Commentary on Salom et al. (2015): Examining the developmental underpinnings of comorbid mental health and substance use outcomes in young adults.
Commentary on Salom et al. (2015): Examining the developmental underpinnings of comorbid mental health and substance use outcomes in young adults.
复制标题
对 Salom 等人的评论。
DOI:
10.1111/add.12810
复制
发表时间:
2015
期刊:
影响因子:
--
通讯作者:
Skinner,Martie
中科院分区:
文献类型:
--
作者:
Herrenkohl,Todd;Skinner,Martie
Research shows that parents’ misuse and abuse of alcohol is predictive of the same behavior in their adolescent and young adult children [1, 2]. A relationship also exists between parents’ alcohol use and later internalizing symptoms and psychological distress in adolescents [1]. Moreover, findings related to the co-occurrence of substance use and mental health problems among adolescents show that these problems can and often do overlap, and that they are influenced not just by parents’ drinking but also low levels of family support and other related risk factors within the family [3]. Whether there are particularly salient risks for similar co-occurring problems in young adult samples is less well understood, and Salom and colleagues [4] have performed an excellent job of attending to these issues in their two-generation longitudinal study. Results of the study show a relatively strong developmental link between co-occurring (comorbid) mental health and alcohol use problems measured in young adults at age 21 and predictors (measured at age 14) of maternal smoking and adolescent drinking, smoking frequency and attention and thought problems. Low maternal warmth, but not measures of family conflict, also predicted comorbidity at age 21. These findings, and others like them, are critical for advancing prevention and intervention programs because they help to isolate malleable targets for universal and more targeted risk reduction strategies [5]. In discussing their results, Salom and colleagues [4] comment on a study that colleagues and I published earlier on 800 youth followed into their late 20s [6]—a study not unlike theirs, in that we investigate the developmental underpinnings of comorbid outcomes in young adults, with a particular focus on the family. We found that family conflict, measured in late childhood (ages 10–12) and adolescence (13–14 and 15–18), was actually more predictive of comorbid mental health and substance use problems (in young adults at age 27) than were other variables in our analyses, including measures of family involvement and parents’ family management practices. Noting what appears to be a discrepancy in our studies’ findings, Salom and colleagues suggest that family conflict was a significant predictor for us, but not them, because our earlier study focused on fewer risk factors and did not include measures of maternal warmth and adolescent behavior. This is a plausible explanation, yet differences may exist for other reasons. For example, what we call ‘family conflict’is not consistent. In the study by Salom et al., the family conflict label refers to three loosely structured measures—one, a multi-item scale of ‘open family communication’; and two, other single-item (yes/no) mother-report measures of parents’ living arrangement and a history of relationship violence. In our previous work, the same label refers to several youth-report, age-based (scaled) indicators of how family members ‘get along’, whether they talk things out, criticize one another or frequently argue or shout. This is an important distinction not only because the same label is used differently, but also because there is an assumed independence among variables in both studies that more probably reflect a constellation of factors that also co-occur and impinge upon youth development, and adult functioning by extension. Rather than seeing our findings, therefore, as being in contrast, we see them as pointing similarly at the damage inflicted on young people that comes from having experienced unsupportive and hostile relationships with parents and siblings. Upon review of the two studies, it is also apparent that our measures of comorbidity differ. For …