Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Oct 9.
Published in final edited form as: J Contemp Crim Justice. 2018 Nov 8;35(1):7–20. doi: 10.1177/1043986218810578

Romantic Partner Alcohol Misuse Interacts with GABRA2 Genotype to Predict Frequency of Drunkenness in Young Adulthood

Jamie M Gajos 1,2, Michael A Russell 1,3, H Harrington Cleveland 4, David J Vandenbergh 3,5, Mark E Feinberg 2
PMCID: PMC6784828  NIHMSID: NIHMS1003036  PMID: 31598057

Abstract

Previous research has identified the importance of romantic partners—including spouses, significant others, and dating partners—for influencing the engagement in health-risking behaviors, such as alcohol misuse during emerging adulthood. Although genetic factors are known to play a role in the development of young adult alcohol misuse, little research has examined whether genetic factors affect young adults’ susceptibility to their romantic partners’ alcohol misusing behaviors. The current study tests whether a single nucleotide polymorphism in the GABRA2 gene (rs279845) moderates the relationship between romantic partner alcohol misuse and frequency of drunkenness in young adulthood. Results revealed differential risk associated with romantic partner alcohol misuse and young adult drunk behavior according to GABRA2 genotype, such that individuals with the TT genotype displayed an elevated risk for frequency of drunkenness when romantic partner alcohol misuse was also high (IRR = 1.06, p ≤ 0.05). The findings demonstrate the potential for genetic factors to moderate the influence of romantic partners’ alcohol misuse on drunk behavior during the transition to young adulthood.

Keywords: romantic partners, alcohol misuse, young adulthood, GABRA2

Introduction

Romantic partnerships encompass one of the most salient social environments individuals experience during the transition to young adulthood, due in part to their potential to either increase or reduce engagement in risky behaviors. Romantic partners may increase the likelihood of engaging in criminal behaviors (Herrera, Wiersma, & Cleveland, 2011) and alcohol misuse (Leonard & Mudar, 2003). Given the role that alcohol misuse plays in the development of criminal behaviors, such as both violent and property crimes (Fergusson & Horwood, 2000), romantic partnerships may also be characterized as a social mechanism that links alcohol misuse to delinquent and criminal behaviors. Entering into romantic partnerships has also been linked to reductions in both substance misuse (Siennick et al., 2014) and deviant behaviors (Sampson & Laub, 1990, 1993). Those who are married or cohabitating with a serious romantic partner tend to show a reduction in heavy alcohol use relative to those who are single, as well as relative to themselves upon becoming married (Fleming, White, & Catalano, 2010). Romantic partnerships may serve as a protective factor against the continuation of alcohol misuse (Chassin, Hussong, & Beltran, 2009), yet the influence of romantic partners’ own alcohol misuse on young adult heavy drinking should not be ignored. The exposure to alcohol-using partners and peers has been linked to increases in young adult binge drinking behaviors (Andrews, Tildesley, Hops, & Li, 2002) and romantic partners tend to show homophily on measures of problem drinking (Boutwell, Beaver, & Barnes, 2012; Windle, 1997).

The mechanisms by which romantic partners may influence alcohol misuse are attributed to two processes: socialization, in which the exposure to partners’ alcohol misuse influences individuals’ own alcohol misuse; and selection, in which individuals who engage in alcohol misuse tend to associate with—or “select into” relationships with—partners who also engage in heavy drinking. Arguments exist for both processes (Akers, Krohn, Lanza-Kaduce, & Radosevich, 1979; Glueck & Glueck, 1950), but the two mechanisms are not mutually exclusive and may co-occur across developmental periods (Rhule-Louie & McMahon, 2007). Genetically informed research designs can help shed light on the synergistic relationship between an individual and their social environment in fostering alcohol misuse. Findings from behavioral genetic research designs reveal that genetic influences on alcohol misuse may become increasingly important in adulthood (Rose, Dick, Viken, & Kaprio, 2001), where approximately 50–60% of the variance in adult alcohol dependence has been attributed to genetic factors (Dick, Prescott, & McGue, 2009). However, social environments may moderate genetic influences on young adult heavy drinking (Barr et al., 2017; Jarnecke & South, 2014), thereby suggesting evidence of a gene-environment interaction (GxE).

