key: cord-0813435-iufzidhi
authors: Hunter, Janine; van Blerk, Lorraine; Shand, Wayne
title: The influence of peer relationships on young people’s sexual health in Sub-Saharan African street contexts
date: 2020-08-14
journal: Social Science & Medicine
DOI: 10.1016/j.socscimed.2020.113285
sha: a639b6f307b02b96049d7bb57e7fe6a2e0f69538
doc_id: 813435
cord_uid: iufzidhi

Abstract This paper explores the interaction between peer relationships and sexual health among street youth in three Sub-Saharan African cities: Accra (Ghana), Bukavu (Democratic Republic of Congo), and Harare (Zimbabwe). It begins by conceptualising peer relationships for youth globally and considers why these are pivotal for young people living in street settings. The paper reconceptualizes street peer relationships not as replacement families, but as sharing ‘social anchorage’ in the street space. It draws on qualitative ethnographic data from Growing up on the Streets, a longitudinal research project with a participatory methodology undertaken between 2012-2016 and engaging street youth (aged 14-20 at project outset) trained in ethnographic observations as research assistants (n=18), following a network of ten peers (n=229 by 2016), reporting their experiences in weekly interviews with facilitators. A wider network attended focus groups (n=399). The project engaged a ‘capability’ approach, with ten capabilities defined by street youth as key to their daily lives. Empirical evidence is from a subset of data qualitatively coded (using NVivo) against capabilities ‘Health and Wellbeing’ and ‘Friendship’, across all interviews, focus groups and cities (n=212 sources). In exploring this intersection, the paper demonstrates beneficial and adverse impacts of peer influence on sexual health, including misinformation about contraceptives and death from an informal sector abortion; highlighting findings from across the three cities around primacy of same-sex peer relations, mistrust between genders and in healthcare providers. The paper finds that while street youth remain subject to cultural norms around gender identities, street peer relationships hold a persuasive power; contributing to both everyday survival and moments of acute need. It concludes that recognising the right of young people to live and seek livelihoods in urban settings, and adopting the social networks they create to advance street youth’s sexual health has become even more relevant in a (post)pandemic world.

The migration of young people from rural to urban environments has resulted in an explosion of youth living in dynamic, informal, city environments with fragile economies and healthcare systems (Le Roux & Smith, 1998) . In Sub-Saharan Africa (SSA), 43% of the population is under the age of 15, compared to 17% in countries defined as high-income (UNDESA, 2017) . Of an estimated population of 189 million living in SSA's informal settlements (UNDESA, 2019a) , 81 million are therefore growing up amid urban expansion, bringing new challenges to the delivery of healthcare and health education. This paper will examine qualitative evidence for the significance of peer relationships upon the sexual health of young people living on the streets, drawing on longitudinal ethnographic research in three African cities: Accra, Ghana; Bukavu, Democratic Republic of Congo (DRC); and Harare, Zimbabwe. The term 'street' is used here to encapsulate the public spaces inhabited by homeless young people; where they live, work, eat, sleep and conduct their relationships. Across the three countries in our research, populations are predominantly young, with around a third of the population aged between 10 and 24 (UNFPA, 2019). The median age for DRC, Zimbabwe and Ghana is 17, 18.7 and 21.5 respectively, well below the average of 29 for low-income countries (UNDESA, 2019a) . For those living in informal settlements, the median age of sexual debut is 15 for both young women and young men, in comparison to 18 and 17 in other urban environments (Klett-Davies, 2017) . The adolescent birth rate among young women aged 15-19 ranges from 76 per 1,000 women in Ghana to 138 per 1,000 women in DRC (figures for UNFPA, 2019) , in a region that sees 61% of all new global HIV infections; around 3050 every day (UNAIDS, 2019) .

In the cities of SSA, street youth frequently live and work outside intergenerational family settings. Social peer networks become intrinsic to daily life by contributing to livelihoods and socialisation, with young people learning social expectations and behaviours from their peers (Brown & Larson, 2009) . 'Peer J o u r n a l P r e -p r o o f relationships', encapsulate relationships between individuals within monogenerational groups, and include friendship, enmity, intimacy, partnership and sex (Stoebenau et al., 2016; Shand et al., 2015) . Both young men and women can engage in sexual relationships from early to mid-teens (Klett-Davies, 2017; Kabiru et al., 2010) , often in concurrency, in sexual exchanges for food, shelter, safety, or money (Hunter, 2002) .

Ostensibly free from prevailing traditional moral regimes of extended familial settings (Ungruhe, 2014) , peer relationships fill voids of emotional and physical care. Relationships with peers can have benign influence on the lives of street youth, including everyday provision of moral and financial support, or assistance at times of acute need; adverse impacts include the creation of sub-cultures where peers fail to support long-term relationships or protective behaviours, sharing deviant behaviours such as drug abuse (Rice, 2005) or misinformation on the side-effects of contraception (Agyemang, 2019) .

