Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Dec 1.
Published in final edited form as: J Res Adolesc. 2014 Jul 14;25(4):638–651. doi: 10.1111/jora.12154

Sibling Facilitation Mediates the Association Between Older and Younger Sibling Alcohol Use in Late Adolescence

Diana R Samek 1, Matt McGue 1, Margaret Keyes 1, William G Iacono 1
PMCID: PMC4667370  NIHMSID: NIHMS603786  PMID: 26640355

Abstract

Previous research has shown adolescent siblings are similar in their alcohol use and that this similarity is largely due to their shared environment. Using a genetically-informed sibling sample (196 full-biological pairs, 384 genetically unrelated pairs), we confirmed that the extent to which older siblings facilitate younger siblings’ alcohol use (i.e., help them get alcohol) was one factor contributing to this shared environmental association. All analyses controlled for parent and peer influences. Findings were not moderated by sibling differences in genetic relatedness, gender, or ethnicity. Proximity in sibling age strengthened these associations, somewhat. Results were especially strong for sibling pairs where the older sibling was of legal drinking age. Implications for prevention and intervention are discussed.

Keywords: Adolescent Alcohol Use, Shared Environment, Sibling Relationships

A large body of literature has shown that adolescent siblings are often similar in their alcohol and substance use (Boyle, Sanford, Szatmari, Merikangas, & Offord, 2001; Fagan & Najman, 2005; Fisher, Miles, Austin, Camargo, & Colditz, 2007; Harakeh, Engels, Vermulst, de Vries, & Scholte, 2007; Rowe & Gulley, 1992) and that this similarity reflects more shared environmental rather than shared genetic influences (Hopfer, Crowley, & Hewitt, 2003; Koopsman & Boomsma, 1996; McGue, Elkins, & Iacono, 2000; Rende, Slomkowski, Lloyd-Richardson, & Niaura, 2005; Rose, Dick, Viken, Pulkkinen, & Kaprio, 2001). For example, working with a twin sample, Koopmans and Boomsma (1996) found that shared environmental effects accounted for 58–88% of the total variance in whether adolescents ever used alcohol, although this study was not designed to identify the specific environmental factors that contribute to this effect. Other research has similarly confirmed substantial shared environmental influences (20–38%) on adolescent alcohol use more broadly defined (e.g., frequency, quantity of drinking; Fowler, Shelton, Lifford, Rice, McBride, Nikolov, et al., 2007; Pagan, Rose, Viken, Pulkkinen, Kaprio, & Dick, 2006; Samek, Keyes, Iacono, & McGue, 2013). Given the high prevalence of adolescent alcohol and substance use (Centers for Disease and Control, 2012), it is important to identify the specific components of the shared environment that are influencing sibling similarity in substance use.

Various environmental pathways have been proposed to explain how siblings influence each other’s alcohol and substance use behaviors. Most tend to focus on aspects of modeling and the potential shared peer network between siblings, but few have incorporated the genetically informed designs necessary to evaluate the hypothesized environmental contributions. For example, research has long demonstrated that the sibling correlation in substance use tends to be consistently greater if siblings are close in age versus further apart (Samek & Rueter, 2011; Trim, Leuthe, & Chassin, 2006) and, to a lesser extent, if they are the same versus opposite gender (Boyle et al., 2001; Rowe & Gulley, 1992). These results suggest that siblings might be more likely to identify with one another or possibly share friends if they are close in age or are the same gender (Bandura, 1969). In fact, there is research that suggests that younger siblings are more likely to report modeling their older sibling’s substance behavior when they share friends with the older sibling (Whiteman, Jensen, & Maggs, 2013). The degree of sibling collusion (defined as observed antisocial and substance use talk between siblings) is also strongly related to younger sibling substance use (Low, Shortt, & Snyder, 2012), and appears to contribute to adolescent delinquent behavior above and beyond the influence of deviant peers (Bullock & Dishion, 2002).

Behavioral genetic research has indicated that when siblings are high in social connection (e.g., share friends, hang out together), sibling similarity in alcohol and substance use is substantially influenced by shared environmental factors. On the other hand, when siblings are low in social connection, their similarity is predominately explained by shared genes (Slomkowski, Rende, Novak, Lloyd-Richardson, & Niaura, Rende et al., 2005). Taken together with other investigations demonstrating substantial shared environmental influences on adolescent substance use overall (e.g., Hopfer et al. 2003), this body of research suggests that processes of modeling and peer-like sibling relationships reflect shared environmental influences on adolescent alcohol and substance use behaviors. Notably, research has shown that siblings tend to share more friends based on their degree of genetic similarity, supporting gene-environment correlation in peer selection (McGuire & Segal, 2013). For example, monozygotic twins (who share 100% of their DNA) have about 82% of their peer group in common, compared to 67% for same-sex dizygotic twins, 39% for same-sex full sibling pairs, and 27% for opposite-sex full-sibling pairs (McGuire & Segal, 2013).

Of course, sibling relationships may be qualitatively different from peer relationships because they occur within a family context and siblings can differ markedly in age. Thus, an older sibling may serve as an attachment figure as well as be one of the first non-adult role models for substance use behavior for their younger siblings. In this vein, Samek and Rueter (2011) demonstrated that in addition to the substantial sibling correlation in adolescent substance use, the more warmth and closeness younger siblings reported they felt toward their older sibling, the less likely they were to use substances three years later. Notably, this effect was not moderated by the older sibling’s substance use, suggesting independent risk and protective effects of the sibling relationship in predicting adolescent substance use. This finding was somewhat inconsistent with previous research demonstrating substantial shared environmental influences on adolescent substance use when siblings are high in social connection (Rende et al., 2005; Slomkowski et al., 2005), perhaps because the reports of warmth and companionship tapped into dimensions of attachment and bonding rather than shared peer networks and collusion in deviant activities.

In addition to modeling and attachment influences, a third environmental pathway linking elder and younger sibling substance use that has long been hypothesized concerns an older sibling’s facilitation of a younger sibling’s substance use (Needle et al., 1986; Rowe & Gulley, 1992; Windle, 2000). This includes the extent to which an older sibling helps the younger sibling obtain alcohol or drugs, and the degree to which an older sibling approves of their use. For example, McGue and Iacono (2009) evaluated sibling facilitation as younger siblings’ perceptions of whether their older siblings would help obtain drugs (tobacco, marijuana, or other drugs) or generally support younger siblings’ drug use (e.g., “My sibling would be upset if I used tobacco”). They found that sibling facilitation of substance use fully mediated the relationship between elder and younger siblings’ report of the number of psychoactive substances used. Because these findings did not differ whether siblings were genetically related or unrelated, the authors concluded that sibling similarity in substance use is due in part to the environmental effect of older siblings’ modeling and encouragement of drug use.

To extend these findings and further evaluate the sibling facilitation hypothesis, we examined whether older sibling facilitation of alcohol use could account for adolescent sibling similarity in adolescent drinking, specifically. We focused on alcohol because it remains the most highly prevalent form of substance use during adolescence (Centers for Disease and Control, 2012), and because adolescent alcohol use is associated with diverse adverse outcomes in adulthood (Duncan, Alpert, Duncan, & Hops, 1997; Flory, Lynam, Milich, Leukefeld, & Clayton, 2004; Hicks, Iacono, & McGue, 2010). A second novel contribution of this study constituted an evaluation of the correspondence between the older siblings’ report of their alcohol use and younger siblings’ perceptions of their older siblings’ alcohol use in relation to younger siblings’ alcohol use. Research has shown that people tend to estimate the behaviors of others based on their own behaviors (Borsari & Carey, 2001; 2003; Iannoti & Bush, 1992; Lewis, Litt, Blayney, Lostutter, Granato, Kilmer, et al., 2011; Martens, Page, Mowry, Damann, Taylor, & Cimini, 2010), such that those who drink heavily tend to believe those around them drink heavily (even if they do not). Much of the literature on this topic relates to perceptions of peers’ drinking behaviors and social norms in college-aged samples (Borsari & Carey, 2001; 2003), but less research has examined perceptions of sibling drinking (D’Amico & Fromme, 1997).

