Scoping review of risk factors of and interventions for adolescent repeat pregnancies: A public health perspective

Background Adolescent repeat pregnancy is of importance in public health because the birth of a second child to an adolescent mother compounds the adverse medical, educational, socio-economic and parenting outcomes. Repeat pregnancy in adolescence is not only an international phenomenon but also a local concern as it also occurs in South Africa. The prevalence of adolescent repeat pregnancy in Durban, KwaZulu-Natal, was reported as 17.6% in 2013. Aim This review aimed to gather relevant information from national and international sources to inform practice and to provide an understanding of what is known about the risk factors of and the interventions for adolescent repeat pregnancy. Methods A scoping review was undertaken using the Arksey and O’Malley framework. An electronic search was conducted using PubMed, Medline, Science Direct, Ebscohost, Sage and Wiley Online and Google Scholar. Results The search identified 3032 citations. After a review of the full text articles, 26 articles met the inclusion criteria. Risk factors pertaining to adolescent repeat pregnancy are categorised according to individual factors, partner relationship factors, family factors, peer factors, and social and community factors. Interventions to reduce adolescent repeat pregnancy have been largely influenced by the ecological framework. Across studies, adolescent mothers who received medical, psychosocial, educational, and family planning support experienced lower rates of repeat pregnancy. Conclusion A single ‘one-size-fits-all’ intervention for adolescent repeat pregnancy prevention is unlikely as different strategies were employed by the intervention programmes in this scoping review.


Introduction
Adolescent childbearing and repeat pregnancy are both public health and social problems. 1 Darroch et al. 2 reported that worldwide approximately 21 million girls in the 15-19 years age category became pregnant in 2016. 2 Furthermore, adolescent pregnancy and childbirth complications are a leading cause of the global burden of poor maternal health conditions and death in the 15-19 years age category. 3 The prevalence of adolescent repeat pregnancy is alarming. Approximately 12% -49% of adolescent repeat pregnancies in the United States of America (USA) occur within 1 year of the previous pregnancy. 4 Supporting data from Australia and Canada indicate that the prevalence of adolescent repeat pregnancy in these countries is 33% and 15.2%, respectively. 5,6 In the United Kingdom, one-fifth of births to adolescents under 18 years of age are repeat pregnancies. 7 Repeat pregnancy amongst adolescents is not only an international phenomenon, but it also occurs in South Africa. Though the national prevalence of adolescent repeat pregnancy is not known in South Africa, the prevalence in Durban, KwaZulu-Natal was reported to be 17.6% in 2013. 8 Adolescent repeat pregnancy is of particular importance in public health because the birth of a second child to an adolescent mother compounds the adverse medical, educational, socioeconomic and parenting outcomes. 9,10 In addition, girls who have repeat adolescent pregnancies generally experience suicidal ideation, depression and anxiety. 11 Adolescent parenting stress also increases with rapid repeat pregnancy, which could result in neglect of the second child and negative parenting behaviour. 12 Adolescent parents bearing two or more children within a period of 5 years are more likely to rely on social grants, drop out of school and experience additional childbearing within shorter birth intervals. 4,10,13, In the USA, taxpayers contribute almost $7 billion Background: Adolescent repeat pregnancy is of importance in public health because the birth of a second child to an adolescent mother compounds the adverse medical, educational, socioeconomic and parenting outcomes. Repeat pregnancy in adolescence is not only an international phenomenon but also a local concern as it also occurs in South Africa. The prevalence of adolescent repeat pregnancy in Durban, KwaZulu-Natal, was reported as 17.6% in 2013.
towards the burden of adolescent childbearing which includes health care, criminal justice, foster care, and public social and economic assistance. 13 Researchers have often referred to the topic of adolescent childbearing and rapid repeat pregnancy as a phenomenon or an enigma. 7,12,14 A range of opinions have been forwarded amongst public health professionals, clinicians, social scientists, advocacy groups and the media regarding the issue of adolescent pregnancy. The current debate centres on the conceptualisation of adolescent pregnancy as a public health problem versus adolescent pregnancy as a reproductive choice and process. Adolescent pregnancy is culturally accepted in different parts of the world despite being labelled a public health problem. Some adolescent health experts have refuted the claim that adolescent pregnancy results in catastrophe for the mother and her infant. 15, 16 Macleod 15 argues that the moral judgment of adolescent pregnancy has now been substituted by 'scientific scrutiny' (p. 59).
A study by Smith and Pell 17 found a causal association between repeat adolescent pregnancy and poor birth outcomes. However, these findings were dismissed on the stance that adolescent repeat pregnancy is not a public health problem and that birth outcomes are largely influenced by confounders such as socio-economic circumstances. 18 Criticising Lawlor and Shaw's 18 argument that adolescent pregnancy is not a public health problem, Scally 19 argues that adolescent pregnancy requires integrated public health action through several sectors to help adolescents prevent unwanted pregnancies and manage the health, economic and social consequences of the pregnancy and birth. In support of Scally's 19 assertions, Rich-Edwards 20 advances the explanation that poverty is a risk factor for adolescent pregnancy and that premature parenthood influences the cycle of future poverty, making this a public health problem.
On the basis of these arguments, the literature review was essential in facilitating a better understanding of the public health approach towards adolescent repeat pregnancy. Understanding the risk factors of and the interventions for adolescent repeat pregnancy can provide guidance to health practitioners and decision makers. This scoping review was designed with the purpose of gathering relevant information from national and international sources to inform practice and to provide an understanding of what is known about the risk factors of and the interventions for adolescent repeat pregnancy.

