Sexual behaviour and related knowledge among a representative sample of secondary school students between 1997 and 2008

Abstract

Objective: This paper reports on the sexual health knowledge and risk behaviours of year 10 and 12 students between 1997 and 2008.

Method: Data were from nationally representative cross-sectional stratified cluster samples of year 10 and 12 students in the Australian secondary school system collected at three intervals — 1997, 2002 and 2008. A number of methods were used to analyse students' sexual health knowledge and behavioural data over time.

Results: Student knowledge of HIV/AIDS has stabilised since the 2002 survey and remains high. Between 1997 and 2008 there has been an increase in student knowledge with respect to STIs and hepatitis. Although the proportion of students reporting ever having had sexual intercourse has increased over time, the increases between 2002 and 2008 were not significant. There has been a significant increase in the number of sexual partnerships reported by students over a year, particularly for those in year 12.

Conclusion: Student knowledge of HIV/AIDS remains high and there has been some improvement in STI and hepatitis knowledge, although knowledge in these domains is still moderate. Sexually active students were more likely to report having sex with multiple partners compared to previous surveys. Consistency of condom use has not increased since 1997.

Implications: Although consistent condom use remains moderately high, it is of some concern that condom use has not increased since 1997 despite related increases in sexual activity among adolescents who have experienced sexual intercourse and increased rates of STIs among this group.

The sexual behaviour and sexual health-related knowledge of young people present as important domains of research given the adverse public health outcomes that unsafe and uninformed sexual practice can yield. In Australia, rates of teenage pregnancy are among the highest in the developed world,¹ and the prevalence of sexually transmitted infections (STIs) among young adult populations is high and has increased considerably over the past decade.²

In the US, the national Youth Risk Behaviour Survey (YRBS) monitors the health-related risk-taking behaviour of secondary school students aged between 14 and 18 years at two-yearly intervals.³ The most recent representative data for the UK population comes from the National Survey of Sexual Attitudes and Lifestyles (NATSAL) study, conducted twice during 1990/91 and in 2000/01 periods and a study by Henderson et al. in Scotland. The NATSAL study is a national stratified probability survey of men and women aged between 16 and 44 years in Britain, and collects data pertaining to sexual behaviour and related sexual health.⁴ The Henderson et al. study was based on a representative sample of 7,630 young Scottish non-denominational secondary school students aged between 14 and 15 years of age.⁵

This paper presents data from a nationally representative repeated cross-sectional study of Australian secondary school students aged between 13 and 18 years. The first administration of the study was in 1992, with subsequent studies in 1997, 2002 and 2008.⁶ In this article we present data from 1997 to 2008 concerning students' sexual behaviour, related knowledge and condom use, and explore how student behaviour has developed over this period. We believe these data are significant and the findings important given this is the only large-scale representative Australian study that focuses on adolescent sexual health and knowledge.

Method

Sample

This study uses representative cross-sectional samples based on Australian Bureau of Statistics (ABS) data on school populations collected in 1997, 2002 and 2008. Complex samples were drawn at each administration of the survey and involved a two-stage selection process. In the first stage, schools were randomly selected with a probability proportional to the size of the target population stratified by state/territory. The smaller states/territories were over-sampled to improve the precision of the results in these strata. At the second stage of sampling, a class of Year 10 students and a class of Year 12 students were randomly selected from all classes at each year level at the school. Where the class size was fewer than 20 students, an additional class at that year level was randomly selected at the same school.

A total sample of 300 schools across government, independent and Catholic school sectors was selected: 50 schools in Victoria, New South Wales and Queensland; and 30 schools in the smaller states/territories of Tasmania, South Australia, Western Australia, the Northern Territory and the Australian Capital Territory. Given school response rates and average cluster (class) sizes from previous administrations of the study, it was expected that a final sample of approximately 150 schools and 3,000 students would be achieved.

In order to counter the poor school response rates encountered in the 2008 study (26%), some variations to sampling methodology were required. First, a second sample was drawn to expedite school and student recruitment, with all randomly selected schools approached simultaneously. In addition to this, given the low school participation rates, the second stage of the sampling method was varied so that whole year groups in schools were selected.