Based on this review, we hypothesized that compared to the older sibling’s report of alcohol use, younger sibling’s perceptions of their older siblings’ alcohol use would relate more strongly to both younger sibling’s perception of sibling facilitation and younger sibling’s report of alcohol use. Distinguishing the predictive ability of older sibling’s report of alcohol use versus younger sibling’s perception of older sibling’s alcohol use in relation to younger sibling’s alcohol use might implicate which characteristics of the sibling relationship are particularly important to assess in adolescent alcohol prevention programs. Specifically, simply assessing the younger sibling’s perception of older sibling’s alcohol use and willingness of their older sibling to facilitate their alcohol use may prove to be an adequate, if not more appropriate, measure than older sibling’s direct report of alcohol use.

A third way we are extending prior research on the topic of sibling facilitation of substance use is by analyzing moderating influences of several contextual variables. In addition to comparing results across genetically related and unrelated pairs to confirm the shared environmental influence that has been demonstrated previously (e.g., Hopfer et al., 2003; McGue & Iacono, 2009), we also examined moderators such as sibling differences in gender, age, and ethnicity (our study contained adoptive families and siblings in these families were not always of the same ethnicity, making such a comparison possible). We expected effects to be stronger when siblings were the same sex, close in age, or shared the same ethnicity because they would be more likely to identify with one another under these conditions (consistent with expectations from a social learning perspective; Bandura, 1969).

Figure 1 depicts the conceptual model on which our hypotheses are based. As prior research has demonstrated important parent and peer influences on adolescent alcohol use (Needle et al., 1986; Rowe & Gulley, 1992; Windle, 2000), we include mother and father alcohol use, relationship quality, and a measure of peer deviance as covariates in our evaluation of whether sibling facilitation of alcohol use mediates the link between elder and younger sibling alcohol use. We expected that elder sibling facilitation of alcohol use would fully mediate the association between elder and younger sibling reported alcohol use (paths c and e), but partially mediate the association between younger siblings’ perceptions of elder siblings’ alcohol use and younger siblings’ alcohol use (paths d and e). To determine whether these associations varied as a result of important sibling contexts, we tested for moderation by sibling pair genetic similarity (genetically related versus unrelated), age difference (within 2 years, 2–3 years, 3+ years), gender similarity (same versus mixed), and ethnic similarity (same versus mixed).

Figure 1. Conceptual Model Depicting Study Hypotheses.

Figure 1

Open in a new tab

RQ = Relationship Quality. Our goal was to evaluate a full mediation model of perceived sibling facilitation in the association between elder and younger sibling alcohol use (path a) and in the association between perceived elder sibling alcohol use and younger sibling alcohol use (path b), after controlling for parenting and peer influences (shown in gray for clarity of presentation). Direct associations are shown for paths a and b. Indirect associations are shown by paths c and e, and d and e. Significant coefficients for both direct and indirect effects suggest partial mediation. Significant indirect, but not direct coefficients, suggests complete mediation. To test for contextual influences, follow-up analyses examined moderation of all paths as a function of genetic relatedness (genetically related versus unrelated), sibling age difference (within 2 years, 2–3 years, 3+ years), sibling gender composition (same versus mixed), and shared ethnicity (same versus mixed).

Method

The Sibling Interaction and Behavior Study (SIBS) was designed to evaluate shared environmental influences on adolescent substance use and related psychopathology (McGue, Keyes, Sharma, Elkins, Legrand, Johnson, et al., 2007). Two types of families were recruited: families with genetically-unrelated, adopted siblings and families with full biological siblings. Adoptive families were recruited through three large Minnesota (MN) adoption agencies. Eligibility included having an adopted child between the ages of 11 and 21 years at the time of the intake assessment who had been permanently placed into the adoptive home prior to age 2 (Mage = 4.7 months, SD = 3.4 months, 96% were placed prior to 1 year of age), and a second adolescent in the home who was not biologically related to the adopted adolescent. Notably, the majority of adoptees were internationally adopted (74%); international adoptees were predominantly female (61%) and of Asian ancestry (90%); domestic adoptees were predominately male (59%) and of European ancestry (79%). Families with biologically related children were ascertained through publicly available MN birth certificates and were recruited to match the adoptive sample in terms of age and sex but not ethnicity (matching MN demographics in the relevant birth years, 95% of the biologically-related children were of European ancestry). For all families, sibling pairs were required to be no more than 5 years apart in age. Participation rates were not significantly different between non-adoptive (57%) and adoptive (63%) families. Comparisons of non-adoptive parents’ education and marital status to 2000 census data show that they were generally representative of MN families with two children. A detailed overview of study recruitment and participation has been provided elsewhere; please see McGue et al. (2007) for further information.

Data from the first follow-up assessment, conducted 3.5 years after intake, was used for this study. This was due, in part, to the availability of the sibling facilitation measure (discussed below) at this assessment, and because alcohol use is more prevalent in late compared to early adolescence (CDC, 2012). Possible participants included 1,158 adolescents (94% retention, 563 complete pairs, 15 elder sibling singletons, and 17 younger sibling singletons) that completed the follow-up assessment. The sample size used for path analysis was N = 580 sibling pairs (missing values analysis described in detail in the Analysis Plan, below). This included 196 genetically related and 384 genetically unrelated pairs (where at least one member of the pair was adopted), 352 same-sex and 228 mixed-sex pairs, and 464 same-ethnicity and 116 mixed-ethnicity pairs. Same-sex pairs consisted of slightly more sisters (57%) than brothers (42%). While the sample contained ethnic diversity, it was primarily limited to European and Asian ethnic ancestry only. Specifically, the same-ethnicity pairs were predominately of European (56%) and Asian (36%) ancestry. Of the mixed ethnicity sibling pairs, the majority (77%) consisted of pairs where one sibling was of European ancestry and the other sibling was of Asian ancestry. On average, younger siblings were 17.0 years old (SD = 1.8) and elder siblings were 19.4 years old (SD = 1.7). The average age difference in sibling pairs was 2.4 years (SD = .88).

Procedure

Participating family members visited the lab for a day-long assessment. All provided written informed assent or consent, including parental consent for minor participation, as appropriate. A variety of measures were completed, including computerized assessments and surveys. Approximately, 14% of participants could not visit the lab and were instead given a shortened assessment via telephone. These participants did not complete the Computerized Substance Use Assessment (described below) and had missing alcohol use data (missing values analysis described in detail in the Analysis Plan, below). Chi-square analyses revealed no statistically significant differences between those with and without computerized data across key demographic variables (adoptive status, sex, ethnicity) or alcohol abuse or dependence diagnoses at the Intake assessment.

Measures

Adolescent alcohol use (last 12 months)

Adolescent alcohol use was assessed using the Computerized Substance Use Assessment. In a private room, adolescents answered a series of questions relating to alcohol and substance use via the computer. This study’s measure of alcohol pertained to the last 12 months only. Items included frequency of alcohol use, frequency of intoxications (both answered on a scale of 0 = Never to 9 = Every day or nearly every day), maximum number of drinks on a single occasion (answered on a scale of 0 = 0 drinks to 10 = 10+ drinks), and the proportion of times they were intoxicated when they drank (answered on a scale of 0 = Never to 5 = Nearly every time or every time). These 4 items were standardized and summed to create a composite score of elder sibling alcohol use (α = .96) and younger sibling alcohol use (α = .97).

