Skip Navigation

Link to  the National Institutes of Health  
The Science of Drug Abuse and Addiction from the National Institute on Drug Abuse Archives of the National Institute on Drug Abuse web site
Go to the Home page

National Institute on Drug Abuse - NIDA NOTES
Volume 10, Number 1
January/February 1995

Early Childhood Behavior and Temperament Predict Later Substance Use

By Neil Swan, NIDA NOTES Contributing Writer

By the first grade, or earlier, children show temperament and behavior traits that are powerful indicators of their inclination to use and abuse drugs in their teenage and adult years. Researchers have identified not only common childhood risk factors and behaviors that predict drug abuse potential but also protective factors that shield some children from influences to use drugs.

A number of long-range NIDA-funded studies have traced at-risk children into adulthood and parenthood, trying to determine why some children are able to resist persistent influences to use substances of abuse. Studies have zeroed in on several important factors in predicting a first-grader's subsequent use of substances: shyness, aggressiveness, rebelliousness, and gender. External risk factors include substance use among peers, drug use by parents, and troubles with the police. Protective factors include achievement in school or after-school activities and close family ties. The researchers are now designing drug abuse prevention and intervention strategies based on these findings made over 20 or more years.

Some of the earliest studies, by Dr. Margaret E. Ensminger and Dr. Sheppard G. Kellam and colleagues of Johns Hopkins University, started in the 1960s with first-graders and their families in Woodlawn, a poor, urban African-American community on the South Side of Chicago.

Today the researchers are following about 1,000 of the 1,242 original first-graders to continue to identify and monitor early childhood factors affecting later drug use and symptoms of psychiatric problems. These first-graders are now 32 or 33 years old.

Interviews in 1993 show that key risk factors such as aggressive behavior and shy-aggressive behavior identified 26 years ago continue to hold and are valid predictors of the subjects' current levels of cocaine use as adults.

For both sexes, higher scores on first-grade
IQ and readiness-for-school tests were associated
with higher levels of beer or wine, hard liquor,
and marijuana use 10 years later.

During their studies, Dr. Ensminger and her colleagues rated each first-grader's mental health using two criteria: social adaptation and psychological health. To measure pupils' social adaptation to school, researchers used teachers' ratings of children's classroom social performance and intelligence as well as the results of standardized tests. Psychological health was determined by a number of criteria, including psychological symptoms, abnormal behavior, and level of self-esteem.

Two important risk factors identified as predictors of later drug use are shy behavior--described by the teachers as sitting alone, having few friends, and not speaking up in class--and aggression--described as fighting with others or breaking rules. Shyness and aggression are types of poor social adaptation distinct from symptoms of anxiety or depression, which are internal feelings, noted the researchers.

While shyness and aggressiveness are key predictors of drug use, a complex relationship exists between the two factors. Among boys, aggressive behavior in the first grade leads to increased teenage substance abuse, while first-grade shyness alone without aggressiveness leads to lower levels of substance abuse as teenagers. However, the combination of shyness and aggressiveness leads to even higher levels of adolescent substance use among boys than aggressiveness without shyness does, the studies found.

Boys whose teachers said they had problems concentrating in class had higher levels of later substance abuse because concentration problems appear to be closely related to aggressiveness, the studies reported. By contrast, neither aggressiveness, nor shyness, nor concentration problems in the first grade were associated with later substance use among girls.


A number of long range NIDA-funded studies have
traced at-risk children into adolescence and beyond
trying to determine why some of them are able
to resist influences to use substances of abuse

Understanding Gender Differences

Shyness and aggressiveness may be less important predictors of substance use among females than among males because girls' peer groups are smaller and less important to them, she adds. These gender-based considerations are now being studied in drug abuse prevention programs, says Dr. Ensminger, who was among the first researchers to urge colleagues to stop dismissing gender as a possible key consideration in predicting children's subsequent drug use.

When the first-graders reached age 16 or 17, girls used smaller amounts of beer, wine, liquor, and marijuana and other illicit drugs but not cigarettes than boys did. For both sexes, higher scores on first-grade IQ and readiness-for-school tests were associated with higher levels of beer or wine, hard liquor, and marijuana use 10 years later, the researchers found. "This tells us that the children who are most ready for school are also those most ready to experiment with drugs," says Dr. Ensminger.