Active consent was obtained from schools, students and their parents/guardians. Data was collected using a paper-based questionnaire and, to protect confidentiality, the survey was generally carried out under exam conditions supervised by teachers. Completed surveys were sealed in an envelope for collection. Students were aware that they could withdraw from the survey at any time and at the completion of the survey were given answers to the sexual health knowledge questions. The study was approved by La Trobe University's human ethics committee.

The sample for this paper comprised 333 schools drawn from government, independent and Catholic school sectors (1997: 118; 2002: 110; 2008: 105). It should be noted that Catholic and independent school sectors were included in the sample from 2002 onwards. Analysis for any effects school sector may have imparted on knowledge and behaviour over time was undertaken. Each of the models (logistic, multinomial and OLS regression) measuring the significance of change across survey administration were specified for both a sample comprising government students only and a sample including students from independent and Catholic school sectors. The patterns observed when independent and Catholic school sector data were included were consistent with analyses of government school data only.

A total of 8,864 students were surveyed across the three surveys (1997: 3,550; 2002: 2,388; 2008: 2,926) and of these 58% were female (1997: 56%; 2002: 55%; 2008: 62%) and 63% were in year 10 (1997: 50%; 2002: 58%; 2008: 83%). The median ages of the students in years 10 and years 12 were 15 and 17 years, respectively. The data were weighted for student year level based on ABS secondary school populations.

Measures

Knowledge was measured across three domains, HIV/AIDS, STIs and hepatitis. In terms of sexual practice, we examined students' experience of sexual intercourse and for those who were sexually active, the number of sexual partners, and consistency of condom use in the past year and at the last sexual encounter.

Knowledge of HIV/AIDS (11 items, Cronbach a = 0.51), STIs (11 items, Cronbach a = 0.70) and hepatitis (7 items, Cronbach a = 0.64) was assessed by responses to true/false questions/statements relating to each infection aggregated to form composite scales. Questions/statements included: “Could a man get HIV through having sex with a woman?” (HIV/AIDS), “Chlamydia is a sexually transmissible infection that affects only women?” (STIs), and “Hepatitis C can be transmitted by tattooing and body piercing” (hepatitis). On each of these scales, a point was given to students for each correct answer. Both the HIV/AIDS and STI knowledge scales ranged from zero to 11, with the Hepatitis knowledge scale ranging from zero to seven. Knowledge scales have face validity in that items relate directly to the sexual health education syllabus in secondary schools nationally.

For knowledge measures, imputation of missing data was undertaken using mean substitution. Scale scores for students with partial responses across sets of knowledge items were computed using their average scores from valid responses. In order to reduce error associated with missing data imputation for knowledge scales, students with substantial missing data (i. e. missing data on more than half the items comprising a scale) on sets of items comprising scales were not imputed scale scores. Thus, students with missing data on six or more items of HIV (n=11) and STI (n=65) scales and missing data on five or more items of the hepatitis scale (n=86) were excluded from analyses.

Student experience of sexual intercourse was measured by responses to the question, “Have you ever had sex?”, cross-referenced with responses to categorical questions regarding the students age at first experience of sex, with or without a condom. Number of sexual partners was assessed by asking students to recall the number of people they had sexual intercourse with in the past year. In responding to this question, students chose one the following categories; ‘I have not had sexual intercourse in the past year, ‘1 person’, ‘2 people’, ‘3 to 5 people’, ‘6 to 10 people’, or ‘11 or more people’.

Consistency of condom use was measured both in terms of the last sexual encounter and over the past 12 months. Students were asked to respond either ‘yes’ or ‘no’ to the question; “Was a condom used the last time you had sex?” In terms of condom use over the past year students were asked; “When you had sex with people in the last year, how often were condoms used?” Students answered: ‘I didn't have sex’, ‘Always used condoms’, ‘Sometimes used condoms’, or ‘Never used condoms’.

Analysis

Data have been weighted to correct for over-sampling in the sample design and for differential response rates across states/territories and schools. As students were selected on the basis of their school class membership, the primary sample unit of the study was defined as a class, and statistical analyses took this clustering into account. Also, variance estimations in all analyses took the stratification of the sample into account. STATA Version 8.1 statistical software was used for statistical analyses.⁷

Comparison of means and contingency table analyses were employed for descriptive analyses, and the results reported by survey administration, gender and year level. For testing the significance of change over time, linear regression, binary and multinomial logistic regression statistical methods were used depending on the outcome variable's level of measurement. Where comparisons were made across the three time-points, survey administration was modeled as a set of dichotomous indicator variables with the 1997 survey as the omitted reference category. Where linear trends were found over time, survey administration was treated as a continuous variable and modeled accordingly.

