«Suggested citation: Panday, S., Makiwane, M., Ranchod, C., & Letsoalo, T. (2009). Teenage pregnancy in South Africa - with a speciﬁc focus on ...»
In the following section, learner pregnancy rates are disaggregated by selected variables that characterise the school environment. As stated above, EMIS data records are incomplete. The table below shows the extent of missing data for the selected factors. As a result of the incompleteness of data, the analysis below must be read with caution.
Table 6: Selected variables and the proportion of missing data
As a census of pregnancies in all schools was not done, a single estimate of the number of pregnancies recorded for every 1000 female learners cannot be provided. Because the analysis outlined below is applicable to a sample of schools, the single value estimate of learner pregnancies is supported by a wider interval that is 41 Teenage pregnancy in South Africa - with a speciﬁc focus on school-going learners likely to include the ‘true’ estimate. This is referred to in the table as the 95% Conﬁdence Interval for the Mean.
For example, the estimated level of pregnancies for primary schools is 57.42 (55.39-59.45); for intermediate schools the estimate is 56.41 (54.06-58.75) and for combined schools the estimate is 78.13 (76.71-79.54). As the conﬁdence interval of primary and intermediate schools overlap, we make the conclusion that there are no signiﬁcant differences in the mean number of pregnancies between primary and intermediate schools. On the other hand, as the conﬁdence interval for combined schools does not overlap with the conﬁdence interval for either primary or intermediate schools, we are 95% conﬁdent that combined schools have higher learner pregnancy rates.
Most schools in SA are exclusively primary schools (grade 1 to grade 7) or secondary schools (grade 8 to grade 12). A minority of schools, however, combine both primary and secondary grades, and are known as combined schools. The analysis below shows differences in pregnancy rates according to these classiﬁcations.
Table 7: Distribution of learner pregnancy by institutional phase
The table above shows that pregnancy rates do not differ signiﬁcantly between learners in either primary, intermediate or secondary schools. The major difference is found in combined schools, where the pregnancy rate tends to be signiﬁcantly higher than in other schools.
The second attribute used to differentiate pregnancy rates was based on the level of specialisation of the school. Ordinary schools provide academic programmes; ‘comprehensive’ schools provide both academic and vocational training; and ‘specialised’ schools cater for learners with some form of disability.
Table 8: Distribution of learner pregnancy by level of specialisation
The above table disaggregates schools according to whether they are a fee or a no fee school. Generally, schools that are situated in poorer socio-economic neighbourhoods have been proclaimed in SA as no fee schools. The above table shows that schools in poor neighbourhoods have a higher pregnancy rate than those in more afﬂuent areas.
Land ownership Table 10: Distribution of learner pregnancy by school land ownership
Learners in schools that have land that is publicly owned have a lower chance of experiencing an early pregnancy than those in schools that have land privately owned. Schools on independently owned land are most likely to be farm schools. Higher pregnancy rates in these schools may be a reﬂection of higher levels of poverty in these areas.
The above analysis shows that pregnancy rates in schools vary signiﬁcantly according to the school characteristics identiﬁed above. However, as indicated earlier, results must be read with caution due to the high proportion of missing data. An indicator of no fee schools was used as a marker of the socio-economic status of the neighbourhood. Speciﬁcally, no fee schools are located in poor neighbourhoods and these schools tended to report higher pregnancy rates. This ﬁnding concurs with the secondary analysis conducted on the HSRC 2003 Status of the Youth Survey conﬁrming a link between poverty and risk for early pregnancy. In addition, the determinants section outlines the extent to which community factors such as poverty and low socio-economic status increase the risk for early pregnancy.
The second set of ﬁndings is more interesting. The results outlined above show that learners from combined schools have higher pregnancy rates. Combined schools have learners who have a vast difference in age. This condition is thought to skew power relations among learners, and likely to result in higher pregnancy rates.
Girls who engage in sex with older partners are more likely to become pregnant (Kirby, 2007). A big gap in age in the same institution is associated with conditions that give rise to higher pregnancy rates.
Provincial data Meaningful analysis of pregnancy rates reported at provincial level was not possible due to the following
• Some provincial reports only provide number of pregnancies without enrolment rates. These ﬁgures are thus unusable, as number of pregnancies can only be analysed against the background of the number of women who were at risk of experiencing a pregnancy.
• The fertility rates ﬂuctuate widely from one year to another. The general trend is that the number of pregnancies reported increases with time. All indications are that this is due to the improvement in the reporting mechanism rather than an escalation in learner pregnancy. Firstly, the earlier pregnancy rates suggested from school records are considerable lower than those reported in national surveys. Secondly, all national surveys have been reporting a decline in pregnancy rates during this period.
• The level of missing information signiﬁcantly compromises meaningful analyses. In addition, the level of missing information varies between provinces. The impression is that the ﬁdelity with which vital statistics are collected at provincial level varies greatly.
Suggestions for improving data collection A revision in the choice of data collected is strongly suggested. The following variables would allow for
meaningful differentiation of pregnancy rates at the school and community level:
• Variables related to the nature of the school: failure rate, school dropout rates, learner-teacher ratio etc.
• Characteristics of the school neighbourhood: school fee structure, urban-rural status, formal-informal environment, historical racial categorisation.
• Variables that locate schools within Statistics South Africa enumeration areas would allow a link between the school and other variables found in national data sets.
• In order to assess the effectiveness of the school policy that allows mothers/fathers to return to school post-pregnancy, data is also required on number of girls/boys who dropout of school because of pregnancy, educational attainment of pregnant teens, number who return post-pregnancy, number of teen parents who complete secondary schooling, academic performance of teen parents etc.
