Associations between substance use disorders and major depression in parents and late adolescent-emerging adult offspring: an adoption study.

Substance use disorders (SUDs) are highly comorbid and also frequently occur with major depressive disorder (MDD; 1,2). This study examined how SUDs— nicotine dependence(ND), alcohol use disorders(AUDs), and cannabis use disorders(CUDs)—and MDD in parents were associated in a sample of late adolescent-emerging adult offspringand whether these associations differed according to whether the offspringwere adopted or non-adopted. Significant associations between parents and adopted youth would imply that environmental influences are accounting for the similarity. A pattern of stronger associations among non-adopted, compared to adopted, offspringwould imply that genetic influences were contributing to these associations. Because adoptive parents are environmentally but not genetically related to their adopted children, this adoption design allows for direct estimation of shared environmental influences on these disorders. SUDs run in families (8), and comorbidity among SUDs can be partially explained by their occurrence on a broad, heritable externalizing spectrum (3,4). The majority of genetic variance among SUDs is shared (9). Genetic risk factors for substance dependencefall into two highly intercorrelated factors—one for vulnerability to illicit drug dependence and a second for licit drug dependence—and a portion of the genetic influences on ND may be substance-specific (5). Findings regarding genetic and environmental influences on the comorbidity of MDD with SUDs have been less consistent. Support for a shared, heritable factor increasing risk for smoking and MDD is strong (10–15). For alcohol, evidence for shared familial risk is mixed (6,16). For cannabis, the nature and pattern of aggregation of CUDs and MDD within families remains unclear (17, 18). The goal of this study was to examine SUDs and MDD in parents and adopted and non-adopted late adolescent-emerging adult offspring. We addressed three issues: (1) within-disorder transmission (whether associations between disorders in parents and the same disorders in their offspringdiffered for adopted and non-adopted youth); (2) transmission across types of SUDs (the degree to which each SUD in parents was associated with different SUDs among their adopted and non-adopted offspring); and (3) associations across “disorder classes” (the extent to which SUDs in parents predicted MDD in offspring, and the extent to which MDD in parents predicted SUDs among offspring). Three studies of this same sample during early adolescence examined related issues. Tully et al. (19) found that parental MDD was associated with increased risk for MDD among both adopted and non-adopted adolescents (indicating that it was an environmental risk factor), though parental MDD was not associated with increased risk for offspringSUDs. King et al. (20) found that parental alcohol dependence was associated with increased risk for a broad “ disinhibition” construct among non-adopted offspring, but a specific risk for alcohol use among adoptees. Keyes et al. (21) found that parental smoking represented an environmental risk factor for tobacco and marijuana use and a genetic risk factor for substance use among adolescents. The present investigation extended these studies in several ways: (1) by examining the youth during late adolescence and emerging adulthood (with the average age of the sample five years older than in these initial investigations), during a stage when substance use escalates and they are more likely to have experienced the onset of MDD and SUDs; (2) by examining both within-disorder and cross-disorder effects for different SUDs (ND, AUDs, and CUDs) among both parents and offspring; (3) by focusing on SUDs, as opposed to substance use, among offspring; and (4) by examining associations between MDD and each SUD in parents and offspring. When within-disorder associations were examined, we expected to find an environmental effect for parental MDD on MDD in youth (19), though we considered the possibility that effects for non-adopted offspringwould be relatively stronger than those for adoptees. We expected to find significant associations between each parental SUD and that same SUD in nonadopted youth, but reduced associations for adopted youth (consistent with a genetic influence on SUDs; 3,4). Turning to cross-disorder associations, we expected that each SUD would have genetic effects on other SUDs (9). Considering associations between SUDs and MDD, we expected to find genetically-based (i.e., stronger for non-adopted than for adopted adolescents) associations between ND and MDD and between CUDs and MDD (7). Based on the mixed evidence indicating shared familial risk for MDD and AUDs (6,16), we did not predict differential risk for the other disorder among adopted versus non-adopted offspring. Finally, we conducted analyses adjusting for the effects of within-person comorbidity (e.g., when examining parental CUDs and offspringMDD, adjusting for parental MDD and offspringCUDs) in order to statistically isolate the effects of each disorder (e.g., if CUDs and MDD occur together among parents at rates greater than chance, a finding of a parental CUD- offspringMDD association could simply be due to the fact that many parents with CUDs have MDD and MDD in parents and offspringis correlated). We expected that these associations would diminish but remain significant when we adjusted for within-person comorbidity