"I know of no other single food product whose elimination can produce this degree of weight change," said the study's leader, Dr. David Ludwig of Boston Children's Hospital and the Harvard School of Public Health.
The weight difference between the two groups narrowed to 2 pounds in the second year of the study, when drinks were no longer being provided. That showed at least some lasting beneficial effect on kids' habits. The study was funded mostly by government grants.
A second study involved 641 normal-weight children ages 4 to 12 in the Netherlands who regularly drank sugar-sweetened beverages. They were randomly assigned to get either a sugary drink or a sugar-free one during morning break at their schools, and were not told what kind they were given.
After 18 months, the sugary-drink group weighed 2 pounds more on average than the other group.
The studies "provide strong impetus" for policies urged by the Institute of Medicine, the American Heart Association and others to limit sugary drink consumption, Dr. Sonia Caprino of the Yale School of Medicine wrote in an editorial in the journal.
The American Beverage Association disagreed.
"Obesity is not uniquely caused by any single food or beverage," it said in a statement. "Studies and opinion pieces that focus solely on sugar-sweetened beverages, or any other single source of calories, do nothing meaningful to help address this serious issue."
The genetic research was part of a much larger set of health studies that have gone on for decades across the U.S., led by the Harvard School of Public Health.
Researchers checked for 32 gene variants that have previously been tied to weight. Because we inherit two copies of each gene, everyone has 64 opportunities for these risk genes. The study participants had 29 on average.
Every four years, these people answered detailed surveys about their eating and drinking habits as well as things like smoking and exercise. Researchers analyzed these over several decades.
A clear pattern emerged: The more sugary drinks someone consumed, the greater the impact of the genes on the person's weight and risk of becoming obese.
For every 10 risk genes someone had, the risk of obesity rose in proportion to how many sweet drinks the person regularly consumed. Overall calorie intake and lifestyle factors such as exercise did not account for the differences researchers saw.
This means that people with genes that predispose them to be obese are more susceptible to the harmful effects of sugary drinks on their weight, said one of the study leaders, Harvard's Dr. Frank Hu. The opposite also was true — avoiding these drinks can minimize the effect of obesity genes.
"Two bad things can act together and their combined effects are even greater than either effect alone," Hu said. "The flip side of this is everyone has some genetic risk of obesity, but the genetic effects can be offset by healthier beverage choices. It's certainly not our destiny" to be fat, even if we carry genes that raise this risk.
The study was funded mostly by federal grants, with support from two drug companies for the genetic analysis.
Obesity info: http://www.cdc.gov/obesity/data/trends.html
BMI calculator: http://www.nhlbisupport.com/bmi/bminojs.htm
New England Journal: http://www.nejm.org
Marilynn Marchione can be followed at http://twitter.com/MMarchioneAP