The Relationship between GABRA2, Romantic Partnerships, and Alcohol Misuse

The association between variants within the GABRA2 gene and alcohol misuse behaviors has been examined in candidate gene studies and genome-wide association studies (GWAS) (Bierut et al., 2010; Dick, Agrawal, et al., 2006). A recent meta-analysis examined the relationship between variants within the GABRA2 gene and alcohol misuse and found evidence in support of GABRA2 variants for contributing to alcohol misuse behaviors (Li et al., 2014). Associations between variants in the GABRA2 gene and alcohol use have been shown to vary across different developmental periods (Dick et al., 2014; Dick, Bierut, et al., 2006). Dick and colleagues (2006) found variation in a single nucleotide polymorphism (SNP) in GABRA2 (rs279871) to be significantly related to alcohol dependence during the transition to young adulthood, where those with at least one copy of the risk allele displayed an elevated risk for alcohol dependence. The age of onset for alcohol dependence and the age at first intoxication did not vary according to GABRA2 genotype. Similar findings have been reported, where variation across six different SNPs in GABRA2 were associated with growth trajectories of drunk behavior between the ages of 18–19 (Dick et al., 2014).

The influence of GABRA2 may partially depend on the exposure to relevant environments in adulthood. One environment that may interact with GABRA2 is romantic partnerships. Dick, Agrawal, and colleagues (2006) examined the association between GABRA2 and marital status in predicting alcohol dependence and found variation in GABRA2 and marital status to have independent effects on the risk for alcohol dependence and to interact with one another. Differences in alcohol dependence did not vary according to genotype among unmarried individuals, but GABRA2 contributed to differences in alcohol dependence among married respondents, with those with a high-risk genotype being at a greater risk for alcohol dependence. GABRA2 was also associated with marital status, such that individuals with a high-risk genotype were more likely to report never being married and less likely to experience stable marriages.

Examining the potential for a genetic variant located in the GABRA2 gene to moderate the influence of romantic partners’ heavy drinking on young adult drunkenness may also shed light on the pathways to alcohol misuse during emerging adulthood. The GABRA2 gene codes for the alpha-2 subunit of GABAA receptors and belongs to a group of GABAA receptor subunit-encoding genes located on chromosome 4 (McLean, Farb, & Russek, 1995). The GABAA alpha-2 subunit-containing receptors are highly expressed in reward-sensitive areas of the brain. Variants within the GABRA2 gene may influence differential sensitivity to environments, resulting in greater responsiveness to both positive and negative environments (Russell et al., 2018). GABRA2 variants may help account for differential responses to romantic partners’ heavy alcohol use during emerging adulthood.

Current Study

The current study analyzes the relationship between romantic partner alcohol misuse and the GABRA2 genotype. Although previous research has examined the potential for romantic partnerships and/or social support mechanisms to moderate the impact of genetic risk factors on alcohol misuse (Barr et al., 2017; Jarnecke & South, 2014), these studies have estimated genetic influences as a latent construct—thereby not allowing for an examination of the specific genes involved in alcohol misuse. Such studies have also been unable to account for the potential impact that romantic partner’s own alcohol misuse behaviors may have on such relationships. The current study extends this area of research in two important ways. First, the GABRA2 genotype and romantic partner alcohol misuse will be used to provide estimates of genetic and developmentally relevant environmental influences on the frequency of drunkenness during the transition to adulthood. The potential for GABRA2 to be associated with romantic partner misuse (i.e., a gene-environment correlation [rGE]) will also be examined. Second, a test of whether the GABRA2 genotype moderates the influence of romantic partner alcohol misuse on young adult drunkenness will be assessed.

Methods

Data

Data for this study were drawn from the Genetics of PROSPER (gPROSPER) sample, which is a subsample of the PROSPER evidence-based intervention program. The PROSPER intervention trial is a cohort-based design, which includes a sample of students from 28 school districts in Iowa and Pennsylvania. Half of the school districts were randomly assigned to receive preventative interventions targeting substance use. The PROSPER intervention was delivered through a school-community-university partnership (Spoth, Greenberg, Bierman, & Redmond, 2004) and consisted of in-school and in-home sequences. Students within each of the school districts were invited to complete an in-school survey, where 90% agreed to participate. During the fall semesters of 2002 and 2003, in-school pre-intervention assessments were conducted for cohorts 1 and 2, respectively. Annual spring follow-up surveys were collected until the 12th grade. A random sample of youth within cohort 2 (n = 2267) were invited to participate in the in-home assessments that were conducted during waves 1–5 and entailed assessments derived from both parent and adolescent questionnaires.

