Sixty percent of all sexually transmitted infections (STIs) worldwide occur in adolescents and young adults between 14 and 24 years of age.1,2 Condom use is associated with significant protection from multiple STIs.3 Adolescents who do not use condoms at sexual debut are less likely to engage in subsequent protective behaviors and experience more infections; therefore, condom use at sexual debut is a good proxy for STI risk in adolescence.4,5 Sexually transmitted infections have numerous lifelong health consequences, which can prove more deleterious for females than males, and include infertility, cervical cancer, pelvic inflammatory disease, or STI transmission from mother to child during childbirth.1,2 However, many teens initiate sex using condoms inappropriately or not using them at all.6,7 Female adolescents, especially, seem to be at risk for STIs, in part, because of their inability to negotiate condom use with males.6,8,9
Education is an important structural determinant of sexual health. Educational attainment, the highest degree of education an individual has completed, is a known protective factor against vulnerability to STIs and condom use.10,11 As educational attainment increases, rates of STI diagnoses decline.12,13 High school performance also influences sexual health.10 Academic difficulties and lower grades in high school are associated with not using condoms at sexual debut.14,15 Higher high school grades have been associated with fewer STI diagnoses.16 Furthermore, youth who drop out of high school are less likely to use condoms and are more likely to have STIs compared with those who stay in school.17 Evidence further shows that adolescents who attend private schools are more likely to use condoms and less likely to report STIs than adolescents who attend public schools.18–20
The extent to which there is a link between the types of school attended before adolescence (i.e., primary school) and sexual risk behavior during adolescence has not been examined. Chile’s universal primary kindergarten through eighth grade educational system offers an ideal opportunity to examine these issues because the law that mandated sex education in schools was enacted in 2010, so differentials observed in sexual risk-taking behaviors by school types before this time are not confounded by school-based sex education.21
The Chilean educational system comprises 3 sectors: free public schools that are widespread throughout the country and serve 53% of students; subsidized charter schools that are privately run, share limited financing with families, are free to establish admission policies, and serve 40% of students; and private schools that serve an additional 7% of students. Public schools serve mostly lower and lower middle-class students; charter schools are predominant among middle and upper middle-class students; private institutions charge high tuition and therefore attract the Chilean economic elite.22,23 Not surprisingly, there is a generalized concern about the educational inequalities linked to a student’s economic resources because substantial differences in academic performance exist across sectors in terms of standardized test scores. All students take a yearly standardized test of math, language, and science in fourth and eighth grades. Test results show that, on average, charter schools score 12% lower than private schools and public schools score 20% lower than private schools.24
The aim of this population-based study was to examine whether primary education is associated with sexual health. We examined the association between having attended public, charter, or private primary schools and condom use at sexual debut among Chilean adolescents. We hypothesized that students, independent of their socioeconomic status (SES), who completed their kindergarten through eighth grade education in public schools would be less likely to use condoms at sexual debut than students who completed primary education in charter or private schools.
MATERIALS AND METHODS
Data were drawn from the 2009 Chilean National Youth Survey.25 This survey used a multistage probability sampling design to select participants who were representative of Chilean youth aged 15 to 29 years. Census estimates indicated that there were 4,024,452 adolescents in this age group. After stratifying by region and urban/rural residence, a complex sampling approach was used to randomly select households, and 1 eligible individual from each household was selected to complete an in-person home interview. Questions regarding sexual behavior were read and answered by the participant in a private room. Of the 7570 participants who answered the survey, 4319 reported onset of sexual activity between the ages of 10 and 18 years and were initially included in the study sample. After excluding individuals who had experienced sexual abuse during sexual debut (n = 31), which were missing data on primary education (n = 39), or on contraceptive use at sexual debut (n = 32), our analytic sample consisted of 4217 participants.
The outcome of interest—condom use at sexual debut—was obtained from 2 survey questions: “Was any method of protection used at first intercourse?” Participants who answered yes were asked: “What method of protection did you use during sexual debut?” Respondents could select 1 or more options including condoms.
The key independent variable was type of primary school attended by each participant. The survey question “In what type of school did you complete your primary education?” was answered: public, charter, or private, by 99.2% of the survey respondents, reflecting the universality of the Chilean primary education system.
The potential confounders considered a priori were demographic characteristics (gender, age at survey, SES, urban/rural residence, and Catholic religion) and sexual behaviors at the time of sexual debut (age of sexual debut, whether the encounter involved a romantic or a casual partner, and whether the age difference with the partner was <3 years or ≥3 years more). Age at survey was categorized into 3 groups: 15 to 19, 20 to 25, and 26 to 29 years. Social class (low, middle, and high SES) was defined in the survey according to the World Association Market Research classification and is based on the highest level of educational attainment obtained in the household and type of employment of the main provider.26 For participants who had a non-economically active provider, a scale of assets was used that included car, computer, microwave oven, hot shower, and cable TV. Religion was assessed using 2 survey questions: “Do you identify with any religion?” and “To what religion do you identify?” Participants were dichotomized as being Catholic or not.