Results

As the mean scale scores in Table 1 show, student knowledge of HIV is high. After a decline between 1997 and 2002 surveys (9.7 vs. 9.3, F_1,326=34.3, p=0.01) student knowledge of HIV remained stable in 2008 (9.3, F_1,326=0.01, p=0.93). Across each survey, year 12 students showed better HIV knowledge than students in year 10 (1997: 10.0 vs. 9.4, F_1,111=130.19, p<0.001; 2002: 9.5 vs. 9.2, F_1,103=15.24, p<0.001; 2008: 9.6 vs. 9.2, F_1,98=8.42, p=0.005). In terms of gender, young women generally reported better knowledge than young men, although the difference in knowledge in the 2008 study was marginal and not statistically significant (1997: 9.8 vs. 9.7, F_1,111=5.1, p=0.027; 2002: 9.4 vs. 9.2, F_1,103=10.9, p=0.001; 2008: 9.4 vs. 9.3, F_1,98=1.76, p=0.188).

Table 1.  Students' sexual health knowledge (mean composite scale scores and 95% CIs).

	Year 10			Year 12
	1997	2002	2008	1997	2002	2008
HIV/AIDS
Men	9.4 (9.3–9.5)	9.0 (8.8–9.2)	9.1 (9.0–9.3)	9.9 (9.8–10.1)	9.3 (9.1–9.5)	9.4 (9.1–9.8)
	n=816	n=601	n=621	n=756	n=436	n=405
Women	9.5 (9.3–9.6)	9.3 (9.1–9.5)	9.2 (9.1–9.3)	10.1 (10.0–10.2)	9.7 (9.5–9.9)	9.6 (9.3–9.9)
	n=967	n=777	n=1032	n=1008	n=574	n=858
STIs
Men	4.5 (4.3–4.8)	5.1 (4.8–5.5)	6.6 (6.4–6.9)	5.7 (5.5–6.0)	5.9 (5.6–6.3)	6.7 (6.0–7.3)
	n=798	n=593	n=619	n=752	n=434	n=411
Women	5.1 (4.9–5.4)	6.3 (6.0–6.7)	7.0 (6.6–7.4)	6.3 (6.1–6.5)	7.2 (6.9–7.5)	8.0 (7.4–8.5)
	n=962	n=776	n=1028	n=1005	n=573	n=858
Hepatitis
Men	2.4 (2.2–2.5)	3.2 (3.0–3.4)	3.5 (3.3–3.7)	2.8 (2.6–3.0)	3.6 (3.4–3.7)	3.8 (3.4–4.2)
	n=792	n=592	n=616	n=749	n=434	n=411
Women	2.5 (2.3–2.6)	3.6 (3.4–3.7)	3.8 (3.7–4.0)	2.8 (2.7–2.9)	3.7 (3.4–4.0)	4.1 (3.9–4.3)
	n=965	n=776	n=1028	n=1003	n=573	n=858

Overall, students mean STI knowledge had increased across time; 1997 (mean = 5.5), 2002 (mean = 6.2) and 2008 (mean = 7.2) (F_1,326=95.1, p<0.001) (Table 1). Between 1997 and 2002 studies, an interaction was found between gender and survey administration where STI knowledge increased for young women (5.7 vs. 6.7, F_1,209=43.5, p<0.001) but remained stable for young men (5.1 vs. 5.5, F_1,208=3.84, p=0.051). This pattern did not persist in the most recent survey, however, with an increase in STI knowledge exhibited by both female (7.4, F_1,195=9.8, p=0.002) and male students (6.6, F_1,177=29.0, p<0.001). As was the case with HIV knowledge, at each survey students in year 12 exhibited better knowledge than those in year 10 (1997: 6.1 vs. 4.8, F_1,111=102.0, p<0.001; 2002: 6.7 vs. 5.8, F_1,103=18.2, p<0.001; 2008: 7.6 vs. 6.8, F_1,98=6.28, p=0.014) and female students better knowledge than young men (1997: 5.7 vs. 5.1, F_1,111=31.8, p<0.001; 2002: 6.7 vs. 5.5, F_1,103=58.2, p<0.001; 2008: 7.4 vs. 6.6, F_1,98=15.1, p<0.001).