The gPROSPER sample includes 2032 youth participating in the larger PROSPER project who were genotyped through the collection of buccal cell samples. Genotypes for rs279845 and additional SNPs for measuring population stratification were determined on an OpenArray system (Life Technologies, Thermo Fisher), utilizing TaqMan genotyping assays that are applied to an array (see Vandenbergh et al., 2016 for more details). Among the 2032 participants who provided DNA, 96% (1952) were successfully genotyped for the GABRA2 rs279845 SNP. The current analytical sample is based on data collected during the wave 9 young adult follow-up assessments (ages 19 and 22). The young adult assessments included 1985 respondents who were randomly selected and recruited from the larger PROSPER project with complete in-class 6th grade pretest assessments and who were enrolled in the same school district during 9th grade. A total of 1006 adults within the young adult sample were successfully genotyped on SNP rs279845. The final analytical sample is comprised of respondents who were successfully genotyped on SNP rs279845 and participated in the young adult wave 9 assessment, as well as who reported being in a current romantic relationship or in a relationship that lasted 6 months or longer within the past year (n = 374).1

Measures

Young Adult Drunk Behavior

During the wave 9 young adult assessment, respondents reported their drinking behaviors. Respondent drunkenness was assessed by asking, “How often do you usually get drunk?” Responses ranged from 0 = Not at all, to 6 = About every day (M (SD) = 1.38 (1.46)).

Partner Alcohol Misuse

Romantic partner alcohol misuse was assessed by respondent reports on how many days their spouse/partner had 5 or more alcoholic drinks in a row within the past month. Responses ranged from 0 = 0 days to 31 = 31 days. Due to the non-normal distribution of the count data, all outliers on the measure of partner alcohol misuse were Winsorized (Kennedy, Lakonishok, & Shaw, 1992). Outliers were Winsorized at the mean + 3 standard deviations and were rounded up to the nearest value (i.e., 12) (M (SD) = 1.19 (2.42)).

GABRA2 Genotype

The rs279845 SNP (A/T) is located within the third intron of GABRA2 and has been linked to alcohol use behaviors (Covault et al., 2004; Dick, Agrawal, et al., 2006). Previous research has identified the minor allele (A) as the risk allele (Dick et al., 2014) and others have identified the major allele (T) as the risk allele (Edenberg et al., 2004). Due to the lack of clear a priori information to inform the coding of the rs279845 SNP, the GABRA2 genotype was coded as a dummy categorical variable where both AA and AT genotypes were coded as 0 and TT genotypes were coded as 1; a coding utilized in previous analyses using this GABRA2 SNP in the gPROSPER sample (see Russell et al., 2018).

Past Drunk Behavior

Previous drunk behavior was assessed with an index comprised of annual assessments of past month drunkenness during waves 2–8. Responses from the baseline assessment (wave 1) were excluded, since the intervention was not administered until wave 2. Respondents reported frequency of being drunk or intoxicated on a 0 to 4 scale (0 = Not at all, to 4 = More than once a week). Responses were standardized and summed to create an overall index of previous drunk behavior (Cronbach’s α = 0.75)

Intervention Status

Within the analytic sample, 48% of the respondents were assigned to the control condition and 52% were assigned to the intervention condition. Intervention status was included as a covariate (0 = Control, 1 = Intervention).2

Control for Population Stratification (PC1)

Differences in genotype and allele frequency distributions across populations increases the potential for spurious associations between alleles and phenotypes (Cardon & Palmer, 2003). To account for genetic differences among ancestral or ethnic populations (known as population stratification), an admixture mapping method utilizing ancestry informative markers (AIMs) was employed, and used 34 AIMs to identify geographic ancestry in a principal coordinate (PC) analysis (see Cleveland et al., 2017). Principal coordinate 1 (PC1) represents a continuous scale of European ancestry, where a higher PC1 score indicated less European ancestry. PC1 was used to control for population stratification in two ways: 1) by including PC1 as a continuous covariate within the models and 2) by reestimating all models among those with PC1 scores indicating European ancestry.3

Sex

Respondent sex was coded as 0 = Female, 1 = Male. In the final analytic sample, 68% were female and 32% were male.

Plan of Analysis

A series of steps were conducted to examine the influence of romantic partner alcohol misuse on young adult drunk behavior and whether the effects of partner alcohol misuse vary by GABRA2 genotype. First, test statistics (e.g., t-tests and chi-square estimates) were calculated to examine the relationship between each of the study variables. Second, the independent effects of romantic partner alcohol misuse and GABRA2 on young adult drunk behavior were estimated in models that also adjusted for the study’s covariates. Frequency of drunkenness was estimated as a count variable within a Poisson regression model.4 Incident-rate ratios (IRRs) were estimated and represent exponentiated unstandardized Poisson regression coefficients. Next, the two-way interaction between romantic partner alcohol misuse and GABRA2 was estimated to test whether the effects of partner alcohol misuse vary by genotype. Within the two-way interaction model, all covariates and partner alcohol misuse were grand mean-centered prior to creating the interaction term. Finally, the predicted exponentiated values for frequency of drunkenness across the number of days romantic partners’ engaged in alcohol misuse were estimated for AA/AT and T/T genotypes.