Defining repeat adolescent pregnancy
The literature distinguishes between adolescent repeat pregnancy and rapid adolescent repeat pregnancy. Adolescent repeat pregnancy is defined as a second pregnancy or additional pregnancies to a woman younger than 20 years of age. 6 Most research to date has focused on rapid repeat pregnancy amongst adolescents. 21,22,23,24,25,26 Rapid adolescent repeat pregnancy is defined as a second birth or pregnancy that occurs within 2 years of the previous pregnancy. 24

Methods
This literature review employed a scoping review methodology based on the framework by Arksey and O'Malley 27 and the recommendations put forward by Levac et al. 28 According to Davies et al., 29 '… scoping involves the synthesis and analysis of a wide range of research and non-research material to provide greater conceptual clarity about a specific topic or field of evidence' (p. 1386). The study followed the stages of the Arksey and O'Malley 27 framework which included: (1) identifying the research question, (2) identifying relevant studies, (3) study selection, (4) charting the data and (5) collating, summarising and reporting the results.

The research question
The research question was generated by our public health concerns about adolescent repeat pregnancy. The scoping review was guided by the research question: 'what are the risk factors of, and the interventions for adolescent repeat pregnancy?' A key aspect of a scoping review is a broad and comprehensive research question to provide breadth of the literature.

The data sources and search strategy
The databases used to conduct the literature search were selected on the basis that the topic of adolescent repeat pregnancy encompasses social science research and public health research. The databases included PubMed, Medline, Science Direct, Ebscohost and Wiley Online. Search engines included Google and Google Scholar. The search terms included 'adolescent pregnancy' and/or 'adolescent repeat pregnancy' and/or 'adolescent parenting' and/or 'adolescent repeat pregnancy risk factors' and/or 'consequences of adolescent pregnancy' and/or 'secondary pregnancy prevention'. The snowballing technique was applied by identifying references in the scrutinised review articles to obtain detailed and relevant information. Searching involved published literature from 1990 to 2016.

Study selection
When the relevant literature was identified, the inclusion and exclusion criteria were established ( Table 1). The papers that were included for this scoping study comprised published peer review studies, review articles and opinion articles. Figure 1; depicts the scoping process which includes the number of publications retrieved and selected from the database search.

