NEW YORK (Reuters Health) - Teenagers who smoke may be more likely than their peers to become problem drinkers, a large U.S. study suggests.

For some people, a drink and a smoke seem to naturally go together. But it's been unclear whether one substance can actually encourage dependence on the other.

In a new survey of nearly 75,000 children and adolescents between12 and 20 years old, researchers found that smokers were 4.5 times more likely to report signs of alcohol abuse or dependence. The risk was particularly pronounced among the youngest smokers -- even with other factors, like race and family income, taken into account.

The higher rate of alcohol problems was not explained by the fact that smokers simply drank more in general.

"What our study shows is that, if you compare smokers and non-smokers who drink the same amount, the smokers have substantially more problems with alcohol than the non-smokers do," explained lead author Dr. Richard A. Grucza, a research assistant professor at Washington University School of Medicine in St. Louis.

There are several potential explanations for this finding, according to Grucza.

One possibility, he told Reuters Health, is that there's a common genetic predisposition that makes some people vulnerable to both alcohol dependence and habitual smoking. Another explanation, particularly in teenagers, is that people tend to pick up certain combinations of habits -- like smoking and drinking -- from their social environment.

But it's also possible that smoking directly influences vulnerability to alcohol problems, according to Grucza.

Animal studies have found that nicotine may affect the still-developing adolescent nervous system in a way that boosts susceptibility to addiction -- though it's still unclear if this applies to humans as well.

The current findings, published in the journal Alcoholism: Clinical & Experimental Research, are based on data from a government survey on drug use and health. Grucza and his colleague Laura J. Bierut focused their analysis on 74,836 survey respondents between the ages of 12 and 20.

Of these, 16 percent said they'd smoked during the past year, while 9.5 percent met the criteria for alcohol abuse or dependence. Smokers had a 4.5 greater risk of developing an alcohol problem compared with nonsmokers, and this pattern held true when the researchers separated the teens according to their reported drinking habits.

Among 12- to 14-year-olds who said they drank between one and eight drinks in the last month, 28.4 percent of smokers had an alcohol problem -- versus only 4.2 percent of non-smokers.

These findings are consistent with what animal research shows regarding the effects of nicotine on the brain, Grucza said.

If that research holds true for humans, he and Bierut report, it would offer a "powerful message" to teenagers about the dangers of smoking.