“Interplay of perceived friend environments with genetics on trajectories of alcohol use across adolescence”
Michelle Zaso, Syracuse University, Department of Psychology
Abstract
While alcohol use, on average, increases across adolescence, there are subgroups of adolescents whose drinking follows distinct patterns, or trajectories, over time. Friend environments are salient in adolescence and associated with drinking trajectories. However, it remains unknown whether associations of friend environments with drinking trajectories differ by genetics. We examined interactive associations of genetics (i.e., composite genetic risk scores for alcohol use based on a recent genome-wide association study) and adolescents’ perceived friend drinking and disruptive behavior at age 15 with drinking trajectories from ages 16 to 20 within a population-based cohort of 5,299 adolescents. Genetic risk scores modulated associations of friend drinking (but not friend disruptive behavior) on likelihood of following a moderately high increasing trajectory (the Middle-High class) relative to a stable low use trajectory (the Low class; OR = 0.55, 95% CI [0.34,0.87], p = .01). Specifically, among adolescents carrying low genetic risks, friend drinking was more strongly associated with likelihood of belonging to the Middle-High relative to Low class. Among adolescents carrying high genetic risks, friend drinking was not associated with drinking trajectories. Findings represent a novel identification of polygenic modulation in friend environmental associations with drinking trajectories and can inform at-risk adolescent alcohol prevention/intervention efforts.
Introduction
Adolescence is a critical developmental period for emergence and rapid escalations in alcohol use and consequences. Over 20% of 8th grade students report drinking in the past year, with rates rising to almost 60% by 12th grade (1). Binge drinking (consuming 5+ drinks in a single occasion over the past 30 days) similarly rose from 5% among 14-15 year olds to almost 30% among 18-20 year olds (2). Adolescent alcohol use results in substantial short-term and long-term individual and societal consequences, including interference with brain development, academic impairment, violence, criminal activity, unintentional injury, risky sexual behavior, traffic accidents, and death (3-8).
While alcohol use increases normatively across adolescence (2,9), developmental theories (10,11) and empirical findings (12-15) indicate that adolescents’ drinking behaviors follow distinct patterns, or trajectories, over time. In contrast to traditional approaches that model average behavior across individuals (14,16), it is crucial to model the heterogeneity in adolescent drinking patterns over time. Adolescent drinking trajectories have notable, differing long-term implications. Prospective research suggests that distinct adolescent drinking trajectories are associated with different educational outcomes, family relations, disruptive behaviors, and later alcohol and/or substance use disorders (12,14,17).
The developmentally-specific factors associated with divergent adolescent drinking trajectories need to be better characterized. Alcohol-promoting peer environments (such as perceived friend drinking or affiliation with deviant friends) are particularly salient in adolescence (10,18,19). Despite well-documented associations of friend environments with adolescent drinking (20), associations of friend environments with adolescent drinking trajectories remain unexplored within large prospective, population-based samples, which could improve generalizability and afford power to identity rare trajectory classes.
Genetics can modulate the degree to which socioenvironments influence behavioral outcomes through gene-environment interactions (G×E) (21-23). Within a social control framework (24), permissive environments with high levels of friend drinking and disruptive behavior may provide opportunities to actualize genetic risks by constraining or enabling access to alcohol or opportunities to engage in risky drinking. Adolescents carrying greater genetic risks may be more likely to follow heavier than relatively lighter drinking trajectories when in environments with high friend drinking and disruptive behavior. Genetic differences may be afforded by variants across the genome. Existing G×E research often has examined associations of single candidate genes (25). In contrast, composite genetic risk scores model the aggregate impact of genetic variants on behavior (26) and permit more comprehensive understanding of the genetic underpinnings to alcohol use. Research has yet to apply a composite genetic approach to investigate interaction among genetic risk scores with perceived friend environments on adolescent drinking trajectories within a population-based sample.
This project identified distinct trajectories of adolescent drinking and characterized interactive associations of genetic risk scores and perceived friend environments with trajectory membership. It was hypothesized that associations of friend environments with drinking trajectories would differ as a function of genetics. Specifically, alcohol-conducive environments (with greater friend drinking and disruptive behavior) were hypothesized to magnify genetic risks and increase the likelihood that such adolescents would belong to trajectories of frequent, increasing drinking over the course of adolescence.
