Sex Among Adolescents Chile

7/29/2019 Sex Among Adolescents Chile

1/8

Rev Panam Salud Publica 28(4), 2010 267

Sexual intercourse among adolescents inSantiago, Chile: a study of individual andparenting factors

Ninive Sanchez,1 Andrew Grogan-Kaylor,1 Marcela Castillo,2

Gabriela Caballero,2 and Jorge Delva1

Objective. To examine a range of individual, parenting, and family factors associated withsexual intercourse among a community sample of youth and their families in Santiago, Chile.Methods. Data were taken from the Santiago Longitudinal Study conducted in January2008November 2009. Participants were 766 youth (mean age = 14.03 years, 51% male) frommunicipalities of low- to mid-socioeconomic status. Variables included emotional and behav-ioral subscales from the Child Behavior Checklists Youth Self Report, parental monitoring,

family involvement, parental control and autonomy, relationship with each parent, and sexualactivity. Bivariate and multivariate logistic regression models were used to examine the oddsof sexual intercourse initiation.Results. Seventy (9.14%) youth reported having had sex in their lifetime; the average

age of first sexual intercourse among this group was 13.5 years (Standard Deviation [SD] =1.74) for males and 14.08 (SD = 1.40) for females. Having sex was inversely associatedwith withdrawn-depressed symptoms (Odds Ratio [OR] = 0.84, Confidence Interval [CI] =0.720.97), but positively associated with somatic complaints (OR = 1.20, CI = 1.041.38)and rule breaking behavior (OR = 1.21, CI = 1.081.36), after adjusting for demographic andother individual and parenting variables. The majority (80%) of the youth who had had sex re-ported using protection at the time of last intercourse.Conclusions. Findings highlight the role that mental health problemssome of them notcommonly associated with onset of sexual activitymay play in a youths decision to have sex.The potential protective effects of several parenting and family characteristics disappeared with

youth age and youth behavioral problems.

Sexual behavior; adolescent; adolescent behavior; pregnancy in adolescence; Chile.

ABSTRACT

Research among youth in Chile hasfound that some adolescents perceivesexual intercourse as an opportunity toexperience positive outcomes, such as

providing love and affection to a partnerand appeasing ones curiosity (1). How-ever, sexual intercourse among youth,especially early sexual onset, can leadto a considerable number of problem-atic outcomes, ranging from decreasedacademic achievement (2) to sexually-transmitted diseases (STDs) (3) and teenpregnancy (1). Pregnant students inChile who dropped out of school re-

ported two major reasons for discontin-uing their education: being ashamed of

being pregnant and pregnancy-relatedcomplications (4). And although school-level interventions aimed at preventingadolescent pregnancy have been triedand tested (5, 6), overall effectivenessappears to be temporary (6). Moreover,the school curricula in Latin Americaoften lacks information on various as-

Key words

Investigacin original / Original research

Sanchez N, Grogan-Kaylor A, Castillo M, Caballero G, Delva J. Sexual intercourse among adolescentsin Santiago, Chile: a study of individual and parenting factors. Rev Panam Salud Publica. 2010;28(4):26774.

Suggested citation

1 School of Social Work, University of Michigan,Ann Arbor, Michigan, United States of America.Send correspondence to Ninive Sanchez, [email protected].

2 Instituto de Nutricin y Tecnologa de los Alimen-tos, Universidad de Chile, Santiago, Chile.


2/8

pects of sexual behavior, including HIVprevention (7).

The current situation has led somehealth professionals in Chile to suggestthat family interventions, such as homevisits, may be promising ways to man-age teen pregnancy and other adolescentsexual behaviors affecting families (8).

Targeting the home requires learningmore about the role that parents andfamilies play in preventing adolescentpregnancy. To better inform potentialfamily interventions for promotinghealthy sexual behavior among adoles-cents, this study examined a number ofindividual, parenting, and family vari-ables and how they relate to adolescentsexual intercourse.

A review of the literature suggeststhere is a gap in the research of the day-to-day family context regarding adoles-cent sexual behavior in Chile. Of the

studies available, one found that youthwho perceived their families as dysfunc-tional were more likely to have sex andfeel insecure, sad, and anxious (9). Themajority of research on sexual behaviorat the family level has focused on thespecifics of what parents tell their chil-dren about sex, rather than examiningthe larger family context in which thistype of communication occurs (10). Forexample, results of a study conducted inChile indicate that a mothers disap-proval of premarital sex, along with agood mother-daughter relationship, de-

creased the likelihood of early sexualonset (6). However, that study excludedmales and measured only a general sat-isfaction with the mother-daughter rela-tionship, not other potentially importantfamily and parenting variables, such asparental involvement, monitoring, and/or control. Other research among fami-lies in Chile found that a fathers pres-ence in the household decreased the like-lihood of sexual onset for females, butnot males (11). Again, the study did notexamine other parenting and family fac-tors or the relationship between fathers

and their children.Expanding on the current understand-

ing of this topic, the present study exam-ines individual factors (i.e., behavioralproblems) and parenting and family fac-tors that may be associated with sexualintercourse among youth in Chile. TheSantiago Longitudinal Study (SLS) is anon-going collaboration between the Uni-versity of Michigan (Ann Arbor, Michi-gan, United States) and the Institute of

Nutrition and Food Technology (Univer-sity of Chile, Santiago, Chile [INTA]),with funding from the National Instituteon Drug Abuse (Bethesda, Maryland,United States). More details about thestudy are provided elsewhere (12). Ap-proval was granted by the institutionalreview boards of both the University of

Michigan and INTA.