Description of studies
The online searches yielded 3032 relevant citations of which 2698 were eligible for title and abstract review. Of these, a total of 87 articles were retrieved for full text screening. After review of the full text articles, 26 articles met the inclusion criteria (illustrated in Figure 1). The articles were summarised into themes of risk factors and of interventions for adolescent repeat pregnancy. Nine of the 26 articles addressed the risk factors of adolescent repeat pregnancy. The majority of the studies (67%) addressing the risk factors of adolescent repeat pregnancy were from the USA. The description of included studies for the risk factors of adolescent repeat pregnancy is presented in Table 2. Seventeen articles in this scoping review addressed interventions on adolescent repeat pregnancy. Details of these 17 articles included for the analysis and discussion of the interventions on adolescent repeat pregnancy are presented in Table 3.
Rowland 30 demonstrates a comprehensive ecological model for the predictors of rapid repeat pregnancy amongst adolescent mothers. For adolescents experiencing a repeat pregnancy, the risk factors are categorised according to individual factors, partner relationship factors, family factors, peer and school factors and social and community factors ( Table 4).
The individual factors relating to adolescent repeat pregnancy include race, ethnicity, age of first pregnancy, int. ended first pregnancy and attitudes towards early childbearing 10,30,31 In addition, cognitive functioning and a history of depression have been linked to adolescent repeat pregnancies. 14 Lewis et al. 22 found that repeat pregnancy was more likely in indigenous Australian adolescents (odds ratio 2.38, 95% confidence interval (CI) 1. 38-4.11). Poor obstetrical outcomes such as miscarriages have been associated with rapid repeat adolescent pregnancy. Partner relationship factors such as living with an older partner, being married at a young age, having a new partner and intimate partner violence are more common in those adolescents with a repeat pregnancy. 30,31 Family factors that are related to subsequent adolescent repeat pregnancies include dysfunctional mother-daughter relationships, having mothers with low educational levels, intergenerational adolescent pregnancy, and lack of family support. 30,31 Equally important, an adolescent mother who lives with her mother after the birth of her first child and relies on her mother for financial and social support is more likely to experience a repeat adolescent pregnancy. 30,31 Peer factors in particular, such as association with friends who are adolescent parents, postpartum school enrolment and low educational ambition, have been linked with adolescent repeat pregnancy. 30,32 The social and community factors associated with adolescent repeat pregnancy include low socio-economic status, low educational status, and society norms that accept adolescent childbearing. 30,31,32 These ecological factors have been tested in studies conducted by Lewis et al., 22 Jacoby et al., 23 and Raneri and Wiemann. 24 Jacoby et al. 23 found a strong association between adolescent rapid repeat pregnancy and interpersonal violence (p < 0.00001; odds ratio 22.6). Similarly, Raneri and Wiemann 24 demonstrated that adolescents with repeat pregnancies had been physically abused by their boyfriends or husbands

Inclusion Exclusion
• Articles written in English.
• Articles not written in English.
• Articles involving a secondary or repeat pregnancy prevention intervention for adolescents.
• Articles focusing on primary pregnancy prevention for adolescents.
• Study participants were defined as adolescents (13 to 19 years of age).
• Study participants who were not defined as adolescents.
• Articles identifying risk factors of repeat adolescent pregnancy.
• Intervention studies with baseline and post-intervention data.
In the South African context, Mphatswe et al. 8 found that adolescents with repeat pregnancies were involved with partners who were 5 years or older than them, or they had had multiple sexual partners in the previous 12 months. The prevalence of HIV was also higher in South African adolescent repeat pregnancies than in first time adolescent pregnancies.
According to Raneri and Wiemann 24 the strongest predictor of adolescent repeat pregnancy is not using a long-acting contraceptive postpartum. Steven-Simons et al. 26 investigated the adoption of the contraceptive implant to avoid adolescent repeat pregnancy. The authors found that the rate of repeat adolescent pregnancy was significantly lower for early implant users (12%, p < 0.0001) than for other adolescent mothers using alternative methods of contraception (46%). Lewis et al. 22 demonstrated that this claim was warranted and advocated for the provision of long-acting contraceptives to adolescent mothers.