Restrictive laws and policies concerning abortion, particularly in developing countries, have resulted in many unwanted pregnancies and an escalation in obstetric complications and maternal deaths due to botched ‘backstreet’ procedures (WHO, 2004). Unsafe abortions pose a signiﬁcant risk to the health of young women in developing countries (NRC & IOM, 2005). Every year, between 2.2-4 million unsafe abortions are undertaken by adolescents in developing countries (UNFPA, 2007). While abortion on request is legally available in the US and UK and other developed countries, in Africa and in many developing countries abortion is still illegal and fraught with taboos and negative social perceptions. SA is one of a few countries in Africa where abortion is available on request in the ﬁrst trimester. Other countries are Mozambique (although with restrictions), Cape Verde, Ethiopia and Tunisia (IPAS, 2009).
44 Teenage pregnancy in South Africa - with a speciﬁc focus on school-going learners Prior to 1996, illegal abortions were common in SA and the majority of these were undertaken in the Black communities (Morroni, Myer & Tibazarwa, 2006). A study conducted in 1994 reported that of 44 868 women attending public hospitals with incomplete abortion, at least a third presented with signs and symptoms of unsafe procedures (Rees et al., 1997). These ﬁndings, together with a commitment to promote the reproductive rights of women motivated the passing of The Choice of Termination of Pregnancy (TOP) Act in SA in 1996.
The Act permits woman of any age to terminate a pregnancy during the ﬁrst 12 weeks of gestation without the consent of her partner, under restrictions between 13-20 weeks and under special circumstances beyond 20 weeks. Minors (women aged less than 18) can terminate pregnancies without the consent of parents.
Termination of pregnancy services are conducted at designated facilities, can be conducted by midwives and registered nurses and are available free of charge.
Despite legalizing abortion in 1996, statistics on abortions in SA are still difﬁcult to obtain. Three years after enacting legislation the health system was providing 40 000 legal terminations each year (Reproductive Rights Alliance, 1999). By 2003, this had increased to about 70 000 per year (Makiwane, 2009). Since the introduction of the TOP legislation in 1997 about 529 410 women have had safe and legal abortions in South Africa (IPAS, 2009). This is compared to about 800-1000 legal abortions conducted annually in Apartheid SA, mainly granted to White women (Anon, 1991). In 1999, only 292 facilities or less than three percent of public health facilities were designated to provide abortion services, and only a third was actually providing services (Dickson et al., 2003). By 2007/8, 70% of hospitals approved to provide TOP were actually providing services (DOH, 2008a).
Survey data on termination of pregnancy among young women indicate that a very small percentage make use of these services (3%) even though two thirds of pregnancies are unwanted (Pettifor et al., 2005). This is in contrast to data from the Department of Health that shows that 30% of abortions conducted in 2003 were for women aged 15-19 (Makiwane, 2009). Qualitative research indicates that abortion is common among school girls and university students, although legal termination is seldom a consideration (Kaufman, de Wet & Stadler, 2001; Varga, 2002). In fact, a case control study among pregnant adolescents in Cape Town reported that over a third (34.1%) had considered termination of pregnancy (Jewkes, Vundule, Maforah & Jordaan, 2001).
Barriers to legal abortion
Although abortion is recognised as morally and religiously objectionable, young women in SA apply a ‘relative morality’ to justify decisions about abortion (Varga, 2002). Stigma about abortion is tempered by the context in which girls ﬁnd themselves. To circumvent the social and ﬁnancial hardships associated with unplanned pregnancies and to protect educational opportunities, young women consider it a viable option. Yet two qualitative studies have suggested that legal termination may not be an option (Kaufman et al., 2001; Varga, 2002). Young people, particularly in rural areas have poor knowledge about the legality and cost of abortion services. The 1998 SADHS showed that knowledge about the TOP Act was low among teenagers (40.1%), and particularly, among uneducated women and women living in rural areas (DOH, MRC & Measure DHS, 2002). Focus group discussions conducted among adolescents in rural Limpopo province also reported low levels of knowledge about TOP (Ratlabala, Makofane & Jali, 2007). Although a more recent clinic-based study in KZN showed that knowledge levels had improved generally among women since the 1998 SADHS, a third of women were not aware of the legislation (Morroni et al., 2006). Furthermore, of those who were aware of the legislation about half were not aware of the time restriction for legally terminating a pregnancy. Due to a lack of support and information about the Act, many young women only seek abortion late into their pregnancy (NRC & IOM, 2005), forcing them to resort to illegal means.
45 Teenage pregnancy in South Africa - with a speciﬁc focus on school-going learners The double stigma of an early pregnancy and termination also serves as a deterrent for young women to seek abortion at a clinic or hospital (Varga, 2002). This is compounded by the negative attitude of health staff both towards early pregnancy and termination (Jewkes et al., 2005). Adolescents in Limpopo province also commented on the ‘harsh treatment that pregnant adolescents receive from health providers’ and the lack of conﬁdentiality in local clinics and hospitals (Ratlabala et al., 2007). Termination of pregnancy is seldom an individual decision. Mothers or older females in the family play a signiﬁcant role in deciding about termination often motivated by the need to protect the good name of the family (Varga, 2002). Partners also inﬂuence the decision based on their own concerns for school disruption and ﬁnancial reasons. However, one of the most critical factors that determine if an early pregnancy will be aborted in the Black community is the acceptance or rejection of paternity (Varga, 2002). Acceptance of paternity means ﬁnancial support, as well as social and cultural connection for the child. Rejection, on the other hand, compromises a girl and her families’ moral standing in the community. In the latter case, abortion, albeit illegal, becomes a real option.