Exploring condom use decision-making of adolescents: The role of affective and rational processes

Condom use remains the most effective behavioural method for the prevention of HIV and unplanned pregnancies. However, condom use remains inconsistent among young people. Exploring the condom use decision-making processes that adolescent engage in might provide information that would assist in the prevention of many challenges related to poor sexual and reproductive health outcomes. This study, therefore, aimed to explore the factors that inuenced decision-making about condom use of adolescents from two schools in the Western Cape, South Africa. A sample of 16 adolescents were selected using stratied random sampling. Data were collected using semi-structured, individual interviews. Thematic analysis was used to analyse the data generated.

estimated to be 60% [5]. In South Africa, condom use is inconsistent, even though many young people are sexually active [6], with the reported rate of condom use inconsistency being 46-55% [7,8].
Adolescents are at risk of poor sexual and reproductive health outcomes often due to a combination of early, unprotected, or coerced sex, and the lack of adolescent-friendly sexual and reproductive health services [9][10][11]. These outcomes include unintended pregnancies, abortion, as well as sexually transmitted infections (STIs) including HIV [2]. HIV acquisition and transmission among adolescents most often results from unprotected sex or inconsistent condom use. While it is di cult to pinpoint exactly what shapes the choice to engage in unprotected sex or inconsistent condom use, we know that condom use decision-making happens in an environment that is changing rapidly [2,12]. These changes include the increase in access to and use of technology and social media which is often associated with exposure to sexually-explicit social media accounts and sexually-orientated reality television predicting willingness to engage in casual sex [13], diverse perspectives on childbearing and marriage including shifts in cultural, sociological and demographic views of childbearing and marriage has seen an increase in births outside of marriage [14], as well as the onset of puberty and changing social roles such as changes in the age of onset which in uences adolescent risk behaviour [15].
Adolescent decision-making processes have primarily been examined from two theoretical perspectives -(i) the affective decision-making (referred to as emotion-based decision-making) and (ii) a rational decision-making (referred to as fact / information-based decision-making) perspective as outlined in Fig. 1 [16]. The affective perspective on decision-making is concerned with the role of emotions and intuition in the decision-making process, while the rational perspective examines the role of negotiation and reasoning in coming to a decision [16,17]. RA Ferrer and WB Mendes [18] have highlighted that the role of affective states in making and shaping decisions, particularly related to health and behaviour, remains poorly understood. The process of making decisions, whether informed by affective states, available information, or both, may of course differ from individual to individual -which is often referred to as the decision-making style [19].
An adolescent's decision-making style could be protective against adverse health-related consequences.
Decision-making styles could also increase the risk of adverse health-related consequences. It is important to acknowledge that there may be situations in which an adolescent's agency is constrained in such a way that does not allow them to make choices. For example, choices related to condom use would be constrained in situations where sex is coerced or intimate partner/sexual violence is occurring.
Adolescent choices might be shaped by sociocultural factors which would impact their decisions related to sexual and reproductive health -such as gender, power, cultural and religious views. Empowering adolescents to make good decisions to promote sexual and reproductive health is important, but the social and structural factors that constrain their agency must also be considered when exploring how best to encourage protective behaviours in this sphere. A more careful understanding of decision-making would assist in identifying the processes or mechanisms to improve decision-making.
Exploring condom use decision-making processes might provide information that would assist in preventing many challenges related to poor sexual and reproductive health outcomes of adolescents. The process of knowing when to 'start, stop or change' condoms and other contraceptive methods is complex for adolescents [20]. The complexity which emerges in these decisions is as a result of the interpersonal, community and macro-social in uences [21].
Interventions leading to favourable behavioural outcomes that promote sexual and reproductive health are needed, particularly as young people make up a large proportion of the global population [2]. Many sexual and reproductive health interventions have focused on increasing knowledge about sexual and reproductive health risks, which is still important as there is an urgent need to inform young people, but it will not be su cient. Importantly, few sexual and reproductive health studies have qualitatively examined what drives adolescents' decision-making process before a behavioural outcome is selected. Examining the decision-making process and its in uences on sexual behaviour could inform health-related interventions. Given these knowledge gaps, the current study aimed to explore the factors that in uence adolescent's condom use decision-making in the Western Cape, South Africa, with a particular focus on the affective and rational factors.

