Risk-taking Sexual Behavior And Depression

Blackwell Science, LtdOxford, UKCCHDChild: Care, Health and Development1365-2214Blackwell Publishing Ltd, 20032003295337344Original ArticleRisktaking sexual behaviour and depressionE. Kosunen et al.

Original Article

Risk-taking sexual behaviour and self-reported depression in middle adolescence – a school-based survey E. Kosunen,* R. Kaltiala-Heino,†‡ M. Rimpelä§ and P. Laippala†¶ *Medical School, Department of General Practice, University of Tampere, †School of Public Health, University of Tampere, ‡Department of Psychiatry, Tampere University Hospital, §National Research and Development Centre for Welfare and Health (STAKES), Helsinki, and ¶Research Unit, Tampere University Hospital, Tampere, Finland Accepted for publication 15 May 2003

Abstract

Keywords adolescence, depression, sexual behaviour, contraception Correspondence: Elise Kosunen, Senior Lecturer of General Practice, Medical School, Department of General Practice, University of Tampere, FIN-33 014, Finland E-mail:[email protected]

Background Early sexual activity has been widely studied in the context of pregnancies, substance use and antisocial behaviour, but the aspects of psychosexual health have received less attention. Aim To study the associations of early sexual activity and self-reported depression. Setting A school survey in Finland in 1999 and 2000 in the eighth and ninth grades. Methods Adolescents with experience of sexual intercourse were studied (11 793 girls and 10 443 boys, mean age 15.5 years). Scores of 8 or more in the Beck Depression Inventory were regarded as indicative of self-reported depression. Associations with sexual behaviour variables were analysed using logistic regression models. Results In both genders, self-reported depression increased in proportion to the number of sexual partners and with the non-use of contraception. A higher number of coital experiences correlated with depression only among boys. Adjusting for age and age at menarche/oigarche did not affect the associations detected. In stepwise logistic regression, an increasing number of partners increased the risk for self-reported depression [for boys with at least five partners odds ratio (OR) 2.5, 95% confidence intervals (CI) 2.2–3.0, and for girls OR 2.7, 95% CI 2.3–3.2]. Boys and girls who did not use contraception showed roughly twice as high a risk as contraceptive users. However, girls with five or more coital experiences had a significantly lower risk for depression compared to girls with only one sexual intercourse. Conclusions Multiple sexual partners and non-use of contraception may reflect a depressive disorder in both genders. While adolescent health service providers should be aware of the risk for depression among sexually active adolescents, the sexual health of depressed adolescents also warrants special attention.

Introduction Adolescent sexual behaviour is most often described in terms of problem behaviour or risk

behaviour (Udry & Bearman 1998): early sexual activity is considered to reflect problems in adolescent development rather than successful adolescent passage. However, a remarkable proportion of

Original Article © 2003 Blackwell Publishing Ltd

337

338

E. Kosunen et al.

teenagers in Western countries engage in sexual intercourse at an age that can be called ‘early’. In many European countries around 30% of adolescents experience their first sexual intercourse before the age of 16 (Ross & Wyatt 2000), and even higher figures are reported from the USA, Canada and Australia (Cheesbrough et al. 1999). Does this mean that all these adolescents are at risk of unfavourable development? In Scandinavia, attitudes to adolescent sexuality are quite permissive: sexual activity is considered normative behaviour during the transition period from adolescence to adulthood (Davis 1989). However, there is also evidence that at least some teenagers engage in sexual activity before they are mature enough, with reasons ranging from peer pressure and an unfavourable family background to substance use (Goodson et al. 1997; Kinsman et al. 1998). According to recent British studies, a large proportion of adolescents who engaged in sexual intercourse at age 16 or below regretted their early initiation (Dickson et al. 1998; Wight et al. 2000). Adolescent sexual health is most typically described from the vantage point of physical health, using such indicators as the incidence of pregnancies, abortions and sexually transmitted infections. The viewpoint of psychosexual health is largely ignored. For instance, there are only very few studies that have discussed the associations between depressive disorders and sexual behaviour (Rubin et al. 1992; Tubman et al. 1996), even though depression is a common and increasing problem in the population, including adolescents (Leon et al. 1993; Prosser & McArdle 1996). Our previous study based on a major school survey was concerned with associations between adolescents’ self-reported depression and sexual events in the chain of advancing sexual experiences (kissing, fondling, petting, sexual intercourse) in the adolescent population aged 14–16 years. We found an association between intimate sexual experiences (petting, sexual intercourse) and depression in girls (Kaltiala-Heino et al. 2003). In boys, depression was associated with experiences of sexual intercourse only. In this report, we focus on adolescents who have experienced at least one sexual intercourse and their psychosexual health. We