Methods
Participants. Data was obtained from the Avon Longitudinal Study of Parents and Children (ALSPAC), a prospective, population-based study (27,28). Pregnant women in the Avon area of Great Britain with an expected delivery date between April 1, 1991 and December 31, 1992 were invited to participate. Mothers, their partners, and their children arising from the index pregnancy were followed up with postal questionnaires and clinic visits. Ethical approval was obtained from the ALSPAC Ethics and Law Committee and the Local Research Ethics Committees. The current study used data from 5,299 Caucasian adolescents with any data on adolescent alcohol use and complete data for the genetic variants examined.
Measures. Adolescents reported on their own drinking frequency at ages 16, 17, 18, and 20 as well as on their friends’ drinking and general disruptive behavior (e.g., “Have any of your friends stolen something?”) at age 15. Drinking frequency was used as a main outcome given it was assessed consistently across adolescence in ALSPAC and has demonstrated high sensitivity and specificity in detecting alcohol use disorder among youth (29). Genetic variants reaching genome-wide significance in associations with weekly drinking within a recent genome-wide association study (30) were used to generate composite genetic risk scores. Adolescent male sex and maternal and/or maternal partner’s heavy drinking were included as covariates.
Data Analytic Strategies: Trajectories of drinking frequency. Mixture modeling was conducted in Mplus, version 7.4 (31) to identify trajectories of drinking frequency from ages 16 through 20. Mixture modeling estimated unobserved subgroups, or classes, within the longitudinal data (32,33). Maximum likelihood estimation with robust standard errors dealt with missing data in the outcome by determining the parameters that maximized the probability of the sample data based on all available data without imputing missing data (34,35). One- through 7-class latent growth mixture models and latent class growth analyses (LCGA) were conducted. The unconditional LCGA with the intercept and linear slope growth factors was retained as the final mixture model. The best-fitting LCGA solution was selected based on low Akaike information criterion (AIC), low sample size adjusted BIC (aBIC), supportive bootstrap likelihood ratio tests (BLRT) comparing k to k-1 class models (36), and interpretability.
Data Analytic Strategies: Genetic and environmental associations with drinking trajectories. Multinomial logistic regression was conducted in Mplus to examine associations of genetic risk score and perceived friend environments on trajectory membership. Using a standard three step method (37), class memberships from the final unconditional LCGA were saved and analyzed in subsequent logistic regression. Two-way interactions of each covariate with genetic and environmental predictors were included to control for confounding effects (38).
Results
Adolescents were 51% male, and most reported any drinking by their friends at 15 years. On average, adolescents increased their drinking frequency over time, from approximately monthly or less at 16 years to 2-4 times a month at 20 years. Greater perceived friend drinking and disruptive behavior at age 15 were significantly associated with more frequent personal drinking, with relatively stronger associations earlier (at 16 years, rs = .24 – .29, p < .001) as compared to later in adolescence (at 20 years, rs = .08 – .14, p < .001).
Trajectories of Drinking Frequency. Regarding unconditional LCGA to identify drinking frequency trajectories across adolescence, model fit indices and comparison tests supported a 5-class solution. The 5-class solution minimized AIC and aBIC, with aBIC reductions from the 4- to 5-class model (∆BIC = 23.65) suggesting very strong preference for the 5-class over the 4-class solution (39). The BLRT also supported the 5-class over the 4-class solution (p < .001). Entropy values suggested medium separation (entropy = 0.63) between the latent classes (33).
Classes were designated as “Low” (n = 193, 4%), “Middle-Low” (n = 1,481, 28%), “Middle-High” (n = 3,173, 60%), “Increaser” (n = 73, 1%), and “High” (n = 379, 7%). Adolescents in the “Low” class generally abstained from drinking or engaged in non-frequent drinking, while adolescents in the “High” class drank approximately 2-3 times each week throughout adolescence. Adolescents in the “Middle-Low” and “Middle-High” classes drank monthly or less or 2-4 times a month at baseline, respectively, increasing their drinking slightly over time. Adolescents in the “Increaser” class drank approximately weekly at baseline yet noticeably increased their drinking to 4 or more times each week by 18 and 20 years.