Design
The study employed a qualitative descriptive approach to examine the factors that in uence adolescent condom use decision-making.

Recruitment & Participants
Participants for the study were recruited from two public schools in the Western Cape province of South Africa. An initial list of all public schools was generated and strati ed on the basis of socio-economic status, using school fees as a proxy. Three schools were randomly selected in each of the strata (no school fees, ZAR 1 -ZAR 1500 per annum and above ZAR 1500 per annum), of which one school in each stratum were invited to participate in the study. Of the nine schools invited three initially agreed, but ultimately only two schools agreed to participate.
The nal sample included sixteen participants from these two public schools. Participants were purposively selected to include an equal split between sex, socio-economic status and developmental phases (summarised in Table 1 below). The developmental phase was categorised into early and late adolescence, using both age and educational level/grade as an indicator. Participants who were in the eighth grade in secondary school were categorised as being in the developmental phase of early adolescence, and those in the eleventh grade as late adolescence. The rationale behind this selection of participants was to see if there were differences in responses about condom use decision-making on the basis of these demographic descriptors. (WCED) to access schools within the Western Cape (WCED Reference: 20170706-2719). After appropriate approvals were granted, the rst author (ELD) made contact with principals at the selected schools and set up initial meetings with the principal and relevant teaching staff at the selected schools to invite them to partake in the study. Students were then invited to participate in the study and given an information sheet, parental consent, and student assent forms. Following receipt of the completed consent and assent forms, dates and times were selected by the school to ensure that data collection would be least disruptive to the normal school day.

Data generation
In-depth, semi-structured interviews were conducted by two researchers. The rst author conducted all interviews with English-speaking participants (n = 10), and an independent second researcher (SM) conducted all interviews with isiXhosa-speaking participants (n = 6). All interviews were audio-recorded, and the length of the interviews ranged between 45 minutes to one hour. Various vignettes about adolescents facing various health-related decisions were used to guide the interviews but this article focussed on decisions related to sexual debut and condom use. Using a vignette afforded participants the opportunity to consider their hypothetical responses even if they had never engaged in sexual intercourse, condom use negotiation and/or other health-related decisions (see Box 1 for vignette and examples of speci c questions related to condom use decision-making). During interviews, both researchers conducting the interviews kept brief notes and also met to debrief following each interview to discuss their emerging thoughts and interpretations.
Box 1 Vignette and speci c questions used to elicit information about condom use decision-making Vignette Craig and Ntombi have been friends for a long time. They have decided to take their relationship to the next level. One night at a party Ntombi is offered vodka and beers to drink. She sees all the other friends at the party drinking vodka, and decides to drink the vodka instead of the beer. After having several rounds of vodka she decides to go outside at the party and get some fresh air. When Ntombi gets outside she is offered a smoke -she isn't exactly sure what it is but takes it and takes a few puffs. Later she realises that it is dagga that she was smoking. As the party goes on -she starts feeling a bit funny.

Data analysis
English interviews were transcribed verbatim by an independent transcriber and checked by the rst author. Interviews conducted in isiXhosa was translated and transcribed into English by an independent bilingual transcriber, and checked by the isiXhosa-speaking interviewer for accuracy.
The transcribed interviews were analysed by the rst author, using a combination of manual coding and NVivo 11 using the steps as outlined by V Braun and V Clarke [22] to conduct thematic analysis. The transcribed interviews were coded using inductive coding, allowing codes and themes to emerge from the data guided by the process of thematic analysis as outlined by V Braun and V Clarke [22]. Twenty-eight per cent (n = 5) of the transcribed interviews were coded and checked by a senior qualitative researcher (YZ). The codes and themes emerged through the analyses by the rst author and the senior qualitative researcher were then compared and discussed. Themes and the substantiated quotations from the transcribed interviews were also discussed and re ned among peers in a structured, peer-reviewed plenary and with two other senior authors (AS, CM).