also elaborate on the association between selfreported depression and indicators of sexual behaviour: the total number of coital experiences, the number of sexual partners and contraceptive use.

Materials and methods The School Health Promotion Study is an anonymous classroom survey concerned with the health, health behaviour and school experiences of adolescents in the eighth and ninth grades of secondary school. The survey has been carried out annually since 1995. The data for the present study were derived from the surveys in 1999 (188 municipalities mainly from western Finland) and 2000 (160 municipalities mainly from eastern Finland). A total of 348 out of the country’s 448 municipalities took part in the survey. Less than 1% of the forms returned were rejected because of poor data quality, leaving us with 47 952 responses from the eighth grade (mean age 14.8 years, SD ± 0.3 years) and 47 214 from the ninth grade (mean age 15.8 years, SD ± 0.3 years). Pupils absent on the survey day were not contacted. Earlier data collections have taught us that each day, 10–15% of all pupils are absent. The material for the present study comprises those eighth and ninth grade pupils who reported having experienced at least one sexual intercourse. The number of girls was 11 793, the number of boys 10 443. Their mean age was 15.5 (SD ± 0.58) years. The majority had stable living conditions: 63.6% lived with both parents, 73.8% had lived in the same municipality for 10 years or more, and 60.8% reported stable parental employment (neither father nor mother unemployed during the past 12 months). One-quarter of the respondents (28.1%) said that their parents (one or both) had academic qualifications. Self-reported depression was measured by the 13-item short form of the Beck Depression Inventory (BDI) (Beck & Beck 1972; Beck et al. 1974; Kaltiala-Heino et al. 1999a). The Finnish version of the instrument has been shown to have good psychometric properties among 14- to 16-year-old adolescents (Kaltiala-Heino et al. 1999a). The 13item BDI measures the respondents’ own percep-

© 2003 Blackwell Publishing Ltd, Child: Care, Health & Development, 29, 5, 337–344

Risk-taking sexual behaviour and depression

tion of their depressive symptoms, but it is not a diagnostic instrument for depressive disorders. The 13 items consist of statements which show an increasing intensity of depressive emotions and cognitions, scoring 0–3 each. The maximum score is thus 39. Scores 0–4 are classified as no depression, 5–7 as mild, 8–15 as moderate and 16 + as severe depression (Beck & Beck 1972). Mood changes are part of normal adolescent development, but severe depressive symptoms are likely to be relatively persistent (Charman 1994). To avoid bias, we did not take into account those BDI scores that indicated mild depression, but dichotomized the depression score (moderately to severely depressed vs. no/mildly depressed). In this study, we refer to high BDI scores (scores of 8 or more, which signifies moderate to severe depression) as self-reported depression. The first School Health Promotion Studies used the 13-item BDI in full without any modifications, but in 1998 the item concerning suicidal ideation had to be removed because of concerns at the Ministry of Education that asking the question might provoke suicidality in adolescents. We analysed a previous sample (Kaltiala-Heino et al. 1999b; 2001) and noted that even with one item less in the scale, the original cut-point between no/mild (0–7 points) and moderate/severe (8 or more) depression caused only a minimal transfer of cases from one class to another: as few as 0.7% of the respondents were transferred from being cases to being non-symptomatic subjects. The psychometric properties of the 12-item scale corresponded to those of the original scale (Kaltiala-Heino et al. 1999a). Onset of menstruation (ejaculations) was elicited by asking, ‘How old were you when you had your first menstruation (ejaculation)?’ [‘10 years or younger/11 years/12 years/13 years/14 years/15 years or older/I have not yet experienced menstruation (ejaculation)’]. The two lowest classes (10 years or younger and 11 years) were combined because of the small number of cases in the lowest category. Adolescents who had experienced sexual intercourse were asked to provide more data on their sexual behaviour. First, the number of intercourse experiences was enquired by the question, ‘How many times have you had intercourse?’ (once/2–4

