Economic planners often build new roads in remote areas to promote urbanization and accommodate industries such as logging. The roads, though, bring not only economic development but also, it seems, diarrhea.

Joseph Eisenberg, an epidemiologist at the University of Michigan, and his colleagues found that remote Ecuadorian villages had nearly one third as many cases of diarrhea than their more accessible counterparts. Diarrheal disease caused by Escherichia coli, rotavirus and the parasite Giardia spreads through contaminated food, water and by person-to-person contact. "Roads have been very important in allowing and facilitating the spread of everything one needs to set up an infection," says Mary Wilson, a professor of population and international health at Harvard University, who was not involved with this study.

In 2003 construction was completed on a highway connecting the town of Borb¿n in northern Ecuador with eastern villages closer to the Andes and Colombia. Previously, only rivers had connected the towns. Since 2003 secondary roads, used mainly for logging, continue to be constructed.

The researchers randomly selected 21 villages near Borb¿n, the most urbanized town in the region. Some villages were closer to Borb¿n and to the road, others a little farther away and some were remote. During the study, the researchers visited each village three times over a period of 15 days. While there, they collected information about current cases of diarrhea by calling on each household, taking samples and asking the villagers about their social contacts.

Borb¿n, the researchers report in this week's issue of Proceedings of the National Academy of Sciences USA, had twice as many diarrheal cases than the study communities. For pathogenic E. coli, Borb¿n residents' risk of disease was 16 times higher than the villagers. The farther from Borb¿n and the farther from the new road a community was, the lower the number of diarrheal cases. But villages nearer to the road had three times more diarrheal disease than remote ones. "It's surprising how significant the results are across different pathogens," Eisenberg says.

Many factors play a role in contracting these diseases, Wilson notes. "E. coli is something found in people's fecal flora," she says. "Giardia can be found in people who don't know they have it. And rotavirus follows a seasonal pattern." Roads, Wilson says, might not be the only answer to the increased spread. They may, however, amplify these factors by disrupting traditional communities, Eisenberg speculates.

Social cohesion in remote villages, like these Ecuadorian ones, creates support for building infrastructure such as water sanitation, Eisenberg says. But roads disrupt close-knit communities by altering the flow of people. "It starts into motion social changes. Changes drive transmission. That doesn't happen immediately; it takes years, decades for social processes to unfold," Eisenberg remarks. And, he notes, these sorts of effects should be considered when planning how to go about developing an area. "Sanitation interventions, water quality interventions are going to be important when developing roads," Eisenberg says.