cornille-Brogger R, Mathews HM, Storey J, Ashkar TS, Brogger S, Molineaux L. 1978. Changing patterns in the humoral immune response to malaria before, during, and after the application of control measures: A longitudinal study in the West African savanna . Bulletin of the World Health Organization 56(4):579-600. [PMC free article: PMC2395642] [PubMed: 365386]
  • Goodman CA, Coleman PG, Mills AJ. 2000. Economic Analysis of Malaria Control in Sub-Saharan Africa . Geneva: Global Forum for Health Research.
  • Laxminarayan R. Forthcoming. ACT now or later: Economics of malaria resistance. American Journal of Tropical Medicine and Hygiene . [PubMed: 15331837]
  • Narasimhan V, Attaran A. 2003. Roll Back Malaria? The scarcity of international aid for malaria control. Malaria Journal 2(1):8. [PMC free article: PMC156896] [PubMed: 12787469]
  • Pringle G, Avery-Jones S. 1966. Observations on the early course of untreated falciparum malaria in semi-immune African children following a short period of protection. Bulletin of the World Health Organization 34(2):269-272. [PMC free article: PMC2475940] [PubMed: 5328493]
  • WHO. 1996. Investing in Health Research and Development: Report of the Ad Hoc Committee on Health Research Relating to Future Intervention Options. TDR/Gen/96.1 . Geneva: World Health Organization.
  • WHO. 2002. World Health Report 2002 . Geneva: World Health Organization.



The rationale for combining two or more drugs is that doing so dramatically reduces the odds that malaria strains resistant to any of the drugs in the combination would survive to be transmitted. In this report, ACT is used mainly to describe a drug “coformulation” (i.e., two drugs in one pill), and not just two separate drugs taken at the same time.


There are several companion drugs with which the artemisinin may be combined, and the development work on these new combinations is due to be completed within the next 2 years.


Coartem has some patent protection, but Novartis is selling it in developing countries for less than production costs, i.e., at a loss.


According to MSF, these are not preferential prices for that organization, but prices that any major buyer would be offered.


All link cost-effectiveness judgments to some measure of national wealth. Cost-effectiveness is a relative concept: what is considered a bargain in a high-income country may be completely out of range for a low-income country.


The DALY is a summary measure of disease burden that incorporates years lost by premature death with years of life lived with disability. It allows comparison of the influence of different diseases whose effects on mortality, morbidity, and disability differ. One DALY can be considered as one year of “healthy” life.


The hypothetical RDT is 100% accurate, which does not exist in reality. There are places where RDTs are in use, but in most African settings, they would not be appropriate for a variety of reasons. This analysis is included as a window on a possible future scenario.

Copyright 2004 by the National Academy of Sciences. All rights reserved.
Bookshelf ID: NBK215621


Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...