In global public health, disease-fighting tools that are cheap, available and sustainable are the Holy Grail. It might be hard to top the one being tested in Tanzania as a way to prevent malaria: smelly socks.

Experiments in three villages where people get about 350 bites a year from malaria-infected mosquitoes are using dirty socks to lure the insects into traps, where they become contaminated with poisons and ultimately die.

Researchers hope that if the strategy works, it will eventually complement insecticide-treated bed nets as a low-tech way to prevent malaria, which kills nearly 900,000 people a year worldwide, most of them children.

“It’s a bold idea. Who would have thought there was a life-saving technology working in your laundry basket?” said Peter A. Singer, a physician who heads Grand Challenges Canada, a development agency of the Canadian government that is helping fund the research.

Previous lab studies have shown that smelly socks work well in attracting mosquitoes. Field experiments have shown that synthetic bait is more attractive than people, at least until the insects get close enough to realize there’s no blood waiting for them.

The new experiments, however, are the first head-to-head field tests of footwear vs. chemistry. The researchers hope the footwear wins.

“It is simply a cost issue and an expediency issue,” said Fredros O. Okumu, the Tanzanian entomologist leading the research. “Socks are more readily available, and you don’t have to mix any chemicals. It is the sort of thing that could be set up in a cottage factory.”

The traps are square boxes that look a little like commercial beehives. Some will contain the human-odor bait, which consists of simple chemicals (including lactic acid, ammonia and propionic acid) that are exuded by people, especially from the legs and feet. Some will contain socks worn for a day by adults. Others will contain cotton pads that schoolchildren will put inside their socks for a day and then deliver to researchers.

The researchers will compare the number of mosquitoes caught with each method.

Earlier work by Okumu and his colleagues at the Ifakara Health Institute in Tanzania showed that the chemical bait attracted four times as many mosquitoes as live people and that dirty socks worked just as well, at least in the lab. If the sock pads prove adequate, they will be the preferred bait.

The inside surfaces of some traps are coated with an organophosphate pesticide. Mosquitoes that land there will die within 24 hours. Other traps contain a fungus that infects the insects and kills them in five days — roughly half the time needed for the complicated cycle that enables a newly infected mosquito to transmit the malaria parasite to a person.

The bait-and-kill strategy is a new one in malaria prevention efforts.

Normally, attempts to prevent malaria by controlling mosquitoes, known as vector control, have aimed at driving the insects away from people or killing them once natural attraction has brought them into proximity.

Insecticide-treated bed nets, millions of which have been sold or given away in Africa in the past decade, have a long-acting repellant, permethrin. In many malaria-endemic areas, people spray the inside walls of dwellings with insecticide that kills mosquitoes when they land.

Bed nets have cut childhood deaths by about 20 percent in malaria-endemic areas. Modeling suggests that traps could reduce malaria transmission about as much as bed nets do in villages where half the households use them.

Despite its low-tech appearance, the strategy Okumu is testing is far more complicated, and potentially fraught with hazard, than it seems.

A key question is where to place traps. They need to be close enough to dwellings to attract mosquitoes, but not so close that they will increase people’s exposure to the disease-carrying insects. Okumu’s research suggests that the traps should be at least 100 feet from houses.

Another question is how many traps a village might need. Okumu has calculated the minimum number at 20 per 1,000 people, although in places where malaria transmission is especially intense and in certain village configurations, 130 traps per 1,000 people might be needed.

Despite these challenges, Singer said, projects such as these are what Grand Challenges Canada is looking to support. It is providing $388,000 for the research, and the Bill and Melinda Gates Foundation is providing a similar amount. The Gates Foundation gave Okumu $100,000 for preliminary studies, as well.

“We are inspired by people like Fredros,” Singer said. “We strongly believe that innovators in low-income countries are best situated to solve their own problems. He is an African researcher with an African innovation for an African problem.”

Okumu, who is a doctoral candidate at the London School of Hygiene and Tropical Medicine, said he is “working on the premise that this is a global problem — a global problem in a flat world.”

He said he doubted that there might be an application for his strategy — should it prove successful — in non-malarious places such as the United States. A pair of socks from a recent 10K run at the corner of a patio will only briefly divert mosquitoes. They’ll soon find the bare legs under the picnic table.

“Mosquitoes are still fairly clever animals,” he said. “What they are looking for is blood. They might be attracted to the socks, but they will not spend much time there.”