All analyses were weighted to reflect a nationally representative sample using survey estimation commands in Stata version 12.1 that account for complex study design.27
Bivariate analysis using Rao-Scott adjustment for χ2 test on weighted samples was used to examine the association between the type of primary schooling and the sociodemographic and sexual behavior covariates and condom use.27 A multiple logistic regression was then used to examine the association between type of primary schooling and condom use at sexual debut while controlling for covariates.
All covariates were retained for the multiple logistic models because they had a P value smaller than 0.2 and changed the main outcome coefficient by more than 10%. Age of sexual debut was modeled as a categorical variable because lowess plots of log odds were not linear or quadratic. The nine participants who were 10 years old did not use condoms and were excluded from the models because of a serious concern of unrevealed sexual abuse where no volition for condom use exists.
Demographic characteristics—gender, SES, and Catholic religion—were identified in the literature as variables that may modify the association between type of primary schooling and condom use at sexual debut. Wald tests were performed using cross products of sex, SES, Catholic religion, and type of primary education in restricted versus full models. Only interaction terms for Catholic religion were significant (P > 0.10) and were retained in the final model.
Sixty percent of the participants had attended a public primary school; 30.3%, a charter primary school; and 9.6%, a private primary school. As shown in Table 1, private school attendees were more likely than charter or public school attendees to be males, have a higher SES, and have urban residence. They were also less likely to identify as Catholic compared with public school attendees. Compared with public school attendees, charter school attendees were more likely to have a higher SES and an urban residence. Regarding sexual debut characteristics, private school attendees were more likely than both public and charter school attendees to be older at sexual debut. Across types of primary schools, there were no differences in whether the first sexual partner was romantic or casual. Public school attendees were more likely than private and charter school attendees to debut with a partner with an age difference of 3 or more years.
Although condoms were used at sexual debut by 60% of participants who had attended private schools, only 52% and 37% of those who had attended charter schools and public schools, respectively, used a condom at sexual initiation (P < 0.001; data not shown). Condom use was higher among males than females, among those who were 15 to 19 years old at the time of responding to the survey, among those who had higher SES, and among those who resided in urban centers (Table 2). There were no differences in condom use by Catholic identification. Although condom use increased with age of sexual debut, overall, condoms were used less than 50% of the time at sexual debut in all age groups. Furthermore, condom use at sexual debut was more likely among adolescents engaging in sex with a romantic partner than with a casual partner and among those who had an age difference less than 3 years.
Results from multiple logistic regression analyses (Table 3) showed that, when adjusting for key covariates including SES, participants who completed their education in public schools had 1.85 (95% confidence interval [CI], 1.12–3.04) higher odds of not using condoms at sexual debut compared with participants who completed primary education in a private school and 1.67 (95% CI, 1.26–2.23) higher odds compared with participants who completed primary education in charter schools. Participants who attended charter schools did not differ in condom use with participants from private schools. Because there were few participants living in rural areas that attended private schools and/or had high SES, we further restricted the analysis to the 3383 urban participants. Compared with urban students who completed their primary education in private or charter schools, urban students who completed their primary education in public schools had 1.56 (95% CI, 1.06–2.30) and 1.56 (95% CI, 1.25–1.96) higher odds, respectively, of not using condoms at sexual debut. Urban participants who attended charter school also did not differ in condom use with urban participants form private schools.
Females had 1.97 (95% CI, 1.60–2.42) higher odds than males of not using a condom at sexual debut. Participants who were older at the time of the survey had lower SES, reported a younger age at sexual debut, and had their first sexual encounter with a casual partner had increased odds of not using condoms at sexual debut (Table 3).
There was a significant interaction between Catholic identification and type of primary school attended with condom use at sexual debut (P = 0.0014). Non-Catholic private school attendees were more likely to use condoms at sexual debut than Catholic private school attendees (Fig. 1). This was not the case for charter and public school attendees for whom Catholic identification was not associated with condom use.
In this population-based study of Chilean adolescents, we found that less than half of the youth reported using condoms at sexual debut. Although this study corroborates evidence that condom use is strongly associated with SES, our findings underscore the relationship between primary education and sexual behaviors independently of SES. We found that students who completed their primary education in public schools were less likely to use condoms at sexual debut than those who completed their primary education in private or charter schools. These disparities persisted after controlling for sociodemographic and sexual behavior characteristics at the time of sexual debut. Because there were few respondents living in rural areas who had attended private primary schools and/or were affluent, our findings may only be generalizable to urban Chilean youth. Neither gender nor SES modified the association between type of primary education and condom use, which suggests that this association operates similarly for males and females and across varying social classes.
