Malaria in emergencies: treatment, diagnosis and vulnerable groups
by Christa Hook, MSF September 2005

The nature of humanitarian crises, and the parts of the world in which they frequently occur, make the management of malaria of critical and urgent importance, and systems for malaria disease management must be set up as a priority in humanitarian emergencies in malaria-endemic regions. A frequent cause of tension in an emergency can be the difference between the needs of malaria control presented by that emergency, and the needs being addressed by national malaria control mechanisms in the affected country. In a stable situation, health workers may have been trained to use clinical algorithms adapted to the level of malaria transmission and partial immunity in their particular region. For many reasons, drug policy may not yet have moved away from older drugs; the drugs needed for the emergency may not be registered for use in the country, and may be unfamiliar to the people being treated, and to the health workers treating them.

In a crisis situation, the medicine given for malaria must be as near to 100% efficacious as possible. For many reasons, it is likely that, for this episode of illness, there will be one contact, and one only, with the health services. If malaria is only partially treated at the first contact, it is likely to recur in a more severe form, requiring more complex treatment, and probably at some distance away.

Older drugs no longer work

The parasite that causes the lethal form of malaria, Plasmodium falciparum, has developed resistance to chloroquine, sulphadoxine-pyrimethamine (Fansidar) and amodiaquine, the standard first- and second-line drugs. This resistance has spread very rapidly, but evenly, through almost all areas where malaria is transmitted. Resistance is caused by random mutations in the parasite. When the right combination of mutations occurs, that parasite is able to survive in the presence of an antimalarial drug. Meanwhile, all the susceptible parasites are killed, favouring the survival of the resistant parasite, which then goes on to multiply, infect the mosquito and get passed on to another person. Factors favouring the development of resistance include any situation in which parasites are exposed to a low level of antimalarial drug, for instance where a full course of treatment is not completed, where poor-quality drugs with inadequate active ingredient are used, or where long-acting drugs are administered, which are eliminated slowly from the body. Resistance develops more easily when drugs are used singly, rather than in combination.

In many places, resistance is now over 50%, and as much as 70%, which means that fewer than half the people ill with malaria will fully recover. Resistance is not always obvious either to the patient or to the health worker, as there may be a partial response, and because chloroquine has an antipyretic effect, bringing down a patient’s temperature even when the parasites are not destroyed. The initial apparent response will be followed by a recurrence of disease, which may become severe with a high risk of death, or by persistence of the parasite, leading to increasingly severe and life-threatening anaemia.


Currently, the only reliably effective antimalarial drugs are artemisinin-based combination treatments (ACTs). ACTs treat malaria using a combination of two drugs, one of which must be a derivative of artemisinin – the most effective anti-malarial ever discovered. It acts very quickly, and has negligible adverse effects. The most widely used derivatives are artesunate and artemether. For use as a first-line drug for uncomplicated malaria, one of these two has to be combined with another longer-acting antimalarial, one known to have low resistance in the geographical region in question. Only four ACTs are currently in use, although more are in development:

  1. Artesunate plus sulphadoxine/pyrimethamine (SP). This has limited utility because of the high level of resistance to SP in most countries. But in places where resistance to SP is low, such as Afghanistan and north Sudan, this combination is easy to administer and relatively cheap.
  2. Artesunate plus amodiaquine (AQ). In many countries resistance to AQ is low, mainly because it has not been used a great deal in monotherapy (single drug administration). There can be cross-resistance with chloroquine, although this is variable and cannot be assumed. This combination is available in blister packs, offering improved adherence over the use of separate pills; it will be available in co-formulation (both drugs made into one pill) in 2006.
  3. Artesunate plus mefloquine has been used for some years in South-East Asia, with very good outcomes, including falling death rates and reduced transmission. It has not yet been used in programmes in Africa.
  4. Artemether plus lumefantrine (AL) Coartem. Since lumefantrine has never been used as a single drug against malaria, this combination has no recorded resistance (except very recently in Cambodia), and is therefore the drug of choice where resistance is not known, especially in emergencies. It is only available co-formulated, an advantage in delaying resistance, but it has to be taken twice a day, instead of the once-daily dosage of the other combinations. It is only properly absorbed if taken with fatty food. Because it is almost wholly efficacious, it is replacing chloroquine and SP in the new Interagency Emergency Health Kit.

There are few absolute contra-indications to the use of these drugs. Most restrictions stem from a lack of evidence, rather than known problems. However, artemisinins should not be used in the first trimester of pregnancy. In the second and third trimesters, women who are sick with malaria must have effective treatment. Any ACT known to be effective can be used, since adherence to the only alternative, quinine, is notoriously bad. AL is not yet licensed for children weighing under 5kg, but the other combinations can be used in very young children. Full details of drug dosages are in the new WHO Treatment Guidelines (2005) and in the Interagency Handbook on Malaria Control in Complex Emergencies (2005).

At the time of writing there is a shortage of ACTs, in particular of artemether-lumefantrine. But even when the supply was more or less keeping up with demand, availability in an emergency was often a significant problem due to cumbersome supply mechanisms. The shortage is now being addressed by increased production of Coartem, and by the development of a generic version. Meanwhile, other ACTs are more readily available and can be used where resistance to the companion drug is known to be low. Production of the raw material is also being stepped up, but because the drug is plant-based, the time from sowing the seed to having pills on the market is determined by the time it takes for the plant to grow. Shortages have come about not because there is an inherent difficulty in growing the plant and extracting the active ingredient, but because national and international agencies delayed making funding available up-front, and in placing firm orders in time to meet predicted and anticipated needs. This is a man-made problem, to which there are man-made solutions.