Student's knowledge of hepatitis increased steadily between 1997, 2002 and 2008 (2.6 vs. 3.5 vs. 3.8, F_1,326=95.1, p<0.001) (Table 1). With the exception of the 1997 survey, where differences were not statistically significant, young women typically demonstrated better hepatitis knowledge than young men (1997: 2.6 vs. 2.5, F_1,111=1.1, p=0.31; 2002: 3.6 vs. 3.4, F_1,103=6.7, p=0.01; 2008: 4.0 vs. 3.6, F_1,98=7.5, p=0.007). Generally, year 12 students showed higher levels of hepatitis knowledge compared to their year 10 counterparts, however the difference in the 2002 survey was not statistically significant (1997: 2.8 vs. 2.4, F_1,111=27.73, p<0.001; 2002: 3.7 vs. 3.4, F_1,103=3.51, p=0.064; 2008: 4.0 vs. 3.7, F_1,98=4.42, p=0.038).

Table 2 shows the proportion of students who had experienced sexual intercourse at each survey administration. In 1997, 34% of students reported having sexual intercourse and this figure had increased marginally in the 2002 study, to 35%, but was not statistically significant (F_1,326=0.16, p=0.689). In 2008, more students (40%) reported having sexual intercourse, however compared to the 2002 study the difference again was not statistically significant (F_1,326=1.43, p=0.233). Students in year 12 were considerably more likely to have experienced sexual intercourse compared to year 10 students in each of the three survey waves (1997: 48% vs. 19%, F_1,111=151.42, p<0.001; 2002: 47% vs. 26%, F_1,103=59.42, p<0.001; 2008: 56% vs. 27%, F_1,98=18.6, p<0.001). There were no significant differences in sexual experience by gender; however, interestingly in the 2008 study female students reported higher rates of sexual intercourse than young men (43% vs 34%%, F_1,98=2.25, p=0.137) — a reverse of the trend observed in previous surveys (1997: 33% vs. 35%, F_1,111=1.5, p=0.225; 2002: 33% vs. 36%, F_1,103=1.26, p=0.264).

Table 2.  Students who have ever had sex (per cent and 95% CIs).

	Year 10			Year 12
	1997	2002	2008	1997	2002	2008
Men	23.4 (19.7–27.6)	27.8 (22.7–33.5)	27.1 (20.1–34.2)	47.4 (42.0–52.9)	48.3 (42.4–54.2)	44.4 (28.2–61.9)
	n=812	n=600	n=624	n=754	n=436	n=411
Women	16.6 (13.7–19.9)	24.2 (19.3–29.8)	27.6 (22.0–34.0)	48.1 (44.2–52.1)	45.7 (38.5–53.0)	61.7 (46.6–74.9)
	n=965	n=777	n=1033	n=1008	n=574	n=858

Generally, most sexually active students reported having sex with one person in the 12 months prior to being interviewed (1997: 57%; 2002: 56%; 2008: 52%), however significant proportions of students report multiple sex partners in the past year (Table 3). In the 2008 survey, for instance, 17% of students reported four or more sexual partners in the previous 12-month period. Multinomial logistic regression using ‘one partner’ as the reference category showed that between 1997 and 2008 surveys, the likelihood of students having sex with three or more people had increased considerably. Between 1997 and 2002 the proportion of students reporting sex with three or more people increased from 16% to 20%, although the difference was not statistically significant (F_1,313=2.38, p<0.124). In the 2008 survey, this figure had risen appreciably to 30% (F_1,313=4.32, p=0.039) and, as Table 3 shows, the increase was attributable largely to students in year 12 (Year 10: 27% vs. 29%, F_1,269=0.10, p=0.758; Year 12: 14% vs. 30%, F_1,207=6.68, p=0.01). Between 1997 and 2008, there were no significant differences in the proportion of students reporting either no sexual intercourse (1997–2002: F_1,313=0.00, p=0.969; 2002–2008: F_1,313=2.59, p=0.109; 1997–2008: F_1,313=3.18, p=0.08) or sexual intercourse with two people (1997–2002: F_1,313=0.66, p=0.417; 2002–2008: F_1,313=0.17, p=0.679; 1997–2008: F_1,313=1.46, p=0.228) in the previous 12 months. At each survey administration, there were no significant differences between young men and young women in terms of reporting three or more sexual partners in the year prior to the study (1997: 18% vs. 13%, F_1,109=0.06, p=0.803; 2002: 23% vs. 17%, F_1,100=2.19, p=0.142; 2008: 37% vs. 27%, F_1,90=0.04, p=0.837.