The paper begins by creating a conceptualisation of peer relationships, before examining why these relationships are critical in SSA street contexts. The Growing up on the Streets research is described, including the methodological approach used in the large-scale qualitative project. An extract from the Growing up on the Streets data, an in-depth case study from Accra exemplifying peer experiences and the impact of peer advice on sexual health, is presented. In order to explore the complexity of the issue, this is contrasted with data from Bukavu and Harare to show similarities for street youth in different contexts. This is followed by a discussion placing our findings within cultural and health contexts and their implications for health service delivery and what this tells us conceptually about peer relationships.

With the transition to adolescence, peer relations grow more complex, embedded and pertinent in young peoples' lives. More time is spent outside of familial or institutional settings and within groups where youth seek new identities and mechanisms of affirmation via peer group formation among their peers in age and social status (Brown & Larson, 2009; Rice, 2005) . Such processes are socially normative, subject to contextual and cultural influences, as well as biological (Crone & Dahl, 2012) . This co-creation of distinct social micro-worlds among adolescent peer groups requires time and proximity rarely found outside this J o u r n a l P r e -p r o o f period in the lifecourse. Adults can feel excluded by the shift of allegiances and wary of new external influences as young people seek to define their own aspirations and belief-systems (Brown, 1990) . Peer influence is often perceived as the negative form of peer relations; linked to behavioural change (Hartup, 2005) or an "antisocial" capital characterised as a threat to community wellbeing as a whole (Lee, 2001) . Such perceptions further segment the role of young people in communities and reinforces their positionality as "out of place" and unwelcome in public spaces (Connolly & Ennew, 1996) .

Reviewing several studies, Brown & Larson (2009) found that peer influence was built around reciprocity with influence flowing bilaterally, as opposed to an adultist perception of unilateral influence; adults having something to teach young people, but nothing to learn in return. For young people growing up in high-income countries, it is reasonable to expect that peer relations sit within a holistic collection of bilateral social relations amongst peers and "significant others" (Ibid: 78). Family members, teachers, sport coaches and community members create an intergenerational social network of role models, advisors and supporters.

While relationships with families tend to be conceptualised as a source of financial and emotional support (Parker & Mayock, 2019; Gillies 2000) , in low-income countries, and in particular for young people living in street contexts, kin relationships are "stretched across time and space" (van Blerk, 2012: 323) as young people migrate to the city seeking new livelihoods. This intra-and inter-national migration from rural familial settings to independence in urban settings is well documented (e.g. Young, 2004) . Ongoing familial relationships may remain important, if distinct from nuclear conceptions of family (van Blerk, 2012) .

Homeless or street youth conduct intergenerational relationships outside of family; they include those with formal actors (e.g. social workers, healthcare practitioners, police) and non-formal actors (e.g. informal employers such as market traders, fishermen, sex work clients). Both, to a greater or lesser extent, hold positions of power or influence in their lives. While extant, these intergenerational relationships are primarily transactional, meaning that multifaceted peer relationships are amplified in importance with positive and negative results (Petering et al., 2014; Langevang, 2008; Rice, 2005) . It is within the "everyday practices or actions" (Parker & Mayock, 2019: 555) of peer relationships, anchored on the street, that street youth build "life worlds" which are "based on reciprocal companionships in everyday life and times of crises" (Ungruhe, 2019: 52) .

J o u r n a l P r e -p r o o f

In studies of young people in SSA street settings, it has been hard to resist the urge to cast street youth into one of the "diverging trajectories of iconic child victims" (Poretti et al., 2014: 22) , such as AIDS-orphan, victim of violence, or as they get older, delinquent. Friendship groups on the street are often characterised as "gangs" (Corburn et al., 2020; Heinonen, 2011) , associated with criminal activity and delinquency, as well as deviancy from cultural norms (Aptekar & Stoecklin, 2014) . In other youth communities where kin relationships may be partially or wholly disrupted, or spatially or emotionally remote, peer groups replace kin relations with "chosen families" of peers who share lifestyles, sexualities or beliefs; for example in the case of

LGBTQ youth of color in the US (Greene, 2018), LGBTQ2S youth in Canada (COH, 2016), or drug-using homeless youth (Rice, 2005) . For young people living in liminal city spaces, street peer groups to some degree replace or replicate family support (Aptekar & Stoecklin, 2014) . However, for street youth the concept of "family" itself is troublesome, a social construct (Parker & Mayock, 2019) with wide variations across cultural contexts. Peer relations map on to street youth's experiences of familial relationships; being flawed and problematic, conflicted and coercive, as well as nourishing and supportive.

More useful here may be the, as yet, underdeveloped concept of "social anchorage" (Due et al., 1999: 663) , which chimes with the significance of place for street relationships. This concept emphasises that the social networks are anchored to the spaces they create together while they work, sleep and are at their leisure; similar to a 'home'. Social anchorage can recognise the right of street youth to occupy street spaces and their shared social connections which replace "the family as a source of emotional and economic support" (Le Roux & Smith, 1998: 1). Thus street peer relationships are based on anchorage in the street space, sharing everyday strategies of income-generation and survival, and involving continuous negotiations over power, obligations and exchanges between street youth and with the broader community. For example, in a study of street children in Colombia, Aptekar describes groups of street children who are associates for mutual financial gain, while others establish "chumships", making them "better able to deal with the demands of street life" (1989: 791).

