Bednets are widely believed to be an effective way of controlling malaria. But why are they not used more often where they are needed? In this policy brief, Chris Curtis outlines the challenges faced by health workers in getting larger numbers of poor Africans to use bednets, and describes methods that have proved effective.
In 1897, Ronald Ross discovered that malaria is transmitted through the bites of Anopheles mosquitoes. Malaria parasites (Plasmodium) are picked up when a mosquito bites an infected person, they mature inside the mosquito and are passed on to a new victim when the mosquito bites again. Soon after making this discovery, Ross realised that because these mosquitoes bite at night, bednets should be a useful protection against infection. But this seemingly simple idea has proved difficult to implement effectively. In villages in Africa, nets frequently tear but are not often repaired — although the rate at which nets are repaired in The Gambia has been increased by an information campaign emphasising how important this is.
In the 1980s, insecticides began to be applied to bednets, which greatly increased their effectiveness — even when torn. Yet, despite the enormous toll that the disease takes on millions of poor Africans, treated bednets are still used far too little.
This policy brief discusses the practicalities of making and distributing insecticide-treated nets. It also looks at the financial and biological barriers to scaling-up the numbers of insecticidal bednets being used, and the efforts to increase their use. Such efforts include providing entire communities with free net treatment, social marketing campaigns, and subsidies to help pregnant women buy nets as they, and their newborn babies, are particularly vulnerable to malaria.
Using bednets to protect against malaria-carrying mosquitoes
Fast-acting synthetic insecticides that are safe for humans have been available since the late 1970s. Treating nets with one of these insecticides adds a chemical barrier to the net's imperfect physical barrier.
There is now plenty of evidence that even if a treated net has holes, it can provide good, if not perfect, protection. This is because the insecticide kills, incapacitates or drives away most mosquitoes before they find the holes.
But using torn, untreated nets is as useless as not using nets.
Several methods can be used to test the effectiveness of treated nets.
For example, in a room with everyone sleeping under insecticide-treated nets, far fewer blood-fed mosquitoes are found in the morning than in a room where no nets are used. Canvas sheets placed on the floor overnight, with provision made to keep out scavenging ants, collect the corpses of many mosquitoes, most killed before they managed to bite as shown by an absence of blood in their abdomens.
In carrying out such counts it is important to distinguish Anopheles (malaria) mosquitoes from Culex and others, which can transmit the worms that cause filaraisis (elephantiasis) but not malaria. Tropical Anopheles have spotting along the leading edge of the wings and they generally stand with their head and body nearly in a straight line and their 'tails in the air'. In contrast, Culex has unspotted wings and its abdomen bent down towards the ground. Anyone can quickly spot these differences but, for more exact identification of species, someone with more entomological experience should be consulted.
How to make an insecticide-treated bednet
Many tests have been made of the 'bite-preventing' and killing power of different chemicals at various dosages Those which work well include deltamethrin (K-othrine) at 0.025 grams per square metre of netting, alphacypermethrin (Fendona) at 0.02 grams per square metre, lambdacyhalothrin (Icon) at 0.01 grams per square metre, and etofenprox (Vectron) at 0.20 grams per square metre. The World Health Organization has certified all of these as safe.
To work out the concentration of the mixture in which to dip nets to produce these dosages, dip a measured area, e.g. a square metre, of the locally available type of netting (e.g. polyester) in a measured volume of water (e.g. 500 millilitres) in a bowl.
Wring out the netting, allowing the drips to fall back into the bowl. Measure the volume remaining in the bowl and subtract it from the original volume. This will determine the volume of water retained by a square metre of the netting (the answer will probably be about 50 millilitres).
Then mix the appropriate number of grams of pure insecticide (as indicated in the above list) with that volume of water. Make allowance for the concentration of insecticide as supplied in the bottle. For example if the label states that this is ten per cent, multiply the target number of grams of insecticide by ten to obtain the number of millilitres of insecticide concentrate to be mixed with the water you would need to wet a square metre of netting (a millilitre volume of liquid is approximately equivalent to a gram in weight).
Then scale up the amount of concentrate to take from the bottle to make, say, 20 litres of mixture in a bowl.
The quickest way to treat or re-treat a batch of nets is to dip one after the other in the bowl, wringing each net out over the bowl wearing strong rubber gloves. Nets being re-treated after a year of domestic use should be washed and dried the day before treatment so that the bulk mix does not become dirty nor diluted by the water in nets which are still wet from washing.
Re-treating all the nets in a village on a pre-arranged day each year can ensure that virtually all the nets are adequately insecticidal for the coming year. The Vietnamese government provides this service for the ten million people in the parts of the country most affected by malaria, and this has reduced their malaria problem remarkably.