MATERIALS AND METHODS

Participants and data collection

This study used cross-sectional datafrom the SLS, a study of community-dwelling adolescents in Santiago, Chile(12). Study participants were recruitedfrom a convenience sample of approxi-mately 1 100 families that had partici-pated in a study of iron and nutritionalstatus when the youth were infants, 5,

and 10 years of age (13). The presentstudy reports findings based on the 766youth (mean age = 14.03 years, 51% male)who completed a comprehensive assess-ment of substance use in January 2008November 2009 and had no missing data.The participants were from municipali-ties with low- to mid-level socioeconomicstatus. The study was specifically de-signed to understand youth develop-ment among those who may experiencemore social and economic disadvantages.

Study instrument

Adolescent participants completed a2-hour interviewer-administered ques-tionnaire with standardized measuresthat were pilot tested and validated withthe population under investigation priorto conducting the present study. Inter-views were conducted in Spanish in aprivate office at INTA. Interviewerswere Chilean psychologists trained inthe administration of standardized in-struments. Youth assent and parentalconsent were obtained by the psycholo-gists prior to commencing the inter-

views. Parents were not present whenyouth were interviewed.

The questionnaire, which consists ofnearly 900 questions, was created bycombining standardized instrumentscommonly used in the United States andin Chile to assess the constructs mea-sured in the study (see the Measures sec-tion below). To ensure language andconceptual equivalence, instrumentsthat only existed in the English language

were first translated into Spanish bythree Chilean educators who reside inthe United States, one of whom was thestudys Principal Investigator. Eachtranslated instruction and stem questionwas subsequently reviewed in Chile bythe Co-Investigator and three of thestudy interviewers. Following modifica-

tions, the entire instrument was pilottested with 30 youth (representative ofthe study participants) and finalized

based on their feedback.Topics assessed measured the adoles-

cents relationship with parents, the ado-lescents perceptions of self, their behav-ior, health status, any substance use, andmore.

Measures

Sexual intercourse. The dependent vari-able in this studywhether or not the

youth had had sexwas assessed fromthe question, Have you ever had sexualintercourse (made love or gone all theway)? This information was used to cre-ate a dichotomous indicator of whether ornot a youth had had sex.

Demographic characteristics. Demo-graphic variables included the youthsself-reported age and gender, and a par-ents report of the familys socioeco-nomic status (SES). The SES instrumentwas completed by the parental figurewho accompanied the youth to the inter-

view site. This instrument was separatefrom the youth questionnaire. The SESvariable was based on a measure sensi-tive to SES circumstances in the develop-ing world (14) and consisted of 86 ques-tions, of which, 13 were specifically usedto obtain a composite score of SES. Ex-amples of items assessed were totalnumber of adults in the same house,type of job held by the head of house-hold, and fathers level of education(Cronbachs = 0.88).

Youth behavior. Youth assessed their be-

havior during the preceding 6 monthsusing the Child Behavior Checklists 112-item Youth Self Report (CBCL/YSR) (15).The CBCL/YSR uses the following re-sponse scale: 0 (not true), 1 (somewhator sometimes true, and 2 (very true oroften true). Subscale scores are based onthe sum of the items in that scale. Alleight subscales were used in this study:anxious-depressed (Cronbachs = 0.74),withdrawn-depressed (0.68), somatic

268 Rev Panam Salud Publica 28(4), 2010

Original research Sanchez et al. Sexual intercourse among adolescents in Chile


3/8

complaints (0.66), social problems (0.61),thought problems (0.62), attention prob-lems (0.64), rule breaking behavior (0.70),and aggressive behavior (0.81).

Youths relationship with each parent.Youth rated their relationships withtheir mother and father, separately,

using the same 17-item scale for eachparent (16, 17). The stem question was:When you and your mother (or father)spend time talking or doing things to-gether, how often does she (he) andwas paired with items such as . . . helpyou do something that is important toyou? and . . . listen carefully to yourpoint of view? Items were rated on a 4-point scale (1 = never, 2 = sometimes, 3 =often, 4 = always). Higher scores repre-sented a more positive relationship withthe parent (Cronbachs = 0.88). TheCronbachs alpha was the same for both

the relationship with mother scale andthe relationship with father scale.