Interventions to prevent adolescent repeat pregnancy
Interventions to reduce adolescent repeat pregnancy have been largely influenced by the ecological framework of individual, partner, familial, peer, and community factors ( Table 5). First time adolescent mothers who attended the Queens Hospital community-based programme in the USA, which provided medical, psychosocial, educational and family planning support through a multidisciplinary team (i.e. a team including a gynaecologist, paediatrician, social worker and health educator), had a lower repeat pregnancy rate (9%) than the control group (70%). 33 Similarly, O'Sullivan and Jacobsen 34 found that only 12% of the adolescent mothers who participated in the health care programme for first time adolescent mothers (behavioural skills development, contraceptive education, life option enhancement, sexuality, HIV and sexually transmitted infections (STI) education, infant care and education support attainment) had a repeat pregnancy in comparison to 28% of adolescent mothers in the control group at 18 months postpartum.
A study carried out by Cox et al. 21 in the USA also evaluated a hospital community-based medical home model for adolescent mothers and their children in which the outcomes of programme retention, health care utilisation, infant immunisation status, contraceptive use, repeat pregnancy, depressive symptoms, social support, school attendance and employment status were measured. The study design was a prospective single cohort study with the following programme components; infant care education, life orientation, social services support, family support groups, contraceptive education, day care and community outreach services. As there was no control group, benchmark comparison data were used through a systematic literature review. The adolescent repeat pregnancy rate was 14.7% at 12 months and 24.6% at 24 months. Cox et al. 21 compared the findings of their study to those of Barnet et al., 9 who found that the repeat pregnancy rate amongst adolescent mothers in the intervention group was 45% at 24 months.   • The duration of the intervention was 18 months. Numerous studies have focused on specific community-based health care centres for adolescent mothers with emphasis on secondary pregnancy prevention. For example, Solomon and Liefeld 25 reported on the success of a family growth centre (FGC) where the repeat pregnancy rate amongst adolescent mothers in the intervention group was 12% in comparison to 41% in the control group in a 3-year longitudinal study. The FGC was designed using the Ecological Model and the Family Support Model, taking into consideration that adolescent pregnancy does not occur in a vacuum. Similarly, the Women's Centre of Jamaica Foundation (WCJF) Programme for adolescent mothers found that the programme participants' risk of a repeat pregnancy was reduced by 45% 4 years post evaluation. 35 The WCJF programme components included community-based pregnancy prevention, parenting education, family planning services, life skills orientation, job training and placement, counselling services for adolescent fathers, and school support services.

Rabin and Seltzer
By employing the ecological framework and social cognitive theory (SCT), Ford et al. 36 evaluated a peer-centred prenatal programme for adolescent mothers. The study focused on birth weight, repeat pregnancy and educational outcomes at 12 months postpartum. The intervention included group therapy and prenatal education whilst the control group received the usual individual prenatal care. Whilst it can be argued that the repeat pregnancy differences between the intervention and control groups were not statistically significant, the repeat pregnancy rate in the peer-centred prenatal programme was lower than in the control group (13.4% vs. 15.9%) Some authors argue that school-based programmes for adolescent mothers are also effective in reducing repeat pregnancies. 37,38 For example, in a South Carolina high school, a school-based programme that provided comprehensive care found that the repeat pregnancy rate was 17% in the treatment group in comparison to 29% in the control group at 3 years after the previous birth. 37 The school-based programme included medical services, social services, peerbased education, contraceptive education, parenting skills and educational attainment support for adolescent mothers.
In a similar school-based programme, at the Polly T McCabe Centre, adolescent mothers who stayed for more than 7 weeks at the school-based centre had a lower repeat pregnancy rate at 24 months versus the adolescent mothers who stayed at the centre for 7 weeks only (12%, 6/50 vs. 36%, 19/52, p = 0.005). 38 In addition, Corocan and Pillai's 39 meta-analysis of 16 studies focusing on adolescent repeat pregnancy prevention programs, which included schoolbased programs, revealed a 50% reduction in the likelihood of repeat pregnancies.
Apart from community and school-based interventions, home nurse visiting and motivational interviewing (MI) have also proven to be effective in delaying and reducing repeat pregnancies amongst adolescent mothers. 37,40,41 The effects of a home-based mentoring programme on repeat pregnancy prevention for first time adolescent mothers was evaluated