Perceived older sibling alcohol use

Younger sibling perception of older sibling alcohol use was assessed using younger sibling’s report on the Sibling Relationship Supplement. The 3-item scale (α = .89) included: “Has your sibling ever used alcohol without your parent’s permission?” answered on a scale of 1 (My sibling has never tried alcohol) to 4 (My sibling has tried alcohol and used it a lot in the past 12 months), “ In the past 12 months, how often has your sibling used alcohol?”, answered on a scale of 1 (Every day or nearly every day) to 5 (My sibling has not used alcohol in the past 12 months), and “In the past 12 months, how many times have you seen your sibling use alcohol?”, answered on a scale of 1 (Never) to 3 (Many times). These 3 items were coded so a higher score indicated more alcohol use, then were standardized and summed to create a composite score of perceived older sibling alcohol use.

Perceived sibling facilitation of alcohol use

Younger sibling report of elder sibling facilitation of alcohol use was assessed using the younger sibling’s report on the Sibling Relationship Supplement. Three items were used: “My sibling would help me get alcohol if I wanted some,” “My sibling’s friends would help me get alcohol if I wanted some,” both answered on a scale of 1 (Strongly Agree) to 4 (Strongly Disagree) and “Has your sibling ever given you alcohol,” answered on a scale of 1 = No, 2 = Yes, but just a few times, 3 = Yes, many times. These three items were coded so that a higher scored indicated greater facilitation, then were standardized and summed to create a composite score of perceived sibling facilitation of alcohol use, α = .84.

Parent alcohol use (last 12 months)

Parent report of alcohol use was measured using the Parent Substance Use Screen, which assesses tobacco, alcohol, and illicit substance use behavior for the last 12 months. Only one parent visited at the follow-up assessment (usually the mother); however, the visiting parent reported on the non-visiting parent’s substance use behavior in a separate interview. Thus, mother and father alcohol use was assessed using the frequency of alcohol in the last 12 months (answered on a scale of 0 = Never to 10 = 3 or more times a day) as reported by the visiting parent.

Parent-child relationship quality

Younger sibling’s report of parent-child relationship quality was measured using the Parental Environment Questionnaire (PEQ; Elkins, McGue, Iacono, & Tellegen, 1997). Adolescents rated statements concerning the parent-child relationship (e.g., “I feel close to my [Mother/Father”) on a scale of 1 (Definitely true) to 4 (Definitely false). The PEQ assess major dimensions of parent-child relationship quality, including conflict, involvement, and positive regard (α’s ranged from .80 to .89; assessed using 8–12 items each). The average across the PEQ scales was computed to assess mother-child and father-child relationship quality, coded so that the higher the score, the better the parent-child relationship quality.

Deviant peer affiliation

Younger sibling’s report of affiliation with substance-using peers was measured using the Friends Survey (Walden, McGue, Iacono, Burt, & Elkins, 2004; Samek et al., 2013). Adolescents rated friendships on a scale of 1 (All my friends are like that) to 4 (None of my friends are like that). An 8-item scale was used to assess deviant peer affiliation, e.g., “My friends drink alcohol or beer” (α = .88.) Items were coded so that a higher score indicated a greater level of deviant peer affiliation.

Analysis Plan

Figure 1 describes the path analytic model, which was analyzed using Mplus, 6.12 (Muthén & Muthén, 1998–2013). As SIBS is diverse in terms of demographics (described above), the effects of adolescent sex (1 = Male, 2= Female), age, ethnicity (1 = White, 0 =African-, Asian-, Hispanic- American, Mixed, Other), and family SES (standardized composite of mother’s and father’s education level, highest occupational prestige rating, and family income) were regressed out of all composite scales prior to analysis. We used the default maximum likelihood (ML) estimator for all analyses. After demonstrating a significant association between the independent variables (elder sibling alcohol use, perceived elder sibling alcohol use) and the mediator (perceived sibling facilitation of alcohol use) in relation to the dependent variable (younger sibling alcohol use), we tested the full mediation model. As illustrated in Figure 1, we tested for the extent to which the mediator (perceived sibling facilitation of alcohol) carried the influence of the independent variable to the dependent variable by using MODEL INDIRECT statements (this provides an estimate of indirect effect size by taking the product of mediating paths, e.g., paths c and e in or d and e in Figure 1; see Preacher & Kelly, 2011; Hayes, 2009). Significant coefficients for both direct and indirect effects suggest partial mediation. Significant indirect but not direct coefficients suggest complete mediation.

Missing data were handled using full information maximum likelihood for all analyses. The proportion of data present in all possible pairs of variables (i.e., covariance coverage) ranged from 65% to 95%. Valid data on at least one of the nine variables (see Figure 1) was necessary for inclusion in the path analysis; n =33 of the eligible sample of 613 pairs (5.4%) were missing on all nine variables and were therefore excluded from the analysis. Thus, the final sample size consisted of 580 sibling pairs.

Finally, we tested for moderation of the full model across several contextual variables using multiple group modeling. Siblings were either genetically related (n = 196) or genetically unrelated (n = 384). Sibling age difference ranged from 1 to 5 years. Due to small numbers of siblings pairs less than 1 year of age apart or more than 3 years of age apart), sibling age difference was defined as those within 2 years of age (n = 229), 2–3 years of age (n = 227) and 3+ years of age (n = 124). Those classified by gender composition were either same sex (n = 352) or mixed sex (n = 228). Ethnic similarity was classified by same ethnicity (n = 464) or mixed ethnicity (n = 116).

In multiple group modeling, all paths were allowed to vary across the different levels of the moderator (e.g., genetically related siblings versus genetically unrelated siblings). Then, all paths were constrained across groups. Differences in the total variant and invariant models were analyzed using the chi-square difference test. If a significant difference was found, paths were constrained one at a time to determine which path was significantly different across groups.

Results

Table 1 describes the overall levels of alcohol use for both elder and younger siblings. In general, the sample was comparable to national statistics on adolescent alcohol use. For example, about 48% of youth (through grade 12) ever used alcohol in 2013 (Johnston, O’Malley, Miech, Bachman, & Shulenberg, 2014). This is generally comparable to the 54% of younger siblings (M age = 17.0, SD = 1.8). As one would expect given their older age (M age = 19.4, SD = 1.7), 76% of elder siblings reported ever using alcohol in this sample. Follow this, 32% of youth (through grade 12) have been intoxicated in 2013 (Johnston et al., 2014), which is comparable to the 39% of younger siblings and 65% of elder siblings that have reported ever being intoxicated in this sample.

Table 1.

Means (Standard Deviations) on Elder and Younger Sibling Alcohol Use Measures

Range Elder Sibling Younger
Sibling		 χ2 on 1 df p-value
Ever used alcohol 0 – 1 .76 (.43) .54 (.50) 62.41 <.001
Frequency of alcohol usea 0 – 9 3.32 (2.51) 1.89 (2.22) 97.63 <.001
Ever intoxicated 0 – 1 .65 (.48) .39 (.49) 71.26 <.001
Frequency of intoxicationsa 0 – 9 2.49 (2.36) 1.35 (2.04) 72.79 <.001
How often drink to get drunkb 0 – 5 2.41 (2.05) 1.40 (1.95) 68.89 <.001
Max drinks in one sittingc 0 – 10 5.05 (3.78) 3.13 (3.76) 74.43 <.001
Open in a new tab

NOTE:

a

Frequency items answered on a scale of 0 = Never to 9 = Every day or nearly every day,

b

how often they drank to get drunk answered on a scale of 0 = Never to 5 = Nearly every time or every time, and

c

max drinks on a single occasion answered on a scale of 0 = 0 drinks to 10 = 10+ drinks. Significant differences between elder and younger siblings were tested by constraining the parameters to be equal across groups and utilizing the chi-square difference test (χ2) on 1 degree of freedom (df) change to compare the chi-square value from the free and constrained models.