Genetic and Environmental Associations with Drinking Trajectories. Genetic risk scores modulated associations of friend drinking with membership in the Middle-High (OR = 0.55, 95% CI [0.34,0.87], p = .01) relative to Low class. Specifically, friend drinking was more strongly associated with likelihood of belonging to the Middle-High versus Low class among adolescents carrying lower (b = 2.27, p < .001) compared to higher (b = 1.06, p = .003) genetic risks. In contrast to friend drinking, associations of friend disruptive behavior with class membership did not differ by genetics. Adolescents reporting greater friend disruptive behavior were more likely to belong to the Middle-Low (OR = 1.47, 95% CI [1.30,2.09], p = .03), Middle-High (OR = 2.13, 95% CI [1.51,3.00], p < .001), High (OR = 3.03, 95% CI [2.08,4.43], p < .001), and Increaser (OR = 2.07, 95% CI [1.20,3.57], p = .01) classes than the Low class.
Discussion
This study represented a significant advancement in the application of a developmental framework to G×E research. Specifically, it was the first to identify composite genetic risk scores modulating associations of friend environments to differentiate adolescents with moderately high, slightly increasing drinking frequency (the Middle-High class) from those with stable non-frequent drinking (the Low class) within a large, population-based sample.
Among adolescents carrying low genetic risks, greater friend drinking was strongly associated with likelihood of belonging to the Middle-High relative to Low class. Magnified genetic differences in low versus high alcohol-conducive environments has been shown in candidate G×E investigations (40-42), although often interpreted as stronger genetic protection in low risk environments. Such findings could tentatively be viewed within a differential susceptibility framework (43), where adolescents carrying certain genotypes are more sensitive to environmental influences on drinking. However, research on cumulative genetic plasticity is sparse, and theoretically driven study designs disentangling differential susceptibility and other G×E frameworks are necessary (44). Among adolescents carrying high genetic risks, friend drinking was less strongly associated with class membership. Alcohol-conducive environments may be more influential among high genetic risk youth during early adolescence and diminish with increasing genetic influences by later adolescence as in this investigation, and future research into any temporal patterns is needed.
Adolescents reporting greater friend drinking were more likely to belong to the Middle-Low, High, and Increaser classes than the Low class regardless of genetics. Peer environments have dynamic, reciprocal associations with drinking over time, with adolescents matching their friends’ drinking (peer socialization) and selecting friends compatible with their own behavior (peer selection) (20,45,46). Future investigations into genetic differences in these reciprocal peer socialization and peer selection processes in adolescent drinking patterns are needed.
In contrast to friend drinking, friend disruptive behavior did not modulate genetic associations. This study’s genetic variants were associated with alcohol use, and genes associated with broad externalizing behavior may instead modulate friend disruptive behavior associations. Alternatively, G×E influences on drinking trajectories may require alcohol in the environment – such as physiological responses upon seeing or tasting alcohol in social contexts.
Findings should be interpreted within the context of several limitations. First, causal effects of friend environments on drinking trajectories cannot be inferred. Second, the extent that actual versus perceived friend drinking drives adolescent drinking over time remains unknown, and externally validated measures may be helpful. Third, entropy values suggested some overlap among the latent classes, particularly the Increaser and High classes. Fourth, the genetic risk score included variants reaching genome-wide significance with alcohol use, although a less conservative threshold may have accounted for greater variability in drinking. Finally, future investigations should assess replicability and generalizability.
This study was the first to demonstrate genetic risk scores modulating associations of developmentally salient perceived friend environments with membership in drinking trajectories. Results suggest G×E contributing to divergent drinking patterns, and such work may eventually accelerate intervention efforts by suggesting the utility of developmentally discriminant interventions for at risk subgroups of adolescents.
Impact Statement
Adolescence is a critical developmental period for emergence and rapid escalation of alcohol use and misuse, which has significant short-term and long-term psychosocial, interpersonal, and health consequences. Using longitudinal data from a large, population-based cohort, this project is the first to demonstrate that genetic risk scores can modulate associations of perceived friend environments with membership in divergent drinking patterns across five years of adolescence. Findings help to identify subgroups of adolescents at increased risk for problematic drinking patterns and inform developmentally specific intervention points in adolescent alcohol risk pathways based on genetics and friend environments. Such work can eventually help accelerate the design of prevention and intervention efforts to curtail underage drinking and its grave consequences.
References
1. Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE, Miech RA. Monitoring the Future national survey results on drug use, 1975-2014: Volume 2, College students and adults ages 19-55. Ann Arbor: Institute for Social Research, The University of Michigan; 2015.
2. Substance Abuse and Mental Health Services Administration. Results from the 2013 National Survey on Drug Use and Health: Summary of national findings (NSDUH Series H-48, HHS Publication No. (SMA) 14-4863). Rockville, MD: Substance Abuse and Mental Health Services Administration; 2014.
