Bringing science and development together through news and analysis

Treating malaria with artemisinin combinations: challenges for policymakers


One of the few compounds that can successfully treat malaria is artemisinin — derived from the Chinese shrub Artemisia annua. But, despite global endorsement, the use of artemisinin combination therapies is limited in many African countries. This policy brief explores the potential of these therapies, and the challenges that still need to be overcome.


One of the few options for successfully treating malaria today is using artemisinin — a compound derived from the shrub Artemisia annua — in combination with other anti-malarial drugs. Artemisinin-based combination therapies, or ACTs, are highly effective against malaria parasites that have become resistant to other drugs.

One of the Millennium Development Goals, set in 2000 to tackle disease worldwide, is to stop the spread of malaria by 2015 and to begin to reduce the numbers of new cases. ACTs offer the hope of achieving these aims.

At least 28 countries, half of them in sub-Saharan Africa, have recently adopted a policy of treating malaria with ACTs, and an increasing number are in the process of changing to ACTs. But many malaria-endemic countries have yet to replace their failing malaria treatment policies. The time is therefore right to encourage a switch to ACTs in the remaining countries and to maximise the public health impact of this move.

This policy brief explores the opportunities created, and challenges that still need to be overcome, in using ACTs as the first-line treatment of uncomplicated malaria caused by the Plasmodium falciparum parasite.

The potential of ACTs

So far, ACTs have amply shown how they can benefit both individuals and communities. It cures more people than other widely available treatments, thus reducing the risk that the patient will get severe malaria or die. And at a community level, ACTs have helped to stem transmission of the disease by lowering the number of gametocytes — the malaria parasites in infected people taken in by Anopheles mosquitoes when they feed, and then transmitted to other humans. 

But this is not all. As with combination therapies against tuberculosis and HIV/AIDS, ACTs also delay antimalarial resistance by eradicating organisms that may be resistant to one of the drugs in the combination, through the action of the other. 

 ACTs thus have real and proven potential to combat the deadly form of malaria. But to optimise cure rates, four particularly challenging requirements must be fulfilled: the patient must complete the entire course of treatment; the drugs need to be good-quality; they also need to be appropriately selected; and the vast majority of malaria infections need to be treated with ACTs. We will look at the first challenge here.

Ensuring patients stay the course

In countries where malaria is endemic, many patients fail to complete the full three-day course of ACTs. The main reason is that the resolution of symptoms, which often happens within 24 to 48 hours, lulls them into a false sense of security. In poor areas with few resources, patients may also be motivated to 'save' the remaining tablets for when they (or a family member) next have malaria. And often, patients seeking treatment in the private or informal sectors are only able buy as many tablets as they can afford — even if this does not add up to a complete treatment.

The risk of not completing the course is 'recrudescence' — a later reappearance of the same malaria infection. But as this can be easily perceived as a new malaria infection, many patients and communities never actually understand the consequences of discontinuing their treatment early.

To ensure that infected people complete their treatment course, intensive ongoing patient information, community education and communication programmes need to be in place. ACTs also need to be made more accessible and easier to use: costs of the therapy to patients should be lowered, and the drugs should be packaged in age-specific packs and in fixed-dose combinations.

Making the most efficient ACTs

The quality of the drugs in ACTs also needs to be addressed if cure rates are to be improved. Poor-quality anti-malarials with too little active ingredient can seriously compromise the treatment. This amounts to underdosing, which, in the case of another drug sulfadoxine-pyrimethamine, may have ultimately encouraged the malaria parasite's resistance to it in Africa. This could have happened by letting parasites that are partly resistant to the drug having a better chance of surviving than those that were killed by it.

Ensuring a supply of good-quality ACTs that are appropriately stored and managed is therefore vital for optimising the public health benefits of it. Such improvements in drug management could have a beneficial effect on the management of other medicines, and thus strengthen the broader healthcare system.

