Sigvardsson · Archives of general psychiatry 1996 · adoption cohort study · n=1237

Replication of the Stockholm Adoption Study of alcoholism. Confirmatory cross-fostering analysis.

Cited 273 times in the scientific literature.

Level 3 - non-randomized controlled study

Retrospective adoption cohort study (non-randomized cross-fostering design)

PubMed 8694681 · doi:10.1001/archpsyc.1996.01830080033007 · record verified 2026-08-26

What was done

This study attempted to replicate the findings of the Stockholm Adoption Study regarding type 1 and type 2 alcoholism. Researchers evaluated an independent sample of 1,237 individuals (577 men and 660 women) born in Gothenburg, Sweden, who were adopted at an early age by nonrelatives. Genetic and postnatal environmental risk backgrounds were classified using the discriminant analysis procedures established in the original study to perform a cross-fostering analysis.

What was found

In male adoptees, the lifetime risk of severe type 1 alcoholism increased nearly 4-fold when both genetic and environmental risk factors were present compared to those without both factors (11.4% vs 3.0%). Neither genetic predisposition nor postnatal environment alone was sufficient to significantly increase type 1 risk. In contrast, risk of type 2 alcoholism increased more than 5-fold to 6-fold in men with a type 2 genetic background compared to others (10.7% vs 2.0%), irrespective of postnatal environment. Type 2 genetic background did not confer excess risk for type 1 alcoholism and vice versa. No increased risk was observed for mild alcohol abuse based on genetic or environmental classification.

Why it matters

This replication confirms that type 1 and type 2 alcoholism represent clinically and etiologically distinct subtypes, establishing that type 1 involves complex gene-environment interaction whereas type 2 is strongly heritable and relatively insensitive to postnatal environment.

Limits

The abstract reports quantitative outcome data only for male adoptees, leaving results for the 660 female adoptees unstated. Findings rely on Swedish adoption registries from a specific historical cohort, which may not generalize to other populations or contemporary social settings. The classification relies on specific registry-derived proxies for genetic and environmental risk rather than direct molecular genetic or detailed prospective environmental measurements.

Cited by