Race
Having a mother with low educational status. Dropping out of school after first pregnancy.
Having an older partner (three or more years older).
Lower socio-economic status.

Ethnicity
Intergenerational adolescent pregnancy. Adolescent mother fails to return to school six months postpartum.
Intimate partner violence.
Low education levels.

Depression and suicidal ideation
Lack of family support.
Having more friends who are adolescent mothers or have experienced adolescent pregnancy.
Failed relationship with the partner responsible for index pregnancy. Cultural and societal norms regarding childbearing and parenting.
Religiosity Dysfunctional family-poor mother-daughter relationship. Low educational motivation.
In relationship with new partner postpartum.
-Risky sexual behaviour or recommencement of sexual intercourse less than six months postpartum or ongoing sexual intercourse for more than three months post-delivery.
Abuse and neglect. Friends place emphasis on childbearing and motherhood.
Married at a young age (adolescent).
-Having more than one sexual partner in the last 12 months postpartum.
Domestic violence.
-Partner pregnancy intention (wants a repeat pregnancy). Child reared by grandmother who provides socio-economic support.
---by Black et al. 41 The mentorship was based on a 'big sister approach' and delivered over 19 home-based lessons. At 2 years postpartum, the adolescent mothers in the intervention group had a lower repeat pregnancy rate at 11% versus 24% in the control group. As with previous studies that have been discussed, the loss of participants at follow-up was significant because 18% of the mothers did not complete the 24-month evaluation.
Barnet et al. 42 conducted a randomised trial in Baltimore, MD, USA, to evaluate the effectiveness of MI through a computerassisted motivational intervention (CAMI) to prevent adolescent repeat pregnancies. Motivational Interviewing is a counselling approach that was used to help adolescent mothers find the motivation to facilitate positive behaviour change. 42 The study recruited and trained counsellors on the computer-assisted MI. These counsellors provided motivation Author (year) Adolescent repeat pregnancy Contraception uptake Educational attainment or school enrolment Barnet et al. (2007) • No significant impact on repeat pregnancy. • Repeat pregnancy in the intervention group was 45% (14/31) at 24 months whilst control group had a repeat pregnancy rate of 38% (12/32).
• CAMI only group repeat pregnancy rate was 17.2%. The usual care group (control) had a repeat pregnancy rate of 25.0%. -- • Intention to treat analysis revealed that adolescent mothers in the control group were 2.5 times more likely to have a repeat pregnancy (24% vs. 11%, p = 0.05).
-No differences were noted with regards to maternal educational attainment.
Cohen et al.