Previous studies have reported an association between education and adolescent condom use and STIs; however, most assessed educational status during adolescence14–20 or after high school.12,13 To our knowledge, this is the first study to examine an educational measure before adolescence and before the onset of sexual activity. Our findings suggest the importance of focusing upstream and prompt future studies to test the hypothesis that primary education is associated with sexual behaviors.
What might explain the association between type of primary school (whether private, charter, or public) and condom use at sexual debut? A potential pathway lies in the quality of education, which may differ by school type. Students who receive a higher-quality education are more likely to have higher educational aspirations, which in turn might influence their perceptions of the cost of risky sexual behaviors.10,28 Moreover, it is possible that higher quality schools equip students with more tools to acquire knowledge and skills to negotiate condom use and prevent STIs.19 The role of quality of education on sexual behaviors has not received sufficient attention and merits further research.
Other risk factors for not using condoms at sexual debut were being female, belonging to the older cohorts of youth, lower SES, and being younger and having a casual partner at sexual debut. Our results corroborate previous studies showing that males use condoms more than females.6,8,20 This could be explained by less condom inhibition, more effective condom negotiation practices, and the fact that condoms are ultimately under physical control of the male partner. The striking magnitude of the sex differences reported here, where females report 96% less condom use than males at sexual debut, requires further investigation.
The finding that older participants were less likely to use condom at sexual debut, a birth cohort effect, reflects a worldwide trend that has been attributed to the success of condom education among newer generations.6,8,25 Since 1991, the Chilean Ministry of Health has undertaken yearly HIV prevention media communication campaigns that include TV, radio, street signs, and volunteers to raise awareness of HIV risk and disseminate sexual transmission prevention strategies like condom use. Socioeconomic status inequality in condom use could be strongly affected by economic access. Chile does not distribute free condoms in schools, and the average price of a condom is as expensive as US$1. In our sample, 14% of the condoms used at sexual debut were provided by health clinics, and the remaining 86% were bought in drugstores.25 The younger the age adolescents experience sexual debut, the less likely they are to use condoms, especially if younger than 14 years. This underscores the extremely vulnerable conditions in which sexual encounters with minors occur.6,13 Furthermore, sexual debut with a casual partner may account for lower condom use due to lack of planning.8,17
Although the Chilean school system may not be generalizable to other settings, many countries (including the United States) have introduced charter schools in addition to having open access public schools and paid private schools. We had no information whether participants attended a religious school. By law, no public school can impart religious education, and 37% of private and charter schools in Chile are Catholic.29 We found that Catholic identification was associated with lower condom use only in private schools. This finding suggests that Catholic students in private schools might be exposed to more abstinence only messages, which may inhibit condom use and negotiation practices at the time of sexual debut.
Our study had several limitations. In this cross-sectional study, causality cannot be established. There was limited SES variability within school settings, and few rural students attended private schools, which might reflect selection bias. Because SES was measured at the time of the survey, misclassification may have occurred if participants changed their SES status after leaving their parental home. Self-reported measures of the exposure and outcome reflect recall of such events by participants. Nonetheless, sexual debut is likely to be a salient experience, which is usually recalled accurately.7 Study participants may have overestimated condom use because of social desirability bias. However, students answered questions about sexual behaviors in private, therefore ensuring confidentiality and reducing bias. Unmeasured covariates like parental and family factors could affect the observed differences. We lacked information on the number of students who may have switched between types of schools during their primary schooling years. Estimates from other studies, however, show that the number of switchers is low (5%–7%).23
Our study has several strengths. Data were drawn from a large population-based survey conducted in 2009. A unique natural experiment allowed us to examine the association between primary school type and sexual health before the implementation of sex education curricula in 2010. It is unlikely that schools were providing sex education before the mandate.
Describing the association between the type of primary school and condom use in sexual debut is an important step in identifying structural determinants of health that impact adolescent condom use. Public health STI interventions should be combined with larger efforts tailored to address structural factors that are associated with protective sexual health behaviors, such as condom use. One way in which this prevention might be accomplished is by improving the quality of primary education. Further prospective research using quasi-experimental methods or randomized controlled designs (perhaps through lottery methods to offer charter schooling) is needed to determine whether type of primary school education affects sexual behaviors as well as other health-related outcomes in adolescence, such as teenage pregnancy and drug or alcohol use, and if so, whether these effects can be modified by offering higher educational quality to lower SES youths. It is worth considering that by optimizing primary education, we might achieve improvements in youth health.