339

times/5–9 times/10 or more times). Second, the number of sexual partners they had had was inquired by asking, ‘With how many different partners have you had intercourse?’ (one/two/three or four/five or more). Finally, use of contraception was evaluated by asking, ‘What kind of contraception did you use in the most recent intercourse?’ [none/condom/oral contraceptives/condom and oral contraceptives/other method (please describe)]. The responses in the category ‘other methods’ virtually includes only methods of natural family planning (withdrawal, calendar method) and the proportion of this category is from zero to three percentages in the Finnish adolescent surveys (Kosunen et al. 2000). For the analyses, the response options ‘none’ and ‘other method’ were grouped together to form the category ‘no contraception’, compared to having used contraception (condom, oral contraceptives or both)’. Age was calculated from date of birth and the survey date and used in bivariate analyses classified into years (14/15/16) and in multivariate analyses as a continuous variable.

Statistical analysis Bivariate associations between sexual behaviour variables and self-reported depression were studied using the chi-square test. Multivariate analysis was performed using stepwise logistic regression, where self-reported depression (yes/no) was the dependent variable. First, the bivariate odds ratios (OR) were calculated for each sexual behaviour variable, forcing age and age at menarche/oigarche into the model in order to adjust for their confounding effect. Then, the sexual behaviour variables were tested as independent variables simultaneously, again forcing age and age at menarche/oigarche into the model. This final model was additionally tested separately for those who were currently dating (6823 girls and 4658 boys) and for those who were not (4934 girls and 5726 boys).

Results The proportion of girls having had at least 10 coital experiences was 45% (of girls having experienced at least one sexual intercourse). The figure was 37%

© 2003 Blackwell Publishing Ltd, Child: Care, Health & Development, 29, 5, 337–344

340

E. Kosunen et al.

in boys, respectively (Table 1). Around a half of adolescents had had only one sexual partner, but 10% of girls and 17% of boys reported at least five partners. Four out of five adolescents had used effective contraceptive methods (oral contraceptives or condoms) at the most recent intercourse. Among both girls and boys, self-reported depression increased in proportion to the number of reported sexual partners. Furthermore, selfreported depression was associated with the nonuse of contraception at the most recent intercourse.

Table 1. Total number of coital experiences, number of sexual partners and contraceptive use (oral contraceptives or condom) among 14- to 16-year-old adolescents who had experienced at least one sexual intercourse. Distributions (%) by gender Girls (n = 11 793) % Total number of coital experiences 1 18.1 2–4 23.5 5–9 13.4 10 or more 45.1 Number of sexual partners 1 50.8 2 22.1 3–4 17.4 5 or more 9.6 Contraception at the most recent intercourse Yes 82.3 No 17.7