Corocan and
Pillai (2007) • According to the meta-analysis, secondary repeat pregnancy interventions in the first 19.13 months yielded a 50% reduction in the odds of pregnancy in comparison to the usual care or control groups. -- • The cumulative repeat pregnancy rate at 12 and 24 months was 14.7% and 24.6%, respectively. The repeat pregnancy rate at 24 months was lower in comparison to benchmark data. -- • At 8 weeks postpartum, the initiation of LARC decreased the odds of rapid repeat pregnancy within 24 months (OR = 0.018, 95% CI: 0.035-0.307). • The repeat pregnancy rate at 24 months was 35%. -- • The rate of repeat pregnancy rate was 37% (32/87) amongst WCJF participants versus 60% (104/173) of non-participants. The participants in the WCJF reduced their risk of one or more repeat pregnancies by 45%.
The contraception use was higher amongst WCJF participants (80 of 87) in comparison to 147 of 173 non-participants (p = 0.04).
Ford et al.
• At 12 months, the repeat pregnancy rate in the experimental group was 13.4% versus 15.9% in the control group.
-There were no differences in maternal educational outcomes.
Key et al.
• The rate of repeat pregnancy amongst participants in the school-based intervention was 17% versus 33% in the control group (p = 0.001).
-There were no differences in maternal educational outcomes.
Lewis et al.
• At 24 months, 35% of adolescents had a repeat pregnancy. Adolescents who chose Implanon experienced a repeat pregnancy at 23.8 months (95% CI: 22.2-25.5) whilst adolescents who chose COCP and/or DMPA experienced a repeat pregnancy rate at 18.1 months. • Implanon users were less likely to experience subsequent pregnancies.
Adolescents who chose Implanon were more likely to continue the usage of this contraceptive method at 24 months in comparison to those using COCP and/or DMPA (p = 0.001).
-O'Sullivan and Jacobsen (1992) • At 18 months, the repeat pregnancy rate was 12% (13/108) in the intervention group versus 28% (32/118) in the control group. Adolescents in the health care programme were less likely to experience a repeat pregnancy.
-There were no differences in maternal educational attainment. Rabin and Seltzer (1991) • There was a 9% repeat pregnancy rate in the intervention group versus 70% in the comparison group.
Eighty-five percent of adolescent mothers in the intervention group were more likely to use contraception versus 22% in the comparison group (p ≤ 0.0001).
The school attendance rate amongst adolescent mothers in the intervention group was 77% versus 38% in the control group (p ≤ 0.0001). Seitz and Apfel (1993) • Adolescent mothers who stayed at the Polly T McCabe Centre for more than 7 weeks had a lower repeat pregnancy rate at 24 months versus the adolescent mothers who stayed at the centre for 7 weeks only (12%, 6/50 vs. 36%, 19/52, p = 0.005).
-Of the adolescent mothers who had not experienced a repeat pregnancy at 24 months, 69% completed high school compared to the 35% of the repeat adolescent mothers (p ≤ 0.005). Solomon and Leifeld (1998) • At 24 months, 3 repeat pregnancies (10%) occurred in the intervention group in comparison to 11 (33%) in the control group (p = 0.006).
-At 24 months, only 3 of 43 mothers in the intervention group dropped out of school in comparison to 12 of 29 in the control group (p = 0.002).
Steven-Simons et al.
• At 24 months, the repeat pregnancy rate amongst implant users was 12% versus 46% in the control group ( p ≤ 0.0001).
• At 6 months, there were no repeat pregnancies in the intervention group IPI in comparison to 9.9% in the control group. • The IPI continuation rates at 6 and 12 months were 96.9% and 86.3%, respectively. • At 12 months, the repeat pregnancy rate in the intervention group was 2.6% (4/153) in comparison to 18.6% (38/204) 43 found reduced repeat pregnancy rates at 24 months amongst adolescent mothers who had initiated a long-acting reversible contraceptive within 8 weeks of delivery. A recent study by Cohen et al. 44 also revealed that the repeat pregnancy rate is lower amongst adolescent mothers who initiate postpartum long-acting reversible contraceptives (LARC). In this regard, contraceptive implants could be an important intervention for adolescent repeat pregnancy.