Results

The means, standard deviations, t-statistics, and chi-square estimates for the study variables are reported in Table 1.5 The mean of respondents’ frequency of drunkenness did not significantly differ from the mean of partners’ alcohol misuse (t-statistic = −0.19), nor did the mean of respondents’ drunkenness differ by GABRA2 genotype (t-statistic = 0.85). The mean of romantic partners’ alcohol misuse did not significantly differ by respondents’ GABRA2 genotype (t-statistic = −0.43), providing no evidence for a gene-environment correlation.

Table 1:

Means, Standard Deviations, and Test Statistics for Study Variables

Test Statistics†
M (SD)a Numberb 1 2 3 4 5 6 7
1. Drunk behavior 1.38 (1.46) 988 --
2. Partner alcohol misuse 1.19 (2.42) 414 −0.19 --
3. GABRA2 rs279845 0.31 (0.46) 1006 0.85 −0.43 --
4. Past drunk behavior 0.29 (0.50) 1005 13.26** 6.88** −0.54 --
5. Intervention condition 0.53 (0.50) 1006 −0.90 −0.56 2.25 0.66 --
6. Sex 0.40 (0.49) 1006 −3.45** 2.78** 1.47 −1.13 0.00 --
7. PC1 score −0.02 (0.00) 980 17.97** 9.80** 1.70 13.24** 0.70 −0.10 --
Open in a new tab
*

p ≤ 0.05;

**

p ≤ 0.01

a

Means (and SDs) for each variable are calculated among those successfully genotyped on GABRA2 rs279845.

b

Based on the number of participants in the analytic sample with valid genotypic data and who were included in the young adult assessment.

†

Estimates reflect t-tests (t-statistics) for continuous-continuous and continuous-binary measures and chi-square estimates (χ2) for two binary measures among the full analytic sample.

The results for the main effects of partner alcohol misuse and GABRA2 genotype on young adult drunk behavior are presented in Model 1 of Table 2. Romantic partner alcohol misuse was significantly associated with an increase in young adult drunk behavior (IRR = 1.14, p ≤ 0.01), while controlling for the other covariates within the model. The GABRA2 genotype was not associated with young adult drunk behavior. Within the two-way interaction model (Model 2), GABRA2 was found to significantly moderate the impact of romantic partner alcohol misuse on respondents’ frequency of drunkenness (IRR = 1.06, p ≤ 0.05).

Table 2:

Association between Romantic Partner Alcohol Misuse, GABRA2 Genotype, and Drunk Behavior in Young Adulthood†

Model 1 Model 2
IRR (95% CI) IRR (95% CI)
Main effects
 Partner alcohol misuse 1.14** (1.11, 1.17) 1.12** (1.09, 1.16)
 GABRA2 rs279845 0.94 (0.77, 1.16) 0.88 (0.71, 1.10)
 Past drunk behavior 1.44** (1.25, 1.65) 1.45** (1.26, 1.66)
 Intervention status 1.05 (0.87, 1.27) 1.04 (0.86, 1.25)
 Sex 1.58** (1.30, 1.93) 1.62** (1.33, 1.99)
Two-way interaction
 Partner alcohol misuse x GABRA2 rs279845 -- 1.06* (1.00, 1.12)
Open in a new tab
*

p ≤ 0.05;

**

p ≤0.01

†

Models adjust for PC1 score.

IRR= Incidence-rate ratio.

The robustness of the findings from the main effects and two-way interaction models were examined by including a number of other relevant social variables, such as relationship quality, full time work and/or school status, and parenthood. These factors were not consistently and significantly correlated with romantic partner alcohol misuse, nor with young adult drunk behavior (results available upon request).6 The potential moderating role of respondent gender was also examined, but did not yield significant results (available upon request).

The predicted exponentiated values for frequency of drunkenness were estimated by levels of romantic partners’ alcohol misuse. Figure 1 presents the predicted drunk behavior when romantic partners engage in alcohol misuse behaviors during 0, 4, 8, and 12 days within the past month. Carriers of the GABRA2 TT genotype display an increased risk of engaging in drunk behavior themselves, once their romantic partner begins to engage in binge drinking behaviors during 8 or more days out of the month. Predicted drunk behavior among A carriers also increases as romantic partner alcohol misuse increases, but does not show the same rate of growth in predicted drunkenness as the TT genotype. Follow-up analyses of the simple effects of romantic partner alcohol misuse on young adult drunkenness also varied between the TT genotype (IRR = 1.19, 95% CI: 1.14–1.24, p ≤ 0.01) and A carriers (IRR = 1.12, 95% CI: 1.09–1.16, p ≤ 0.01).