Thus, while understanding street peer relationships as a new form of family may be useful, we propose a realignment of relational belonging which more accurately reflects the experiences of SSA street youth. Social anchorage recognises the importance of street spaces as the backdrop for everyday practices of lived survival within it and a unifying context to peer social networks that is something other than the socio-emotional constructs of (substitute) family. Peers do not seek to perform the role of family, though they may at times fulfil aspects of familial roles. This is partially due to the quantity of peer interactions; numerous relatively shallow peer relationships, sharing social anchorage in a street space, with a few deeper friendships. Unlike most familial relationships, peer relationships are subject to greater temporal variation, and unlike familial patterns of parenting, love, obligation, and negotiation, peer relations engage in mutually beneficial exchange ranging from unconditional assistance in times of need to lucrative or dependent transactional relationships. It is this complex interplay of relations which we are reconceptualizing not as street family but as street peer relationships; socio-spatially co-created by young people living on the street.

Literature around adult sexualities in diverse African contexts has tended to overlook young people's sexual relationships (Bhana, 2017) . Studies focusing on street-living young people cover a spectrum of experiences including early sexual debut (Kabiru et al., 2010) , informal sector abortions (Chemlal & Russo, 2019) , and rape (Swart-Kruger & Richter, 1997) . Most adhere to a "storm and stress" model of adolescence (Gillies, 2000: 213) , where "teenage years contain the most intense physical, emotional and mental experiences of our lives" (Heinonen, 2011: 131) . The representation of childhoods in situations of precarity, and African contexts in particular, as "calamitous and catastrophic" (Bhana, 2017: 244) fails to recognise youth agency as they negotiate their own sexualities while navigating both the transition into adulthood alongside their daily survival, to which sex is both contributor and threat (Swart-Kruger & Richter, 1997) .

Street youth grow into sexual maturity surrounded by peers in mixed urban settings such as informal settlements, market places and transport depots. Negotiating sexual relationships is, for most young people, a marker of growing up; sex is also a means of subsistence for young women and men on the streets (Hunter, 2002) . Transactional and concurrent relationships are conducted with street peers, using multiple sexual J o u r n a l P r e -p r o o f partnerships or relational multiplexity as a means of survival (Stoebenau et al., 2016) and contributing to the disproportionate impact of STIs on street youth (Anarfi, 1997) .

While African countries are heterogeneous, they share legacies of colonial, cultural and religious influence which sees the reproduction of attitudes and practices; for example, polarised gender identities underpinned by prevailing dominant masculinities, resulting in double standards between genders (Tamale, 2017; Bhana, 2016; Ratele, 2011; Evans, 2006) . Street youth live outside of this cultural norm, yet conduct sexual relationships subject to these culturally normative femininities and masculinities. Their sexualities are an important expression of gender identity, but are circumscribed by a "dominant order" of masculinity where the performance and proclamation of sex is the expression and essence of heteronormative masculinity (Ratele, 2011; Evans, 2006) . Female sexualities are subordinate to male in a context where boys and young men express expectations of regular sexual encounters and tell tales of their exploits to achieve acceptance or status in peer groups; girls and young women are socially ostracised if they express or act upon sexual desires or become pregnant (Sathiparsad, 2010; Izugbara, 2004) .

Being subject to societal and community discrimination for both being young and living on the street, relationships for street youth are funnelled towards peer sub-cultures which contribute to survival but require compromising individual values for the benefits of network membership (Evans, 2006) . For female street youth especially, street peer sub-cultures can reproduce the assumptions and prejudices of the wider community and be less rather than more supportive of its members. Fearon et al. (2015) , in a systematic review of young people's sexual behaviours in SSA, found inconclusive quantitative evidence of peer influence but acknowledged the critical role of socio-cultural context. Performances of femininities are variable for young women who for disparate reasons place themselves in street settings, and girls and women often engage in circular geographical and emotional journeys between street boyfriends.

The reconceptualization above of street peer relationships demonstrates their centrality in street settings, and that sexual relationships among street youth are subject to heteronormative gender identities. In our largescale qualitative study, sexual relationships are an everyday part of peer relations, including transactional relationships based on the exchange of money or goods; friendships and partnerships where young people explore their sexualities and commit emotionally and financially and co-parent children. However, contexts of J o u r n a l P r e -p r o o f poverty, few elder role models, minimal interaction with health services and cultural norms make this challenging, particularly for young women. The following section describes the methodological basis of the research. Then, by providing an emblematic empirical example typifying street peer relationships and their relation to gendered performances of intimate relationships, the remainder of the paper will analyse the significance of street peer relations on sexual relationships and sexual health.

