Sugary drinks could be causing women to start their menstrual cycle prematurely, according to a new study.
The discovery is particularly important because earlier onset periods have been linked to an increased risk of breast cancer later in life.
Sugary drinks are also said to be contributing to childhood obesity.
A study of girls aged nine to 14 found that those drinking more than 1.5 sugar-sweetened beverages a day had their first period 2.7 months earlier than those consuming two a week or fewer.
US lead researcher Dr Karin Michels, from Harvard Medical School, said: "Our study adds to increasing concern about the widespread consumption of sugar-sweetened drinks among children and adolescents in the USA and elsewhere.
"The main concern is about childhood obesity, but our study suggests that age of first menstruation (menarche) occurred earlier, independently of body mass index, among girls with the highest consumption of drinks sweetened with added sugar.
"These findings are important in the context of earlier puberty onset among girls, which has been observed in developed countries and for which the reason is largely unknown."
SEE ALSO:
Comedy Group's Spoof Video Shows Why Men Should Never Make Tampon Adverts
Curb The Period Pain, Boost Your Energy And Make Life Easier With These Bloody Good Period Hacks
A one year reduction in age at first period is estimated to raise the risk of breast cancer by 5%.
The impact of bringing forward menarche by 2.7 months was likely to be "modest" said the scientists. But they expected some girls to be consuming more than the quantities of sugary drinks seen in the study.
Writing in the journal Human Reproduction, the researchers concluded: "The amount of SSBs consumed by girls in our highest category of consumption, more than 1.5 servings per day ... is likely low compared with consumption in certain other populations, in which we would expect an even more dramatic decrease in age at menarche.
"Most importantly, the public health significance of SSB consumption at age at menarche, and possibly breast cancer, should not be overlooked, since, unlike most other predictors of menarche, SSB consumption can be modified."
The girls included in the research were participants in the Growing Up Today study, which is following the progress of 16,875 children across the US.
Story continues below...
Questionnaires were used to ask the girls about their diets at several points in time between 1996 and 2001, and collect information about the kind of drinks they consumed.
A drink "serving" was defined as either a can or a glass. Sugary drinks included sodas, fruit drinks, lemonade and iced tea.
By the end of the five year period, all but 3% of the girls had started menstruating.
The average first period age for girls consuming the most sugary drinks was 12.8 years compared with 13 years for those drinking the least.
After adjusting the results to take account of Body Mass Index (BMI), girls drinking the most SSBs were 22% more likely to start their period in the month after being questioned about their diet than the lowest consumers.
The researchers found that artificially added sugar was chiefly responsible for their findings rather than natural sugars in drinks such as fruit juices.
Drinks with added sugar have a higher "glycaemic index" than naturally sweetened drinks, leading to rapid "spikes" of the hormone insulin.
Boosted insulin can result in higher concentrations of sex hormones, and large changes in the levels of these have been linked to periods starting earlier, said the scientists.
Commenting on the research, Dr Ken Ong, from the Medical Research Council Epidemiology Unit at Cambridge University, said: "This is a very large study, which was representative across the USA, and the findings are strongly statistically significant. From previous research we know that sugar-sweetened beverages (SSB) intake promotes weight gain, and that weight gain promotes early puberty in girls - so a 'SSB-to-puberty' link is expected.
"The surprise here is the claim that the association is independent of childhood size - ie that there is a more direct effect of SSB on puberty. This is unexpected."