Figure 1: Predicted Drunk Behavior across levels of Partner Alcohol Misuse and by GABRA2 rs279845.

Figure 1:

Open in a new tab

Note: The y-axis represents exponentiated values of predicted drunk behavior.

Discussion

The findings suggest that the influence of romantic partner alcohol misuse on young adult drunk behavior varies according to GABRA2 genotype. Not only did respondents with alcohol misusing partners engaged in more drunk behavior, but those with the TT GABRA2 genotype exhibited elevated levels of heavy drinking in comparison to A carriers. There was no observed significant main effect of GABRA2 on young adult drunk behavior, but the T allele was related to an increased risk for alcohol misuse. Although romantic partnerships may attenuate genetic influences on drinking habits (Heath, Jarden, & Martin, 1989), alcohol misuse of romantic partners themselves is relevant for understanding the conditions under which GABRA2 has implications for young adult heavy drinking. Marriage has been shown to exert a protective effect among those with the greatest genetic risk, but these associations are limited to individuals who are partnered with spouses with no lifetime history of alcohol use disorders (Kendler, Lönn, Salvatore, Sundquist, & Sundquist, 2016).

Other young adult life transitions may be important for preventing the onset or continuation of alcohol-related problems (Chassin et al., 2009). Additional analyses accounting for such factors revealed that relationship quality, attending work or school full time, and parenting were not consistently and significantly correlated with romantic partner alcohol misuse, nor with young adult drunk behavior. Future research will need to decipher which aspects of romantic partnerships are the most relevant for moderating genetic risk on young adult alcohol misuse (Barr et al., 2017). Romantic partner effects have also been reported to vary by gender (Rhule-Louie & McMahon, 2007), such that social support mechanisms may actually increase the genetic influences on alcohol misuse among females (Barr et al., 2017). Upon testing for the moderation of gender, the findings remained in the expected direction with similar magnitude, although the two-way interactions between GABRA2 and romantic partner alcohol misuse were not significant among male- and female-only samples. Reduced sample size as a result of splitting the sample by gender may have reduced the likelihood to detect significant effects. Romantic partner alcohol misuse continued to exert significant main and conditional effects on young adult drunk behavior among both male and female samples.

The study is not without limitations. First, although examining the association between a single genetic variant and romantic partner alcohol misuse demonstrates the importance of considering how genes and contexts transact to contribute to young adult drinking, the use of a single gene approach offers an incomplete insight into these processes. Future research may benefit from assessing the overall impact of the GABAergic system when estimating the relationship between genetic risk factors and alcohol-related problems in young adulthood. Other concerns surrounding candidate gene research—such as small sample sizes resulting in the limited power to detect effects in contrast to high-powered GWAS—have also been expressed (Dick et al., 2015; Duncan & Keller, 2011). The current study is not immune to such criticisms, but these limitations should be acknowledged with the understanding that candidate gene studies are designed to address very different research questions than those driven by GWAS. Rather than seeking to identify all of the possibly associated genetic variants with a phenotype, candidate gene studies are interested in examining the biological pathways/processes related to the phenotype under study (Moore, 2017). Candidate gene research has tended to examine well-characterized and well-understood genetic markers (e.g., SLC6A4, DRD4, MAOA, CHRNA5) for which there have been clear replications across multiple methodologies (see Caspi, Hariri, Holmes, Uher, & Moffitt, 2010 with regard to SLC6A4). The SNP within the GABRA2 gene has shown consistent associations with alcohol use phenotypes (Lind et al., 2008) and consistency with regard to gene-by-environment interactions (Russell et al., 2018). Candidate gene studies also apply “deep phenotyping” approaches to examine the mechanisms underlying phenotypic variation, therefore, smaller sample sizes are typically employed (in comparison to GWAS) to obtain the necessary information on the phenotype. Applying this approach to examine the moderating role of environmental factors among a sample rivaling that of GWAS is currently challenging due to the cost of collecting the data. Hypothesis-free approaches (GWAS) and hypothesis-driven approaches (candidate gene studies) should both be utilized in gene discovery efforts because they offer distinct advantages and neither should be dismissed solely on the basis of their limitations (Schlomer, Cleveland, Vandenbergh, Fosco, & Feinberg, 2015).