Parental control and autonomy. Youthrated their perception of parental controland autonomy using an 8-item scale withitems that asked them to report how deci-sions in their families were made (1719).The stem question: In your family,how do you make most of the decisionsabout . . . : was followed by topic items,such as . . . how late you can stay up ona school night, . . . which friends youcan spend time with, and . . . how you

dress. Items were rated on a 5-pointscale (1 = my parents decide, 2 = my par-ents decide after discussing it with me,3 = we decide together, 4 = I decide afterdiscussing it with my parents, 5 = I de-cide all by myself) ( = 0.70).

Family involvement. This measure,taken from the Child Health and IllnessProfileAdolescent Edition (CHIP-AE)(20, 21), refers to the level of parentalmonitoring by those whom the adoles-cent considers to be parent figures. Youthrated their perceptions of family involve-

ment using a 5-item scale. The stem ques-tion: Thinking about your family, abouthow many days in the past 4 weeks didyour parents or other adults in your fam-ily. . . was followed by: . . . talk withyou or listen to your opinions and ideas,. . . spend time with you doing some-thing fun, . . . eat meals with you. Thescale also included, In the past 4 weeks,on how many days did you like being amember of your family? and . . . did

you and your family get along? Itemswere rated on a 5-point scale (1 = 0 days,2 = 13 days, 3 = 46 days, 4 = 714 days,5 = 1528 days). Higher scores repre-sented more family involvement (Cron-

bachs = 0.72).

Parental monitoring. Youth rated their

perceptions of their parents monitoringusing a 7-item scale (22) that includeditems such as: How often would yourmom/dad/guardian know if you camehome an hour late on weekends?; Arethere kids your mom/dad/guardiansdont allow you to hang out with?; andHow often, before you go out, do you tellyour mom/dad/guardian when you will

be back? Items were rated on a 5-pointscale (1 = never, 2 = hardly ever, 3 = some-times, 4 = most times, 5 = always). Higherscores represented more parental moni-toring (Cronbachs = 0.68).

Sexual activity. Sexual activity was mea-sured with questions taken from theCHIP-AE (20, 21), with the followingquestions: How old were you when youhad sexual intercourse for the firsttime?; How many people of the oppo-site sex have you had sex with?; andHow many people of the same sex haveyou had sex with?

As a follow-up question, those youthwho had sexual intercourse with oppo-site sex partners were asked, Did you oryour partner use something to prevent

pregnancy or sexually-transmitted dis-eases the last time you had sexual inter-course? Those youth who respondedyes were asked, Which of the follow-ing things did you use to prevent preg-nancy or sexually-transmitted diseasesthe last time you had sexual inter-course? Youth responded with yes orno to each of the following choices:the pill/Norplant/Depo provera; foam/cream/jelly/suppository; diaphragm/sponge; rubber/condom; withdrawal/pulling out; the morning after pill; and,something else. In addition, adolescents

were asked, Have you ever beenpregnant/gotten someone pregnant?

Finally, youth were asked, Has a doc-tor ever told you that you had a sexually-transmitted disease or venereal diseaselike gonorrhea, syphilis, Chlamydia,genital warts, or genital herpes? Itemswere rated on a 3-point scale (1 = no,never; 2 = yes, but no problems with it inthe last 12 months; 3 = yes, and had aproblem with it in the last 12 months).

Statistical analyses

Four sets of analyses were tested usinglogistic regression. First, the bivariate as-sociations of each independent variablewith the dependent variable (ever hadsex) were examined. Model 1 examinedthe multivariate association of ever hav-

ing had sex with the behavioral problemvariables while statistically adjusting forthe demographic characteristics of studyparticipants, using multivariate logisticregression.

Model 2 examined the multivariate as-sociation of ever having had sex withthe parenting and family variables alsoadjusting for demographics, using multi-variate logistic regression.

The Full Model examined the associa-tion of ever having had sex with the indi-vidual, parenting, and family variablesentered simultaneously while adjusting

for demographic characteristics.Analyses were conducted with STATA

10.0 (Stata Corporation, College Station,Texas, United States). As a follow-up tothis set of logistic regression models, thesexual behavior of youth who reportedhaving had sexual intercourse was exam-ined, to provide a more complete de-scription of these behaviors. However,the relatively small number of youth whoreported having had sex precluded test-ing of models using multivariate statis-tics. Only descriptive statistics could beemployed in these follow-up analyses.

RESULTS

The study analyses included only the766 youth who had both a mother andfather figure in their lives. The majorityhad a biological mother (80.6%). Theanalyses also excluded any youth withmissing data. The average age of thestudy participants was 14 years (SD =1.2). About half of the sample was male(51.2%). The average score on the SESscale was 32.7 (Table 1).

Seventy (approximately 9%) of the 766

youth reported having had sex in theirlifetime. The average age of youth whoreported having had sex (15.5 years,SD = 1.2) was significantly higher thanthat of youth who reported never havinghad sex (13.9 years, SD = 1.1, P < 0.001).However, of the 9% who had had sex,the percent of males (48.5%) and females(51.4%) was fairly similar. In addition,the difference in the average age of firstsexual intercourse between males (13.5


Sanchez et al. Sexual intercourse among adolescents in Chile Original research


4/8

years, SD = 1.7) and females (14.1 years,SD = 1.4) was not statistically different.

