ARTICLE Parental Influence on Substance Use in Adolescent ...

1 ARTICLEP arental Influence on Substance Use in AdolescentSocial NetworksHolly B. Shakya, PhD; Nicholas A. Christakis, MD, PhD; James H. Fowler, PhDObjective:To evaluate the relationship between the par-enting style of an Adolescent s peers parents and an ado-lescent s Substance :Longitudinal :Adolescents across the United States were in-terviewed at school and at :Nationally representative sample of ado-lescents in the United Exposure:Authoritative vs neglectful parentingstyle of Adolescent s parents and Adolescent s friends par-ents and Adolescent Substance Outcome Measures: Adolescent alcohol abuse,smoking, marijuana use, and binge :If an Adolescent had a friend whose mother wasauthoritative, that Adolescent was 40% (95% CI, 12%-58%) less likely to drink to the point of drunkenness, 38%(95% CI, 5%-59%) less likely to binge drink, 39% (95%CI, 12%-58%) less likely to smoke cigarettes, and 43%(95% CI, 1%-67%) less likely to use marijuana than anadolescent whose friend s mother was neglectful, con-trolling for the parenting style of the Adolescent s ownmother, school-level fixed effects, and results were only partially mediated by peer sub-stance :Social network influences may extend be-yond the homogeneous dimensions of own peer or ownparent to include extradyadic influences of the wider net-work.

2 The value of parenting interventions should be re-assessed to take into account these spillover effects in thegreater Pediatr Adolesc Med. 2012;166(12) online October 8, ON ADOLESCENTand adult social networkshas focused on the impactof peers on risk behaviorsinvolving drugs, tobacco,and alcohol may influ-ence individual Substance use behavior viathe prevalence of Substance use within thenetwork as well as the interpersonal dy-namics among network ,10 Theseeffects may have serious consequences; forexample, the probability of a future over-dose is related to both the number of mem-bers of an individual s social network usingdrugs and the degree of conflict within the same time, there is evidence thatparents may Influence adolescents via theirstyle of parenting stylesframework encompasses 4 distinct par-enting categories that are derived from 2dimensions of interaction: (1) parentalcontrol (how much a parent intervenes intheir Adolescent child s life) and (2) pa-rental warmth (how much positive affecta parent shows for their Adolescent ).

3 Au-thoritative parents are warm and commu-nicative, but they also exert appropriatecontrol. Neglectful parents exhibit nei-ther warmth nor control. Authoritarianparents exert control while lackingwarmth, while permissive parents showwarmth but do not exert control. Studiesof these 4 parenting styles suggest that theauthoritative parenting style is optimal,with long-term benefits including aca-demic success, positive peer relation-ships, minimal delinquent behavior, riskavoidance, and positive psychosocial ad-justment, including higher levels of psy-chological withauthoritative parents are also less likely tohave delinquent peer , we explore the possibility thatparenting matters not only because of thedirectandproximaleffectofparentonchil dAuthor AffilDepartmentof MedicineMedical GenPolitical Scie(Dr Fowler)California,SHarvard FacSciences andSchool, Bost(Dr ChristakAuthor Affiliations:Department of Medicine, Schoolof Medicine (Dr Shakya), andMedical Genetics Division andPolitical Science Department(Dr Fowler), University ofCalifornia, San Diego.)

4 AndHarvard Faculty of Arts andSciences and Harvard MedicalSchool, Boston, Massachusetts(Dr Christakis).ARCH PEDIATR ADOLESC MED/VOL 166 (NO. 12), DEC 2012 American Medical Association. All rights From: by a University of California - San Diego User on 01/13/2013but also because of the indirect and more distal relation-shipbetweenparentsandtheiradole scentchildren ,dothebenefitsofgoodparentingspillover,s preading from person to person and affecting multipleadolescentsinanetwork? ,22networkauthoritativeness(anaverageoft hedegree to which the parents of an Adolescent s peers usedauthoritative parenting ) was correlated with a decreasedpropensitytowarddelinquency,low erlevelsofsubstanceabuse, and greater psychosocial competence. To investi-gatethisquestionmorethoroughlyus inglongitudinalanaly-sesandcompletenetwo rkdata,weusetheNationalLon-gitudinalStud yofAdolescentHealth(AddHealth),asourceof data that contains information about Adolescent socialnetworks,theirparents stylesofparenting,andself-reportedmeasur es of Substance abuse.