Second, the measure of young adult drunkenness did not include an anchoring point (e.g., past month drunkenness).7 Utilizing a general measure of young adult drunkenness avoided potential biases related to restricting the measure of drunkenness to specific times of the year based on when the survey was administered.8 Third, estimates of romantic partner alcohol misuse was derived from respondent reports on romantic partners’ drunk behaviors instead of from self-reports. Prior work suggests that respondent reports of partners’ and peers’ substance use behaviors may actually inflate the true relationship between young adult and partner/peer substance use (Bauman & Ennett, 1996), yet the relationship between the two remains significant—albeit becomes attenuated—when peer/partner responses are directly reported (Chassin et al., 2009). Preliminary models helped account for the potential for respondent characteristics (e.g., GABRA2 genotype) to confound the relationship between romantic partner alcohol misuse and young adult drunk behavior. The examined variant within the GABRA2 gene was not predictive of romantic partner alcohol misuse and the effect of romantic partner alcohol misuse on young adult drunk behavior remained, despite controlling for respondents’ previous drunk behaviors from 6th to 12th grade.9 These findings illustrate the complex interaction between romantic partner alcohol misuse and genetic risk for predicting young adult drunk behavior. Rather than examining the genetic and environmental effects on alcohol use in isolation, future research will benefit from identifying how genetic and environmental risk factors work together to influence alcohol use misuse behaviors across the life course.

Acknowledgements:

This research was supported by the Prevention and Methodology Training Program (T32 DA017629, PI: L. M. Collins; P50 DA039838 PI: L. M. Collins). PROSPER and gPROSPER projects were supported by grant numbers DA013709 (PI: Spoth) and DA030389 (M-PIs: Cleveland and Vandenbergh) from the National Institute on Drug Abuse, respectively. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Drug Abuse or the National Institutes of Health.

Footnotes

1

The final analytic sample size also reflects the sample utilized for complete case analysis (i.e., only cases with non-missing data on all study variables were included in the analyses). The final analytic sample also reflects the n size for the analyses that controlled for population stratification, as detailed in the Methods section below.

2

Preliminary analyses tested for intervention status as a moderator, but findings were not significant.

3

All of the reported model results are representative of the analyses that controlled for population stratification. Results did not differ when employing either strategy to control for population stratification (i.e., when either including PC1 as a continuous covariate or by only restricting the sample to those with European ancestry).

4

Due to the non-normality of the residuals, linear regression was not used to estimate the outcome. Subsequent investigations revealed that a count model was a better alternative, and that Poisson regression should be applied. We also compared the fit of the negative binomial model to the Poisson model but the likelihood ratio test revealed that alpha was not significantly different from zero (prob >= chi2 = 0.134), suggesting that the Poisson model was appropriate.

5

Paired t-tests were used to assess the relationship between two continuous variables and independent group t-tests were used to assess the relationship between continuous variables across binary variables.

6

Two exceptions were found: full time student status was positively correlated with young adult drunk behavior (rₛ = 0.13, p ≤ 0.05) and being a parent was negatively correlated with partner alcohol misuse (rₛ = −0.13, p ≤ 0.01). After including these covariates within the models, the findings remained significant.

7

The lack of a defined recall period for young adult drunkenness may introduce challenges for establishing temporal order in regard to determining whether romantic partner alcohol misuse overlaps with young adult drunkenness or not.

8

A general measure of partner alcohol misuse was not available; therefore, past month alcohol misuse was examined.

9

Including a measure of respondents’ past drunk behavior may also address issues related to causal order in regard to whether partner alcohol misuse precedes or follows respondents’ drunk behavior. Although the results cannot determine whether partners’ past month alcohol misuse precedes respondents’ general drunkenness with certainty, we believe accounting for a measure of respondents’ past drunkenness may help to address some of these concerns.

