«Suggested citation: Panday, S., Makiwane, M., Ranchod, C., & Letsoalo, T. (2009). Teenage pregnancy in South Africa - with a speciﬁc focus on ...»
Although strong arguments exist for the thrust of interventions for adolescents to focus on prevention, the unabated and increasing levels of marginalisation of young people across a range of domains (education, employment, health and wellbeing and civic engagement) provide impetus for a more systematic focus on treatment, care and support. Termed, second chances, by the 2007 World Development Report – Development and the Next Generation (World Bank, 2006), the approach recognises the extent to which policy and programmatic failures, rather than poorly informed choices by youth alone, have borne a heavy burden on young people resulting in their increased marginalisation (World Bank, 2006). Even if we instituted the most rigorous prevention programmes, some young women will experience early pregnant. Although remediation is costly and difﬁcult to achieve, it far outweighs the costs to a society of lost human capital potential. A systematic and more formalised system of support is required for those who do become pregnant. Such programmes of
support need to reach beyond addressing the health consequences of pregnancy to mitigating the educational, economic, educative and social challenges that young mothers and fathers face.
The following section outlines both prevention and second chance policies and programmes for teenage pregnancy. Clearly, given the strong prevention focus among adolescents, the weight of the evidence will be biased towards prevention. However, examples of second chance programmes will be outlined where they exist. Within the prevention realm, strongest evidence is available for curriculum-based intervention and most of these have been instituted in the school setting. Given that the report is being prepared for the Department of Education, the review will focus primarily on school-based or school-linked interventions. It must be noted, however, that while substantive evidence for the effectiveness of health policy and programmes are available from the US, as indicated earlier, the empirical evidence supporting the effectiveness of interventions in developing countries is thin at best and supports only tentative conclusions about effectiveness (NRC & IOM, 2005). In addition, while a few studies are focused on unintended pregnancy, most are in fact focused on STIs and HIV.
Prevention programmes School-based sex education The high enrolment rates of adolescents in the school setting, provides an important access point for interventions on sexual and reproductive health. Most young people who are engaged in school have not initiated sexual activity and for some, schooling coincides with the onset of sex. As such, the high coverage of adolescents in the school setting provides an important leverage point to delay the onset of sex and to ensure that those who are sexually active are able to adequately protect themselves (Kirby, Obasi & Laris, 2006).
A review of 56 curriculum-based programmes in the US – half of which were implemented in the school setting, reported that there is strong evidence that sex education can both delay and promote safe sex (Kirby, 2007).
Two thirds of the programmes reviewed had a signiﬁcant impact on at least one aspect of sexual behaviour, or lowered rates of pregnancy, childbearing, or STIS. While the 48 comprehensive programmes that focused on both abstinence and contraceptive use showed strong positive effects on sexual behaviour and importantly, did not increase sexual behaviour, the eight abstinence only programmes showed no impact in delaying sexual initiation. About two thirds of the comprehensive programmes, on the other hand, delayed the initiation of sex, reduced the frequency of sex and number of partners, increased condom and contraceptive use or reduced risky sexual behaviour. As such there is strong empirical support for comprehensive programmes that focus on both abstinence and contraceptive use rather than abstinence only programmes.
But the programmes reported only modest effects on sexual behaviour, pregnancy or STI rates. The most effective programmes reduced risky sexual behaviour by about a third. In addition, durability of programme effects was not sufﬁciently demonstrated. While effects of a few programmes lasted for years, most did not measure long-term impact. What is noteworthy, however, is that comprehensive programmes were equally effective and replicable with a variety of communities in different settings (e.g., urban vs. rural, low vs. middle income and boys vs. girls). While education programmes are not the panacea for pregnancy, STIs and HIV, they can play an important role as part of a comprehensive strategy to positively inﬂuence sexual behaviour (Kirby, 2007).
In reviewing 19 of the most effective curricula for sex-based education, Kirby (2007) identiﬁed 17 characteristics that were common to these programmes. These characteristics, presented in Table 17, can be clustered into
three categories: the process of developing the curriculum, the content of the curriculum and the process of implementation.
Table 16: Characteristics of effective curriculum-based programmes
A few reviews have been undertaken on sex-based (or speciﬁcally school-based) education programmes in developing countries (see for example Kaaya, Mukoma, Flisher & Klepp, 2002; Kirby, Laris & Rolleri, 2005; Gallant & Maticka-Tyndale, 2004; Speizer, Magnani & Colvin, 2003). While these reviews were able to demonstrate the positive effect of school-based programmes in improving knowledge, they did not demonstrate strong effects on improving skills, changing values and norms, and changing behaviour. More recently, Kirby, Obasi and Laris (2006), undertook a systematic review of 22 interventions that used either experimental or quasi-experimental designs with the purpose of ﬁlling the gaps of previous reviews.
There was strong evidence for the effect of school-based sex education and HIV education interventions on adolescent sexual behaviour. While the interventions did not increase sexual activity, they did report positive effects in delaying sexual activity, reducing the number of sexual partners, reducing the frequency of sex and increasing condom and contraceptive use. There was also ample evidence for the effect of programmes on knowledge but less consistent effects of the programmes on improving skills or changing values, attitudes and peer norms. The strongest evidence for programmes that had an impact on behaviour was curriculumbased and led by adults (either teachers or other adults such as health workers). Only two curriculum-based programmes were implemented by peers, one of which showed some evidence for positive impact on sexual behaviour. In line with evidence from developed countries (Kirby, 2007), the most effective curriculum-based interventions incorporated four ﬁfths of the 17 criteria identiﬁed for effective sex education in Table 17 above. In addition, like developed countries, ﬁndings were shown to be robust, in that they were equally effective among different subgroups in different settings. Despite the strong effect of interventions on sexual behaviour, only 1 study measured the effect of programmes on STI rates and pregnancy and showed no impact. The review concluded that given the strong evidence of programmes in improving knowledge and reducing sexual risk behaviour, school-based interventions that are curriculum-based, led by adults and incorporate most of the characteristics of effective programmes described in Table 17 should be taken to scale. However, there is a need for many more rigorous evaluation studies of school-based interventions involving large enough sample sizes to detect effects, and that speciﬁcally measure impact of programmes on biological outcomes such as STI, HIV and pregnancy rates.