This paper draws on three years of ethnographic fieldwork from the Growing up on the Streets research (2012) (2013) (2014) (2015) (2016) in Accra, Ghana; Bukavu, DRC; and Harare, Zimbabwe. Embedded in participants' lives on the street, the research design recognised that street youth are experts in their own lives. The research approach involved a capability framework, recognising the abilities of youth to make positive, if constrained, choices as they live their lives in street settings; as opposed to assumptions of vulnerability to external forces (Sen, 1999; Nussbaum, 2000; Shand, 2014; van Blerk, et al., 2015) . In the project's pilot phase, participants defined ten capabilities, in brief: food, shelter, friendship, health and wellbeing, play, safe movement, building assets, future plans, resilience and meeting basic needs.

The project developed an ethical approach with local partners and approval was gained from the University of XXX Research Ethics Committee. As most participants could not read or write, facilitators and researchers were trained in explaining ethnographic principles and ethics and gained participants' verbal consent through workshops. Participants could withdraw themselves or their data at any time.

The project set out to involve a core group of street youth as network members and RAs (n=198), by the end of data collection involving 116% of its intended participant group (n=229); with a wider network attending focus groups, or followed by RAs in weekly ethnographic reports (n=399). Gender parity varied, but broadly reflected the balance on the streets, with Accra having the highest proportion of female participants (Table 1) . Most participants lack birth certificates or other identity documents, but participants' given ages remained largely consistent across four annual surveys (Table 2) . We use the term 'street youth' as participants' mean age across the three cities was 16.79 y at baseline. Participants, who all lived in street settings, were informed about the research by local partners and volunteered to take part. The project created bespoke ethnographic training (GUOTS, 2014) for potential key informants. Some who took part in training did not wish to be involved further. Of those remaining, six young people in each city (all living on the streets) were selected as voluntary research assistants (RAs; n=18).

Others who participated in training replaced RAs who dropped out due to gaining employment (n=1), migration (n=1), being sponsored to go to school (n=1), serious injury (n=1), or lengthy imprisonment (n=2).

RAs used their training to engage in ethnographic observations on a network of ten peers, reporting their experiences across the capability framework in weekly ethnographic reports (interviews) with project facilitators; social workers aligned with non-governmental organisations (NGOs). Further training sessions were held throughout the project for RAs and participants.

In total, 2,478 interviews were conducted, recorded, transcribed, translated by local team members. 856 from Accra, 798 from Bukavu and 824 from Harare. Interviews were triangulated with 198 focus groups on each capability, which took place across the cities throughout the duration of the project and involved the wider network of street youth.

Data coding was undertaken in the UK using NVivo 10. The capability framework formed ten 'parent node' headings with 66 'child' nodes created as themes emerged from the data. This paper uses a subset of the J o u r n a l P r e -p r o o f complete data set outlined above; data qualitatively coded against relevant child nodes of the capabilities for Health and Wellbeing and for Friendship across all interviews and focus groups. Where the two capabilities intersect, the positive and negative effects of peer influence emerged as having a significant impact on sexual health in particular. To identify specific instances of this intersection a coding query was conducted in interviews and focus groups across the three cities, resulting in data from 212 sources (>50,000 words) ( Table   3) . The young people quoted here are representational of the consistent themes which emerged across the data around peer relationships and their impact on sexual health. However, an ethnographic report from 'Constance' (all participant names are pseudonyms) encapsulated many experiences and implications of peer influence on the sexual health of street youth. The empirical section below will use this extract as a framework to tease out the strands of peer influence and the impact on individuals' lives across this data subset. Our intention is not to conflate findings across three culturally distinct settings, but highlight broader themes raised by participants around the role of peer relationships in sexual health, with potential relevance to peer sub- So is that the same medicine she used that killed the girl?

Yes, they said some of her womb got torn.

Why do you think she decided to abort the pregnancy?

She didn't tell her boyfriend that she is pregnant, so when he heard it some of his friends told him that he is not responsible for the pregnancy; the guy started arguing with the girl with every little thing so that is why the girl bought the medicine to abort it. She didn't want to abort it but it is because most of the guys were saying the girl has been sleeping with other guys.

Awunyo told her that if she can get pregnant after the child, it means she has been having an affair with another guy.

No, it was because of what the boys were saying that is why she did it; and they even told Awunyo not to accept responsibility so she did it out of desperation.

Ok, these things happen to these young girls, they get pregnant and abort it and through that they die. Why?

Sir it is poverty; sometimes the guys impregnate you and abandon you, or he tells you to abort it. So if you don't go to hospital this is what happens, because the medicine might not be good for everyone.

Sir, it is too expensive to do it at the hospital; one month's pregnancy costs 150 Ghana Cedis (GBP£21/USD$28); so just imagine if the pregnancy is about three or four months. So if the person gets 50

Ghana Cedis (GBP£7/USD$9); she can get medicine which is not prescribed from the hospital and that is what brings the problem.

We don't tell anybody, not even your boyfriend; unless she has a close friend, then she will give her advice.

We normally feel shy to discuss it with adults because they might tell someone later, so we do it in secret till everything is over before you tell people. She didn't do it well and people got to know about it.