Correlations among age, sex, ethnicity, and SES-corrected study variables relevant to the hypothesized conceptual model (Figure 1) are shown in Table 2. All independent variables were significantly correlated with the dependent variable, younger sibling alcohol use (see bottom row of table, all p’s < .01). Additionally, all independent variables were significantly correlated with each other, with the exception of parent alcohol use and parent-child relationship quality variables and elder sibling alcohol use and father-child relationship quality (see Table 2 for details).

Table 2.

Correlations Among the Composite Variables Modeled in Figure 1

1 2 3 4 5 6 7 8 9
1. Mother Alcohol Use --
2. Father Alcohol Use .45** --
3. Mother-Child Relationship Quality −.01 −.04 --
4. Father-Child Relationship Quality −.02 −.05 .71** --
5. Peer Deviance .14** .13** −.39** −.35** --
6. Elder Sibling Alcohol Use .10* .15** −.10* −.07 .24** --
7. Perceived Elder Sibling Alcohol Use .18** .18** −.21** −.17** .38** .59** --
8. Perceived Sibling Facilitation of Alcohol .14** .11* −.19** −.15** .42** .46** .57** --
9. Younger Sibling Alcohol Use .14** .18** −.26** −.21** .61** .29** .45** .50** --
Open in a new tab

NOTE: All variables were adjusted for age, sex, ethnicity, and SES prior to analysis. All variables were standardized prior to analysis and adjusted for age, sex, SES, and ethnicity. For all variables, means = .00 and standard deviations = 1.0. Significance is denoted by

*

p < .01,

**

p < .05.

Before evaluating the hypothesized mediation model, we first evaluated the significance of the direct associations among study variables. As expected given correlations presented in Table 2, elder sibling’s report of alcohol use significantly predicted younger sibling’s report of alcohol use (β = .27, t(456) = 6.34, p < .001), as did younger sibling’s perceptions of elder sibling alcohol use (β = .45, t(402) = 11.23, p < .001), and perceived sibling facilitation (β = .50, t(495) = 15.73, p < .001). With evidence of direct associations, we went on to test the hypothesized mediation model (Figure 1) using path analysis.

Mediation results

Our central question was whether perceived sibling facilitation of alcohol use mediated the direct association between a) elder and younger sibling alcohol use and b) younger sibling’s perceptions of elder sibling’s alcohol use and younger sibling’s alcohol use. Results are shown in Figure 2. Evidence of significant indirect but not direct effects of elder sibling alcohol use supported our expectations and showed complete mediation between elder and younger sibling’s alcohol use by perceived sibling facilitation (indirect effect size: β = .05, t(580) = 3.16, p = .002). As evidenced by both the significant direct and indirect effects of younger sibling’s perceptions of elder sibling’s alcohol use, there was partial mediation between younger siblings’ perceptions of elder sibling alcohol and their own alcohol use by perceived sibling facilitation of alcohol use (indirect effect size: β = .11, t(580) = 4.76, p < .001).

Figure 2. Sibling Facilitation of Adolescent Alcohol use: Path Analysis Results (N= 580).

Figure 2

Open in a new tab

Standardized coefficients are shown for each path Significance is denoted by ** p < .01, * p < .05. For ease of presentation, standardized coefficients are not shown for correlations among independent variables that are non-significant from zero and all non-significant paths are dashed. Proportion of variance explained (R2) are shown for all dependent variables. In total, results suggest that the association between elder and younger sibling alcohol use was completely mediated by perceived sibling facilitation of alcohol use, and the association between perceived elder sibling use and younger sibling alcohol use was partially mediated by this variable. Fit statistics and indirect effect sizes are reported in text.

Notably, 46% of the total variance in younger sibling alcohol use was explained the combination of parent, peer, and sibling factors tested. The model fit adequately by SRMR (.03), CFI (.93), and (TLI .80) criteria, but somewhat poorly by RMSEA criteria (.11; 90% Confidence Interval: .08, .15); χ2 (5, N = 580) = 41.28, p < .001. As this first analysis demonstrated significant sibling influences after controlling for parent and peer influences, we conducted a separate more parsimonious analysis which excluded all parenting and peer covariates, as well as the non-significant direct path of elder sibling alcohol use on younger sibling alcohol use. This more parsimonious model showed excellent fit by all fit criteria: χ2 (1, N =580) = .76, p = .38, RMSEA = .00; 90% CI: .00, .11, CFI = 1.0, TLI = 1.0, SRMR = .01. In this parsimonious model, 28% of the variance in younger sibling alcohol use was explained by factors related to elder sibling alcohol use, perceived elder sibling alcohol use, and perceived facilitation, alone. Altogether, results from the full mediation model shown in Figure 2 and this more parsimonious model suggest substantial influence of sibling facilitation in the relationship between elder and younger sibling alcohol use.

Of note, there were 12 cases where the younger sibling was 21 or older at the time of data collection; excluding these cases did not impact study results (i.e., there was evidence of complete mediation by sibling facilitation for path a, as described in Figure 1, but not for path b – for both the full mediation model described in Figure 1 and the parsimonious model that excluded parenting and peer covariates). There were 97 cases where the elder sibling was 21 or older; excluding these cases also did not impact study results (by the same criteria reviewed above).We also ran an analysis only including the 97 pairs where the elder sibling was 21 or older and legally able to buy alcohol. The full mediation model included a sample size of 96 (as 1 case was missing on all 9 variables). Results showed a different pattern; here there was evidence of complete mediation by sibling facilitation for path a, and path b (as described in Figure 1) for both: 1) the full mediation model shown in Figure 2; and 2) the more parsimonious model that excluded parenting and peer covariates (described above). These results suggest that sibling facilitation of alcohol use may be particularly influential if the elder sibling is of legal drinking age. Because of this finding, we added whether or not elder sibling was of legal drinking age to our list of moderators to evaluate, described next.

Moderation results

We evaluated whether paths were moderated by several sibling contextual variables. Some zero-order correlations were stronger in magnitude among some context groups compared to others, but few were significantly different from one another at p < .05 (see Table 3). For example, the correlation between elder and younger sibling alcohol use was stronger for genetically related siblings (r = .37) than unrelated siblings (r =.22), but not significantly so; χ2 (1, N = 516) = 3.10, p = .08. The only significant differences that were found were for sibling age difference (the correlations between siblings 3+ years in age apart tended to be lower than siblings within 2 years of age or 2–3 years in age) and elder sibling legal drinking age (the correlation between elder sibling report of alcohol use and younger sibling’s perceptions of elder sibling alcohol use was less correlated for pairs where the sibling was 21 years of age or older compared to pairs where the sibling was less than 21 years of age; no other significant differences were found). Please see Table 3 for details.

Table 3.

