1. Introduction

Adolescent girls and young women aged 15–19 living in Sub-Saharan Africa are underserved by existing sexual and reproductive health (SRH) service delivery [1]. Research consistently demonstrates adolescents’ disproportionate risk of HIV acquisition and unintended pregnancy [2, 3], and the related poor outcomes of maternal morbidity and mortality and complications of abortion under unsafe conditions [4, 5]. Forty-five percent of pregnancies among African adolescents are estimated to be unintended (63% in Kenya [6]), and nearly two-thirds are classified as having unmet need for contraception—meaning they are sexually active and report not wanting a child for at least 2 years, but are not using a contraceptive method [3]. In response to these indicators and the unique developmental and social circumstances facing adolescents, researchers, governments, and funders have advocated for a renewed emphasis on improving adolescent SRH [7]. With the goal of promoting adolescent SRH, the global public health discourse endorses targeting this population with interventions to increase contraceptive prevalence—or reduce unmet need—through voluntary and tailored family planning services [8].

Despite substantial investments in adolescent SRH and high contraceptive availability in many countries [8, 9], however, contraceptive prevalence has been slow to rise among adolescents [10]. Multiple sociocultural and health systems barriers intersect to influence adolescents’ contraceptive access and use [11]. Adolescent sexuality and contraceptive use are highly stigmatized in many communities [12–14], and discriminatory treatment from healthcare providers further deters SRH care-seeking [15, 16]. Adolescents are susceptible to misconceptions about contraceptive risks, especially future fertility concerns, which are rooted in gender and social norms [17–19].

Furthermore, it is clear that unmet need for contraception—a frequently-used concept and indicator referring to fecund women whose stated desire is to delay or limit childbearing but are not using a contraceptive method [20]—does not equate with desire to use contraception [21]. The public health research and policy narratives have focused on the former, leading to a gap in knowledge about how adolescents in sub-Saharan Africa who are most likely to experience unintended pregnancy perceive their own pregnancy risk and contraceptive need, or make decisions about contraception within their lived realities and social networks. Thus, the reproductive priorities of adolescents classified as having an unmet need for contraception, and influences on their contraceptive use or nonuse, have not been adequately characterized. These needs and drivers of SRH must be framed by the complex and often non-binary nature of reproductive preferences. A growing body of literature has highlighted the limits of using the pregnancy planning paradigm, or dichotomous categories of planned/unplanned and intended/unintended, to understand contraceptive needs [22]. More reproductive preference data and measures specific to African contexts and adolescent populations are needed to guide interventions to support contraceptive decision-making. These research gaps are particularly germane to adolescents younger than 18; younger, unmarried adolescents are underrepresented in SRH research due to ethical constraints and parental consent requirements, though studies suggest they are even more vulnerable to poor health outcomes [23, 24].

This study explores how Kenyan adolescent girls and young women aged 15–19 perceive their own contraceptive needs and make contraceptive decisions within their social contexts. We examine the social influences on contraceptive decision-making among adolescents at risk for pregnancy, aiming to provide nuanced insights into adolescents’ contraceptive behaviors. Our research questions and approach were guided by constructs of Social Cognitive Theory [25] and Social Norms Theory [26]. Specifically, we focused on conceptualizing the dynamic interplay between adolescent behavior and their social environment, and adolescents’ perceived self-efficacy in making contraceptive decisions [27]. We drew on insights of literature exploring how social norms affect reproductive health behaviors [26, 28, 29]. Through advancing understandings of adolescent contraceptive decision-making, this study aims to inform adolescent-centered efforts to better align reproductive preferences and outcomes for this underserved population.

2. Materials and methods

2.3 Data collection

The primary IDI/FGD domains represented in the semi-structured interview guide are listed in Fig 1. Each participant completed a brief tablet-based socio-demographic questionnaire, which was administered verbally by study staff. IDIs were primarily conducted in private homes; when privacy could not be assured or when preferred by the participant, study offices located at two county health facilities were used. FGDs were held with permission in private areas of county health facilities. Staff were fluent in written and spoken English, Luo, and Kiswahili; IDIs and FGDs were conducted in the language preferred by participants. IDIs and FGDs were digitally audio-recorded and simultaneously transcribed and translated into English by the interviewers themselves; a different staff member reviewed each transcript for accuracy of transcription and translation.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Primary in-depth interview/focus group discussion domains and sample questions.

https://doi.org/10.1371/journal.pone.0255954.g001

2.4 Data analysis

We used an inductive, thematic approach to analyzing the qualitative data. Our iterative analytic strategy drew on contemporary grounded theory methods, described by Charmaz as “systematic, yet flexible guidelines” for qualitative analysis and theory-building [32]. The co-PIs agreed on a set of initial codes designed to reflect a priori domains of interest from the interview guides and exposure to the raw transcripts and field notes. The lead author (EH), who has over 10 years of experience working in Kenya, and a Kenya-based expert qualitative data analyst (EC) coded 5 IDIs in parallel, after which they compared code application and meaning, adding new codes as needed to identify emerging themes. After a second round of parallel coding, investigators constructed a final codebook for both IDIs and FGDs, and re-coded or coded all remaining transcripts. Two-thirds of transcripts were double-coded, and investigators met regularly to discuss coding discrepancies and emerging themes. The analytic team identified the most significant codes, grouped similar and contrasting excerpts within and between codes, and wrote analytic summaries for each major theme. Investigators conducted all coding in Dedoose (Version 8.0.35, SocioCultural Research Consultants, LLC, www.dedoose.com), an online qualitative analysis program facilitating data management and collaborative coding. The conceptual model (Fig 2) was developed during the analytic process in order to organize and interpret themes emerging from the data.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Conceptual model: Social influences on adolescent contraceptive decision-making.