She was saying that family planning is not good because it makes your stomach big; but I know if the five 

She heard it but she listened to people and ignored it because she did the five years family plan after she gave birth to her first child and her friends deceived her, so she removed it and the moment she removed it she gave birth to [became pregnant with] the second child; so the nurse told her to do the three months but she told the nurse that she will not do the family planning again. The nurse prescribed a medicine for her to take before having sex but she didn't buy it.

His eyes are red. (Constance, Accra, 10 November 2014)

This extended extract incorporates many aspects of the complexity of peer influence around sexual relationships on the street, in a negative synergy of gender dynamics, peer influence, and service providers' failure to build relationships of trust with street youth. The following section will examine these in more detail, drawing on data from across the Growing up on the Streets cities to illustrate the complexity of peer influence on sexual health and demonstrating that this example, while noteworthy in itself, is not atypical.

Same-gender peer relations are invested with more persuasive power than trust between partners on the street.

Awunyo and his girlfriend have two children and live together, yet the word of his male peers holds more value in his estimation than hers. The primacy of same-gender relationships and masculine peer groups in particular has been shown elsewhere (Sathiparsad, 2010) . Long-term relationships are novel in street settings, meaning that street youth have few role models and often fail to support friends in partner relationships.

Gender dynamics in the urban informal settlements where Awunyo and his girlfriend lived are complex and variable depending on location, temporality, individual agency and age. The absence of trust between young men and women and the value placed on same-gender peer perspectives sits in a context where violence, including 'domestic' violence in the absence of 'home' (Petering et al., 2014) are part of a performance of gender roles bounded by contexts of geography and inequality. Sexually transmitted infections (STIs) affect both genders; pregnancy is both physically and economically carried by young women (Sathiparsad, 2010) .

Constance says: "we don't tell anybody, not even your boyfriend" (emphasis added). This may explain why Awunyo's girlfriend did not share her secret with him. Rather than concealing infidelity, her reasons related to iniquities of power in street settings and the fear of abandonment. Awunyo's peers encourage him to abandon his family in a context where men and boys commonly state mistrust in their paternity as a reason to leave their partner, despite having unprotected sex. Questioning female fidelity is a defence mechanism against the expense of raising a child.

Similar accounts of deficient trust in fidelity or paternity appear among data from the other two research cities.

Estelle, RA for the girls' group in Bukavu, reports that Pinganaye is pregnant, but "the author does not agree that he is responsible for that pregnancy". In this case Estelle advises her friend against abortion because "most of us have got one or two children and we keep on dealing with our business [transactional sex] on the street" (15 February 2016). From a male perspective, female peer influence can also be perceived as negative.

Goodwill, an RA in Harare, describes how Verity moved from the streets with her child and lives in an area of poor housing near friends, whom he sees as threatening Verity's partner relationship, as when she: "meets with her friends; that is when she starts to be engaged into old habits or be linked with old boyfriends. But if she is at a place where others [peers] are not, it will be ok" (5 March 2014). Peer influence is stronger in same-J o u r n a l P r e -p r o o f gender relations; enhancing normative gender roles and expectations, contributing to inter-gender mistrust, and setting up partner relationships to fail.

Constance describes a shyness to discuss abortion: "we do it in secret" with the secret shared only with a "close friend". This is in contrast to peers openly sharing opinions on contraceptives. Awunyo's girlfriend attempts to keep the secret of her pregnancy but cannot escape the observations of those around her. She secretly attempts an abortion but "didn't do it well and people got to know about it."

Maintaining privacy and keeping secrets is important to street youth in an environment where peers live together in close proximity. The perceptions of peers are a constant presence in accounts of transactional sex, STIs and pregnancy from across the three cities. Inhabiting public spaces almost all of the time serves to emphasise the distinction between public and private displays of self. Also in Accra, Araba attempts to conceal her return to sex work from her peers, but Josephine is informed: "whatever it is, someone will see Socio-spatially anchored in a panoptic environment (Alsayyad & Roy, 2006) , participants fear the disclosure of sources of shame or embarrassment, despite being situated outside of the moral parameters of traditional family settings. Street youth fear the judgment of peers around activities which could be deemed morally or culturally questionable, and the impact of this upon their fragile self-esteem. As a result, they have few friends in which they place genuine trust.

4.3 Gender, sex and peer morality: "family planning is not good"

It may be peer morality which lends weight to Awunyo's judgement that his girlfriend's pregnancy is caused by unfaithfulness to him. In turn, her perception that "family planning is not good" may be influenced by a J o u r n a l P r e -p r o o f belief that contraception is against Bible teaching, expressed elsewhere in data from Accra. Cultural subtexts in qualitative interviews can be as important as what is said (Knapik, 2006) , and both genders engage in unspoken rules around relationships which are embodied in contraceptive use. Trust between sexual partners is fragile; seeking to use a condom in sexual encounters with girlfriends or boyfriends is seen as either an admission of infection (and therefore of promiscuity) or lacking trust in your partner. In this case Awunyo's girlfriend had opted out of intrauterine contraceptives, and declined others, possibly due to cost.

