Background: This article is the first report of the Irish Affected Sib Pair Study of Alcohol Dependence, whose goal is to detect the genomic location of susceptibility loci for alcohol dependence (AD). This article describes phenotypic characteristics of the probands, siblings, and parents included in the sample and examines agreement among different sources of diagnostic information, including the validity of family history (FH) assessment.

Methods: Structured diagnostic interviews were conducted with 1414 individuals from 591 families ascertained in Ireland. AD was assessed among 1201 probands and affected siblings with use of the Semi-Structured Assessment for the Genetics of Alcoholism and among 213 parents with use of a modified version of the Structured Clinical Interview for DSM. Probands and siblings were also assessed for drinking history, comorbid disorders, and other clinical characteristics. FH reports based on FH-Research Diagnostic Criteria were obtained for 1113 of these individuals as well as for 3652 first-degree relatives who were not interviewed.

Results: Sample characteristics confirm the severity of AD among the affected individuals. Agreement between FH ratings and diagnoses based on direct interviews was high for both parent-offspring and sibling-sibling comparisons (e.g., positive and negative predictive values > 80% for a range of cutoffs). Agreement among individuals about their family members was also high for a single item (1 month or more of drinking problems, tetrachoric r = 0.86-0.98), the total number of DSM-IV AD symptoms (polychoric r = 0.86-0.96), and classifications based on a range of cutoffs ([kappa] = 0.75-0.80). Use of multiple informants improved classification accuracy only slightly (6-10%).

Conclusions: The authors successfully collected data for a large sample of affected sibling pairs for molecular genetic analysis of AD. Individuals with AD were able to provide accurate evaluations of alcoholism symptoms in their parents and adult siblings. A single screening item performed nearly as well as the full scale. Collecting information from multiple informants may not be cost effective for the gain in predictive accuracy. FH information collected from affected informants can be a valuable source of diagnostic information for family studies of alcoholism.