Boys (n = 10 443) % 23.2 27.2 12.5 37.1 48.0 20.3 14.4 17.3 79.4 20.6

Among boys, self-reported depression also increased in line with the number of reported coital experiences, but no such association was found among girls (Table 2). Adjusting for age and age at menarche/oigarche did not change the associations detected between the variables related to sexual behaviour and selfreported depression, when the sexual behaviour variables were tested independently in turn in the logistic regression models (Table 3). Next, the association between self-reported depression and sexual behaviour variables was tested by entering all the sexual behaviour variables in one model (age and age at menarche/oigarche forced). When using the stepwise forward selection procedure in logistic regression, all the sexual behaviour variables studied as well as age and age at menarche emerged in the final model among girls (Table 4). A growing number of partners increased the risk for self-reported depression, and non-use of contraception at the most recent intercourse was likewise associated with a greater risk [OR 1.7, 95% confidence intervals (CI) 1.6–1.9]. However, if a girl had had five or more coital experiences, the risk of depression was significantly lower compared to girls with only one sexual intercourse. Among boys, OR for self-reported depression was 2.5 (95% CI 2.2–3.0) for those who had had at least five sexual partners compared to those with only one partner. Boys who did not use any contraception at the most recent intercourse had twice

Table 2. Self-reported depression (%) by total number of coital experiences, number of sexual partners and contraceptive use among 14- to 16-year-old adolescents who had experienced at least one sexual intercourse Girls (n = 11 793) % Total number of coital experiences 1 2–4 5–9 10 or more Number of sexual partners 1 2 3–4 5 or more Contraception at the most recent intercourse Yes No

Boys (n = 10 443) P-value

%

P-value

23.6 23.4 23.4 21.6

0.134

9.2 10.8 12.5 19.8

<0.0001

19.0 22.2 26.3 35.8

<0.0001

9.6 11.4 14.0 29.5

<0.0001

20.4 32.9

<0.0001

10.6 26.9

<0.0001

© 2003 Blackwell Publishing Ltd, Child: Care, Health & Development, 29, 5, 337–344

Risk-taking sexual behaviour and depression

341

Table 3. Bivariate odd ratios [OR, 95% confidence intervals (CI)] for self-reported depression according to total number of

coital experiences (Model 1), number of sexual partners (Model 2), and use of contraception (Model 3) at the most recent intercourse, controlled for age and age at menarche/oigarche, among 14- to 16-year-old adolescents

Model 1: total number of coital experiences 1 2–4 5–9 10 or more Model 2: number of sexual partners 1 2 3–4 5 or more Model 3: use of contraception at the most recent intercourse Yes No

Girls OR (95% CI)

Boys OR (95% CI)

NS

Ref 1.3 (1.0–1.6) 1.4 (1.1–1.8) 2.2 (1.8–2.6)

Ref 1.2 (1.1–1.4) 1.6 (1.4–1.8) 2.4 (2.1–2.8)

Ref 1.2 (1.0–1.4) 1.5 (1.3–1.9) 3.2 (2.7–3.7)

Ref 1.9 (1.7–2.1)

Ref 2.6 (2.3–2.9)

NS, not significant.

Table 4. Multivariate odds ratios [OR, 95% confidence intervals (CI)] for self-reported depression according to total number of coital experiences, number of sexual partners, and use of contraception at the most recent intercourse, controlled for age and age at menarche/oigarche, among 14- to 16-year-old adolescents (forward stepwise logistic regression)

Total number of coital experiences 1 2–4 5–9 10 or more Number of sexual partners 1 2 3–4 5 or more Use of contraception at the most recent intercourse Yes No

Girls OR (95% CI)

Boys OR (95% CI)

Ref 0.8 (0.7–1.0) 0.7 (0.6–0.9) 0.7 (0.6–0.8)

NS

Ref 1.3 (1.2–1.5) 1.7 (1.5–2.0) 2.7 (2.3–3.2)

Ref 1.2 (1.1–1.4) 1.5 (1.3–1.8) 2.5 (2.2–3.0)

Ref 1.7 (1.6–1.9)

Ref 2.1 (1.8–2.4)

NS, not significant.

as high a risk for depression as contraceptive users (Table 4). When stepwise modelling was carried out separately for adolescents who were currently dating and for those who were not, the results remained almost the same as in Table 4, except that the total number of coital experiences did not enter the model among girls who were not dating.