In Bukavu and Harare, condoms are the primary method of contraception among participants, but, as has been shown among homeless youth elsewhere, accessibility and behavioural factors such as alcohol and drug use mean that condom use is not consistent (e.g. Barman-Adhikari, 2017). As a result, participants describe contracting STIs including chlamydia, gonorrhoea and HIV (see also Winskell et al., 2011) . Baba, an RA in Bukavu, advises his peers to use condoms in transactional sex, but they refuse, exhibiting behaviour characterized as adolescent risk-taking and sensation-seeking (e.g. Crone and Dahl, 2012) . The downside to peer group belonging is a compromise of personal values to conform to negative group behaviours that are detrimental to health. In exhibiting protective behaviours contrary to the group's, Baba (who also eschews alcohol and drugs) risks becoming outcast. Perhaps because of his status as de facto group leader, Baba is able to show self-determination despite reporting being "mocked" by group members (8 October 2015). This is an interesting contrast to Awunyo's girlfriend, who finds it harder to resist peer pressure. While gender identities may be a factor, in Harare, health focus group members also discussed "peer pressure" and being excluded: "if you do not take drugs they will be calling you stupid or they may say 'we do not walk with you...'" (Goodwill's group, 9 May 2013).

The interplay of peer group membership, poverty, and cultural norms limit the livelihood strategies for girls and women. When asked why his peers have transactional sex, Baba blames: "the influence of friends, the environment and drunkenness" (31 March 2016). Didier suggests that a shared street peer identity transcends gender identity, so that "boys" pay "girls" for sex, who "accept that little money because they know that we all street children" (5 May 2016). Yet although the street peer identity is shared, it is also divided in terms of masculinity and femininity and gendered norms. There are unequal outcomes for street youth based on gender, with relational multiplexity used by girls as a means of survival. With multiple partners, where unprotected J o u r n a l P r e -p r o o f sex is the norm, risk of infection and pregnancy increases, but for girls this is a trade-off as mixed gender groups or brief partnerships provide some safety and protection.

Constance's account demonstrates that the challenge to women in accessing healthcare, in this case contraception and abortion, is due to three factors. Firstly, charges at the point of access are prohibitive.

Awunyo and his girlfriend worked in informal trading and are unlikely to have savings. Secondly, Constance describes regular area visits by nurses who "educate us about the family planning". On-street health education seems well designed to reach a cohort typically educated only at primary level (Bose & Heymann, 2019) .

However, brief appearances of specialists fail to establish bonds of trust, as indicated by Constance's statement that young women are "shy to discuss it with adults because they might tell someone later"

(emphasis added). The nurses are perceived as both inter-generationally distinct, untrustworthy, and a threat to secrecy and self-esteem in the panoptic environment. Thirdly, the power of everyday peer persuasion leads Awunyo's girlfriend to conclude that "she will not do the family planning again". Constance describes this as a "deceit" on the part of peers; we can presume that peers did not set out with malicious intent, nor could they foresee its consequences. But as Awunyo's prioritises male peer advice, for his girlfriend, ill-informed female peer advice holds greater authority than advice from nurses.

Inadequate trust in healthcare professionals is a contributing factor to the failure to use contraceptives even when available. In all three cities, there are accounts of both care and discrimination from healthcare providers. Baba explains that as "the nurses do not accept to treat street children". It is his group who provide care because "we have love for one another […] when one of us falls ill, we have to manage to find how to treat him and he recovers" (Bukavu, 31 March 2016).

Due to scant service provision, exclusion and discrimination, it is in peers that street youth place their trust.

An RA in Accra describes how peers helped a 19-year-old woman after she gave birth in a market place,

where every night many street youth sleep. The situation was "very dangerous" says Jonathan, "we all know childbirth is life and death". He describes bravery as a stranger carried her from the market place "because the girl was soaked with blood […] the baby's placenta was hanging; it had not been cut." In order to take the J o u r n a l P r e -p r o o f young woman and child to hospital, everyone "had to quickly contribute some money" and they were put "in a taxi straight to the hospital" (12 May 2015). In street contexts, at moments of life or death, it is peer assistance that is immediate and potentially life-saving.

Adolescence is a definitive time for human physical and emotional development. Adolescence experienced on the streets exhibits a distinctive intensity, with daily survival challenges and a paucity of intergenerational input, state education and care. As we have seen, peers' social anchorage places them in the place and emotional space to deal with life and death on the streets. Their involvement in the lives of their peers is an inevitable result of both their presence and the absence of reliable formal housing and healthcare. Street youth are not always caring, generous or providing, and their input into the lives of others is often flawed and damaging, but peer care is always present in some form. Situating and understanding peer relations, as socially anchored within street contexts, and the implications for service provision will be examined in the discussion below. We close this section with some final words from Awunyo, who at the end of a focus group, spoke about his predicament: "For me, I have totally lost hope because I don't have any girlfriend and I have two children I have to take care of; but I am alone so I have no hope." His co-participants responded: "yes, but we are here with you, and we will support you." (Constance's group discussing Building Assets, 21 January 2015).

Street peers have a disproportionate role in daily life, with evidence presented in this paper revealing the criticality of street contexts in exaggerating peer influence. In the examination above, two key themes in particular have emerged: the locus of pregnancy as an expression of gender relations; and scant trust in health provision around sexual health.