Zero-order Correlations (95% Confidence Intervals) for all Pairs of Composite Variables Across Sibling Contextual Groups

Elder Sibling
Alcohol Use
with Younger
Sibling Alcohol
Use		 Perceived Elder
Sibling Alcohol
use with
Younger Sibling
Alcohol Use		 Elder Sibling
Alcohol Use
With Perceived
Sibling
Facilitation of
Alcohol Use		 Perceived Elder
Sibling Alcohol
Use with
Perceived
Sibling
Facilitation of
Alcohol Use		 Perceived
Sibling
Facilitation of
Alcohol Use
with Younger
Sibling Alcohol
Use		 Elder Sibling
Alcohol Use
with Perceived
Elder Sibling
Alcohol Use
Genetic Similarity
  Related .37 (.24, .50) .42 (.29, .56) .53 (.42, .64) .60 (.50, .70) .56 (.46, .67) .68 (.59, .77)
  Unrelated .22 (.11, .33) .45 (.36, .55) .41 (.31, .51) .56 (.48, .63) .47 (.39, .55) .56 (.47, .64)
Age Difference
  Within 2 years .31 (.18, .44)a .42 (.30, .54) .55 (.45, .65)a .54 (.44, .64) .54 (.44, .64)a .65 (.57, .74)
  Within 2–3 years .34 (.21, .48)a .53 (.42, .64)a .47 (.35, .59)a .64 (.56, .73)a .52 (.41, .62)a .60 (.50, .70)
  3+ Years apart .05 (-.14, .25)b .30 (.12, .49)b .23 (.05, .41)b .51 (.37, .65)b .41 (.26, .56)b .50 (.34, .66)
Gender Similarity
  Same gender .35 (.24, .45) .48 (.38, .57) .48 (.39, .57) .59 (.52, .67) .57 (.49, .65) .61 (.53, .69)
  Mixed gender .17 (.03, .30) .38 (.26, .51) .41 (.29, .53) .55 (.45, .64) .40 (.28, .52) .58 (.48, .69)
Ethnic Similarity
  Same ethnicity .29 (.19, .38) .42 (.33, .51) .43 (.35, .52) .58 (.52, .64) .51 (.44, .58) .60 (.53, .67)
  Mixed ethnicity .23 (.04, .42) .52 (.36, .67) .53 (.38, .68) .53 (.39, .67) .49 (.34, .63) .60 (.46, .74)
Elder Sibling Legal Drinking Age
  Elder sibling is ≥ 21 years .29 (.06, .52) .37 (.17, .57) .44 (.25, .63) .52 (.36, .67) .51 (.34, .67) .38 (.17, .66)a
  Elder sibling is < 21 years .28 (.19, .67) .45 (.37, .54) .48 (.40, .55) .60 (.53, .66) .50 (.42, .67) .64 (.58, .70)b
Open in a new tab

NOTE: All variables were adjusted for age, sex, ethnicity, and SES prior to analysis. Correlations were significant at p < .05 if the 95% confidence did not overlap zero. Significant differences in correlations across groups (calculated by the chi-square difference test on 1 df, p < .05) are indicated if the correlations have different subscripts.

Next, we tested whether the entire path analysis model (Figure 2) was significantly moderated by these sibling contextual variables. Following expectations given the correlations reported in Table 3, there was no significant moderation of the model by genetic relatedness (related versus not: χ2 (6, N = 580) = 7.48, p = .28), gender composition (same versus mixed: χ2 (6, N = 580) = 10.24, p = .12), ethnic similarity (same versus mixed: χ2 (6, N = 580) = 4.33, p = .63), or elder sibling legal drinking age (χ2 (6; N = 564) = 11.58, p = .07).

The only significant differences involved sibling age difference (χ2 (12, N = 580) = 26.64, p = .009). Follow-up analyses that constrained one path at a time showed the only significant difference in paths was for the association between elder sibling’s report of alcohol use and perceived sibling facilitation of alcohol, which was stronger for those within 2 years of age (β = .35, t(229) = 4.87, p < .001) compared to those 3+ years of age a part (β = .00, t(124)= .03, p = .98; χ2 (1; N = 580) = 8.95, p = .003). No other significant differences were found. In total, the indirect effect size of sibling facilitation as a mediator on the relationship between elder sibling’s alcohol use and younger sibling’s alcohol use was stronger for siblings within 2 years (β = .13, t(229) = 3.40, p < .001) compared to siblings 2–3 years of age apart (β = .03, t(227) = 1.51, p = .13), and siblings 3 or more years of age apart (β = .00, t(124) = .15, p =.88).

Based on the pattern of correlations shown in Table 3, we also evaluated for differences across age groups of siblings less than 3 years of age apart (n = 456) versus 3 or more years of age apart (n = 124). The same pattern of results was found, where the only significant difference across groups was for the association between elder sibling’s report of alcohol use and perceived sibling facilitation, which was stronger for siblings within 3 years of age (β = .25, t(456) = 4.66, p < .001) compared to siblings 3 or more years of age apart (β = .02, t(124) = .15, p = .88). The indirect effect size of sibling facilitation as a mediator on the relationship between elder sibling’s alcohol use and younger sibling’s alcohol use was stronger for siblings within 3 years (β = .06, t(456) = 3.67, p = .001) compared to siblings 3 or more years of age apart (β = .00, t(124) = .15, p = .89). In total, results suggest fairly consistent evidence that associations hold up across a variety of sibling contexts, perhaps more so for siblings within 2 years of age compared to siblings further apart in age.

Discussion

Previous research has consistently demonstrated that adolescent siblings are often similar in their report of alcohol and substance use (Fagan & Najman, 2005; Fisher et al., 2007; Rowe & Gulley, 1992), primarily due to shared environmental rather than shared genetic factors (Hopfer et al., 2003; Koopsman & Boomsma, 1996; McGue et al., 2000; Rende et al., 2005). This study provided evidence that perceived sibling facilitation of alcohol use is one specific shared environmental factor linking older and younger sibling alcohol use in late adolescence. Mediation of perceived sibling facilitation of alcohol use in adolescent sibling pairs was consistent across a variety of sibling contexts, including whether siblings were genetically related or unrelated (confirming shared environmental influence), the same or opposite gender, and the same or of different ethnicity. Consistent with expectations from previous literature (Samek & Rueter, 2011; Trim et al., 2006), there was some evidence the associations were stronger for siblings closer in age compared to those further apart in age. All of these findings were significant above and beyond the effect of parent and peer influences, suggesting substantial older sibling influence on late adolescent alcohol use. In total, this study’s findings extend previous research by showing that one possible route through which an older sibling influences the drinking of a younger sibling is through facilitation of alcohol use by the older to the younger sibling. While the data are cross-sectional and longitudinal investigation is needed to support the timing of these effects, this study provides initial support for the notion that older siblings may increase or reduce the likelihood of younger sibling’s teen drinking based on the extent to which they facilitate alcohol use directly.

We hypothesized, and our results supported, two pathways by which elder sibling alcohol use might influence younger sibling alcohol use. First, results showed full mediation of sibling facilitation in the association between elder sibling’s reported alcohol use and younger sibling’s reported alcohol use. This was particularly salient for siblings within 2 years of age compared to siblings further apart in age. These results are consistent with the notion that older siblings who drink appear to be more likely than those who do not drink to create an environment that encourages and facilitates younger sibling alcohol use, and that younger siblings may be more likely to model their older siblings drinking behavior, particularly when siblings are close in age.

Second, we found younger sibling’s perceptions of elder sibling’s alcohol use was a significant predictor of their own alcohol use even after accounting for sibling facilitation of alcohol use (i.e., evidence of partial but not complete mediation in this pathway). This is likely because perceptions are more relevant to adolescent alcohol use than the actual alcohol use behavior of their older sibling, as has been demonstrated in prior research on peers (Borsari & Carey, 2001; 2003) and siblings (D’Amico & Fromme, 1997).

On the other hand, when evaluating sibling pairs where the older sibling was of legal drinking age, results showed complete mediation of both pathways (actual report and perception), although results were not significantly different across pairs where the elder sibling was 21 years of age or older compared to less than 21 years. While caution should be warranted in generalizing these finding due to a small sub-sample comparison of pairs where the elder sibling was 21 years of age or older (n = 96), this indicates that sibling facilitation of alcohol use is relevant for sibling pairs whether the older sibling can legally buy alcohol or not.