1. Shrier L. Sexually transmitted infections: Chlamydia, gonorrhea, pelvic inflammatory disease, and syphilis. In: Emans SJ, Laufer MR, Goldstein DP, eds. Pediatric and Adolescent Gynecology. Philadelphia, PA: Lippincott Williams & Wilkins, 2012: Kindle Location 12667.
3. Holmes KK, Levine R, Weaver M. Effectiveness of condoms in preventing sexually transmitted infections. Bull World Health Organ 2004; 82: 454–461.
4. Shafii T, Stovel K, Holmes K. Association between condom use at sexual debut and subsequent sexual trajectories: A longitudinal study using biomarkers. Am J Public Health 2007; 97: 1090–1095.
5. Shafii T, Stovel K, Davis R, et al. Is condom use habit forming? Condom use at sexual debut and subsequent condom use. Sex Transm Dis 2004; 31: 366–372.
6. Centers for Disease Control and Prevention (CDC). Trends in HIV- and STD-related risk behaviors among high school students—United States, 1991–2007. MMWR Morb Mortal Wkly Rep 2008; 57: 817–822.
7. Gebhardt W, Kuyper L, Dusseldorp E. Condom use at first intercourse with a new partner in female adolescents and young adults: the role of cognitive planning and motives for having sex. Arch Sex Behav 2006; 35: 213–219.
8. Feldman BS, Shtarkshall RA, Ankol OE, et al. Diminishing gender differences in condom use among a national sample of young Israeli men and women between 1993 and 2005. J Adolesc Health 2012; 50: 311–314.
9. Tschann JM, Flores E, de Groat CL, et al. Condom negotiation strategies and actual condom use among Latino youth. J Adolesc Health 2010; 47: 254–262.
10. Bradley BJ, Greene AC. Do health and education agencies in the United States share responsibility for academic achievement and health? A review of 25 years of evidence about the relationship of adolescents’ academic achievement and health behaviors. J Adolesc Health 2013; 52: 523–532.
11. Zuilkowski SS, Jukes MH. The impact of education on sexual behavior in sub-Saharan Africa: A review of the evidence. AIDS Care 2012; 24: 562–576.
12. Painter JE, Wingood GM, Di Clemente RJ, et al. College graduation reduces vulnerability to STIs/HIV among African-American young adult women. Womens Health Issues 2012; 22: e303–e310.
13. Annang L, Walsemann KM, Maitra D, et al. Does education matter? Examining racial differences in the association between education and STI diagnosis among black and white young adult females in the U.S. Public Health Rep 2010; 125 (suppl 4): 110–121.
14. Marsiglio W. Adolescent males’ orientation toward paternity and contraception. Fam Plann Perspecti 1993; 25: 22–31.
15. Doljanac RF, Zimmerman MA. Psychosocial factors and high-risk sexual behavior: race differences among urban adolescents. J Behav Med 1998; 21: 451–467.
16. Ford CA, Pence BW, Miller WC, et al. Predicting adolescents’ longitudinal risk for sexually transmitted infection: Results from the National Longitudinal Study of Adolescent Health. Arch Pediatr Adolesc Med 2005; 159: 657–664.
17. Crosby RA, DiClemente RJ, Wingood GM, et al. The protective value of school enrolment against sexually transmitted disease: A study of high-risk African American adolescent females. Sex Transm Infect 2007; 83: 223–227.
18. Upchurch DM, Mason WM, Kusunoki Y, et al. Social and behavioral determinants of self-reported STD Among adolescents. Perspect Sex Reprod Health 2004; 36: 276–287.
19. Martins LB, da Costa-Paiva LH, Osis MJ, et al. Factors associated with condom use and knowledge about STD/AIDS among teenagers in public and private schools in São Paulo, Brazil. Cad Saude Publica 2006; 22: 315–323.
20. Oliveira-Campos M, Giatti L, Malta D, et al. Contextual factors associated with sexual behavior among Brazilian adolescents [published online ahead of print]. Ann Epidemiol 2013 doi:10.1016/j.annepidem.2013.03.009.
22. Mizala A, Torch F. School discipline in Chile. In: Arum R, Velez M, eds. Improving Learning Environments: School Discipline and Student Achievement in Comparative Perspective. Stanford, CA: Stanford University Press, 2012: 74–103.
23. Raczynski D, Muñoz G. Reforma Educacional Chilena: El Dificil Equilibrio entre la Macro y la Micro Política. REICE 2007; 5: 40–83.
27. STATA 12.1. [computer program] Version 12.1. College Station TX. StataCorp LP.
28. Harris KM, Duncan GJ, Boisjoly J. Evaluating the role of “nothing to lose” attitudes on risky behavior in adolescence. Soc Forces 2002; 80: 1005–1039.