A systematic evaluation of school-based sex/HIV education programmes in SA was undertaken by Mukoma and Flisher (2008). Although not directly focused on teenage pregnancy, life skills programmes introduced in schools in response to the explosion of the HIV epidemic in the 1990s have the potential to inﬂuence the trajectory of teenage pregnancy. Life skills programmes were introduced in schools to increase learner’s knowledge of HIV, improve their skills for engaging in health relationships by improving communication and decision-making ability and to shift attitudes about people living with HIV and AIDS. A range of factors limit the number of ﬁrm conclusions that can be reached about the overall effectiveness of programmes in SA.
Mukoma & Flisher (2008) were only able to identify a small number (12) of evaluation studies, only two of which were conducted in the past ﬁve years and most of which (5) were undertaken in the Western Cape.
They generally involved small sample sizes of students and schools and the variability in how and by whom they were designed and implemented, their duration, intensity, and the questions used in the evaluation, make comparison across studies difﬁcult. In addition, the weak design of evaluation studies (very few involving control groups and lack of follow up post intervention beyond 2 months) make it difﬁcult to assign changes in behaviour to the interventions. In general, while programmes did show positive effects on knowledge, attitudes and communication about sexuality, they had little or no effect on behaviour.
While the lack of effects may in part be attributed to the design of the programme itself and how the evaluation studies were conducted, the degree to which the programmes were implemented as intended also needs to
be investigated. This could not be thoroughly assessed in the review by Mukoma and Flisher (2008) as most studies focused on outcomes evaluation to the neglect of process evaluation that would provide such insight.
However, a few South African studies have alluded to the failure to implement programmes fully in schools as one of the reasons for their limited effects (James et al., 2006; Magnani et al., 2005; Visser, 2005). James et al., (2006) showed only an increase in knowledge about HIV in their evaluation of a life skills programme in KwaZulu-Natal. However, when their analysis was restricted to schools that had full implemented the programme, more positive effects were reported on perceptions of sexual behaviour, social connectedness, levels of sexual activity and condom use.
Failure to implement programmes as intended may be related to teachers’ attitude, skill and preparedness to teach sex/ HIV education as well as the level of priority afforded to, and acceptance of sex education within the education system. A recent evaluation of teacher training on an AIDS prevention programme in schools showed that despite intensive training, teachers lacked adequate and accurate knowledge on HIV, they struggled with non-didactic methods of teaching required for skills development such as role play, and they showed discomfort in teaching areas of the curriculum (such as safe sex practices) that conﬂicted with their own value system (Ahmed et al., 2006). The study recommended that sexuality education should be integrated into undergraduate teacher training courses to improve their knowledge, skills and comfort in teaching about HIV (Ahmed et al., 2006. In addition, Visser (2005) also reported on the lack of commitment towards the programme by teachers and principals, organisational problems in schools (lack of allocated time and human resources) and competing priorities in the school system that contributed towards limited implementation of programmes. A shift in the climate of the education system that is supportive of, and views sex education as a priority, is required for programmes to be effectively implemented (Visser, 2005).
Although the evidence for the effectiveness of sex education in SA is not as convincing, the available international experience suggests that the question is not whether sex education should be provided but how its effects can be optimised (Mukoma & Flisher, 2006). Incorporating as many of the characteristics that make for good programmes may assist in improving the outcomes of sex/HIV education in SA. In addition, well designed evaluation studies are required in SA to demonstrate and improve the effectiveness of school-based sex education programmes. To ensure a distinct focus on pregnancy, life skills evaluation studies should also include responsibilities of parenthood, knowledge and skills required for successful parenthood, together with the importance of planning for, and timing of parenthood (NRC & IOM, 2005).
School based/linked health services
A limited number of studies have evaluated the effectiveness of school-based / school-linked health services in the US as well as the distribution of condoms in schools and the evidence is mixed (Kirby, 2007). Schoolbased clinics are either available on the school premises or are located close to the school. While more than 80% of school clinics provide a range of reproductive health services, only about 25% dispense hormonal contraception or condoms (Kirby, 2007). Providing contraceptives through school-based clinics and making available condoms in schools do not increase the onset or frequency of sex. Instead, when contraception is available through school clinics and condoms can be obtained easily and conﬁdentially on school premises, many sexually active students make use of these services. But obtaining contraception from school-based sources is dependent on how widely it is available in the community. Studies that showed an increase in condoms obtained from school sources also reported a concomitant decrease in condoms obtained from other sources in the community. As such, a substitution effect occurred. In other words, schools can be an important source of condoms when they are not readily available from other sources in the community.
84 Teenage pregnancy in South Africa - with a speciﬁc focus on school-going learners The evidence from the US for increased contraceptive use and decreased childbearing as a result of schoolbased or school-linked health services is both weak and mixed (Kirby, 2007). More promising results were reported in two school and health facility-based programmes in developing countries (NRC & IOM, 2005).