In many developing countries, ensuring a supply of good-quality drugs is made even harder by the prevalence of fake artemisinins that are difficult to tell from the true product. At the moment this is a bigger problem in Asia than in Africa, but there are fears that the practice may spread, along with the increase in demand for ACTs in sub-Saharan Africa and South America.

One way of countering the trend would be to introducing an international subsidy, allowing the price of real artemisinins to drop substantially — ideally to the price of chloroquine — the world's cheapest antimalarial. This would remove any motivation to manufacture and distribute fakes. It could also increase the number of patients or caregivers able to buy the complete course of treatment. There has yet to be agreement, however, over the nature of such a subsidy. 

Another issue is choosing which drugs to partner the artemisinins. By itself, a three-day regimen of an artemisinin is too short to eradicate a malaria infection, and is likely to lead to treatment failure and resistance. So policymakers need to select an effective antimalarial partner to ensure the efficacy and public health impact of ACTs. The goal is to achieve cure rates well above 90 per cent, and preferably at least 95 per cent, after 28 days of follow-up in the most vulnerable local population – usually children under five years in areas of high-intensity malaria transmission.

For each region, new combinations need to be tested. The largest randomised controlled antimalarial trial conducted so far shows ACTs to have an unacceptably low cure rate in a number of African countries because the longer-acting antimalarial drug in the combination failed to work. The study revealed high levels of resistance to chloroquine and sulfadoxine-pyrimethamine of up to 78 per cent, even though these drugs had been selected on the basis of their local efficacy.

In the few areas where amodiaquine and sulfadoxine-pyrimethamine remain effective, they can be partnered with artemesinins as artesunate-amodiaquine and artesunate/sulfadoxine-pyrimethamine combinations. Where parasite resistance to these partner drugs has built up, artemether-lumefantrine or artesunate-mefloquine are more successful. In South-East Asia, the dihydroartemisinin-piperaquine combination is widely and successfully used, and is currently being developed by the non-profit foundation, Medicines for Malaria Venture (MMV). This treatment is also expected to become the least expensive of available ACTs.

At the moment, the trend in working out which drugs will go into ACTs is simply to exclude those that are ineffective. In future, as new combinations such as artesunate-chlorproguanil-dapsone and artesunate-pyronaridine become available, a more systematic evaluation of ACTs will be needed to decide policies on these drugs.

ACTs and malaria transmission

A great deal of evidence shows that ACTs reduce malaria transmission by curbing the number of gametocytes in the person with the disease. On the northwestern border of Thailand, there was a 47 per cent reduction in the incidence of P falciparum infections in the 12 months after ACTs containing artesunate and mefloquine were introduced. There were no changes to control the Anopheles mosquito in the area during this time. Over the next ten years, there was a further improvement — a sixfold reduction in malaria transmission.

Similarly in Vietnam, general use of artemisinins has led to a sustainable drop in the number of infected people dying from malaria. And in KwaZulu Natal, South Africa, a reduction in malaria cases, admissions and deaths of more than 90 per cent has been sustained over the past three years after control of the mosquito was improved, and an artemether-lumefantrine malaria treatment policy was introduced.

To maximise ACTs' effect on the transmission of malaria, it is vital to treat most infected people, and to avoid anti-malarial drugs that have no effect on (or even increase) gametocyte carriage. Reaching as many patients as possible demands both a reliable supply of ACTs, and the compliance of healthcare workers with the ACT malaria treatment policy. This in turn depends on healthcare workers understanding the motives behind the policy change, and the need for having enough ACTs.

These interventions also need to focus on where the majority of malaria patients in any one area seek treatment. If the public health sector is the primary source of care, reaching a significant number of patients may be more easily achievable. But if most patients go to private, informal or traditional practitioners for their treatment, providing subsidised drugs to the public health sector alone may lead to the major problems we have seen — the failure of drugs to reach those who need them, an increase in the use of fake artemisinins, and the temptation for patients to sell their incomplete treatment courses when they start to feel better.