Results
When exploring the factors that in uence adolescents' condom use decision-making, sexual debut emerged as the most prevalent scenario informing condom use decisions. Results were therefore grouped according to themes that emerged for (i) sexual debut and (ii) condom use decision-making, respectively (see Fig. 2). The sample included 16 adolescents with a mean age of 15 years, ranging from 13-18 years. There was an equal split in terms of sex (Males = 8 participants; Females = 8), socio-economic status (Low socio-economic setting = 8 participants; High socio-economic setting = 8) and developmental phase (Early adolescence = 8 participants; Late adolescence = 8). The themes which emerged showed no evidence of differences for condom use decision-making among adolescents in terms of the demographic descriptors of sex, socio-economic status and developmental phases.

(i) Decision-making about sexual debut
When speaking about condom use decision-making, adolescents often made reference to sexual debut and how they navigated the thought processes about it. Adolescent sexual debut was repeatedly shown to be in uenced by adolescents' belief that relationships were about showing true love for one's romantic partner through sex on the one hand, and on the other hand, that sexual debut should be delayed as a sign of respect for parents' expectations and rules about the appropriate age to have sex. These two themes are further detailed below.
Sex as a symbol/manifestation of true love Adolescence has often been considered as a developmental period of exploration and discovery, with sexual development highlighted as an integral part of the phase. When adolescents referred to relationships and the role of sex within them, it was clear that they considered sex to be an important mechanism by which 'true love' can be shown to a romantic partner. The link between sex and true love framed their decision-making regarding sexual debut: The decision-making style or process in which adolescents engaged when exploring sexual debut was characterized by societal expectations about what a romantic relationship entails. The adolescents appeared to believe that being in a relationship brought with it the expectation that they show and prove their love and affection for their partner. In this scenario, often the idea of love and affection was equated with having sex, which was considered as part of the set of expectations that characterize romantic relationships. One participant indicated this expectation by stating that adolescents should engage in sex because 'they are in a relationship' (Male, Age 14, Low SES, Early Adolescence). Some of the adolescents viewed the decision-making process to be about the hope and the expectation of showing true love within their relationship. Showing true love was viewed as a gateway to sex within the relationship.
Delaying sexual debut as a sign of respect for parents' expectations and rules Some participants did not view sexual debut as being an option for them because of their immature age. Concerns about the future and fear of perceived negative consequences When adolescents explored condom use decision-making, it was clear that this was shaped by anxiety and concern about their future. This is in line with evidence of how affectives states or emotions such as anxiety drive the examination of the best alternatives available to address the situation at hand. When considering condom use, adolescents expressed their concerns about their future and the need to have stability. Stability to these participants took the form of access to nancial resources, pursuing further education and securing a job: The adolescents decision-making process regarding condom use was clearly marked by anxieties about the future, which they thought would be compromised if they engaged in condomless sex and consequently contracted a sexually transmitted infection or became pregnant. One adolescent put it this way: 'You don't want that [referring to becoming a parent or contracting HIV] for yourself … so you will use the best thing, a condom …' (Female, Age 13, High SES, Early Adolescence).
The adolescents' fears that engaging in sex without condoms would lead to sexually transmitted infections was informed by uncertainties about partner delity; this consideration hugely in uenced their decision-making regarding condom use: 'I was going to say [yes to the] condom because you were friends mos [colloquial word for right or OK] at rst and you don't know how many girls did he sleep with. And even now that you are together you don't know how many girls he has outside, so I was going to say condom because I don't want to get infected.
Because he is the only one that knows whom he sleeps with and there are diseases out there. So, I was going to say condom always.' (Female; Age 17; Low SES; Late Adolescence).
The role of fear of undesirable sexual and reproductive consequences in in uencing the condom use decision-making process was apparent in many of the remarks made by participants in the study:

Discussion
The study aimed to explore the factors that in uence adolescent's condom use decision-making in the Western Cape, South Africa. The results suggest sexual debut was central to conversations about condom use decision-making among adolescents. Conversations about sexual debut might have emerged among the adolescents so that they are not seen as already having sex -so that conversations about sex and condom use are more euphemistic. The role of fear, however, was also seen as driving the decision-making process of rationally examining the alternatives when a decision needed to be taken about using a condom or not.
Scholars have called attention to the importance of understanding and exploring adolescent sexual decision-making [23], particularly related to condom use, to inform HIV and pregnancy prevention interventions. Through exploring the data, it became clear that social expectancies, parental in uence and the fear of undesirable future outcomes were integral to adolescent condom use decision-making.
Using the affective decision-making perspective, values, expectancies and emotions are key when examining adolescent risk behaviour [23]. Evaluating positive and negative outcomes of behaviour during the decision-making process, called expectancies, is important to explore as adolescents exhibit agency in their choices around health-related decision-making [23]. Findings in this study con rm what has been written in the literature about the assumption that negative expectancies, such as becoming pregnant, becoming a parent or contracting HIV or other related sexual transmitted infections [23] are often shaped by the values held by adolescents.
In addition to the values and expectancies, the ndings suggest that emotions and affective states (the emotions and affective states which emerged in the decision-making process from the ndings included: hope, love, anticipated regret, and fear) were important factors in adolescent decision-making. Affective states are often ignored in health-related decision-making and thus are not always incorporated into the theoretical frameworks that underpin many health-related interventions [18]. Affective states such as motivation, emotion, and stress, to name but a few are important when making decisions about health [18]. Adolescents are making decisions about emotionally laden issues [18]. They talk about things in relation to how they think or expect they may feel. We tend to ignore emotions in interventions/research because of social constructions of emotions and the self as being vulnerable which is disempowering and perpetuates stigma [24]. Instead, we focus on risk perception, knowledge and attitudes [18], which are important too. We could reshape our interventions/research by understanding the role of emotion and affective states in health-related decision-making and behavioural outcomes.
The themes suggest the important role of an evaluation of the alternatives but also the role of affective states in driving the outcomes of such evaluations, that are often ignored in health promotion and prevention interventions. The themes related to condom use decision-making also highlighted the role of fear in the decision-making process. Fear has often been cited as in uencing the process of making sense of available information which is important to making a decision. Diminished attention is paid to the risk-related information in the presence of fear. However, others have believed that the presence of fear is associated with the awareness of risk related to the potential alternatives to be chosen in the health-related decision-making process [25].
Guilamo-Ramos and colleagues [26,27] found that adolescents were concerned about the implications and consequences that engaging in risky sexual behaviour would have on those they interact with socially, like peers, parents and neighbours. These concerns about the implications and consequences of sexual behaviour were also seen among adolescents in the current study where there were concerns about pregnancy, parenthood and disease infection. Anticipation of regret as an affective state which pointed to concerns about the social implications and consequences of sexual behaviour, was a common underlying factor that in uenced how adolescents thought about sexual debut. These ndings highlight not only the importance that adolescents may place on values as part of their sexual debut and condom use decision-making processes, but also how these values inform some of the expectancies or outcomes regarding condom use or non-use.
Public health interventions aimed at promoting sexual and reproductive health of adolescents should not only consider adolescents as being rational young people who evaluate the available alternatives based on the knowledge shared in many sexual and reproductive health interventions. It is evident that when faced with a situation in which a choice needs to be taken around condom use or non-use, decisions are also guided by various emotions and affective states. Whether condom use decision-making is informed by knowledge related to rational decision-making or emotions as part of affective decision-making only is not possible to answer. The result suggests that these two forms of decision-making (rational and affective) are not mutually exclusive and could be happening simultaneous during the condom use decision-making process. Many sexual and reproductive health interventions that have focused on cognitive and behavioural models have seen an increase in knowledge related to sexual risk [28], but have not seen a change in behavioural outcomes. Perhaps by taking into account affect and how it might work in tandem with knowledge when thinking about adolescent sexual decision-making could be the shift in behavioural outcomes. There is a need for interventions that are based on both rational and affective decision-making processes which are central to the decision-making styles used by adolescents when making decisions around condom use and risky sexual behaviour [18,23,27,29]. Interventions which consider both rational and affective decision-making processes would be categorised as behavioural change interventions stemming from motivation [30].
Behaviour change interventions aimed at improving condom use decision-making should, therefore, be multi-faceted with a combination of education (information to facilitate behaviour change), persuasive (apply communication strategies to prompt emotions or spark action), incentivisation (provide an anticipated reward), coercion (bringing about anticipated punishment or cost for actions against targeted behaviour change), environmental restructuring (foster change within the social or physical environment), modelling (have individuals to emulate or aspire to be) and enablement (provide support to increase the targeted behaviour change) components as adapted from S Michie, MM van Stralen and R West [30] Behaviour Change Wheel. Emotions and affective states have been shown to play an important role in the condom use decision-making process, addressing the gap, as highlighted by RA Ferrer and WB Mendes [18], for better understanding of emotions and affective states in health decision-making to inform intervention development and implementation [18].
In addition, interventions should consider the social ecology of adolescents, as the in uence of parents and social expectations emerged as important factors in the decision making process regarding sexual debut in our study. RJ DiClemente, LF Salazar and RA Crosby [31] have supported the notion of interventions that are ecological -considering the various levels of causation or in uence in sexual debut and condom use choices -suggesting that these are effective and promote the adoption of less risky outcomes. Therefore, complex interventions that consider adolescents, relationships with others, peers, family and the community are needed which inform decision-making processes around condom use decision-making. These complex interventions would also address the different levels of causation or in uence in the decision-making process which is furthermore shaped by society, culture, values, economic and other social factors [31]. Interventions might also consider role-playing situations or behavioural rehearsal in which sex and condom use negotiations take place which might inform adolescent decision-making but also decision-making competence and self-e cacy. Decision-making competence and self-e cacy have been associated with better decisional outcomes [32-34].
Understanding how adolescents make decisions about health-related behavioural outcomes should be further examined to identify and understand whether values, expectancies, anticipated future outcomes and emotions are central to the decision-making process engaged. Being able to understand and identify whether these rational and affective appraisals are part of the decision-making processes across all health-related decisions, could inform future public health interventions which are aimed at primary and secondary prevention.

