their first-degree relatives. (Alcohol dependence must have been
diagnosed as occurring at least 1 year before other drug dependence
[e.g., opioid dependence, cocaine dependence] was
present.)
The high-risk group consisted of 74 children and adolescents
(38 male and 36 female) from high-density (an average of 4.0
first- and second-degree relatives who were alcoholic), multigenerational
alcoholism families (Table 1). Alcoholism tended to
segregate within these families in a pattern consistent with a
major genetic effect (Yuan et al 1986).
LOW-RISK CONTROL SUBJECTS. Community control
subjects were identified through an index case who responded to
a newspaper advertisement. Families were chosen on the basis of
having the same structural characteristics as the high-risk families
(at least two adult brothers). Family members were interviewed
using the same diagnostic procedures used for the
high-risk families. Each potential control family was screened for
the presence of alcohol or drug dependence using the family
history report of the index case. Presence of a definite diagnosis
of alcoholism by Feighner criteria or alcohol or drug dependence
by DSM-III in the index case or his first-degree relatives
disqualified a potential control family. Low-risk families were
included if all first- and second-degree relatives of the index case
were free of alcohol and drug dependence. The study design,
which included obtaining family history and direct interviews of
family members from both sides of the family, assured that the
control children/adolescents came from bilineal low-risk-foralcoholism
pedigrees. Fifty-one low-risk children/adolescents
(28 male and 23 female) were available for follow-up.