5 Using longitudinal dyadicnetworkregressionmodels,wemeasured theassociationbetween adolescents behavior and their friends behav-ior,theirmothers parentingstyle,andtheirfriends moth-ers parenting Health is a nationally representative study that explores mul-tiple facets of Adolescent well-being. Four waves of the Add Healthstudy have been completed: Wave I was conducted in 1994-1995 and included adolescents who were then in seventh through12th grade; Wave II, in 1996; Wave III, in 2001-2002; and WaveIV, in 2007-2008. In Wave I of the Add Health study, research-ers collected an in-school sample of 90 118 adolescents cho-sen from a nationally representative sample of 142 part of the survey, these students named up to 5 maleand 5 female friends who were later identified from school-wide rosters to generate information about each school s com-plete social network.

6 A subset of this group was then chosenfor in-depth follow-up in subsequent waves. This in-home sample was administered longer questionnaires about their so-cial networks, health behaviors, family dynamics, and emotional/developmental outcomes. We drew our information about par-enting and Adolescent Substance abuse from the Wave I and IIin-home data dyads were included in each analysis onlyif the observations for both individuals included data on all mea-sures of interest and if the pair indicated that they were friendsfor both Wave I and Wave II. Furthermore, adolescents whoindicated that they were siblings, either full or half, were re-moved from the sample. Questions on maternal warmth werenot asked of individuals for whom no one was acting in the roleof mother (which could include nonbiological mothers suchas aunts or grandmothers).

7 Table 1provides summary sta-tistics for the sample populations. Adolescents in our sample,compared with those in the complete Add Health Wave IIsample, were less likely to be black (13% vs 23%), slightly lesslikely to be Hispanic (13% vs 17%), and similar in likelihoodto be Asian (8% vs ) and came from marginally wealthierhouseholds (mean income, $48 670 vs $46 000) but had simi-lar levels of Parental education (mean, vs years).MEASURESA dolescents in the Add Health data set responded to a battery ofquestions regarding their parents parenting behavior. Parentalcontrol was assessed using yes/no responses to 7 questions fromwhich we created a composite measure,20based on the averageresponses to all 7 questions (Cronbach of .63).20 Adolescentswhose parents were reported to exert less than the median levelof control were categorized as low control.

8 Those more than orequal to the median were categorized as high control. Maternalwarmth was assessed using responses to 5 questions used in on the 5 questions was .85. Warmth, likecontrol, was categorized by placing those at the median level ofwarmth and higher in the high-warmth parenting category andthose less than the median in the low-warmth parenting cat-egory. The combination of the control and warmth categories al-lowed us to define 4 different parenting types20coded as follows:authoritative: high warmth, high control; authoritarian: lowwarmth, high control; permissive: high warmth, low control; andneglectful: low warmth, low control. While Adolescent re-sponses regarding their parents could be biased because of re-spondent error, Steinberg et al23found that Adolescent report wasless biased than parent self-report because parents tend to err to-ward depicting their own behavior in the most positive 1.

9 Summary Statistics for 1386 Respondentsand 1404 Friends%Wave I Value Wave II ValueDrunk in last yearRespondent2629 Friend2931 Cigarette in last monthRespondent2432 Friend3735 Marijuana use in last monthRespondent1113 Friend1416 Binge drinking in last yearRespondent2630 Friend2831 Neglectful parentingRespondent2428 Friend2533 Permissive parentingRespondent2230 Friend2430 Authoritarian parentingRespondent2422 Friend2318 Respondent age, y, mean (SD) ( )Female51 Household income, 1000s ofdollars, mean (SD) ( )Parent s education, mean (SD) ( )Hispanic13 Black13 Asian8aParent s education is a 10-item scale: 0 = never went to school;1 = eighth grade or less; 2 = more than eighth grade but did not graduatefrom high school; 3 = went to a business, trade, or vocational school insteadof high school; 4 = high school graduate; 5 = completed a General EducationDevelopment program; 6 = went to a business, trade, or vocational schoolafter high school; 7 = went to college but did not graduate; 8 = graduatedfrom a college or university; and 9 = professional training beyond a 4-yearcollege or PEDIATR ADOLESC MED/VOL 166 (NO.)

10 12), DEC 2012 American Medical Association. All rights From: by a University of California - San Diego User on 01/13/2013In a comprehensive section on Substance use, adolescentswere asked a variety of detailed questions about prior and cur-rent Substance use related to alcohol use, cigarette smoking,marijuana use, and binge drinking. We coded 4 separate di-chotomous Substance abuse outcomes from questions asked inWaves I and II to represent either having engaged in the be-havior or not. Details on variable coding are in the eAppendix( ).To identify the networks, we treated each friendship nomi-nation as a directed tie from the namer to the named called interviewed individuals adolescents and the peoplewho they named friends. Dyadic observations were createdso that each observation included data from both an adoles-cent and a friend at Waves I and II for Adolescent -friend pairsobserved in the data.