Of the 70 youth who reported havinghad sex, 56 (80%) reported using protec-tion to prevent pregnancy/STDs at thetime of last sexual intercourse. The meanage of first intercourse for those whoreported using protection was 14 years

(SD = 1.39); versus 12.9 years (SD = 2.04)for those who reported not using protec-tion, a statistically significant difference(P < 0.05). Of the 56 who used protection,29 were male (almost 50%), and 27 werefemale. In addition, of the 56 youth whoused protection, 36 (64%) used one formof contraceptive. Eighteen (32%) youthused two different types of contracep-tives, and two (0.04%) youth used threedifferent types of contraceptives at thetime of last sexual intercourse. The pre-ferred form of contraception was the con-dom, followed by the pill and with-

drawal. Of the youth who reportedhaving had sex none reported ever hav-ing been told by a doctor that they had asexually-transmitted disease (STD).

Sixty-seven of the 70 adolescents(95.7%) reported having had sex with theopposite sex. Most of these adolescents

(52 individuals) also reported having hadsex with one person. Three adolescentsreported having had sex with the samesex. Two of these reported having hadsex with one person. Finally, five totaladolescents reported having ever beenpregnant/gotten someone pregnant.

Bivariate analyses

These analyses examined the associa-tions between youth ever having had sexand demographic characteristics, mentalhealth and behavioral problems, andparenting and family characteristics. Theresults are shown in Table 2. As ex-pected, age was positively associatedwith increased odds of youth ever hav-ing had sex (Odds Ratio [OR] = 3.23, 95%Confidence Interval [CI] = 2.514.15).Other demographic variables were notassociated with youth having had sex.

Several behavioral problems were as-sociated with the likelihood of youthhaving had sexual intercourse. The oddsof ever having had sex increased withyouth having higher levels of somaticcomplaints (OR = 1.13, 95%CI = 1.041.24), thought problems (OR = 1.12,

95%CI = 1.021.24), rule breaking behav-ior (OR = 1.21, 95%CI = 1.131.29), andaggressive behavior (OR = 1.07, 95%CI =1.021.12). The other behavioral prob-lems assessed in this study were not sig-nificantly associated with the dependentvariable.

The majority of the parenting and fam-

ily factors were significantly associated,as expected, with youth having had sex-ual intercourse; that is, the odds of everhaving had sex were lower among youthwho reported having a better relationshipwith their mother (OR = 0.97, 95%CI =0.940.99) and father (OR = 0.95, 95%CI =0.930.98), and more parental monitoring(OR = 0.93, 95%CI = 0.890.97). In addi-tion, youth who experienced more auton-omy and less parental control had higherodds of ever having had sex (OR = 1.12,95%CI = 1.061.17).

Multivariate models

Mental health and behavioral factors.Model 1 included associations of mentalhealth and behavioral factors with youthssexual intercourse while adjusting fordemographic controls (Table 2). The odds



TABLE 1. Self-reported descriptive statistics of a sample of 766 youth from low- and middle-income municipalitieswho participated in a study of substance abuse in Santiago, Chile, 20082009

StandardVariable No. % Mean deviation Minimum Maximum

DemographicsMale 392 51.7Age 14.0 1.22 11.3 17.7

years years years

Socioeconomic statusa 32.75 6.53 18 53

Sexual behaviorEver had sex 70 9.14Age at first sexual intercourse 70 13.8 1.60 9.0 17.0

years years years

Mental health/behavioral problemsAnxious-depressed 6.00 3.71 0 20Withdrawn-depressed 4.17 2.75 0 14Somatic complaints 3.10 2.46 0 14Social problems 4.03 2.70 0 18Thought problems 2.56 2.25 0 16Attention problems 5.61 3.04 0 17

Rule breaking behavior 4.92 3.23 0 22Aggressive behavior 8.04 4.80 0 25

Parenting/FamilyRelationship with mother 55.85 8.03 23 68Relationship with father 54.75 8.99 18 68Parental control and autonomy 29.75 6.17 8 40Family involvement 18.80 4.15 5 25Parental monitoring 27.80 5.23 7 35

a Socioeconomic status was measured using a composite score of the following items: father abandonment, head of households employment, fathersbenefits, fathers education, land arrangement, type of housing, kitchen appliances, type of waste disposal system, type of access to drinking water,frequency of trash pickup, crowdedness in the home, and amount and type of belongings in the home.


5/8

ratios associated with age (OR = 3.57,95%CI = 2.674.78), somatic complaints(OR = 1.20, 95%CI = 1.041.38), and rule-

breaking behavior (OR = 1.22, 95%CI =1.10 1.35) not only remained significant,

but also increased slightly. Thought prob-lems were no longer significantly associ-ated with ever having had sex after demo-

graphics and other behavioral problemswere included in the model. In addition,in this multivariate model, the magnitudeof the coefficient estimating the asso-ciation between ever having had sex andwithdrawn-depressed increased, result-ing in a significant, inverse association(OR = 0.84, 95%CI = 0.730.97). The Chi-square for the overall model was 153.32(P < 0.001).