Discussion This study was carried out in Finnish secondary schools, which cover more than 99% of adolescents at this age. The samples collected in 2 years com-

prised around 80% of 1-year-birth cohorts in both the eighth and the ninth grades. Almost all the pupils present at school returned an acceptable response. Pupils absent on the day were not contacted, which may constitute a source of bias because they probably have more health and psychosocial problems, including depression. Then, it is possible that our results for the prevalence of depression are slight underestimates, but there is no reason to suggest that this in any way affects the associations between the phenomena studied. Against the background of the high and growing prevalence of depressive disorders in the population (Leon et al. 1993; Prosser & McArdle 1996), it

© 2003 Blackwell Publishing Ltd, Child: Care, Health & Development, 29, 5, 337–344

342

E. Kosunen et al.

is surprising how little work has been done to explore the associations of sexual behaviour and depression among adults, let alone adolescents. In the metropolitan New York area, Tubman et al. (1996) found in a school sample of 10th and 11th graders (aged 15–17 years) that depressive symptoms were common among sexually active adolescents. Rubin et al. (1992) studied correlates of depressive affect among high school students (9th12th grades, 13–19 years) and found that life events related to sexuality were the prepotent stressor among girls. For boys, sexuality-related events did not emerge as a risk factor when other domains were controlled for. According to the present study, having had multiple partners increased the risk of depression in both genders. Among boys, a high number of coital experiences was also associated with an increased risk of depression. A previous Finnish study showed that boys with early sexual debut report more sexual partners than girls of the same age (Kosunen et al. 2000). This points at different gender patterns in early sexual relationships: boys tend more often to start with casual partners while girls start with their dating partners. Another Finnish study showed that boys start regular dating later than girls, with the proportions of current dating being lower for boys throughout adolescence, even at the age of 18 (Rimpelä et al. 1992). In boys, this way of commencing sexual activities does seem not to be satisfactory from the point of view of sexual health, but may be an expression of depressive behaviour. The association between the total number of sexual intercourses and depression was more complicated among girls: a high number of coital experiences lowered the risk of self-reported depression. We thought that this connection might result from the fact that young girls more often have sexual intercourse within a steady dating relationship rather than with a casual partner. This dating relationship is probably more satisfying in emotional respects as well. Therefore, we analysed this association separately among currently dating and non-dating adolescents. Among dating girls, the association remained unchanged, but among non-dating girls, the protective effect of sex disappeared. However, the risk of self-reported depres-

sion did not increase with a growing number of coital experiences among non-dating girls, either. The group of non-dating girls is more heterogeneous, including not only girls who had had sex with casual partners, but also girls who had earlier had a steady sexual partner, before the study was carried out. Unfortunately, we were unable to pursue this issue further because our questions only covered current dating, not the number or length of previous dating relationships. Variables describing sexual behaviour, such as the number of partners and the number of intercourse experiences, are probably interwoven with one another in complex ways, and they might also have interactions regarding the risk for depression. We considered the potential interactions between sexual behaviour variables, age and age at menarche/ oigarche. However, the stepwise logistic regression procedure did not support this assumption. Sex education has been an integral part of the curriculum in Finnish secondary schools since 1970. Earlier studies have shown that knowledge of contraception is reasonably good (Kontula et al. 1992). Although there are undoubtedly areas that are not adequately covered in sex education, contraception is certainly not among them in the majority of the schools (Liinamo et al. 2000): at least ignorance cannot be blamed for the non-use of contraceptives. Despite high awareness and the availability of birth control services, there are still those who opt not to use contraceptive methods. The literature on the psychology of contraception suggests that neglecting contraceptive use reflects a developmental stage in early or middle adolescence: a teenager cannot yet plan ahead or anticipate the future consequences of his/her current activities (DuRant et al. 1990). On the other hand, some researches have suggested that neglecting to use contraception is associated with low selfesteem (DuRant et al. 1990) and could be a sign of depressive disorders. This is indeed supported by our findings, because the non-use of contraceptive methods at the most recent intercourse was strongly associated with self-reported depression among both boys and girls. We found in this study that risky sexual behaviour (non-use of contraception and multiple sex partners) was associated with self-reported depres-