Table 3.  Sexually active students' reported number of sexual partners in the previous year (per cent and 95% CIs).

	Year 10			Year 12
	1997	2002	2008	1997	2002	2008
Men
I have not had sex in the past year	5.1 (2.7–9.5)	9.7 (6.2–14.7)	6.5 (4.2–9.9)	9.2 (6.5–12.8)	9.0 (5.2–15.1)	2.6 (0.7–9.2)
1 person	53.7 (45.6–61.6)	42.0 (33.6–50.9)	45.3 (38.7–52.0)	51.3 (45.2–57.5)	56.0 (48.6–63.2)	44.2 (27.5–62.2)
2 people	24.7 (19.1–31.4)	15.7 (10.4–22.9)	12.2 (8.4–17.4)	20.0 (15.8–24.9)	19.4 (13.3–27.5)	15.1 (8.5–25.4)
3 or more people	16.4 (10.9–24.1)	32.7 (25.0–41.4)	36.1 (30.6–41.9)	19.6 (15.9–23.8)	15.6 (10.5–22.5)	38.2 (20.9–59.1)
	n=185	n=161	n=165	n=349	n=209	n=183
Women
I have not had sex in the past year	5.1 (2.7–9.6)	2.9 (1.0–7.9)	4.2 (2.5–7.2)	5.3 (3.6–7.7)	4.3 (2.1–8.6)	2.2 (0.8–5.6)
1 person	55.1 (46.3–63.6)	56.6 (46.3–66.5)	53.3 (48.7–57.9)	63.8 (59.5–67.9)	65.1 (58.1–71.4)	56.2 (47.4–64.6)
2 people	24.6 (17.9–32.8)	18.0 (12.5–25.0)	17.1 (13.6–21.2)	17.9 (15.0–21.3)	17.1 (11.3–25.0)	14.5 (8.0–24.7)
3 or more people	15.2 (10.5–21.4)	22.5 (14.5–33.3)	25.4 (21.4–29.8)	12.9 (10.2–16.3)	13.6 (9.3–19.4)	27.2 (17.6–39.6)
	n=157	n=187	n=281	n=479	n=260	n=529

Generally, most students use condoms consistently when having sex in terms of average use over a year (‘Always’— 1997: 54%; 2002: 52%; 2008: 51%) and at the last sexual encounter (1997: 68%; 2002: 65%; 2008: 64%) (Table 4). In terms of student condom use in the previous year, there has been no significant changes in consistency between 1997 and 2008 surveys. Using ‘always’ as the reference category, multinomial logistic regression analysis showed that since 1997 there has been no significant change in the proportion of students reporting inconsistent condom use (1997: 37%; 2002: 39%, F_1,313=0.33, p=0.566; 2002–2008: 43% F_1,313=0.95, p=0.330) or those not using condoms at all (1997: 9%; 2002: 9%, F_1,313=0.00, p=0.971; 2002–2008: 7% F_1,313=0.60, p=0.439) in the past year.

Table 4.  Sexually active students' reported condom use in the previous year (per cent and 95% CIs).