Another approach is to sell tablets of deltamethrin or sachets of the other insecticides containing the right dose for an average-sized net, so that people can treat their own nets.
Several companies now make long-lasting treated nets (e.g. Olyset or Permanet) in which the insecticide deposit is more resistant to washing than conventionally dipped nets.
To test how well a treated net can kill mosquitoes — and whether this stands up to repeated washing — wrap part of the net round a small wire frame and insert mosquitoes inside it using an entomologist's sucking tube. Remove the mosquitoes after they have walked on the netting for three minutes, place them in a cardboard cup with a non-insecticidal netting top. If the insecticide deposit on the net was in good condition almost all the mosquitoes will absorb a lethal dose and be dead or dying within 15 minutes.
Scaling-up the use of treated nets
African villages are more affected by malaria than anywhere else in the world because the African species Anopheles gambiae and Anopheles funestus concentrate on biting humans (and not animals) and are therefore highly efficient as malaria vectors and they breed profusely in or near villages. The number of beds in African villages with insecticidal nets is inadequate and urgently needs to be increased.
'Social marketing' advertising campaigns have been devised to persuade African villagers to find the money to buy nets and insecticide. However health experts are increasingly realising that if enough poor villagers are to be protected, it will at least be necessary to provide subsidies to help pregnant women buy nets for themselves and their newborn babies. Providing free treated nets for children who attend vaccination clinics has also been very successful.
These schemes are based on the idea that insecticidal nets are especially important as personal protection for pregnant women and children who are particularly vulnerable to malaria. They tend to be more at risk because they have lower levels of the antibodies that give partial immunity to malaria than other members of the community.
Ensuring personal protection of these particularly vulnerable groups is necessary, but there is now much evidence that if an entire community uses treated nets this reduces the population of Plasmodium infected mosquitoes far more than if only pregnant women and children use them.
Compared with villages with few or no nets, there are far fewer cases of malaria fever and anaemia in children in villages where most people have had treated nets for 3-4 years. This applies even to those children who do not themselves have nets.  This indicates that community-wide use of bednets adds considerably to personal protection.
To confirm this, there needs to be a direct comparison between targeting treated nets only to people vulnerable to malaria, and targeting whole villages. However, the data so far strongly suggest that it would be worthwhile to provide the extra treated nets to ensure community-wide coverage and to kill the maximum number of mosquitoes.
It takes a short time for a team to check the number and size of the beds and sleeping mats in each house in a village, to provide a treated net for each, and to collect a householder's signature confirming that these nets have been provided.
In this way, a team of ten people can provide 800 treated nets in a day (enough for all in a medium-sized village). Their work is quick because they do not need to assess who does, and does not, belong to a targeted category of people. Nor do they need to collect and keep safe the many small sums of money from providing subsidised nets.
At these rates of productivity, one team could provide the treated nets needed by a million people over four years, after which the nets would probably be badly torn (if they are made of polyester) and the team would need to start the process of replacing the nets with new ones. An annual visit by a supervisor can take care of the necessary re-treatment of all of a village's nets.
It may be more cost-effective, despite the higher purchase price, to invest in physically durable polyethylene Olyset nets, which would allow each net replacement team to look after considerably more than a million people, because net replacement would be only needed after much longer than four years.
Funding bednet initiatives
In tropical towns, many people already buy nets and insecticide, mainly to protect themselves from the nuisance of Culex mosquitoes which (unlike malaria mosquitoes) breed in open urban drains and cess pits. Thus organised treated bednet programmes need not focus on towns.
It is the lowland rural areas of tropical Africa that bear the main malaria burden, and in which farmers find it difficult to find the money to pay for nets to protect their families. With distribution and re-treatment teams working as outlined above, the total yearly cost of providing treated bednets free of charge to all tropical African villagers is estimated to be US$200-350 million.
Donations by the Global Fund for AIDS, TB and malaria are now rising to the level at which such a continent-wide programme could be supported (in addition to their commitments to providing drugs for these diseases).
Thus, any African government that applies to the fund with a clear and carefully costed programme for full rural provision of treated nets seems now to have a good chance of receiving the necessary funding.
Many malaria experts and policymakers warn of 'donor fatigue'. This refers to the notion that the affluent world's generosity regarding malaria will not last and that African villagers should therefore get used to the idea that they will have to pay to protect their children from malaria themselves.
However, recent estimates that US$4 could provide subsidised treated nets for a year for pregnant women suggest that such systems will be at least as dependent on continuing donor funding as high-productivity system of universal provision for highly malarious villages. 