Overall, results suggest that sibling facilitation of alcohol use is an important environmental influence on adolescent alcohol use. It may be important to incorporate an assessment of sibling influences when working within a substance use prevention or intervention context. Notably, characteristics of the parent-child, peer, and school environments have been widely-assessed within the adolescent substance use prevention and intervention context (for a review, see Griffin & Botvin, 2010). There has been surprisingly little research aimed at using siblings to reduce adolescent alcohol and substance use. In fact, a review of the literature on sibling-based prevention and intervention research suggests that these strategies have been primarily focused on siblings in middle childhood in order to reduce sibling conflict (Kennedy & Kramer, 2008; Kramer & Radey, 1997), have involved educating and providing support for siblings of those with a disability (Lobato & Kao, 2005; Williams, Hanson, Karlin, Ridder, Liebergen, Olson, et al., 1997; Williams, Graff, Hanson, Stanton, Hafeman, Liebergen, et al., 2003), or have targeted siblings to assist in the development of pro-social skills for those with autism (Celiberti & Harris, 1993; Tsao & Odom, 2006).

One exception was a study conducted by Donovan and Levin (2011), in which researchers recruited adolescent siblings of those in substance use treatment. Only those with ADHD were recruited, and the treatment was ADHD medication. While the sample size was too small to adequately model long-term outcomes, this study demonstrated the ability to recruit and retain such a sample. At the very least, results from our study suggest that when treating adolescents with alcohol use problems, a good first step would be to assess the extent to which younger siblings think their older siblings drink, as well as the degree which siblings may be facilitating or modeling substance use (either to the target adolescent, or from the target adolescent to their other siblings).

Notably, Feinberg, Solmeyer, and McHale (2012) introduced the Siblings are Special (SAS) universal prevention program that targets 5th graders and their younger sibling’s with the long-term goal of reducing adolescent substance use and negative adjustment outcomes. Groups of siblings meet for 1.5 hours after school over the course of 12 weeks, where teamwork, emotional regulation, and problem-solving skills are encouraged and practiced in order to promote positive sibling relationship quality. This prosocial sibling relationship dynamic is expected to impact additional prosocial peer relationships and a reduced likelihood of adolescent substance use. It will be important to note the immediate and long-term outcomes of SAS as results become available.

Limitations and Future Research

While this study showed strong associations between perceived sibling facilitation and adolescent alcohol use, we examined these cross-sectional associations in late adolescence; it remains to be determined whether the magnitude of sibling facilitation is just as crucial in early adolescence and how these sibling dynamics may relate to alcohol use in later adulthood. Also, while we adjusted for possible effects of ethnicity and SES in our analyses, children in this study come from families with predominately white, married parents, and adoptive parents had relatively high socio-economic status. It is unclear how effects might be replicated in different family cultures and structures. However, previous research has demonstrated substantial sibling effects in low SES, single-parent, Latino, and African American families (East & Khoo, 2005), and no difference in sibling factors in predicting substance use across African American and white youth (Williams, Ayers, Abbott, Hawkins, & Catalano, 1999).

It is important to note that the majority of adopted adolescents in this sample were of Asian ancestry. Those of Asian ancestry are more likely to have a version of the ALDH2 gene that codes for a deficient form of aldehyde dehydrogenase, important to metabolizing the effects of alcohol. A previous SIBS analysis has confirmed those with this ALDH2 mutation are less likely to report problematic alcohol use, but not other substance use problems (Irons, Iacono, Oetting, & McGue, 2012; Irons, McGue, Iacono, & Oetting, 2007). It remains unclear if the present study’s results would be replicated across sibling pairs, for example, where the older and younger sibling differed on the presence of the ALDH2 mutation. However, prior SIBS research (Irons, McGue, Iacono, & Oetting, 2007) has demonstrated that sibling (but not parent) past year drinking predicted past year drinking among youth with the ALDH2 mutation, suggesting the protective effect of the ALDH2 gene is offset by sibling drinking behavior. Thus, we might expect that the present study’s results would not be affected by sibling differences in the ALDH2 gene. Nonetheless, future studies are necessary to specifically test this hypothesis.

In addition, we analyzed sibling facilitation of alcohol use in terms of younger siblings’ perceptions of older sibling support of their alcohol use and whether their older sibling had ever given them alcohol; using the same reporter has the potential for introducing method error as it inflates associations. It would be useful for future research to obtain the older sibling’s report to evaluate potential differences or similarities in these reports. It would also be useful to understand how siblings may facilitate other forms of specific substance use, such as nonprescription drug misuse or marijuana use.

Conclusion

Using a genetically-informed sibling sample in late adolescence, this study demonstrated at least one possible mechanism for adolescent sibling similarity in alcohol use: older siblings may be facilitating alcohol use to their younger sibling. We found younger sibling’s perception of older sibling’s alcohol use was a stronger predictor of their alcohol use in comparison to older sibling’s report of alcohol use, particularly for siblings within 2 years of age. Results were consistent across a variety of sibling contexts and remained significant after controlling for parenting and peer influences. In total, results indicate sibling facilitation of alcohol use is an important shared environmental influence in the development of adolescent alcohol use. Findings suggest alcohol preventions and interventions might include an assessment of sibling influences on adolescent alcohol use (e.g., the extent younger siblings perceive their older siblings use alcohol or would get it for them) to better understand and potentially treat or reduce the likelihood of adolescent alcohol use and misuse.

Acknowledgements

This research was supported by Grant AA11886 from the National Institute on Alcohol Abuse and Alcoholism, Grant MH66140 from the National Institute of Mental Health. Diana R. Samek was also supported by Grant MH017069 from the National Institute of Mental Health postdoctoral training grant. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

