The obesity epidemic has already reached the youngest demographic: About 17 percent of U.S. children and teens are obese, and about one in three are overweight. These numbers, reflected in many other countries, have risen steadily in recent decades. And researchers are racing to find the most powerful drivers behind these scary figures, as children who carry the extra pounds into adulthood increase their risks for diabetes, heart disease and other chronic diseases.

The most basic cause—no surprise, has to do with too much unhealthful food and not enough physical activity. Looking more closely at the roots of childhood obesity across different populations yields a complex and nuanced puzzle. Lifestyle and environmental factors that lead to obesity differ for a high schooler in the Bronx versus one in rural Alabama.

Many researchers have turned to geographic information systems (GIS) as powerful analysis tools to map this massive health problem and its possible causes. A special report, published online April 10 in the American Journal of Preventive Medicine, presents six new studies in this growing subfield, some with surprising results that could have implications for health, education and urban planning policy.

"To fully understand the multifaceted nature of childhood obesity, a geographic perspective should be considered," Celeste Torio, of the Department of Health and Human Service's Agency for Healthcare Research and Quality, wrote in an essay published online in the same issue of American Journal of Preventive Medicine. "Many of the contributing factors to childhood obesity (particularly social, policy and environmental determinants) have place-to-place variation."

Here's one example. Where do you think children ages 12 to 16 years get most of their daily physical activity? Sports? Gym class? At home? The answer turns out to be largely dependent on where they live. In a study led by Daniel Rainham, of the Environmental Science Program at Dalhousie University, researchers tracked 380 adolescents for a week, using both accelerometers and GPS. They found that teens who lived in rural areas did, in fact, get most of their recommended one-hour daily moderate-to-vigorous physical activity at school. However, kids who lived in suburban and urban environments logged most of their active time getting from place to place. And both groups got a fair amount ( 10 to 25 percent) at home.

Rainham and his colleagues also found that girls were getting more physical activity than boys, and that urban teens seemed to be getting the most overall. "Policy makers should be aware that active transportation is an important source of moderate-to-vigorous physical activity for urban youth," they wrote. And likewise, they contend, education policies should recognize that school-day activities continue to be an important source of exercise for rural teens.

In addition to differences in physical activity, urban and rural adolescents also seem to pick different food options—even when faced with the same smattering of fast food restaurants. Researchers have continued to debate about whether proximity of fast food leads to greater levels of consumption and poorer health outcomes. But another new study, led by Lorna Fraser, of the School of Geography at the University of Nottingham in England, suggests that the culpability of a fast food burger and fries for that extra weight might depend largely on on where teens live.

The researchers measured the number of fast food establishments within one kilometer of the study participants' homes along with each 13 to 15 year old's obesity status and reports from caretakers about how often the teens ate at fast food joints.

Those subjects who ate more fast food meals were, predictably, more likely to be obese. Interestingly, a larger number of fast food restaurants in an urban area did not lead to increased consumption (in fact, it seemed to make teens less likely to eat there), but the trend reversed in more rural environments. The results suggest that "public health interventions that place restrictions on the location of fast food outlets may not uniformly decrease consumption," the researchers noted.

Additional studies like these should help policy makers zero in on the most effective ways to combat childhood and teen obesity at a more fine-grained scale, allowing GIS modeling to analyze “complex reality by breaking it down into layers of information," she wrote in her essay.

The improved spatial resolution will not provide all of the answers about how to fight childhood obesity. "GIS is not a panacea," Stephen Matthews, of the Pennsylvania State University, noted in an essay published online in the same issue of American Journal of Preventive Medicine. "The role of place and how children interact with place is just one part of the obesity puzzle." But she hopes that further study of the spatial environment will help provide new insights into battling childhood obesity.