© 2003 Blackwell Publishing Ltd, Child: Care, Health & Development, 29, 5, 337–344

Risk-taking sexual behaviour and depression

sion. It is noteworthy that these associations remained when age at menarche/oigarche was controlled for. This suggests that at this age, adolescents with early physical maturation are not less prone to emotional problems related to sexual activity, probably because psychosocial maturation is delayed compared to physical maturation. In both genders, having multiple partners is associated with poor psychosexual health, suggesting perhaps a lack of social and emotional support that these individuals are trying to compensate. On the other hand, a large number of coital experiences was a protective factor among girls who had a steady dating relationship, regardless of their young age. To summarize these findings from the point of view of health service providers for adolescents, it is clearly important that each individual’s sexual behaviour is carefully reviewed: psychosexual health cannot be automatically inferred from age or the duration of biological maturity.

Conclusion In middle adolescence, having multiple sexual partners and the non-use of contraception may reflect a depressive disorder. Among boys of this age, multiple experiences of intercourse is in itself a risk factor. Adolescent health service providers should be aware of the risk for depression among teenagers with risk-taking sexual behaviour. On the other hand, it is important to note that the sexual health of depressed adolescents warrants special attention.

Acknowledgements We wish to dedicate this article to the memory of our co-author, Professor Pekka Laippala, who unexpectedly passed away during the process of publishing this article. This study was funded by the Ministry of Social Affairs and Health.

References Beck, A. T. & Beck, R. W. (1972) Screening depressed patients in family practice. A rapid technic. Postgraduate Medicine, 52, 81–85.

343

Beck, A. T., Rial, W. Y. & Rickels, K. (1974) Short form of depression inventory: cross-validation. Psychological Reports, 34, 1184–1186. Charman, T. (1994) The stability of depressed mood in young adolescents: a school-based survey. Journal of Affective Disorders, 30, 109–116. Cheesbrough, S., Ingham, R. & Massey, R. (1999) Reducing the Rate of Teenage Conceptions. A review of the international evidence on preventing and reducing teenage conceptions: the United States, Canada, Australia and New Zealand. Health Education Authority, London. Davis, S. (1989) Pregnancy in adolescents. Pediatric Clinics of North America, 36, 665–680. Dickson, N., Paul, C., Herbison, P. & Silva, P. (1998) First sexual intercourse: age, coercion, and later regrets reported by a birth cohort. British Medical Journal, 316, 29–33. DuRant, R. H., Sanders, J. M. Jr, Jay, S. & Levinson, R. (1990) Adolescent contraceptive risk-taking behavior: a social psychological model of females’ use of and compliance with birth control. Advances in Adolescent Mental Health, 4, 87–106. Goodson, P., Evans, A. & Edmundson, E. (1997) Female adolescents and onset of sexual intercourse: a theorybased review of research from 1984 to 1994. Journal of Adolescent Health, 21, 147–156. Kaltiala-Heino, R., Kosunen, E. & Rimpelä, M. (2003) Pubertal timing, sexual behaviour and self-reported depression in middle adolescence. Journal of Adolescence (in press). Kaltiala-Heino, R., Rimpelä, M., Rantanen, P. & Laippala, P. (1999a) Finnish modification of the 13item Beck Depression Inventory (R-BDI) in screening an adolescent population for depressiveness and positive mood. Nordic Journal of Psychiatry, 53, 453–457. Kaltiala-Heino, R., Rimpelä, M., Marttunen, M., Rimpelä, A. & Rantanen, P. (1999b) Bullying, depression and suicidal ideation in Finnish adolescents: school survey. British Medical Journal, 319, 348–351. Kaltiala-Heino, R., Rimpelä, M., Rantanen, P. & Laippala, P. (2001) Adolescent depression. The role of discontinuities in life course and social support. Journal of Affective Disorders, 64, 155–166. Kinsman, S., Romer, D., Furstenberg, F. & Schwartz, D. (1998) Early sexual initiation: the role of peer norms. American Academy of Pediatrics, 102, 1185–1192. Kontula, O., Rimpelä, M. & Ojanlatva, A. (1992) Sexual knowledge, attitudes, fears and behaviours of adolescents in Finland (the KISS Study). Health Education Research, 7, 69–77. Kosunen, E., Rimpelä, M., Liinamo, A. & Jokela, J. (2000) Suomalaisten nuorten seksuaalikäyttäytymisen muutokset 1990-luvun lopulla (Trends of adolescent sexual