Gender relations in poor urban settings has been discussed elsewhere (e.g. Chant & McIlwaine, 2015; Moser, 2016) . Our findings show that for youth anchored in street settings, the interplay and imbalance of femalemale power in sexual relationships are not disrupted by separations from cultural norms or extended family networks via migratory flux. Within street youth's peer sub-cultures, gender roles remain as played out in broader society and this manifests in similar ways across the three cities.

Accra's informal sector offers more livelihood choices to girls and women than Bukavu, where Estelle's group were almost all sex workers sleeping in bars. However, Awunyo's girlfriend still fears abandonment;

whereas Estelle confidently tells Pinganaye that it is possible to make a living and sole parent a child. As evidenced above, the burden of pregnancy falls on women and girls from prevention, conception, abortion and birth. In the three cities there were examples of young men supporting their children once born, as in Awunyo's case; male single parents are rare and the emotional and physical cost of raising children on the streets primarily falls on the mother. Awunyo's reaction of suspicion and attribution of blame when his girlfriend is pregnant for the third time is indicative of desperation rooted in poverty, leading in turn to his girlfriend's act of desperation. As is shown in the extract, informal sector abortions are commonplace and the outcome here not unusual (Chemlal & Russo, 2019) .

This commonality of participant experience also extends to inequality in healthcare delivery; discrimination and stigma when accessing services, and reliance on peer relationships in place of formal care. Health services are underfunded in all three countries (Ray & Masuka, 2017) with evidence of denial of access to those with a positive HIV status (UNAIDS, 2019). In Harare and Bukavu, NGOs, pharmacies and traditional medicines were relied upon for treatment. After ten years since its inception in 2003, only a third of Ghana's population had enrolled in its health insurance scheme (Kusi et al., 2015) . The most vulnerable, including street youth, face exclusion from formal healthcare (Williams et al., 2017) .

In addition, street youth endure state-sanctioned violence and removal from the streets (de Benítez, 2007 , Rice, 2005 . Their informal homes -in the case of our participants, shacks, cardboard boxes, or sheets -are regularly removed or destroyed by local authorities. Possessions such as health insurance cards are lost, and even if in possession of a card, there are still fees at the point of treatment. In settings where young people's rights to be on the streets are actively challenged, youth are exposed to negative health promotion strategies equating condom use as inhibiting of pleasure, interfering with prized fertility, and for use only in casual relationships (Winskell, 2011; Hunter, 2010) . This has resulted in a pernicious combination of cultural and peer norms, mixed health promotion messages, and poor education indicative of higher unmet need (Bose & Heymann, 2019) . The result is a mistrust of authorities leading to a system of peer care where peers fill the void of formal healthcare.

That said, it must be recognised that in the three countries public health service delivery is challenging with the rise of rural-urban migration (Vearey et al., 2010) . Girls' and women's access to modern contraceptives is of primary importance but the uptake is as low as 23% in DRC (48% in Ghana and 86% in Zimbabwe) (UNDESA, 2019b) . While there are no figures for contraceptive use among street youth, our qualitative data shows variable delivery and uptake for reasons which go beyond challenges or shortfalls of service provision.

As Starfield (2007) observed, "the mere presence of and access to 'health services' is inadequate. The benefits from health services occur only if health services are delivered appropriately" (p. 1360).

Ethnographic accounts such as Constance's can contribute to greater understanding of the role of peer relationships in sexual health. Their experiences show that service provision is often designed in a top-down manner which fails to recognise the alienation and mistrust of those most in need (Moser, 2016; Panter-Brick, 2004) . Harnessing the power of peer relationships socially anchored in the street in the delivery of sexual health services may be beneficial for uptake; recognising that peers have influence is essential. Most importantly, there is a need to build relationships of trust with young people living in street settings in order for health efforts to be successful. This requires investment in longer term bottom-up strategies that involve young people as active participants in their own and their peers' sexual health.

J o u r n a l P r e -p r o o f

In examining qualitative ethnographic data from young participants who live in street settings, the paper has shown the critical role peers play in the sexual wellbeing of young street people. We have explored how relational belonging can be realigned to reconceptualize peer networks not as street family, but as a street peer relationships where street youth are socio-spatially anchored in the street space. Street peer relationships contribute to everyday survival and moments of acute need, but contextual proximity lends them a disproportionate persuasive power. An asymmetrical bias in the make-up of street relations, where monogenerational and same-gender relationships holds primacy, can see the perpetuation of cultural norms around gender identities and countermand efforts of healthcare providers.

This new conceptualization that positions street peer relationships as socio-spatially anchored and disproportionately relevant provides new insights into the lives of homeless youth, and speaks directly to those considering approaches and mechanisms of healthcare provision for harder to reach groups more generally.

Urban informality is increasingly the norm in SSA cities. In the light of Ebola and COVID-19 pandemics, there is recognition that to be able to deliver these basic services at scale requires new partnerships within these communities (Corburn, et al., 2020; Hussein, 2020) . Young people are the majority of those living in urban informality, and as outlined in the discussion, it is essential that in settings with few trusting relationships with formal healthcare providers, the potentiality of youth peer networks is mobilized.