Parenting and family factors. Model 2 in-cluded associations of parenting and fam-ily factors with youths sexual behavior

while adjusting for demographic controls(see Table 2). The associations between allof the parenting variables with ever hav-ing had sex became non-significant withthe exception of parental monitoring.Parental monitoring remained signifi-cantly and inversely associated with theodds of youth ever having had sex (OR =0.94, 95%CI = 0.88 1.00). The Chi-squarefor the model was 123.86 (P < 0.001).

Full Model. The Full Model included de-mographics, and individual, parenting,and family factors. Age continued to bepositively associated with having hadsex (OR = 3.65, 95%CI = 2.69 4.96). Inthis multivariate model, the odds of everhaving had sex remained inversely asso-ciated with being withdrawn-depressed

(OR = 0.84, 95% CI = 0.72 0.97) and pos-itively associated with experiencing so-matic complaints (OR = 1.20, 95%CI =1.04 1.38) and rule breaking behavior(OR = 1.21, 95%CI = 1.081.36). In the fullmodel, none of the parenting variablesremained significantly associated withever having had sex. The Chi square forthis model was 158.47 (P < 0.001).

DISCUSSION

Mental health

The study results suggest that there isa positive association between some ofthe behavioral problems youth may haveand the likelihood of ever having had sex.These findings indicate that youth ex-periencing higher levels of somatic com-plaints and less withdrawn-depressionhad higher odds of having had sex, evenafter accounting for age, other behav-ioral problems, and parenting factors.

The somatic complaints measured in thisstudy were complaints for which nomedical cause could be ascertained.

One could argue that the somatic com-plaints are symptoms of the withdrawn-depression experienced by some youth.However, by including both somaticcomplaints and withdrawn-depression in

the model, this study controlled for po-tential comorbidity. It may be that the so-matic complaints are independent ofwithdrawn-depression. For instance, re-search suggests that children can learn touse somatic complaints to draw attentionfrom family members when they are un-able to draw attention for emotional prob-lems (23). In addition, research suggeststhat somatic complaints may be morelikely to occur in households where it ismore acceptable to express ones somaticcomplaints rather than ones feelings (23).

The finding that youth experiencing

withdrawn-depression had lower oddsof having had sex is somewhat inconsis-tent with what was found in Brazil (24).Researchers found that in Brazil, femaleadolescents 1317 years of age who werepregnant for the first time experiencedsignificantly higher levels of withdrawn-depression (as measured by CBCL/YSR)compared to females who were sexuallyactive, but not pregnant (24).



TABLE 2. Logistic regression analyses of youths likelihood of sexual onset among 766 youth from low- and middle-income municipalities who par-ticipated in a study of substance abuse in Santiago, Chile, 20082009

Bivariate

Odds 95% confidenceratio interval Model 1 Model 2 Full Model

Variable (OR) (CI) OR CI OR CI OR CI

Demographics

Age 3.23 2.514.15 3.57 2.674.78 3.16 2.414.13 3.65 2.694.96Sex 1.12 0.861.83 0.93 0.481.83 0.95 0.531.68 0.97 0.491.92Socioeconomic status 0.98 0.941.01 0.99 0.951.04 0.99 0.951.04 0.99 0.951.04

Mental health/behavioral problemsAnxiousdepressed 1.01 0.941.08 0.96 0.851.08 0.95 0.841.08Withdrawndepressed 0.97 0.891.07 0.84 0.730.97 0.84 0.720.97Somatic complaints 1.13 1.041.24 1.20 1.041.38 1.20 1.041.38Social problems 0.99 0.911.09 0.96 0.821.12 0.96 0.821.12Thought problems 1.12 1.021.24 1.12 0.971.30 1.34 0.981.32Attention problems 1.07 0.981.15 0.99 0.871.13 0.99 0.871.13Rule breaking behavior 1.21 1.131.30 1.22 1.101.35 1.21 1.081.36Aggressive behavior 1.07 1.021.12 0.99 0.911.09 0.99 0.901.08

Parenting/familyRelationship with mother 0.97 0.940.99 1.00 0.961.04 1.02 0.971.06Relationship with father 0.95 0.930.98 0.97 0.941.01 0.97 0.931.00Parental control and autonomy 1.12 1.061.17 1.01 0.951.06 0.98 0.921.04

Family involvement 0.86 0.671.09 1.03 0.951.11 1.03 0.941.12Parental monitoring 0.93 0.890.97 0.94 0.881.00 0.96 0.901.03

Note:Bold cells indicate statistical significance at the P< 0.05 level


6/8

Behavioral problems

Rule breaking behaviors was anothervariable that was consistently associatedwith youth having had sex, even afteraccounting for age, other behavioralproblems, and parenting and family fac-tors. This is consistent with research

using a nationally representative sampleof youth in the United States whichfound that higher levels of externalizing

behavior, including rule breaking, wasassociated with increased likelihood ofsexual onset (25). Youth who receivevery limited support from family andwho engaged in rule breaking behaviorsmight look to others, including romanticpartners, for support. In turn, these part-ners may then influence youths behav-ior and sexual activity. For example,prior research has found that the delin-quent behavior of a youths romantic

partner is associated with the youthsdelinquent behavior (26).