Yet another factor is the extent to which malaria is transmitted. It is primarily in areas where malaria transmission is of low to moderate intensity that a wide-scale use of ACTs has been associated with a drop in cases of malaria. Such successes have yet to be replicated in areas with high rates of malaria transmission, where just under half the people at risk of malaria live; so the effective use of ACTs to curb transmission in these areas is likely to be substantially more challenging.

Why should this be so? In high-intensity transmission areas, fewer people might be taking ACTs, and of those, few might be completing the treatment course. In these areas, people develop partial immunity against the disease. Their infections do not often cause symptoms, and if they have symptoms, these would probably resolve with incomplete treatment.

Furthermore, the higher risk of new malaria infections (re-infections) could lessen the overall impact of ACTs. Re-infections are more common following ACTs than with the combination of two longer-acting antimalarials (such as amodiaquine plus sulfadoxine-pyrimethamine), suggesting that these may have some prophylactic effect. Over time, however, a longer exposure to antimalarials could exacerbate antimalarial drug resistance (because the parasite has more chance to develop resistance). Instead it may be more effective to better control mosquitoes, to reduce the frequency of new infections and enhance the long-term public health impact of ACTs.

There are fewer examples of successful large-scale mosquito control programmes in high-intensity transmission areas. But pilot malaria eradication projects primarily using widespread indoor spraying of insecticides across sub-Saharan Africa from the 1940s to the 1960s recorded significant reductions in new cases of the disease. Other measures that have proven effective include systematic indoor residual insecticide spraying, particularly in southern Africa and island states, and widespread distribution of insecticide-treated bednets. So sustainable vector control might be possible across all areas where malaria is endemic, and could enhance the effects of widespread ACT use.

ACTs and antimalarial resistance

Antimalarial resistance has been one of the major challenges in efforts to tackle malaria. Resistance develops with widespread antimicrobial use (including antimalarial, antiviral, and antibiotic treatment), particularly when these agents are used on their own, taken in ineffectual amounts, are eliminated slowly from the body, or when patients only partially adhere to their treatment regimens.

ACTs counter malaria parasite resistance in the same way that combination therapies work against tuberculosis and HIV/AIDS: any parasites that are resistant to one of the drugs in the combination are likely to be eradicated by the other. ACTs and other effective antimalarial combinations have been shown to have this beneficial effect on antimalarial resistance in areas of low-intensity malaria transmission where medicines are strictly regulated, but similar results have yet to be confirmed in areas of higher-intensity malaria transmission with less well-regulated use of medicines.

At the northwestern border of Thailand, for example, the widespread use of the artesunate-mefloquine combination led to a decreased resistance to mefloquine. But to achieve the same result elsewhere, high coverage with ACT is necessary, along with minimising (or ideally stopping) the use of the longer-acting partner drug on its own.

Reaching as many people as possible with ACTs would eradicate most parasites resistant to the longer-acting partner drug (as these would be killed by the artemisinin), and would curtail the transmission of surviving parasites (as artemisinins reduce the gametocyte stage responsible for transmission of the parasites).

Improving diagnosis to reduce the need for ACTs

For over four decades, malaria diagnosis in most areas of higher-intensity transmission has relied on clinical signs and symptoms, as definitive diagnosis has actually been costlier than chloroquine or sulfadoxine-pyrimethamine treatment. Healthcare providers understandably considered it better to treat several viral (and other) illnesses with an inexpensive antimalarial than to risk missing one potentially fatal yet treatable infection. But now the balance may be tipping in favour of basing treatment decisions on definitive diagnoses.

Malaria control efforts and improved social services and housing have led to a decline in the incidence of malaria, especially in urban areas and in adults. So in cities and adult populations, a strategy for confirming malaria is needed to limit the unnecessary use of ACTs, particularly as they currently cost at least ten times as much as chloroquine or sulfadoxine-pyrimethamine. Microscopy had been the mainstay of malaria diagnosis, but it is seldom available in poorer and rural areas. Simpler diagnostic tools such as rapid diagnostic tests are now being evaluated as a way of extending malaria diagnostic services to the majority of those who need them.