3. American Medical Association. Harmful consequences of alcohol use on the brains of children, adolescents, and college students. Chicago, IL: American Medical Association, Office of Alcohol and Other Drug Abuse; 2010.
4. McCambridge J, McAlaney J, Rowe R. Adult consequences of late adolescent alcohol consumption: a systematic review of cohort studies. PloS Med. 2011;8(2):e1000413.
5. Hingson R, Kenkel D. Social, health, and economic consequences of underage drinking. In: Bonnie RJ, O’Connel ME, eds. Reducing underage drinking: A collective responsibility. Washington, DC: National Academies Press; 2004:351-382.
6. Marshall EJ. Adolescent alcohol use: risks and consequences. Alcohol Alcohol. 2014;49(2):160-164.
7. National Center for Statistics and Analysis. Alcohol impaired driving: 2014 data (Traffic Safety Facts. DOT HS 812 231). Washington, DC: National Highway Traffic Safety Administration; 2015.
8. Miller JW, Naimi TS, Brewer RD, Jones SE. Binge drinking and associated health risk behaviors among high school students. Pediatrics. 2007;119(1):76-85.
9. Miech RA, Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. Monitoring the Future national survey results on drug use, 1975–2015: Volume I, Secondary school students. In. Ann Arbor: Institute for Social Research, The University of Michigan; 2016.
10. Zucker RA, Fizgerald HE, Moses HD. Emergence of alcohol problems and the several alcoholisms: a developmental perspective on etiologic theory and life course trajectory. In: Cicchetti D, Cohen DJ, eds. Developmental psychopathology: Volume 2: Risk, disorder, and adaptation. New York: Wiley; 1995:677-711.
11. Zucker RA, Hicks BM, Heitzeg MM. Alcohol use and the alcohol use disorders over the life course: a cross-level developmental review. In: Cicchetti D, ed. Developmental psychopathology: Volume Three: Maladaptation and psychopathology. Vol 3. Hoboken, NJ: Wiley; 2016:793-832.
12. Chassin L, Pitts SC, Prost J. Binge drinking trajectories from adolescence to emerging adulthood in a high-risk sample: predictors and substance abuse outcomes. J Consult Clin Psychol. 2002;70(1):67-78.
13. Colder CR, Campbell RT, Ruel E, Richardson JL, Flay BR. A finite mixture model of growth trajectories of adolescent alcohol use: Predictors and consequences. J Consult Clin Psychol. 2002;70(4):976-985.
14. Hill KG, White HR, Chung IJ, Hawkins JD, Catalano RF. Early adult outcomes of adolescent binge drinking: person- and variable-centered analyses of binge drinking trajectories. Alcohol Clin Exp Res. 2000;24(6):892-901.
15. Jackson KM, Sher KJ. Similarities and differences of longitudinal phenotypes across alternate indices of alcohol involvement: A methodologic comparison of trajectory approaches. Psychol Addict Behav. 2005;19(4):339-351.
16. Muthén B, Muthén LK. Integrating person-centered and variable-centered analyses: Growth mixture modeling with latent trajectory classes. Alcohol Clin Exp Res. 2000;24(6):882-891.
17. 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. Dev Psychopathol. 2004;16(1):193-213.
18. Jacob T, Loanord K. Family and peer influences in the development of adolescent alcohol abuse. In: Zucker R, Boyd G, Howard J, eds. The development of alcohol problems: Exploring the biopsychosocial matrix of risk. Rockville, MD: Department of Health and Human Services; 1994:123-155.
19. Kandel DB. On processes of peer influences in adolescent drug use: a developmental perspective. Adv Alcohol Subst Abuse. 1985;4(3-4):139-163.
20. Curran PJ, Stice E, Chassin L. The relation between adolescent alcohol use and peer alcohol use: a longitudinal random coefficients model. J Consult Clin Psychol. 1997;65(1):130-140.
21. Brendgen M. Genetics and peer relations: a review. J Res Adolescence. 2012;22(3):419-437.
22. Plomin R, DeFries JC, Loehlin JC. Genotype-environment interaction and correlation in the analysis of human behavior. Psychol Bull. 1977;84(2):309-322.
23. Rutter M, Moffitt TE, Caspi A. Gene-environment interplay and psychopathology: multiple varieties but real effects. J Child Psychol Psychiatry. 2006;47(3-4):226-261.