References

  1. Bandura A. Social-learning theory of identificatory processes. In: Goslin DA, editor. the handbook of socialization theory and research. Chicago, IL: Rand McNally & Company; 1969. pp. 213–261. [Google Scholar]
  2. Borsari B, Carey KB. Peer influences on college drinking: A review of the research. Journal of Substance Abuse. 2001;13:391–424. doi: 10.1016/s0899-3289(01)00098-0. [DOI] [PubMed] [Google Scholar]
  3. Borsari B, Carey KB. Descriptive and injunctive norms in college drinking: A meta-analytic integration. Journal of Studies on Alcohol. 2003;64:331–341. doi: 10.15288/jsa.2003.64.331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Boyle MH, Sanford M, Szatmari P, Merikangas K, Offord DR. Familial influences on substance use by adolescents and young adults. Canadian Journal of Public Health. 2001;92:206–209. doi: 10.1007/BF03404307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Buchanan JP, McGue M, Keyes M, Iacono WG. Are there shared environmental influences on adolescent behavior?. Evidence from a study of adopted siblings. Behavior Genetics. 2009;39:532–540. doi: 10.1007/s10519-009-9283-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bullock BM, Dishion TJ. Sibling collusion and problem behavior in early adolescence: Toward a process model for family mutuality. Journal of Abnormal Child Psychology. 2002;30:143–153. doi: 10.1023/a:1014753232153. [DOI] [PubMed] [Google Scholar]
  7. Celiberti DA, Harris SL. Behavioral intervention for siblings of children with autism: A focus on skills to enhance play. Behavior Therapy. 1993;24:573–599. [Google Scholar]
  8. Center for Diesease and Control (CDC) Youth Risk Behavior Surveillance - United States, 2011. Morbidity and Mortality Weekly Report. 61, 1-162. 2012 Retrieved (April 30, 2013) from http://www.cdc.gov/mmwr/pdf/ss/ss6104.pdf. [PubMed]
  9. D’Amico EJ, Fromme K. Risk behaviors of adolescent and young adult siblings. Health Psychology. 1997;16:426–432. doi: 10.1037//0278-6133.16.5.426. [DOI] [PubMed] [Google Scholar]
  10. Donovan SJ, Levin FR. The “younger-sibling-at-risk design”: A pilot study of adolescents with ADHD and an older sibling with substance use disorder. American Journal of Drug and Alcohol Abuse. 2011;37:235–239. doi: 10.3109/00952990.2011.569805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Duncan SC, Alpert A, Duncan TE, Hops H. Adolescent alcohol use development and young adult outcomes. Drug and Alcohol Dependence. 1997;49:39–48. doi: 10.1016/s0376-8716(97)00137-3. [DOI] [PubMed] [Google Scholar]
  12. Duncan SC, Duncan TE, Hops H. Latent variable modeling of longitudinal and multilevel alcohol use data. Journal of Studies on Alcohol and Drugs. 1998;59:399–408. doi: 10.15288/jsa.1998.59.399. [DOI] [PubMed] [Google Scholar]
  13. East PL, Khoo ST. Longitudinal pathways linking family factors and sibling relationship qualities to adolescent substance use and sexual risk behaviors. Journal of Family Psychology. 2005;19:571–580. doi: 10.1037/0893-3200.19.4.571. [DOI] [PubMed] [Google Scholar]
  14. Elkins IJ, McGue M, Iacono WG, Tellegen A. Genetic and environmental influences on parent-son relationships: Evidence for increasing genetic influence during adolescence. Developmental Psychology. 1997;33:351–363. doi: 10.1037//0012-1649.33.2.351. [DOI] [PubMed] [Google Scholar]
  15. Enders CK, Bandalos DL. The relative performance of full information maximum likelihood estimation for missing data in structural equation models. Structural Equation Modeling. 2001;8:430–457. [PubMed] [Google Scholar]
  16. Fagan AA, Najman JM. The relative contributions of parental and sibling substance use to adolescent tobacco, alcohol, and other drug use. Journal of Drug Issues. 2005;35:869–883. [Google Scholar]
  17. Feinberg ME, Solmeyer AR, McHale SM. The third rail of family systems: Sibling relationships, mental and behavioral health, and preventive intervention in childhood and adolescence. Clinical Child & Family Psychology Review. 2012;15:43–57. doi: 10.1007/s10567-011-0104-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Fisher LB, Miles IW, Austin B, Camargo CA, Colditz GA. Predictors of initiation of alcohol use among US adolescents. Archives of Pediatrics & Adolescent Medicine. 2007;161:959–966. doi: 10.1001/archpedi.161.10.959. [DOI] [PubMed] [Google Scholar]
  19. Flory K, Lynam D, Milich R, Leukefeld C, Clayton R. Early adolescent through young adult alcohol and marijuana use trajectories: Early predictors, young adult outcomes, and predictive utility. Development and Psychopathology. 2004;16:193–213. doi: 10.1017/s0954579404044475. [DOI] [PubMed] [Google Scholar]
  20. Fowler T, Shelton K, Lifford K, Rice F, McBride A, Nikolov I, et al. Genetic and environmental influences on the relationship between peer alcohol use and own alcohol use in adolescents. Addiction. 2007;102:894–903. doi: 10.1111/j.1360-0443.2007.01824.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Griffin KW, Botvin GJ. Evidence-based interventions for preventing substance use disorders in adolescents. Child and Adolescent Psychiatric Clinics of North America. 2010;19:505–526. doi: 10.1016/j.chc.2010.03.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Han C, McGue M, Iacono WG. Lifetime tobacco, alcohol and other substance use in adolescent Minnesota twins: univariate and multivariate behavioral genetic analyses. Addiction. 1999;94:981–993. doi: 10.1046/j.1360-0443.1999.9479814.x. [DOI] [PubMed] [Google Scholar]
  23. Harakeh Z, Engels RCME, Vermulst AA, de Vries H, Scholte RHJ. The influence of best friends on adolescent smoking: A longitudinal study. Journal of Applied Social Psychology. 2007;29:337–367. [Google Scholar]
  24. Hayes AF. Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs. 2009;76:408–420. [Google Scholar]
  25. Hicks BM, Iacono WG, McGue M. Consequences of an adolescent onset and persistent course of alcohol dependence in men: Adolescent risk factors and adult outcomes. Alcoholism: Clinical and Experimental Research. 2010;34:1–15. doi: 10.1111/j.1530-0277.2010.01154.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hingson RW, Heeren T, Winter MR. Age at drinking onset and alcohol dependence: age at onset, duration, and severity. Journal of the American Medical Association, Pediatrics. 2006;160:730–746. doi: 10.1001/archpedi.160.7.739. [DOI] [PubMed] [Google Scholar]
  27. Hollingshead AB. Four Factor Index of Social Status. Unpublished paper. 1975 Retrieved (April 30, 2013) from http://www.yale-university.com/sociology/faculty/docs/hollingshead_socStat4factor.pdf.
  28. Hopfer CJ, Crowley TJ, Hewitt JK. Review of twin and adoption studies of adolescent substance use. Journal of the American Academy of Child and Adolescent Psychiatry. 2003;43:710–719. doi: 10.1097/01.CHI.0000046848.56865.54. [DOI] [PubMed] [Google Scholar]
  29. Iannotti RJ, Bush PM. Perceived vs. actual friends’ use of alcohol, cigarettes, marijuana, and cocaine: Which has the most influence? Journal of Youth and Adolescence. 1992;21:375–389. doi: 10.1007/BF01537024. [DOI] [PubMed] [Google Scholar]
  30. Irons DE, Iacono WG, Oetting WS, McGue M. Developmental trajectory and environmental moderation of the effect of ALDH2 polymorphism on alcohol use. Alcoholism: Clinical and Experimental Research. 2012;36:1882–91. doi: 10.1111/j.1530-0277.2012.01809.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Irons DE, McGue M, Iacono WG, Oetting WS. Mendelian randomization: A novel test of the gateway hypothesis and models of gene-environment interplay. Development and Psychopathology. 2007;19:1181–1195. doi: 10.1017/S0954579407000612. [DOI] [PubMed] [Google Scholar]
  32. Johnston LD, O’Malley PM, Miech RA, Bachman JG, Schulenberg JE. Monitoring the Future national results on drug use: 1975–2013: Overview, Key Findings on Adolescent Drug Use. Ann Arbor. Institute for Social Research, The University of Michigan. 2014 Retrieved (April 22, 2014) from http://www.monitoringthefuture.org/pubs/monographs/mtf-overview2013.pdf.