The use of definitive diagnosis to guide treatment decisions will become more cost-effective as ACTs become more widely deployed and the malaria burden decreases. With the addition of spraying or bednets to control the mosquito vectors, the probability that any fevers seen in a population are malarial decreases. This in turn increases the cost-effectiveness of ACTs, as their use can be limited to confirmed cases. The emphasis on limiting ACT access to confirmed malaria cases must not, however, preclude early access to treatment for the communities affected most by malaria.

If it is to work, a policy of limiting ACT use to confirmed cases will depend very much on healthcare providers and patients accepting the validity of a negative malaria rapid diagnostic test or smear. This requires a substantial change in mindset and practice, and the ability of healthcare workers to diagnose and treat alternative causes of febrile illness. A cautionary note, however, is that as the intensity of malaria transmission increases, the significance of a positive malaria test decreases. This is because of the high prevalence of asymptomatic parasitaemia (the presence of parasites in the blood, but no symptoms of malaria), particularly in older children and adults. So a positive malaria test in a patient with fever may mean that the person has asymptomatic parasitaemia that just happens to coincide with a non-malarial febrile illness, rather than an actual malarial fever.  

Finally, any measures to reduce the number of malaria treatments would also reduce the quantity of ACTs needed, and so make them more affordable. This would have the added benefit of reducing pressure on healthcare facilities. Patients who complete the full ACT regimen, in an adequate dose with a good-quality drug, are less likely to fail treatment (and thus require re-treatment with an ACT) and less likely to progress to severe malaria requiring hospital admission.

Policy implications

There is widespread consensus that ACTs could greatly reduce the burden of malaria by improving cure rates, delaying parasite resistance to drugs, and lowering malaria transmission. The cost of ACTs is one of the main barriers to their widespread use in malaria-endemic countries.

Optimising the public health benefits and affordability of ACTs relies on policymakers considering drug treatment within the broader framework of strengthening their healthcare systems and malaria control programmes. Effective education programmes for healthcare providers, patients and communities are needed to ensure that more people complete the full three-day treatment course, and to limit the use of too low doses and poor-quality or fake ACTs.

There also needs to be a reliable supply of ACTs to facilities where most patients are seeking treatment for malaria-like symptoms. Each country's health ministry needs to consider the role of controlling Anopheles mosquitoes and definitive diagnosis (usually through rapid diagnostic tests) in limiting numbers of malaria cases to be treated, especially as improving mosquito control may enhance the effects of ACTs on malaria transmission.

Achieving these changes in practice will not be easy. It will require intense commitment by all those involved — health ministries, local communities and international agencies — as well as strengthening of healthcare systems and operational research.

A number of unresolved issues remain. One is how to implement and sustain an international subsidy of ACT cost. The second is how far ACT policies are likely to reduce transmission and delay resistance in areas of high-intensity malaria transmission, particularly in the absence of vector control measures.

Also an issue is how cost-effective non-artemisinin-based combinations, such as amodiaquine-sulfadoxine-pyrimethamine, are in the long term in areas of high-intensity transmission, compared with ACTs. And how should the need to restrict ACT use to prevent resistance be balanced with the need to expand access to ACTs into the community to provide effective treatment at or near home before the disease becomes severe? How cost-effective are rapid diagnostic tests to limit ACT use to confirmed cases in high-intensity transmission, and what is the public health impact? Finally, there is the issue of how to achieve widespread use of ACTs through the public health system, when most patients seek treatment in the traditional, private or informal health sectors. 

Several studies are underway to address many of these unresolved issues around ACT deployment, namely the availability of new, cheaper fixed-dose combination ACTs, and reducing the cost of rapid diagnostic tests as demand increases.

Policymakers need to be aware of the complex interaction of the malaria parasite with communities at risk and healthcare systems when they decide on whether or not to deploy ACTs. If we are to meet our commitments to substantially reduce the burden of malaria, we need to consider the broader context into which ACT policy is introduced.