	Year 10			Year 12
	1997	2002	2008	1997	2002	2008
Men
Always	74.0 (65.9–80.8)	69.6 (62.8–75.6)	66.1 (60.1–71.5)	55.8 (49.1–62.3)	52.2 (44.6–59.6)	56.0 (42.4–68.7)
Sometimes	21.2 (15.5–28.3)	24.1 (18.1–31.4)	27.0 (21.3–33.6)	35.2 (29.5–41.3)	36.8 (30.3–43.9)	42.1 (29.9–55.4)
Never	4.7 (2.3–9.4)	6.3 (2.9–12.9)	6.9 (4.3–11.1)	9.1 (6.4–12.7)	11.0 (6.7–17.6)	1.9 (0.4–8.0)
	n=171	n=149	n=161	n=324	n=195	n=179
Women
Always	53.3 (45.3–61.1)	62.8 (55.7–69.3)	50.8 (45.3–56.2)	44.2 (39.5–49.1)	34.0 (28.3–40.2)	43.7 (34.6–53.2)
Sometimes	38.3 (30.6–46.6)	31.6 (25.3–38.7)	40.7 (35.7–45.9)	44.6 (39.2–50.1)	54.5 (45.7–63.0)	48.5 (40.8–56.3)
Never	8.5 (4.8–14.5)	5.7 (2.9–10.7)	8.6 (5.7–12.8)	11.2 (8.6–14.6)	11.5 (7.3–17.7)	7.8 (3.4–16.8)
	n=150	n=182	n=270	n=453	n=250	n=519

Since 1997, young people in year 12 have been more likely than those in year 10 to report inconsistent condom use in the year prior to the survey (1997: 41% vs. 29%, F_1,109=15.87, p<0.001; 2002: 47% vs. 28%, F_1,100=33.67, p<0.001; 2008: 47% vs. 36%, F_1,90=8.64, p=0.004) and were also more likely to report never using condoms when they had sex, although there was no difference in the 2008 survey (1997: 10% vs. 6%, F_1,109=5.92, p=0.016; 2002: 11% vs. 6%, F_1,100=10.14, p=0.001; 2008: 6% vs. 8%, F_1,90=0.01, p=0.914).

Similarly, compared to young men, female students were more likely to use condoms inconsistently (1997: 43% vs. 30%, F_1,109=16.45, p<0.001; 2002: 45% vs. 31%, F_1,100=14.24, p<0.001; 2008: 47% vs. 36%, F_1,90=7.34, p=0.008). In 1997 female students were more likely to report never using condoms when they had sex in the previous 12 months, although in 2002 and 2008 studies this difference was no longer significant (1997: 11% vs. 8%, F_1,109=8.04, p=0.005; 2002: 9% vs. 9%, F_1,100=1.04, p=0.310; 2008: 8% vs. 4%, F_1,90=3.59, p=0.061).

Overall, there has been no significant change in the rate of student condom use at the last sexual encounter across the three survey administrations (68% vs. 65% vs. 64%, respectively; 1997–2002: F_1,314=0.95, p=0.329; 2002–2008: F_1,314=0.06, p=0.800; 1997–2008: F_1,314=1.85, p=0.175) (Table 5). There were variations in condom use at the last sexual encounter by gender and year level, however. Male students were more likely than young women (1997: 75% vs. 62%, F_1,109=18.6, p<0.001; 2002: 74% vs. 58%, F_1,100=13.8, p<0.001; 2008: 74% vs. 60%, F_1,91=7.06, p=0.009) and students in year 10 were more likely than those in year 12 (1997: 79% vs. 63%, F_1,109=17.9, p<0.001; 2002: 72% vs. 60%, F_1,100=11.8, p<0.001; 2008: 69% vs. 61%, F_1,91=4.5, p=0.037) to have used a condom the last time they had sex.

Table 5.  Sexually active students using a condom at last sex (per cent and 95% CIs).

	Year 10			Year 12
	1997	2002	2008	1997	2002	2008
Men	82.4 (74.5–88.2)	76.4 (68.7–82.7)	76.4 (69.9–81.8)	71.1 (65.8–75.9)	72.3 (61.3–81.1)	70.9 (61.3–78.9)
	n=182	n=157	n=167	n=348	n=206	n=177
Women	74.6 (66.9–81.0)	69.0 (61.8–75.4)	65.4 (60.0–70.5)	57.9 (52.9–62.7)	49.8 (41.5–58.1)	57.5 (46.8–67.4)
	n=158	n=186	n=283	n=478	n=262	n=529

Discussion

Since 1997 the number of young people who are sexually active has increased and although the majority of sexually active students continue to report consistent condom use, notably this proportion has not increased for some time.