  33. Kendler KS, Jacobson KC, Gardner CO, Gillespie N, Aggen SA, Prescott CA. Creating a social world: A developmental twin study of peer-group deviance. Archives of General Psychiatry. 2007;64:958–965. doi: 10.1001/archpsyc.64.8.958. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Kennedy DE, Kramer L. Improving emotion regulation and sibling relationship quality: The more fun with sisters and brothers program. Family Relations. 2008;57:567–578. [Google Scholar]
  35. Kenny DA. Measuring Model Fit. 2012 Jul 5; Retrieved February 28, 2013, from http://davidakenny.net/cm/fit.htm.
  36. Kline RB. Principles and practice of structural equation modeling. New York: The Guilford Press; 2005. [Google Scholar]
  37. Koopmans JR, Boomsma DI. Familial resemblances in alcohol use: Genetic or cultural transmission. Journal of Studies on Alcohol. 1996;1:19–28. doi: 10.15288/jsa.1996.57.19. [DOI] [PubMed] [Google Scholar]
  38. Kramer L, Radey C. Sibling relationships among young children: A social skills training model. Family Relations. 1997;46:237–246. [Google Scholar]
  39. Lewis MA, Litt DM, Blayney JA, Lostutter TW, Granato H, Kilmer JR, et al. They drink how much and where? Normative perceptions by drinking contexts and their association to college students’ alcohol consumption. Journal of Studies on Alcohol Drugs. 2011;72:844–853. doi: 10.15288/jsad.2011.72.844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Lobato DJ, Kao BT. Brief report: Family-based group intervention for young siblings of children with chronic illness and developmental disability. Journal of Pediatric Psychology. 2005;30:678–682. doi: 10.1093/jpepsy/jsi054. [DOI] [PubMed] [Google Scholar]
  41. Low S, Shortt JW, Snyder J. Sibling influences on adolescent substance use: The role of modeling, collusion, and conflict. Development and Psychopathology. 2012;24:287–300. doi: 10.1017/S0954579411000836. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Martens MP, Page JC, Mowry ES, Damann KM, Taylor KK, Cimini MD. Differences between actual and perceived student norms: An examination of alcohol use, drug use, and sexual behavior. Journal of American College Health. 2010;54:295–300. doi: 10.3200/JACH.54.5.295-300. [DOI] [PubMed] [Google Scholar]
  43. McGue M, Elkins I, Iacono WG. Genetic and environmental influences on adolescent substance use and abuse. American Journal of Medical Genetics. 2000;96:671–677. doi: 10.1002/1096-8628(20001009)96:5<671::aid-ajmg14>3.0.co;2-w. [DOI] [PubMed] [Google Scholar]
  44. McGue M, Iacono WG. Siblings and the socialization of adolescent deviance: An adoption study approach. In: McCartney K, Weinberg R, editors. Experience and development: A festschrift to honor Sandra W. Scarr. London, U.K.: Taylor & Francis; 2009. pp. 170–201. [Google Scholar]
  45. McGue M, Keyes M, Sharma A, Elkins I, Legrand L, Johnson W, et al. The environments of adopted and non-adopted youth: Evidence on range restriction from the sibling interaction and behavior study (SIBS) Behavior Genetics. 2007;37:449–462. doi: 10.1007/s10519-007-9142-7. [DOI] [PubMed] [Google Scholar]
  46. McGue M, Sharma A, Benson P. Parent and sibling influences on adolescent alcohol use and misuse: Evidence from a U.S. adoption cohort. Journal of Studies on Alcohol and Drugs. 1996;57:8–18. doi: 10.15288/jsa.1996.57.8. [DOI] [PubMed] [Google Scholar]
  47. McGuire S, Segal NL. Peer network overlap in twin, sibling, and friend dyads. Child Development. 2013;84:500–511. doi: 10.1111/j.1467-8624.2012.01855.x. [DOI] [PubMed] [Google Scholar]
  48. Miller JW, Naimi TS, Brewer RD, Jone SE. Binge drinking and associated health risk behaviors among high school students. Pediatrics. 2007;119:76–85. doi: 10.1542/peds.2006-1517. [DOI] [PubMed] [Google Scholar]
  49. Muthén B, Muthén L. Mplus version 6.12. Los Angeles, CA: Muthén & Muthén; 1998-2013. [Google Scholar]
  50. Needle R, McCubbin H, Wilson M, Reineck R, Lazar A, Mederer H. Interpersonal influences in adolescent drug use - the role of older siblings, parents, and peers. Substance Use & Misuse. 1986;21:739–766. doi: 10.3109/10826088609027390. [DOI] [PubMed] [Google Scholar]
  51. Pagan JL, Rose RJ, Viken RJ, Pulkkinen L, Kaprio J, Dick DM. Genetic and environmental influences on stages of alcohol use in adolescence and into young adulthood. Behavior Genetics. 2006;36:483–497. doi: 10.1007/s10519-006-9062-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Preacher KJ, Kelly K. Effect size measures for mediation models: Quantitative strategies for communicating indirect effects. Psychological Methods. 2011;16:93–115. doi: 10.1037/a0022658. [DOI] [PubMed] [Google Scholar]
  53. Rende R, Slomkowski C, Lloyd-Richardson E, Niaura R. Sibling effects on substance use in adolescence: Social contagion and genetic relatedness. Journal of Family Psychology. 2005;19:611–618. doi: 10.1037/0893-3200.19.4.611. [DOI] [PubMed] [Google Scholar]
  54. Rose RJ, Dick DM, Viken RJ, Pulkkinen L, Kaprio J. Drinking or abstaining at age 14? A genetic epidemiological study. Alcoholism Clinical & Experimental Research. 2001;25:1594–1604. [PubMed] [Google Scholar]
  55. Rowe DC, Gulley B. Sibling effects on substance abuse and delinquency. Criminology. 1992;30:217–233. [Google Scholar]
  56. Samek DR, Keyes MA, Iacono WG, McGue M. Peer deviance, alcohol expectancies, and adolescent alcohol use: Explaining shared and nonshared environmental effects using an adoptive sibling pair design. Behavior Genetics. 2013;43:286–296. doi: 10.1007/s10519-013-9595-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Samek DR, Rueter MA. Considerations of elder sibling closeness in predicting younger sibling substance use: Social learning versus social bonding explanations. Journal of Family Psychology. 2011;25:931–941. doi: 10.1037/a0025857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Slomkowski C, Rende R, Novak S, Lloyd-Richardson E, Niaura R. Sibling effects on smoking in adolescence: Evidence for social influence from a genetically informative design. Addiction. 2005;100:430–438. doi: 10.1111/j.1360-0443.2004.00965.x. [DOI] [PubMed] [Google Scholar]
  59. Trim RS, Leuthe E, Chassin L. Sibling influence on alcohol use in a young adult, high-risk sample. Journal of Studies on Alcohol and Drugs. 2006;67:391–398. doi: 10.15288/jsa.2006.67.391. [DOI] [PubMed] [Google Scholar]
  60. Tsao L, Odom SL. Sibling-mediated social interaction intervention for young children with autism. Topics in Early Childhood Special Education. 2006;26:106–123. [Google Scholar]
  61. Walden B, McGue M, Iacono WG, Burt SA, Elkins I. Identifying shared environmental contributions to early substance use: The respective roles of peers and parents. Journal of Abnormal Psychology. 2004;113:440–450. doi: 10.1037/0021-843X.113.3.440. [DOI] [PubMed] [Google Scholar]
  62. Whiteman SD, Jensen AC, Maggs JL. Similarities and differences in adolescent siblings’ alcohol-related attitudes, use, and delinquency: Evidence for convergent and divergent influence processes. Journal of Youth and Adolescence. 2013 doi: 10.1007/s10964-013-9971-z. online publication. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Whiteman SD, McHale SM, Soli A. Theoretical perspectives on sibling relationships. Journal of Family Theory and Review. 2011;3:124–139. doi: 10.1111/j.1756-2589.2011.00087.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Williams JH, Ayers CD, Abbott RD, Hawkins JD, Catalano RF. Racial differences in risk factors for delinquency and substance use among adolescents. Social Work Research. 1999;23:241–256. [Google Scholar]
  65. Williams PD, Hanson S, Karlin R, Ridder L, Liebergen A, Olson J, et al. Outcomes of a nursing intervention for siblings of chronically ill children: A pilot study. Journal for Specialists in Pediatric Nursing. 1997;2:127–137. doi: 10.1111/j.1744-6155.1997.tb00070.x. [DOI] [PubMed] [Google Scholar]
  66. Williams PD, Williams AR, Graff C, Hanson S, Stanton A, Hafeman C, Liebergen A, et al. A community-based intervention for siblings and parents of children with chronic illness or disability: The ISEE study. Journal of Pediatrics. 2003;143:386–393. doi: 10.1067/S0022-3476(03)00391-3. [DOI] [PubMed] [Google Scholar]

RESOURCES