Parenting and family factors

Another interesting finding is that inthe Full Model, parenting and family fac-tors were no longer significantly associ-ated with sexual intercourse. Bivariateresults suggested that youth who hadhad sex had lower mean levels on re-lationship with mother and father,parental monitoring, and parental con-trol, than youth who had not had sex.

However, no measure of relationshipwith families and parents was statisti-cally significant in a multivariate model,suggesting that mental health issues mayplay a larger role in the onset of sexual

behavior than parenting and family fac-tors. However, this does not indicate thatthe family is not an important target forintervention in adolescents sexual activ-ity. Research focused predominantly onyouth in the United States suggests thatpositive parenting practicesparentalmonitoring and high quality parent-child relationshipshave a protective

influence on youths risk-taking behav-iors, including sexual behavior (27). It ispossible that mental health problemsmay mediate the relationship betweenfamilies and sexual activities.

Home visits

Some mental health professionals sug-gest that home visits made by facilitators

to adolescents and their families may bean effective way to encourage families tointeract and discuss issues that are im-portant, including sexual behavior (28).In addition, these facilitators are likely tofacilitate families access to health andhuman services (28).

A program using home-visit facilita-

tors may be a good option for families inChile. In fact, health and mental healthworkers in Chile have found that someteens are scared to access public healthservices, possibly due to the stigma asso-ciated with teen pregnancy (8). Anotherreason for the home-visit option is thatindividuals and families looking for ser-vices sometimes make initial contactwith individuals who are unable or nottrained to engage them at first contact;whereas, trained home-visit facilitatorsmay be able to discuss and provide guid-ance on various issues experienced by

the family (28). Even as it relates to thefindings of the present study, which in-dicated that most of the youth who hadhad sex also reported using protectionand none reported an STD, a home-visitfacilitator could explain the possibilityof an asymptomatic infection. Anothertopic of discussion might be the risks in-volved with having multiple sex part-ners without protection.

Limitations

As is the case with all studies, this

study had some limitations that must benoted. One of the limitations is the cross-sectional design that limits the directionsof the associations found. Further re-search is needed to examine whetheryouths withdrawn-depressive, somatic,and rule breaking behavior follows sex-ual intercourse or precedes it. It is possi-

ble that somatic complaints may resultfrom having sexual intercourse. Youthmay also engage in rule breaking or ex-perience withdrawn-depressive symp-toms as a result of shame, regret, or otheroutcomes that result from the sexual

experience.A second limitation is that some of the

youths mental health and behavior mea-sures had Cronbachs alphas lower than0.70. Measures with lower reliabilitiesmay impede ability to discern statisti-cally significant constructs.

A third limitation has to do with theuse of the CBCL/YSR with the samplein this study. The CBCL/YSR was not

developed solely for use with Chileanyouth, and as such, some of the behav-ioral problems of Chilean youth maynot have been captured by this mea-sure. Nevertheless, the CBCL/YSR is ameasure that is easy to administer andthat includes one of the most compre-hensive sets of behavioral problems

youth may experience. The CBCL/YSRhas been successfully used in a widevariety of international settings, includ-ing with many Spanish-speaking popu-lations, and is broadly considered to bea sufficiently reliable and valid instru-ment for use with diverse populations(29, 30).

The fourth limitation is that all thedata are based on youth self-report.Youth may have either underreported oroverreported sexual behaviors. Youthmay feel more comfortable and reportmore accurately when responding

through audio and computer support(31). However, it is important to notethat these youth and their parents had

been interviewed by INTA psychologistsseveral times in the past and a rapporthad been established. In addition, theseparticipants have reported trusting theseprofessionals and the work affiliatedwith INTA (32).

Finally, the study does not measurewhether the parents are biological, adop-tive, or step-parents, and the family dy-namics of two-parent families may differsubstantially from that of single-parent

families. Future research is needed withdiverse families.

Future research

Notwithstanding these limitations, thestudy findings contribute to the researchon individual and family factors associ-ated with sexual intercourse in a com-munity sample of youth in Chile. Thisstudy also suggests that sexual activityin relation to youths withdrawn-depressed, somatic complaints, and rule-

breaking behavior is an area for further

study. A number of important associa-tions were identified, but a future studycould include other sources, such aspeers, teachers, and/or parents, whowould provide additional perspectivesnot assessed in the present study.

Future research might also examineother factors, such as the importance ofvirginity and satisfying sexual needs(33), along with other measures of par-




7/8

enting among international populations.Research is also needed to examinewhether for some youth, having sex andexperiencing its outcomes, both positiveand negative, provides skills necessaryfor managing other aspects of life. In ad-dition, a greater understanding of theseassociations and behaviors in countries

other than the United States can serve to

enhance the overall understanding ofadolescent sexual development.