Psychological well-being and family relationships in the first grade seemed more important to girls than to boys in terms of influencing psychiatric symptoms 10 years later. Mothers had an important effect on the psychological status of their daughters but not of their sons. Mothers' expectations of how far daughters would go in school and mothers' own psychological health were positive factors in their daughters' psychological well-being 10 years later, the study found. Girls with strong family bonds tend to use drugs less than other girls do, but the same family influence is not so apparent with boys, said Dr. Ensminger.

Results from the Woodlawn study served as the basis for prevention programs started in Baltimore in the 1980s by Dr. Kellam and colleagues. That prevention effort focused on aggressive behavior because of its relationship to later drug use and on underachievement because of its relationship to depressed feelings.

Recent data gathered on the Woodlawn study subjects show that early childhood aggression is still a valid predictor of drug abuse when measured against the now-adult subjects' levels of use of cocaine, Dr. Ensminger reports. Those data are now being prepared for publication.

Examining Protective Factors

Another long-term study of drug-use predictors focuses on children in Northeastern States. For 20 years, Dr. Judith S. Brook of Mt. Sinai School of Medicine has studied risk factors identified in early childhood and in adolescence that are related to drug use during adolescence. She is conducting a study of 1,000 children and their mothers that began in 1975 in two communities representative of the population of the Northeastern United States. In the continuing research, Dr. Brook is examining not only risk factors but also protective factors that help shield children and adolescents from these risk factors.

Dr. Brook and her colleagues have identified a number of risk factors for subsequent drug use such as childhood aggression, which includes anger, aggression toward siblings, noncompliance, temper, and nonconforming behavior. Other risk factors are unconventionality--an attitude of deviance, rebelliousness, and evasion of responsibility--the extent of drug use among peers, and parental sociopathy, that is, parents' problems with drinking, drugs, or the police.

Dr. Brook's group examined the risk factors and their implications during childhood, ages 5 to 10; middle adolescence, ages 13 to 18; and late adolescence, ages 15 to 20. They found that childhood aggression and parental sociopathy predicted increased levels of drug use in late adolescence. They also determined that unconventionality during the early years of adolescence had an "important and pervasive impact on all aspects of middle and late adolescent functioning," including increased drug use.

The research team is now observing the original subjects' children, beginning at age 2, and interviewing both parents of these children to collect data on the new generation. "So we're now studying the third generation--the grandchildren of the mothers [of the original subjects] initially studied in 1975," says Dr. Brook. "And we're finding a great deal of consistency down through the generations in regards to personality and family characteristics," including traits that are drug-use risk factors.

The researchers also have studied interactions among risk factors and their implications for subsequent drug use as the children grow older. In addition to childhood aggression, they found three additional factors that influence late-adolescent drug use--unconventionality and drug use in middle adolescence and parental sociopathy during childhood. They found that little or no drug use in middle adolescence when combined with conventionality during the same age span resulted in the least amount of subsequent drug use.

As expected, parental sociopathy is related to late-teen drug use. Parents who drink or use drugs or both may be the most strict with their children, telling them, "Do as I say, not as I do," according to Dr. Brook's study. These admonitions might be effective in middle adolescence, when children are more likely to be influenced by parental demands, but not in later adolescence, when the family has less control and the parents' own display of negative behaviors becomes a drug-use risk factor, she says.

Dr. Brook agrees with other researchers that there are childhood protective factors that can be very powerful shields to safeguard children and adolescents from the recognized risk factors. These protective factors include achievement, religious commitment, strong family bonds, and a solid attachment to and emulation of a wholesome role model, she says.

"Some of these children are remarkably resilient," says Dr. Brook. Among those who become successful, she found evidence of protective factors such as church attendance, childhood achievement in school or in extracurricular activitties, or close ties to brothers and sisters.

"Many of these kids go on to lead successful, productive lives, yet we tend to focus on the ones that don't," says Dr. Brook. "I want to learn more about what makes those that do well do so."

Family Relationships Critical

Still another continuing study of predictive factors for drug use focuses on a different population segment--at-risk children of white families living in small and medium-sized communities in Oregon. The study by Dr. Hyman Hops and colleagues of the Oregon Research Institute examines family and peer-group influences on adolescent substance use and is now in its 10th year. About 500 subjects were ages 11 through 15 at their first assessment in 1984 and will be 21 through 25 at their last assessment this year.