References

  1. Akers RL, Krohn MD, Lanza-Kaduce L, & Radosevich M (1979). Social learning and deviant behavior: A specific test of a general theory. American Sociological Review, 44, 636–655. [PubMed] [Google Scholar]
  2. Andrews JA, Tildesley E, Hops H, & Li F (2002). The influence of peers on young adult substance use. Health Psychology, 21, 349–357. [DOI] [PubMed] [Google Scholar]
  3. Barr PB, Salvatore JE, Maes HH, Korhonen T, Latvala A, Aliev F, … & Dick DM (2017). Social relationships moderate genetic influences on heavy drinking in young adulthood. Journal of Studies on Alcohol and Drugs, 78, 817–826. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bauman KE, & Ennett ST (1996). On the importance of peer influence for adolescent drug use: Commonly neglected considerations. Addiction, 91, 185–198. [PubMed] [Google Scholar]
  5. Bierut LJ, Agrawal A, Bucholz KK, Doheny KF, Laurie C, Pugh E, … & Rice JP (2010). A genome-wide association study of alcohol dependence. Proceedings of the National Academy of Sciences, 107, 5082–5087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Boutwell BB, Beaver KM, & Barnes JC (2012). More alike than different: Assortative mating and antisocial propensity in adulthood. Criminal Justice and Behavior, 39, 1240–1254. [Google Scholar]
  7. Cardon LR, & Palmer LJ (2003). Population stratification and spurious allelic association. Lancet, 361, 598–604. [DOI] [PubMed] [Google Scholar]
  8. Caspi A, Hariri AR, Holmes A, Uher R, & Moffitt TE (2010). Genetic Sensitivity to the Environment: The Case of the Serotonin Transporter Gene and Its Implications for Studying Complex Diseases and Traits. American Journal of Psychiatry, 167, 509–527. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chassin L, Hussong A, & Beltran I (2009). Adolescent substance use In Lerner R & Steinberg L (Eds.), Handbook of adolescent psychology (3rd ed., pp. 723–763). Hoboken, NJ: Wiley. [Google Scholar]
  10. Cleveland HH, Schlomer GL, Vandenbergh DJ, Wolf PSA, Feinberg ME, Greenberg MT, … & Redmond C (2018). Associations between alcohol dehydrogenase genes and alcohol use across early and middle adolescence: Moderation × Preventive intervention. Development and Psychopathology, 30, 297–313. 10.1017/S0954579417000633 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Covault J, Gelernter J, Hesselbrock V, Nellissery M, & Kranzler HR (2004). Allelic and haplotypic association of GABRA2 with alcohol dependence. American Journal of Medical Genetics Part B (Neuropsychiatric Genetics), 129B, 104–109. [DOI] [PubMed] [Google Scholar]
  12. Dick DM, Agrawal A, Keller MC, Adkins A, Aliev F, Monroe S, … & Sher KJ (2015). Candidate gene-environment interaction research: Reflections and recommendations. Perspectives on Psychological Science, 10, 37–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dick DM, Agrawal A, Schuckit MA, Bierut L, Hinrichs A, Fox L, … & Begleiter H (2006). Marital status, alcohol dependence, and GABRA2: Evidence for gene-environment correlation and interaction. Journal of Studies on Alcohol, 67, 185–194. [DOI] [PubMed] [Google Scholar]
  14. Dick DM, Bierut L, Hinrichs A, Fox L, Bucholz KK, Kramer J, … & Foroud T (2006). The role of GABRA2 in risk for conduct disorder and alcohol and drug dependence across developmental stages. Behavior Genetics, 36, 577–590. [DOI] [PubMed] [Google Scholar]
  15. Dick DM, Cho SB, Latendresse SJ, Aliev F, Nurnberger JI Jr., Edenberg HJ, … & Kuperman S (2014). Genetic influences on alcohol use across stages of development: GABRA2 and longitudinal trajectories of drunkenness from adolescence to young adulthood. Addiction Biology, 19, 1055–1064. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dick DM, Prescott C, & McGue M (2009). The genetics of substance use and substance use disorders In Kim YK (Ed.), Handbook of Behavior Genetics (pp. 433–453). New York, NY: Springer. [Google Scholar]
  17. Duncan LE, & Keller MC (2011). A critical review of the first 10 years of candidate gene-by-environment interaction research in psychiatry. American Journal of Psychiatry, 168, 1041–1049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Edenberg HJ, Dick DM, Xuei X, Tian H, Almasy L, Bauer LO, … & Begleiter H (2004). Variations in GABRA2, encoding the α2 subunit of hthe GABA-A receptor, are associated with alcohol dependence and with brain oscillations. American Journal of Human Genetics, 74, 705–714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fergusson DM, & Horwood LJ (2000). Alcohol abuse and crime: A fixed-effects regression analysis. Addiction, 95, 1525–1536. [DOI] [PubMed] [Google Scholar]
  20. Fleming CB, White HR, & Catalano RF (2010). Romantic relationships and substance use in early adulthood: An examination of the influences of relationship type, partner substance use, and relationship quality. Journal of Health and Social Behavior, 51, 153–167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Glueck S, & Glueck E (1950). Unraveling juvenile delinquency. Cambridge, MA: Harvard University Press. [Google Scholar]