Opportunity exists for both sexual health policy makers and practitioners to examine the persuasive power of street peer networks, working with them to overcome issues such as informal sector abortions and contraceptive misinformation. Recognising the right of young people to live and seek livelihoods in urban settings, and adopting the social networks they create to advance street youth's sexual health, has become even more relevant in a (post)pandemic world.

Contraceptive use and associated factors among sexually active female adolescents in Atwima Kwanwoma District, Ashanti region-Ghana

Medieval modernity: On citizenship and urbanism in a global era

Vulnerability to sexually transmitted disease: street children in Accra

Colombian Street Children: Gamines and Chupagruesos

Street Children and Homeless Youth: A Cross-Cultural Perspective

Condomless Sex Among Homeless Youth: The Role of Multidimensional Social Norms and Gender

Sex isn't better than love': Exploring South African Indian teenage male and female desires beyond danger

Love, Sex and Gender: Missing in African Child and Youth Studies

Effects of tuition-free primary education on women's access to family planning and on health decision-making: A cross-national study

Peer Relationships in Adolescence. Handbook of Adolescent Psychology

At the threshold: The developing adolescent

Analysing gender in cities of the South. Introducing the 'gender-urbanslum-interface'. Cities, slums and gender in the Global South: Towards a feminised urban future

Why do they take the risk? A systematic review of the qualitative literature on informal sector abortions in settings where abortion is legal

Canadian Definition of Youth Homelessness. Toronto: Canadian Observatory on Homelessness

Introduction: Children Out of Place

Slum Health: Arresting COVID-19 and Improving Well-Being in Urban Informal Settlements

Understanding adolescence as a period of social-affective engagement and goal flexibility

State of the world's street children: violence

Social relations: network, support and relational strain

Negotiating social identities: The influence of gender, age and ethnicity on young people's 'street careers' in Tanzania

Is the sexual behaviour of young people in sub-Saharan Africa influenced by their peers? A systematic review

Young People and Family Life: Analysing and Comparing Disciplinary Discourses

Growing up on the Streets: Knowledge Exchange Training Pack

Peer Interaction: What Causes What

Youth Gangs and Street Children: Culture, Nurture and Masculinity in Ethiopia

The Materiality of Everyday Sex: Thinking beyond 'prostitution'

Love in the Time of AIDS: Inequality, Gender, and Rights in South Africa

COVID-19: What implications for sexual and reproductive health and rights globally? Sexual and Reproductive Health Matters

Notions of sex, sexuality and relationships among adolescent boys in rural southeastern Nigeria

Transition into first sex among adolescents in slum and non-slum communities in

Young Female Adolescents in Urban Areas of the Global South, The Challenges of Slums and Potential Solutions

The Qualitative Research Interview: Participants' Responsive Participation in Knowledge Making

Refusal to enrol in Ghana's National Health Insurance Scheme: is affordability the problem?

We are managing!' Uncertain paths to respectable adulthoods in Accra

Childhood and Society: Growing up in an Age of Uncertainty

Causes and Characteristics of the Street Child Phenomenon: A Global Perspective

Gender transformation in a new global urban agenda: challenges for Habitat III and beyond

Creating Capabilities: The Human Development Approach

They're Always Complicated but That's the Meaning of Family in My Eyes": Homeless Youth Making Sense of "Family" and Family Relationships

The Social Networks of Homeless Youth Experiencing Intimate Partner Violence

The rise and fall of icons of 'stolen childhood' since the adoption of the UN Convention on the Rights of the Child

Facilitators and barriers to effective primary health care in Zimbabwe

The Effects of Peer Group Network Properties on Drug Use Among Homeless Youth

Young Rural Males in South Africa Speak on Teenage Pregnancy

Development as Freedom

Growing up on the streets -Understanding the lives of street children and youth in Africa. The Capability Approach from Theory to Practice

Economic Practices of African Street Youth: DRC, Ghana, and Zimbabwe

Pathways of influence on equity in health

Revisiting the understanding of "transactional sex" in sub-Saharan Africa: A review and synthesis of the literature

AIDS-related knowledge, attitudes and behaviour among South African street youth: reflections on power, sexuality and the autonomous self

Exploring the Contours of African Sexualities: Religion, Law and Power

World Urbanization Prospects: The 2018 Revision

World Population Prospects

World Population Prospects: The 2017 Revision, Data Booklet

World Population Dashboard

Migration, Marriage and Modernity: Motives, Impacts and Negotiations of Rural-Urban Circulation amongst Young Women in Northern Ghana

Beyond Agency's Limits. Street Children's' Mobilities in Southern Ghana

Street Children as Researchers: Critical Reflections on a Participatory Methodological Process in the "Growing Up on the Streets

Berg-en-See street boys: merging street and family relations in Cape Town

Urban health in Johannesburg: the importance of place in understanding intra-urban inequalities in a context of migration and HIV