The finding by Virginia Commonwealth University researchers conflicts with the conclusions of at least 15 previous studies.
Instead, the VCU researchers say they found a link between adolescent behavior problems and the mother's own "antisocial behavior" -- which may have caused her to smoke during her pregnancy in the first place.
"You might hypothesize that a woman who smokes during her pregnancy -- not one or two cigarettes a day but a pack or more -- it says something about her," says study author Judy L. Silberg.
"She may have some antisocial traits or conduct disorder traits. She knows it's bad for her kid and she does it anyway," adds Silberg, an assistant professor of human genetics and a researcher at the Virginia Institute for Psychiatric and Behavioral Genetics at VCU.
Smoking during pregnancy has long been associated with a variety of problems, including premature delivery, low birth weight, infant death and obstetric complications. More recently, numerous studies have implicated maternal smoking during pregnancy with behavioral problems that arise during a child's teenage years.
The VCU study, published in the Jan. 15 issue of Biological Psychiatry, refutes that connection. Instead, it contends that antisocial behavior may cause a mother to smoke during pregnancy, and those antisocial tendencies are passed on to the children, causing later conduct problems.
These problems, seen most often in boys, involve repetitive and persistent behaviors such as lying, fighting, setting fires, mutilating animals and a general disregard for other people, Silberg says. She classifies them as "sub-threshold conduct disorders."
"If you were measuring it on a scale and you added all these things up, you would see an alarming pattern," she says.
For the study, Silberg and her colleagues collected data on 538 adolescent twin boys -- aged 12 to 17 -- over a 15-year period to determine the role of genetic and social factors in the development of behavioral problems in children and teens.
What they found was that a child's tendency to have behavioral problems was related more to the mother's own behavioral problems than it was to her smoking during pregnancy.
Silberg emphasizes that her study in no way minimizes the harmful effects of smoking during pregnancy. She says it's important, however, to get to the truth about causes of conduct problems.
"Just because two variables are associated, doesn't mean one causes the other," she says. "The two variables can be explained by a third factor."
Silberg says her study needs to be replicated, especially since the results conflict with at least 15 previous studies. She's also doing additional research to determine whether the tendency toward antisocial behavior is passed from mother to child genetically, or whether it's a result of the home environment.
"We don't know whether it's genetic or environmental. We're now collecting other data that will determine within the context of behavioral and emotional problems whether the transition from parent to child is genetic, environmental or both," she says.
Dr. Ed Christophersen, a professor of pediatrics at Children's Mercy Hospital in Kansas City, Mo., says the results of the VCU study seem reasonable -- even in light of the prior studies' findings.
"Even considering all the other results, this could still be possible because they [the previous researchers] didn't look at this," Christophersen says. "It never occurred to me that there was something else that could explain this behavior in children. The probability that the mother contributed as much to her son's behavior problems as the smoking did makes sense.''
What To Do