© 2003 Blackwell Publishing Ltd, Child: Care, Health & Development, 29, 5, 337–344

344

E. Kosunen et al.

behaviour in Finland in the late 1990s). (Abstract in English.) Sosiaalilääketieteellinen Aikakauslehti (Journal of Social Medicine), 37, 273–282. Leon, A. C., Kleman, G. L. & Wickramaratne, P. (1993) Continuing female predominance in depressive illness. American Journal of Public Health, 83, 754–757. Liinamo, A., Rimpelä, M., Kosunen, E. & Jokela, J. (2000) Vaikuttavatko seksuaaliopetus ja seksuaalivalistuslehti nuorten seksuaalitietoihin? (The impact of sexuality education lessons at school and reading the national sexual education magazine on the sexual health knowledge of the 14–16 year-olds). (Abstract in English.) Sosiaalilääketieteellinen Aikakauslehti (Journal of Social Medicine), 37, 293–303. Prosser, J. & McArdle, P. (1996) The changing mental health of children and adolescents: evidence for a deterioration? Psychological Medicine, 26, 715–725. Rimpelä, A. H., Rimpelä, M. K. & Kosunen, E. A.-L. (1992) Use of oral contraceptives by adolescents and its consequences in Finland 1981–91. British Medical Journal, 305, 1053–1057. Ross, J. & Wyatt, W. (2000) Sexual behaviour. In: Health and Health Behaviour Among Young People. Health

Behaviour in School-Aged Children: A WHOCross-National Study (HBSC). WHO Policy Series for Children and Adolescents, Issue 1. (eds C. Currie, K. Hurrelmann, W. Settertobulte, R. Smith & J. Todd), pp. 115–120. WHO, Copenhagen. Rubin, C., Rubenstein, J., Stechler, G., Heeren, T., Halton, A., Housman, D. & Kasten, L. (1992) Depressive affect in ‘normal’ adolescents: relationships to life stress, family and friends. American Journal of Orthopsychiatry, 63, 430–441. Tubman, J., Windle, M. & Windle, R. (1996) The onset and cross-temporal patterning of sexual intercourse in middle adolescence: prospective relations with behavioral and emotional problems. Child Development, 67, 327–343. Udry, J. R. & Bearman, P. S. (1998) New methods for research on adolescent sexual. In: New Perspectives on Adolescent Risk Behavior (ed. R. Jessor), pp. 000–000. Cambridge University Press, Cambridge. Wight, D., Henderson, M., Raab, G., Abraham, C., Buston, K., Scott, S. & Hart, G. (2000) Extent of regretted sexual intercourse among young teenagers in Scotland: a cross sectional survey. British Medical Journal, 320, 1243–1244.

© 2003 Blackwell Publishing Ltd, Child: Care, Health & Development, 29, 5, 337–344