  22. Heath AC, Jarden R, & Martin NG (1989). Interactive effects of genotype and social environment on alcohol consumption in female twins. Journal of Studies on Alcohol, 50, 38–48. [DOI] [PubMed] [Google Scholar]
  23. Herrera VM, Wiersma JD, & Cleveland HH (2011). Romanic partners’ contribution to the continuity of male and female delinquent and violent behavior. Journal of Research on Adolescence, 21, 608–618. [Google Scholar]
  24. Jarnecke AM, & South SC (2014). Genetic and environmental influences on alcohol use problems: Moderation by romantic partner support, but not family or friend support. Alcoholism: Clinical and Experimental Research, 38, 367–375. [DOI] [PubMed] [Google Scholar]
  25. Kendler KS, Lönn SL, Salvatore J, Sundquist J, & Sundquist K (2016). Effect of marriage on risk for onset of alcohol use disorder: A longitudinal and co-relative analysis in a Swedish national sample. American Journal of Psychiatry, 173, 911–918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kennedy D, Lakonishok J, & Shaw WH (1992). Accommodating outliers and nonlinearity in decision models. Journal of Accounting Auditing & Finance, 7, 161–190. [Google Scholar]
  27. Leonard KE, & Mudar P (2003). Peer and partner drinking and the transition to marriage: A longitudinal examination of selection and influence processes. Psychology of Addictive Behaviors, 17, 115–125. [DOI] [PubMed] [Google Scholar]
  28. Li D, Sulovari A, Cheng C, Zhao H, Kranzler HR, & Gelernter J (2014). Association of gamma-aminobutyric acid A receptor α2 gene (GABRA2) with alcohol use disorder. Neuropsychopharmacology, 39, 907–918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lind PA, Macgregor S, Agrawal A, Montgomery GW Heath AC, Martin NG, & Whitfield JB (2008). The role of GABRA2 in alcohol dependence, smoking, and illicit drug use in an Australian population sample. Alcoholism: Clinical and Experimental Research, 32, 1721–1731. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. McLean PJ, Farb DH, & Russek SJ (1995). Mapping of the α 4 subunit gene (GABRA4) to human chromosome 4 defines an α 2— α 4—β 1—γ 1 gene cluster: Further evidence that modern GABAA receptor gene clusters are derived from an ancestral cluster. Genomics, 26, 580–586. [DOI] [PubMed] [Google Scholar]
  31. Moore SR (2017). Commentary: What is the case for candidate gene approaches in the era of high-throughput genomics? A response to Border and Keller (2017). Journal of Child Psychology and Psychiatry, 58, 331–334. [DOI] [PubMed] [Google Scholar]
  32. Rhule-Louie DM, & McMahon RJ (2007). Problem behavior and romantic relationships: Assortative mating, behavior contagion, and desistance. Clinical Child and Family Psychology Review, 10, 53–100. [DOI] [PubMed] [Google Scholar]
  33. Rose RJ, Dick DM, Viken RJ, & Kaprio J (2001). Gene-environment interaction in patterns of adolescent drinking: Regional residency moderates longitudinal influences on alcohol use. Alcoholism: Clinical and Experimental Research, 25, 637–643. [PubMed] [Google Scholar]
  34. Russell MA, Schlomer GL, Cleveland HH, Feinberg ME, Greenberg MT, Spoth RL, … Vandenbergh DJ (2018). PROSPER intervention effects on adolescents’ alcohol misuse vary by GABRA2 genotype and age. Prevention Science, 19, 27–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sampson RJ, & Laub JH (1990). Crime and desistance over the life course: The salience of adult social bonds. American Sociological Review, 55, 609–627. [Google Scholar]
  36. Sampson RJ, & Laub JH (1993). Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press. [Google Scholar]
  37. Schlomer GL, Cleveland HH, Vandenbergh DJ, Fosco GM, & Feinberg ME (2015). Looking forward in candidate gene research: Concerns and suggestions. Journal of Marriage and Family, 77, 351–354. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Siennick SE, Staff J, Osgood DW, Schulenberg JE, Bachman JG, & VanEseltine M (2014). Partnership transitions and antisocial behavior in young adulthood: A within-person, multi-cohort analysis. Journal of Research in Crime and Delinquency, 51, 735–758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Spoth R, Greenberg M, Bierman K, & Redmond C (2004). PROSPER community-university partnership model for public education systems: Capacity-building for evidence-based, competence-building prevention. Prevention Science, 5, 31–39. [DOI] [PubMed] [Google Scholar]
  40. Vandenbergh DJ, Schlomer GL, Cleveland HH, Schink AE, Hair KL, Feinberg ME, … & Redmond C (2016). An adolescent substance prevention model blocks the effect of CHRNA5 genotype on smoking during high school. Nicotine & Tobacco Research, 18, 212–220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Windle M (1997). Mate similarity, heavy substance use and family history of problem drinking among young adult women. Journal of Studies on Alcohol, 58, 573–580. [DOI] [PubMed] [Google Scholar]

RESOURCES