24. Chartier KG, Karriker-Jaffe KJ, Cummings CR, Kendler KS. Review: Environmental influences on alcohol use: informing research on the joint effects of genes and the environment in diverse U.S. populations. Am J Addict. 2017;26(5):446-460.
25. Milaniak I, Watson B, Jaffee SR. Gene-environment interplay and substance use: a review of recent findings. Curr Addict Rep. 2015;2(4):364-371.
26. Dudbridge F. Power and predictive accuracy of polygenic risk scores. PLoS Genet. 2013;9(3):e1003348.
27. Boyd A, Golding J, Macleod J, et al. Cohort Profile: The ‘children of the 90s’–The index offspring of the Avon Longitudinal Study of Parents and Children. Int J Epidemiol. 2013;42(1):111-127.
28. Fraser A, Macdonald-Wallis C, Tilling K, et al. Cohort Profile: The Avon Longitudinal Study of Parents and Children: ALSPAC mothers cohort. Int J Epidemiol. 2013;42(1):97-110.
29. Chung T, Smith GT, Donovan JE, et al. Drinking frequency as a brief screen for adolescent alcohol problems. Pediatrics. 2012;129(2):205-212.
30. Clarke TK, Adams MJ, Davies G, et al. Genome-wide association study of alcohol consumption and genetic overlap with other health-related traits in UK Biobank (N=112 117). Mol Psychiatr. 2017;22:1376.
31. Muthén L, Muthén B. Mplus user’s guide. 7 ed. Los Angeles, CA: Muthén & Muthén; 2012.
32. Jung T, Wickrama KAS. An introduction to latent class growth analysis and growth mixture modeling. Soc Personal Psychol Compass. 2008;2(1):302-317.
33. Wickrama KAS, Lee TK, O’Neal CW, Lorenz FO. Higher-order growth curves and mixture modeling with Mplus: A practical guide. New York: Routledge; 2016.
34. Enders CK. The impact of nonnormality on full information maximum-likelihood estimation for structural equation models with missing data. Psychol Methods. 2001;6(4):352-370.
35. Graham JW, Cumsille PE, Elek-Fish E. Methods for handling missing data. In: Schinka JA, Velicer WF, eds. Research methods in psychology. New York, NY: Wiley; 2003:446-454.
36. Asparouhov T, Muthén B. Using Mplus TECH11 and TECH14 to test the number of latent classes. 2012.
37. van de Schoot R, Sijbrandij M, Winter SD, Depaoli S, Vermunt JK. The GRoLTS-Checklist: guidelines for reporting on latent trajectory studies. Struct Equ Modeling. 2017;24(3):451-467.
38. Keller MC. Gene × environment interaction studies have not properly controlled for potential confounders: the problem and the (simple) solution. Biol Psychiat. 2014;75(1):18-24.
39. Raftery AE. Bayesian model selection in social research. Sociol Methodol. 1995;25:111-163.
40. Desalu JM, Zaso MJ, Kim J, Belote JM, Park A. Interaction between ADH1B*3 and alcohol-facilitating social environments in alcohol behaviors among college students of African descent. Am J Addict. 2017;26(4):349-356.
41. Irons DE, Iacono WG, Oetting WS, McGue M. Developmental trajectory and environmental moderation of the effect of ALDH2 polymorphism on alcohol use. Alcohol Clin Exp Res. 2012;36(11):1882-1891.
42. Olfson E, Edenberg HJ, Nurnberger J, Jr., et al. An ADH1B variant and peer drinking in progression to adolescent drinking milestones: evidence of a gene-by-environment interaction. Alcohol Clin Exp Res. 2014;38(10):2541-2549.
43. Belsky J, Pluess M. Beyond diathesis stress: differential susceptibility to environmental influences. Psychol Bull. 2009;135(6):885-908.
44. Roisman GI, Newman DA, Fraley RC, Haltigan JD, Groh AM, Haydon KC. Distinguishing differential susceptibility from diathesis-stress: recommendations for evaluating interaction effects. Dev Psychopathol. 2012;24(2):389-409.
45. Burk WJ, van der Vorst H, Kerr M, Stattin H. Alcohol use and friendship dynamics: selection and socialization in early-, middle-, and late-adolescent peer networks. J Stud Alcohol Drugs. 2012;73(1):89-98.
46. Park A, Sher KJ, Wood PK, Krull JL. Dual mechanisms underlying accentuation of risky drinking via fraternity/sorority affiliation: the role of personality, peer norms, and alcohol availability. J Abnorm Psychol. 2009;118(2):241-255.