Among those studied, 90 percent of subjects who progressed from one substance of abuse to another did so in the following sequence--abstinence, alcohol, cigarettes, marijuana, and hard drugs. The most dramatic increase in drug use occurred between the ages of 13 and 14, when adolescents are going from middle to high school.

Parents' use of substances, including cigarettes, is an important predictive factor influencing their children's drug use. Within two-parent families, Dr. Hops found that fathers' drinking appears to have a greater impact than mothers' drinking on both girls and boys, while mothers' drinking has an effect only on adolescents under 14, before they enter high school. Parents who use cigarettes and alcohol may influence not only their children's use of the same substances but illicit substances as well, he says.

Family conflict and strife are strongly associated with increased substance abuse, based on the researchers' direct observations of problem-solving scenarios between parents and adolescents. Their findings suggest that families with substance-abusing children typically are unable to easily resolve problems and that the resulting confrontations negatively affect drug use.

In examining peer influences, the Oregon researchers balanced each study subject's self-reports of levels of substance use against reports of his or her substance use level from the child's best friend. The scientists reported that the amount of both family cohesion and peer encouragement to use drugs was predictive of initial levels of substance abuse. A good family relationship may play a powerful role as a protective factor in middle and late adolescence, they say. On the other hand, peer encouragement to use substances plays a stronger role across the age range and also suggests that early peer influences may encourage higher levels of drug use at later ages.

"These findings underscore the importance of family influences on substance abuse throughout adolescence and suggest greater attention to the family, as well as the peer group, in designing prevention strategies," says Dr. Hops. "You've got to have a healthy family relationship to counter the very powerful peer influences that kids face today."


Andrews, J.A.; Hops, H.; Ary, D.; Tildesley, E.; and Harris, J.
Parental influence on early adolescent substance use: Specific and nonspecific effects. Journal of Early Adolescence 13:285-310, 1993.
Brook, J.S.; Whiteman, M.; Cohen, P.; and Tanaka, J.S.
Childhood precursors of adolescent drug use: A longitudinal analysis. Genetic, Social, and General Psychology Monographs 118(2):195-213, 1991.
Brook, J.S.; Whiteman, M.; Gordon, A.S.; and Brook, D.W.
The psychosocial etiology of adolescent drug use: A family interactional approach. Genetic, Social, and General Psychology Monographs 116(2):113-267, 1990.
Brook, J.S.; Whiteman, M.; Hamburg, B.A.; and Balka, E.B.
African-American and Puerto Rican drug use: Personality, familial, and other environmental risk factors. Genetic, Social, and General Psychology Monographs 118(4):417-438, 1992.
Ensminger, M.E.
Sexual activity and problem behaviors among black, urban adolescents. Child Development 61:2032-2046, 1990.
Ensminger, M.E.; Brown, C.H.; and Kellam, S.G.
Sex differences in antecedents of substance use among adolescents. Journal of Social Issues 38(2):25-42, 1982.
Hops, H.; Tildesley, E.; Lichtenstein, E.; Ary, D.; and Sherman, L.
Parent-adolescent problem-solving interactions and drug use. American Journal of Drug and Alcohol Abuse 16:239-258, 1990.
Kellam, S.G.; Brown, C.H.; Rubin, B.R.; and Ensminger, M.E.
Paths leading to teenage psychiatric symptoms and substance use: Developmental epidemiological studies in Woodlawn. In: Guze, S.B.; Earls, F.J.; and Barrett, J.E., eds. Childhood Psychopathology and Development. New York: Raven, 1983. pp. 17-51.

From NIDA NOTES, January/February, 1995

[Home Page][NIDA NOTES Index][1995 Archive Index]

Archive Home | Accessibility | Privacy | FOIA (NIH) | Current NIDA Home Page
National Institutes of Health logo_Department of Health and Human Services Logo The National Institute on Drug Abuse (NIDA) is part of the National Institutes of Health (NIH) , a component of the U.S. Department of Health and Human Services. Questions? See our Contact Information. . The U.S. government's official web portal