Health economists consider that malaria is a major cause of Africa's slow or negative rate of economic growth. As trade becomes increasingly global, rich countries should realise that it is in no-one's interests for Africa to remain trapped in the vicious circle of malaria and poverty that greatly restricts its ability to import from richer countries.
The sustained investment that rich countries would have to make, for a properly scaled-up programme of treated nets to have a major impact on Africa's malaria burden is in fact quite small compared with the annual US$2.1 billion which Europeans and North Americans are willing to pay to control another blood sucking insect — the fleas on their pet cats and dogs.
Biological factors that could threaten the use of treated nets
Apart from the politics surrounding sustained provision of treated nets, there are two biological factors that need to be considered. The first is a potential loss of immunity. Antibodies that protect against malaria accumulate in young children through repeated malaria infections. Research indicates that these antibodies do not build up as much in people living in villages that have used treated nets for several years, because they are infected far less often.
Thus, some malaria experts fear that children who have early and constant access to bednets would only be partially protected from malaria, and may die from it later in their childhood.
But the results of trials done over sixyears show that infant mortality continues to be reduced and older children show no higher mortality as a result of using treated bednets from infancy.  This is reassuring, and, in the long run, malaria vaccines will hopefully replace the immunity lost through bednet use.
Another concern is that mosquitoes might evolve resistance to the insecticide used to treat nets. The fact that so far all the insecticides used on nets are chemically similar — they are called 'pyrethroids' — is worrying, because resistance to one might mean resistance to all. However, it is important to investigate the full effects of genes that can help insects resist insecticides in field conditions.
In West Africa, laboratory tests show that many Anopheles gambiae mosquitoes have a gene called 'kdr', which enables them to partly resist pyrethroids. However, in areas where this gene is common, nets treated with pyrethroids still kill wild A gambiae when they enter huts and still prevent malaria.
This seems to be because the mosquitoes with the resistance gene are not as irritated by the insecticide as normal mosquitoes. Thus, they sit on treated nets for longer, eventually absorbing enough chemical to kill them. But complacency would be unwise, because there are some wild populations of Culex mosquitoes and bedbugs that have evolved resistance so that they cannot be killed by treated bednets.
It is important to identify insecticides safe enough for close contact with humans that are effective against such forms of resistance in insects. One possibility is the use of fungi that have been commercialised for use against locusts and cockroaches. They have recently been shown to increase the death rate of adult Anopheles mosquitoes. 
Comparison with 'indoor residual spraying'
In the current enthusiasm for scaling up treated bednets, it is important not to lose sight of the fact that indoor residual spraying, or spraying insecticides on the walls inside houses, has had great success against malaria mosquitoes. It has, for example, eradicated malaria in southern Europe and greatly lowered disease rates in southern Africa and Zanzibar. It continues to be used with success in South Africa, Zambia and Mozambique.
It would not be appropriate to switch to treated nets in areas where there are trained spray teams with the logistics to keep their spray pumps operational and to transport them to all villages that need treatment before of the start of the season of rains and malaria.
Some African highland areas are only subject to occasional malaria epidemics linked to exceptional rains. In these areas everyone is vulnerable to malaria because they do not build up partial immunity. Such epidemics are probably best tackled with a 'fire brigade' type of response with a mobile spraying team rather than trying to rush in enough nets for everyone.
Insecticide treated bednets act like traps for malaria mosquitoes that bite at night, which are attracted to the nets by the body odour of the people sleeping under them. Such nets provide partial protection against biting and, if used by a whole community, the number of mosquitoes in the community that survive long enough to develop infective malaria parasites, is greatly reduced.
Health experts and policymakers in Africa will need to find sources of funding — whether nationally or internationally — to support local teams to provide communities in highly malarious rural areas with insecticide-treated nets.
Chris Curtis is professor of medical entomology at the London School of Hygiene and Tropical Medicine, London, UK.
- Maxwell C. A. et al. Effect of community-wide use of insecticide-treated nets for 3-4 years on malarial morbidity in Tanzania. Trop Med Int Hlth 7, 1003-1008 (2002).
- Stevens W. et al. The costs and effects of a nationwide insecticide-treated net programme: the case of Malawi. Malaria J 4, 22 (2005)
- Eisele T. P. et al. Effect of sustained insecticide-treated bed net use on all-cause child mortality in an area of intense perennial malaria transmission in western Kenya. Am J Trop Med Hyg 73, 149-156 (2005)
- Scholte E. J. et al. An entomopathogenic fungus for control of adult African malaria mosquitoes. Science 308, 1641 (2005)