Discussion
Adolescent repeat pregnancy has been studied extensively in other countries but in the African continent, studies on this phenomenon are scarce. With regard to the risk factors for adolescent repeat pregnancy, the ecological framework best explains this phenomenon. The results obtained by Mphatswe et al. 8 on the prevalence and the risk factors on adolescent repeat pregnancy in KwaZulu-Natal, South Africa, were similar to that of other published studies around the globe. The scoping review of the literature demonstrates that interventions on secondary pregnancy prevention have positive outcomes for adolescents. These comprehensive interventions offered a wide range of medical services, education and psychosocial services which targeted repeat pregnancy, school enrolment and contraceptive uptake. Furthermore, the literature indicates that the ecological framework provides guidance on strategies to prevent adolescent repeat pregnancy.
The vast majority of interventions in this scoping review recruited participants from clinics, hospitals and community centres. In this regard, adolescent mothers not seeking health care would have been excluded. The findings of this scoping review illustrate the need for adolescent repeat pregnancy interventions to shift from individuals to the wider social and community context that influences adolescent sexual and reproductive health. Interventions by Tocce et al., 4 Lewis et al. 22 and Steven-Simons et al. 26 drew attention to the use of contraception in adolescents.
In interpreting the results, it is noteworthy that contraception as an intervention plays a vital role in the adolescent repeat pregnancy phenomenon. In this regard, healthcare providers need to support adolescents in the access and use of contraceptives. The reported levels of contraception uptake and use are low amongst both married and unmarried adolescents. 45,46 This implies that, though most married adolescents do not want a pregnancy, their contraceptive use is lower than that of sexually active unmarried adolescents. 45 In a nutshell, there is an unmet need for contraception amongst adolescents in general. For example, a study in Nigeria by Ahanonu 47 found that health care providers did not approve of adolescent premarital sex and that they, therefore, felt that prescribing contraceptives to adolescents was promoting promiscuity. Other factors that influence contraception use include: (1) sexual experiences, (2) psychosocial development, (3) gender issues, (4) previous health care experiences, (5) access to health care services, (6) access to information, (7) health and education structures and (8) social and cultural norms. 46 In the South African context, Ehlers's 48 explorative descriptive survey study found that only 48% of 250 adolescent mothers surveyed had used contraception.
A 'single one-size-fits-all' intervention for adolescent repeat pregnancy prevention is unlikely as different strategies were employed by the intervention programmes in this scoping review. In short, health care services need to be tailored to the needs of pregnant and parenting adolescents. Interventions can be implemented at various sites, including clinics, schools and community centres.

Study limitations
For practical reasons, only articles written in English were considered for this scoping review. The authors acknowledge that important published research may have been omitted using the method outlined in the search methodology. Rather, the review was designed to further illuminate the phenomenon of adolescent repeat pregnancy and to provide an understanding of what is known about the risk factors of and the interventions for adolescent repeat pregnancy.

Conclusion and recommendations
Health services, health care workers and health information are essential building blocks of the public health system. According to the World Health Organization (WHO), 49 the '…strengthening of health systems is everybody's business'. In this regard, the improvement of health care services for adolescent mothers is indeed everybody's business. Based on this review, it can be concluded that health care providers play an important role in the prevention of adolescent repeat pregnancies as it was shown that, across various studies, adolescent mothers who received medical, psychosocial, educational and family planning support experienced lower rates of repeat pregnancies.
To address the shortage of literature on adolescent repeat pregnancy in South Africa, future research needs to focus on this phenomenon and the strengthening of comprehensive health care services for adolescent mothers. Recommended for inclusion in future interventions, the Pinzon and Jones 50 paediatric policy statement for the care of adolescent parents and their children highlights the following points: (1) continuity of care (creation of a medical home model), (2) provision of a multidisciplinary approach to care (nursing, social services, nutritional care, developmental screening services), (3) breastfeeding support, (4) contraception, (5) youth development, (6) promoting academic achievement, (7) promoting a healthy lifestyle, (8) addressing mental health issues, (9) parenting skills, (10) development of support groups and (11) active involvement of adolescent fathers. Adolescent repeat pregnancy is a huge burden on the adolescent mother, her family, her community and her country of residence. 51 According to Morris and Rushwan, 46 health care workers need to 'move away from being part of the problem to being part of the solution' (p. S41).