Although the Australian findings presented here suggest an increase in sexual activity among young people since 1997, the rates of sexually active students are nonetheless lower than those reported in some studies. In the US, the YRBS found that 44% of year 10 and 64% of year 12 students had experienced sexual intercourse at least once in their life³ and this compares with 27% and 56% in the Australian data. Abel and Brunton in a study of year 12 and year 13 secondary schools students (mean age = 16.6 years) in Christchurch, New Zealand, found 49% of students had engaged in sexual intercourse.⁸ The rates of sexual intercourse found in the Australian data were higher than those found by Henderson et al. in Scotland,⁵ however this difference is likely attributable in some part to the lower mean age of the Scottish sample (mean = 14.2 years).

In terms of condom use, consistency of use is comparable across studies. Data from the YRBS showed that 66% of year 10 and 54% of year 12 students used a condom at their last sexual encounter and this compares with 70% and 61%, respectively, in the 2008 Australian data. In terms of a single estimate, Henderson et al. found that 60% of students reported using a condom at their last sexual encounter and the 2008 Australian data this figure was 62%. In the New Zealand study, 45% of students reported ‘always’ using condoms when they had sex, and this compares with just over half the sample (51%) in the Australian data.

Encouragingly, the decline observed in student HIV knowledge appears to have halted in the 2008 study, and knowledge of STIs and Hepatitis has continued to improve since 2002. Despite these increases in knowledge, the finding that students appear to be better informed about HIV than either STIs or hepatitis is still concerning, however, given the higher prevalence of these latter diseases among adolescent populations. The pattern observed in previous surveys whereby young women, compared with young men, report significantly higher levels of knowledge with regards to HIV, STIs and hepatitis continued in 2008 and points to the need for gender-specific sexual health education that addresses this knowledge difference.

Similarly, year 10 students consistently demonstrate lower levels of STI knowledge than students in year 12. This is significant given that year 10 students — although less sexually experienced as a group — have comparable rates of sexual partnerships compared with those in year 12, once they become sexually active. Despite the improvements in STI and hepatitis knowledge, student awareness in these domains continues to be low and represents an area of public health education and promotion requiring more focused attention. Although several studies have found that higher levels of knowledge do not necessarily preclude young people from engaging in risky sexual behaviour,^9–11 nonetheless sound knowledge does provide the basis for young people to make informed decisions regarding their sexual lives (particularly before becoming sexually active) and general health.

The marked increase in students, particularly those in year 12, having sexual intercourse with multiple partners is also significant. In the YRBS, 15% of students reported having sex with four or more partners in their lifetime, and this is less than the estimate (17%) over only a 12-month period in the 2008 Australian data. Factors such as early age of sexual debut and relationship formation, and increased alcohol consumption have been associated with increases in the number of sexual partnerships of adolescents,^12,13 and these may also be symptomatic of a more broader cultural change in adolescent sexuality and identity. In Australia, rates of alcohol consumption among secondary students have increased markedly, as has the proportion of young people engaging in sex while under the influence of alcohol or drugs¹⁴— these factors may be associated with the increases observed in sexual activity here.

There are limitations to this study that should be taken into consideration when interpreting the results presented here. The low school-level participation rates may have introduced some level of bias into the study, particularly where refusal to participate reflected a degree of conservatism on behalf of the school with respect to issues surrounding the sexual health and behaviour of students. Similarly, the study may have under-represented students from non-English-speaking backgrounds as the requirement of parental consent may have adversely affected the participation of students with non-English-speaking parents and those with religious or cultural backgrounds where research on sexual health is considered inappropriate. The exclusion of cases with substantial missing data on knowledge measures should also be noted when interpreting knowledge levels reported in this paper. One should also exercise caution when interpreting statistics for student sexual behaviour given the small cell sizes in some analyses, and be mindful of the confidence intervals around such estimates. Finally, although we found no evidence of any change to the patterns observed over time in terms of the effect of school sector on student sexual behaviour and knowledge, the reader should be aware that when drawing conclusions from these data, students from independent and Catholic schools sectors were not represented in the 1997 sample.

In conclusion, there are both positive and negative aspects to the findings of this study. Encouragingly, student knowledge of HIV/AIDS remains high and the gains in STI and hepatitis knowledge are also positive. Furthermore, students' consistency of condom use when having sex has remained stable, with rates of use higher than young people in other countries. However, given the clear increases in sexual activity and still moderate levels of STI knowledge exhibited by young people, it is of some concern that levels of safe sex practised by adolescents have not increased in Australia since 1997.

Acknowledgements

This project was funded by the Australian Commonwealth Department of Health and Ageing.