Acknowledgments. The authors wishto particularly thank the youth andtheir families for their participationin this study. We are also thankful tothe Santiago Longitudinal Study staff

for their dedication to the project.

This study was supported by a grantfrom the National Institute of DrugAbuse (DA021181) (Bethesda, Mary-land, United States) and received sup-port from the Vivian A. and James L.Curtis School of Social Work Researchand Training Center (Ann Arbor, Michi-gan, United States).



1. Gonzalez EA, Molina, MG, Montero A,Martnez NV, Leyton MCl. Sexual behaviorand gender differences among adolescentsconsulting at a university public health sys-tem. Rev Med Chile. 2007;135:12619.

2. Coker AL, Richter DL, Valois RF, McKeown,RE, Garrison CZ, Vincent ML. Correlates andconsequences of early initiation of sexual inter-

course. Journal Sch Health. 1994;64(9):3726.3. Kaestle CE, Halpern CT, Miller WC, Ford CA.

Young age at first sexual intercourse and sex-ually transmitted infections in adolescentsand young adults. Am J Epidemiol. 2005;161(8):77480.

4.Molina M, Ferrada C, Perez R, Cid L,Casanueva V, Garcia A. The relationship be-tween teenage pregnancy and school deser-tion. Rev Med Chile. 2004;132:6570.

5. Cabezn C, Vigil P, Rojas I, Leiva ME,Riquelme R, Aranda W. Adolescent preg-nancy intervention: an abstinence-centeredrandomized controlled intervention in aChilean public high school. J Adolesc Health.2005;36:649.

6. Silva M, Ross I. Association of perceived

parental attitudes towards premarital sexwith initiation of sexual intercourse in adoles-cence. Psychol Rep. 2002;91:7814.

7. DeMaria LM, Galrraga O, Campero L,Walker DM. Sex education and HIV preven-tion: An evaluation in Latin America and theCaribbean. Rev Panam Salud Publica. 2009;26(6):48593.

8. Chereau BM, Silva CB, Alvarez MA. An inte-grative perspective of adolescent pregnancy:Home visitation program as an interventionstrategy. Rev Psicol. 2001;001:2143.

9. Santander S, Zubarew T, Santelices L, ArgolloMP, Cerda LJ, Brquez PM. Family influenceas a protective factor against risk behaviors inChilean adolescents. Rev Med Chile. 2008;136:31724.

10. Elliot S. Parents constructions of teen sexual-ity: Sex panics, contradictory discourses, andsocial inequality. SI. 2010;33(2):191212.

11. Murray NJ, Zabin LS, Toledo-Dreves, et al.Gender differences in factors influencing firstintercourse among urban students in Chile.Int Fam Plan Perspect. 1998;24(3):13952.

12. Delva J, Castillo M. International Research. InB. Thyer (Ed.), Handbook of Social Work Re-search Methods (2nd ed.). CA: Sage; 2010.

13. Lozoff B, De Andraca I, Castillo, M, Smith, JB,Walter, T, Pino, P. Behavioral and develop-mental effects of preventing iron-deficiencyanemia in healthy full-term infants. Pedi-atrics. 2003;(4)112(4):84654.

14. Graffar, M. Une methode de classification so-ciale dechantillons de population. Courrier.1956;6(8):4559.

15. Achenbach TM, Rescorla LA. Child BehaviorChecklist. Youth Self-Report for Ages 1118(YSR 1118). Manual for the ASEBA school-age forms and profiles. Burlington, VT: Re-search Center for Children, Youth, and Fami-lies; 2001.

16. Conger RD, Ge X. Conflict and cohesion inparent-adolescent relations: changes in emo-tional expression from early to mid-adolescence. In: Conflict and cohesion in fami-lies: Causes and consequences. Cox M, Brooks-Gunn J, eds. Mahwah, NJ: Erlbaum, 1999:185206.

17. National Institute of Child Health andHuman Development. NICHD study of earlychild care and youth development. Phase IVInstrument Documentation. Available from:

https://secc.rti.org/Phase4InstrumentDoc.pdf. Accessed 20 October 2009.

18. Brody GH, Moore K, Glei D. Family processesduring adolescence as predictors of parent-young adult attitude similarity. A 6-year lon-gitudinal analysis. Family Relations. 1994;43:36973.

19. Eccles J, Buchanan CM, Midgley C, FuligniAJ, Flanagan C. Individuation reconsidered:Autonomy and control during early adoles-cence. J Soc Issues. 1991;47:5368.

20. Riley AW, Green BF, Forrest CB, Starfield B,Kang M, Ensminger ME. A taxonomy of ado-lescent health: development of the adolescenthealth profile-types. Med Care. 1998;36(8):122836.

21. Riley AW, Forrest CB, Starfield B, Green B,

Kang M, Ensminger M. Reliability and valid-ity of the adolescent health profile-types. MedCare. 1998;36(8):123748.