Limitations
One of the limitations of the study related to the limitations inherent in the vignettes used to elicit information from the participants about their condom use decision-making: (i) the characters in the vignettes had heterosexual relationships and thus did not portray condom use among adolescents in same-sex relationships, (ii) the vignettes also failed to consider the use of other contraceptives which might have played a role in the perceptions towards condom use and thus impacted on the adolescent's decision-making around condom use, (iii) the estimated control associated with understanding the hypothetical versus the real experience of adolescent agency in making condom use decisions might be di cult to ascertain through the use of vignettes.

Conclusion
When examining the factors that shaped condom use decision-making among adolescents it became evident that sexual debut is central to the conversations about condom use. This study highlighted the importance of the role of affective states as part of the process of examining the alternatives when making a decision about condom use or non-use. Interventions are needed that examine both the role of affective (emotion-based) and rational (knowledge-based) processes as part of the decision-making process to inform adolescent sexual and reproductive health outcomes. The factors which shape adolescent condom use decision-making as highlighted in the study could inform sexual and reproductive health interventions to promote behaviour change and reduce HIV incidence and unintended pregnancies among adolescents.

Consent for publication
Consent for publication was obtained from all study participants and their parents as part of the informed parental consent and learner assent processes outlined above.

Availability of data and materials
The data generated during the study are not publicly available but are available from the corresponding author on reasonable request.

Competing interests
The authors declare no competing interests. Adolescent decision-making is often informed by two theoretical perspectives, namely affective and rational decision-making Adolescent condom use decision-making is often informed by discussions related to sexual debut and condom use. Decisions related to sexual debut is informed by the belief that when in a relationship true love is shown through having sex which is informed by the affective component of hope, love and affection. Sexual debut decisions are also informed by the perceptions of parents related to ageappropriate sex which is shaped by the affective component of fear and regret. Decisions related to condom use is informed by concerns about the future (such as becoming pregnant or becoming a parent) which is informed by the affective component of fear. Condom use decisions are also informed by concerns about partner delity related to contracting STIs which is shaped by the affective component of fear too.