Parent-offspring similarity for drinking: a longitudinal adoption study

Abstract

Parent-offspring resemblance for drinking was investigated in a sample of 409 adopted and 208 non-adopted families participating in the Sibling Interaction and Behavior Study. Drinking data was available for 1,229 offspring, assessed longitudinally up to three times in the age range from 10 to 28 years. A single drinking index was computed from four items measuring quantity, frequency and density of drinking. As expected, the mean drinking index increased with age, was greater in males as compared to females (although not at the younger ages), but did not vary significantly by adoption status. Parent-offspring correlation in drinking did not vary significantly by either offspring or parent gender but did differ significantly by adoption status. In adopted families, the parent-offspring correlation was statistically significant at all ages but decreased for the oldest age group (age 22-28). In non-adopted families, the parent-offspring correlation was statistically significant at all ages and increased in the oldest age group. Findings imply that genetic influences on drinking behavior increase with age while shared family environment influences decline, especially during the transition from late-adolescence to early adulthood.

Figure 1

The relationship between drinking and age in males and females. Plotted is all observed drinking index data, in which case there are up to three observations for participating offspring but only one observation for participating parents. The curves give the lowess-averaged values with the shaded areas giving the associated 95% confidence intervals. The oldest offspring was assessed at age 28 years; the youngest parent at age 35.

Figure 2

Figure 2a. Estimated parent-offspring correlation as a function of offspring age in adopted and non-adopted families. The error bars demarcate one standard error; the p-values below the age-group label give the significance of the difference between the adopted and non-adopted correlations. Figure 2b. Estimated midparent-offspring correlation as a function of offspring age in adopted and non-adopted families. The error bars demarcate one standard error; the p-values below the age-group label give the significance of the difference between the adopted and non-adopted correlations.

Figure 2

Figure 2a. Estimated parent-offspring correlation as a function of offspring age in adopted and non-adopted families. The error bars demarcate one standard error; the p-values below the age-group label give the significance of the difference between the adopted and non-adopted correlations. Figure 2b. Estimated midparent-offspring correlation as a function of offspring age in adopted and non-adopted families. The error bars demarcate one standard error; the p-values below the age-group label give the significance of the difference between the adopted and non-adopted correlations.