22. Patterson GR, Capaldi DM. Psychometricproperties of fourteen latent constructs fromthe Oregon Youth Study. New York Ctiy:Springer-Verlag; 1998.

23. Silber TJ, Pao M. Somatization disorders inchildren and adolescents. Pediatr Rev. 2003;24:25564.

24. Caputo VG, Bordin IA. Mental health prob-lems among pregnant and non-pregnantyouth. Revista de Saude Publica. 2007;41(4):57381.

25. Meier AM. Adolescent first sex and sub-sequent mental health. AJS. 2007;112(6):181147.

26. Lonardo RA, Giordano PC, Longmore MA,

Manning WD. Parents, friends, and romanticpartners: Enmeshment in deviant networksand adolescent delinquency involvement. JYouth Adolesc. 2009;38:36783.

27. DeVore ER, Ginsburg KR. The protective ef-fects of good parenting on adolescents. CurrOpin Pediatr. 2005;17(4):4605.

28. Tapia MI, Schwartz, SJ, Prado G, Lopez B,Pantin H. Parent-centered intervention: Apractical approach for preventing drug abusein Hispanic adolescents. Res Soc Work Pract.2006;16:14665.

29. Gershoff ET, Grogan-Kaylor A, Lansford JE,Chang L, Zelli A, Deater-Deckard K, DodgeKA. Parent discipline practices in an interna-tional sample: Associations with child behav-iors and moderation by perceived normative-

ness. Child Dev. 2010;81(2):48095.30. Achenbach TM, Rescorla LA. Multicultural

understanding of child and adolescent psy-chopathology: implications for mental healthassessment. New York: Guilford Press;2007.

31. Gutirrez JP, Torres-Pereda P. Acceptabilityand reliability of an adolescent risk behaviorquestionnaire administered with audio andcomputer support. Rev Panam Salud Publica.2009;25(5):41822.

32. Horner P, Sanchez N, Castillo M, Delva, J.Parental perceptions of neighborhood effectsin Latino comunas. Subst Use Misuse. [Inpress].

33. Deardorff J, Tschann JM, Flores E, Ozer EJ.Sexual values and risky sexual behaviors

among latino youths. Perspect Sex ReprodHealth. 2010;41(1):2332.

Manuscript received on 6 January 2010. Revised versionaccepted for publication on 9 August 2010.

REFERENCES


8/8

Objetivo. Examinar distintos factores relacionados con el inicio de la actividadsexual que presentan los jvenes, sus padres y su familia en una muestra de jvenesy su respectiva familia tomada de la ciudad de Santiago, Chile.Mtodos. Se tomaron los datos de la primera ronda del Estudio Longitudinal deSantiago, que se llev a cabo de enero del 2008 a noviembre del 2009. La muestra es-tuvo integrada por 766 jvenes (media de edad = 14,03 aos; 51% del sexo masculino)de municipios de nivel socioeconmico bajo a medio. Las variables evaluadas fueronlas subescalas emocionales y conductuales del instrumento de autonotificacin sobrecomportamiento juvenil (Youth Self-Report) que forma parte del inventario de com-portamiento infantil (Child Behavior Checklist), la vigilancia de los padres, la participa-cin familiar, el control y la autonoma de los padres, la relacin con el padre y lamadre, y la actividad sexual. Se emplearon modelos bifactoriales y multifactoriales deregresin logstica para examinar las probabilidades de inicio de la actividad sexual.Resultados. Setenta (9,14%) de los jvenes informaron que ya haban tenido algunarelacin sexual; en este grupo, la edad promedio de iniciacin fue 13,5 aos (des-viacin estndar [DE] = 1,74) en los hombres y 14,08 (DE = 1,40) en las mujeres. Seobserv una relacin inversa entre la actividad sexual y los sntomas de retraimientoy depresin (razn de posibilidades [OR] = 0,84; intervalo de confianza [IC] =0,72 0,97), pero una relacin positiva con los sntomas somticos (OR = 1,20; IC =1,04 1,38) y el comportamiento transgresor (OR = 1,21; IC = 1,081,36), despus deajustar los valores en funcin de las variables demogrficas y otras variables perso-

nales, as como las relativas a la crianza. La mayora de los jvenes (80%) que habantenido relaciones sexuales informaron que haban utilizado algn tipo de proteccinen su ltima relacin.Conclusiones. Estos resultados ponen de manifiesto la importancia que puedentener para los jvenes, a la hora de decidir el inicio de su vida sexual, los problemasde salud mental, algunos de los cuales no suelen estar asociados al inicio de la activi-dad sexual. El efecto protector que potencialmente pueden brindar algunas caracte-rsticas de la crianza y de la familia desapareci durante la juventud y algunos pro-

blemas conductuales de esa etapa.

Conducta sexual; adolescente; conducta del adolescente; embarazo en adolescencia;Chile.

RESUMEN

Relaciones sexualesentre adolescentes de

Santiago, Chile:

un estudio de factoresindividuales y de crianza



Palabras clave

Documents

Sex Among Adolescents Chile