Further reading

Arrow K. J., Panosian C., Gelband H., eds. Saving Lives, Buying Time: Economics of Malaria Drugs in an Age of Resistance. Institute of Medicine, Washington DC (2004).

Barnes, K. I. et al. Effect of artemether-lumefantrine policy and improved vector control on malaria burden in KwaZulu-Natal, South Africa. PLoS Med (in press).

Brockman, A., et al. Plasmodium falciparum antimalarial drug susceptibility on the north-western border of Thailand during five years of extensive use of artesunate-mefloquine. Trans R Soc Trop Med Hyg 94, 537-44 (2000).

Hay, S. I., et al. The global distribution and population at risk of malaria: Past, present and future. Lancet Inf Dis 4, 327-36 (2004).

International Artemisinin Study Group. Artesunate combinations for treatment of malaria: Meta-analysis. Lancet 363, 9-17 (2004).

Kouznetsov, R. L. Malaria control by application of indoor spraying of residual insecticides in tropical Africa and its impact on community health. Trop Doctor 7,: 81-91 (1977).

Lengeler, C. Insecticide-treated bednets and curtains for preventing malaria. Cochrane Database of Systematic Reviews (2000; update 2004).

Mabaso M. L., Sharp B., Lengeler C. Historical review of malaria in southern Africa with emphasis on the use of indoor residual house spraying. Trop Med Int Hlth 9(8): 846-856 (2004).

Malenga G. et al. Antimalarial treatment with artemisinin combination therapy in Africa: Desirable, achievable, but not easy. Br Med J (in press).

Muheki C., McIntyre D., Barnes K. I. Artemisinin-based combination therapy reduces expenditure on malaria treatment in KwaZulu Natal, South Africa. Trop Med Int Health 9, 959-966 (2004).

Newton P. N., et al. Counterfeit artesunate antimalarials in southeast Asia. Lancet 362, 169 (2005). 

Nosten F. et al. Effects of artesunate mefloquine combination on incidence of Plasmodium falciparum malaria and mefloquine resistance in Western Thailand: a prospective study. Lancet 356, 297-302 (2000).

Price R. N. et al. Effects of artemisinin derivatives on malaria transmissibility. Lancet 347, 1654-1658 (1996).

Roll Back Malaria. Facts on ACTs: An Update On Recent Progress In Policy And Access To Treatment.

Roll Back Malaria (2005). Changing Malaria Treatment Policy to Artemisinin-Based Combinations: An Implementation Guide.

Snow W., Trape J., Marsh K. The past, present and future of childhood malaria mortality in Africa. Trends in Parasitol 17, 593-597 (2001).

Sutherland C. J., et al. A six-dose regimen of Co-artemether prevents Gambian children with falciparum malaria from becoming infectious to Anopheles mosquitoes. PLoS Med 2, e92 (2005).

United Nations Millennium Development Goals. Available at: http:

Watkins, W. M., Mosobo M. Treatment of Plasmodium falciparum malaria with pyrimethamine-sulfadoxine: Selective pressure for resistance is a function of long elimination half life. Trans R Soc Trop Med Hyg 87, 75-8 (1993).

Watkins W. M., Sibley C. H., Hastings I. M. The search for effective and sustainable treatments for Plasmodium falciparum malaria in Africa: A model of the selection of resistance by antifolate drugs and their combinations. Am J Trop Med Hyg 72, 163-173 (2005).

White, N. J. et al. Averting a malaria disaster. Lancet 353, 1965-67 (1999).

White, N. J. Malaria. In G. C. Cook and A. Zumla, eds. Manson's Tropical Diseases pp 1242-8. Elsevier Science (2003).

Yeka, A. et al. Artemisinin versus nonartemisinin combination therapy for uncomplicated malaria: Randomised clinical trials from four sites in Uganda. PLoS Med 2, e190 (2005).