Malaria can look like other diseases, and other diseases can masquerade as malaria. Many studies have shown that even the best and most experienced clinicians cannot identify malaria purely by clinical history and physical examination. This is a particular problem in a crisis; health workers and patients may be used to considering every fever as malaria, or may never have encountered malaria before, and may be unaware of the potentially disastrous nature of a fever. The identification of other infectious diseases, such as Kala Azar or Viral Haemorrhagic Fever, has been delayed because fevers have initially been treated as malaria.

It is therefore of critical importance to use biological tests, either microscopy or rapid tests, to ascertain the level of malaria in the community at risk, and to confirm suspected malaria in individuals. It is unusual in an emergency to have adequate numbers of trained microscopists, and it is likely that rapid tests will be more appropriate in the initial stages. Health workers quickly learn the technique, and how to interpret these tests.

Rapid tests detect antigens from the parasite itself. They come as strips or cassettes, labelled with an antibody, to which a drop of the patient’s blood is added, along with a buffer solution. The test is usually read within 15 minutes. The tests most commonly used detect an antigen from Plasmodium falciparum only. Other species are not detected. These HRP II tests are the most sensitive and robust. Many tests are available, with very variable performance. A list is available on the WHO website, but with no indication of quality – only a list of names. MSF and Epicentre (a sister organisation of MSF working in epidemiology and operational research) have verified the sensitivity and specificity of some of the HRP II tests. Other antigens and enzymes are used in different types of tests, but to date these are not sensitive enough for use in an emergency.

In some crisis situations, patient numbers can be very high, and even a 15-minute test can be difficult. It is therefore critically important to ascertain very early on what the level of malaria really is in the population, through a survey in the community or in the clinic. It is always good medical practice to confirm a suspect diagnosis in every patient, but if this is not possible, and it has been ascertained that a very high percentage of young children under five years with suspected malaria do indeed have parasites, then it is reasonable on clinical grounds to treat all such children. However, because things can change quickly in an acute crisis, it is always important to continue to monitor the situation by confirming the diagnosis in a sample of the patients on a regular basis. Another important benefit of confirming malaria is that ACTs are in short supply and are expensive, so should be given for malaria only, and not for every fever.

Vulnerable groups

Young children are always at higher risk from malaria as they have not yet developed even partial immunity, but the risks for a malnourished child are greater. In the first place, severe malnutrition masks the signs of malaria, so an unwary clinician may be falsely reassured by the lack of fever or other signs of malaria. Until recently, it was accepted practice to give presumptive treatment with chloroquine or SP to all children on admission to a therapeutic feeding programme. With new drugs and diagnostic possibilities, this is no longer acceptable. Rapid tests are not dependent on the immune reaction of the patient, so they are reliable in severely malnourished children. Best practice now is to test all children on admission, and to treat all positives with ACT.

In pregnancy, women lose any immunity to malaria that they built up during childhood. This is particularly the case in first and second pregnancies, but continues into later pregnancies in women who are HIV positive. The malaria parasite favours the placenta, causing increased anaemia, low birth weight and the risk of premature labour. In high transmission areas it does not necessarily cause any acute illness in the woman, so it is important to use every opportunity to identify pregnant women, test them for malaria parasites and treat with ACT if positive. Later, when the crisis has stabilised, intermittent presumptive treatment with SP may be appropriate in antenatal services.

Prospects for the future

All international bodies supporting and funding malaria interventions in emergencies accept that ACTs alone are effective enough for these critical situations. Many countries also want to switch to ACTs. Problems arise in the slow pace at which this change is being made, exacerbated by shortages of these drugs. These shortages are being addressed, albeit too late for some people suffering and dying from malaria. But for these new drugs to be used properly, health professionals need training, preferably before the crisis arises. It is of great concern that many international NGOs involved in health care are not well versed in how to use ACTs, or in the need for confirmation of diagnosis. It also seems to be the case that doctors and nurses being trained in tropical health are still not knowledgeable or confident enough to put these approaches into practice when faced with a humanitarian crisis. This must be urgently addressed. Meanwhile, the even-greater task of disseminating this knowledge and training within endemic countries, especially those affected by emergencies, must be tackled and supported by donors.

In future humanitarian crises in a malaria-affected country, we need to know that there is information about drug resistance, that the drugs needed are registered and can be imported quickly, that national and international health workers understand how to use diagnostic methods and can interpret the results, and that they are prepared to implement up-to-date management of malaria from the very beginning. We also need to step up the production of these life-saving drugs and set up mechanisms to ensure their availability in an emergency. Only then will this eminently treatable disease stop exacting such a huge death toll in emergencies.

Dr Christa Hookhas worked for MSF and other NGOs since 1992. She was recently coordinator of MSF’s international malaria working group, and malaria adviser to the Manson Unit in MSF UK.

References and further reading

Pascal Ringwald, Global Report on the Susceptibility of Plasmodium Falciparum to Antimalarial Drugs 1996–2004 (Geneva: WHO, 2005).
M. Amexo et al., ‘Malaria Misdiagnosis: Effects on the Poor and Vulnerable’, The Lancet, 2004, 20; 364(9448): 1896–68.
J. P. Guthmann et al., ‘Validity, Reliability and Ease of Use of Five Rapid Tests for the Diagnosis of Plasmodium Falciparum Malaria in Uganda’, Trans R Soc Trop Med Hyg, May–June 2002; 96(3): 254–57.
Malaria Diagnosis – New Perspectives (Geneva: WHO, 2000).
Malaria Control in Complex Emergencies: An Interagency Handbook (Geneva: WHO, forthcoming, 2005).