https://doi.org/10.1371/journal.pone.0255954.g002

3. Results

Study staff screened 200 adolescent girls and young women in community settings, of whom 115 were eligible and 86 (75%) were enrolled: 40 in IDIs and 46 in FGDs. The median age was 17, with 11 (13%) participants aged 15 or younger (Table 1). The majority were currently students (71%), romantically partnered (81%), and reported ever being sexually active with a male partner (87%). Most (79%) reported ever using a contraceptive method; 71% of these had used male condoms alone. No participants reported a prior pregnancy.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Participant characteristics.

https://doi.org/10.1371/journal.pone.0255954.t001

Our findings synthesize the complex social factors at play as adolescents navigate pregnancy risk and decisions around contraceptive use. In the results and conceptual model (Fig 2) below, we present the primary themes emerging from the data: adolescent pregnancy intentions and consequences, the apparent incongruence between pregnancy desires and contraceptive behaviors, and the major social influences on adolescent contraceptive decision-making, including various sources of stigma.

4. Discussion

Among the 86 adolescents who participated in IDIs and FGDs in this community-based study, we elicited a strong preference to delay pregnancy that was often uncoupled from a desire for contraception. While the majority of participants would be considered to have an “unmet need” for contraception based on the indicator’s definition, their perception of contraceptive need and ultimately their contraceptive decisions were shaped by various factors and priorities within their social contexts. These factors are illustrated in the conceptual model (Fig 2). Informed by social norms and future life aspirations, participants viewed pregnancy in adolescence prior to relative financial independence as an unacceptable outcome. They then had to navigate interrelated “layers” of social messaging that ultimately influenced their contraceptive decision-making and ability to prevent pregnancy. Each “layer” had differing levels of influence on each adolescent. For example, for some, perceived risks of hormonal contraception causing infertility and stigma associated with contraceptive use outweighed pregnancy concerns and motivated a desire for condom use, while unequal power with sexual partners and need for financial support further influenced ability to prevent pregnancy using condoms.

The curved arrows in the model represent the ability of adolescents who desire to delay pregnancy to traverse social norms and narratives, stigma, and unequal power dynamics that may intervene in their ability to prevent pregnancy. Use of a contraceptive method is a primary way that adolescents may exercise agency to prevent pregnancy, and several participants clearly articulated self-efficacy in demanding condom use or using contraception despite dominant social influence. Pulerwitz et al. propose a conceptual framework articulating the relationship between social norms and adolescent SRH that features social and gender norms as a central influence on health outcomes for adolescents [33]. The concept that power can be demonstrated in decisions to “adhere to (or not adhere to) social norms” [33] is relevant to our analysis. Contraceptive use was generally viewed as a transgression of social norms among our participants, but some adolescents made the decision to use a method in spite of social pressures. Furthermore, adolescents’ agentic actions around contraceptive decision-making may not take the form of bold steps [28], but rather smaller and more socially-acceptable steps such as avoiding sex or inconsistent condom use. For example, since condoms are widely considered a mode of HIV prevention rather than a contraceptive method by adolescents, condom use carries less contraceptive stigma than hormonal contraceptive methods, which are known more widely as “family planning.”

While participants frequently reported elements of transactional sexual relationships, few adolescents reported other forms of sexual coercion, such as age-disparate relationships, or sexual assault. We discussed sexual coercion in the interviews, but our study was not focused on examining experiences of coercion and violence in detail. Studies suggest that many Kenyan adolescents have limited interpersonal power in sexual relationships [34], and approximately 20% of Kenyan girls and young women aged 15–24 have experienced sexual violence from an intimate partner in the prior year [35]. Additional research examining how a history of violence affects adolescent contraceptive decision-making is needed.

Stigma emerged as a powerful source of adolescent influence in the present study, consistent with existing research [14], including a qualitative study of postpartum Kenyan adolescents from the same geographic region [17]. “Good girls” abstained from sex, didn’t use contraception, and didn’t get pregnant. The relevant stigmas can be conceptually tied together by participants’ use of the word “spoiled,” directly translated from Luo. Adolescents referred to themselves and peers who had begun sexual activity or were seen as promiscuous as “spoiled,” futures were “spoiled” by adolescent pregnancy, and contraceptive use could “spoil” one’s eggs and make one unable to have children in the future. Indeed, sexually-active adolescents risk experiencing stigma regardless of whether they actively prevent pregnancy through contraceptive use or become pregnant during adolescence. Understanding this context of multidimensional stigma may help to explain adolescents’ apparently incongruent pregnancy preferences and contraceptive behaviors. Our study focused on contextual rather than on health systems factors influencing contraceptive decision-making, which limits our data’s representation of stigma experienced by adolescents in the health care setting. However, prior qualitative and mystery-client studies describe provider bias towards sub-Saharan African adolescents, leading to restriction in method choice due to provider beliefs [36] as well as experiences of shame, embarrassment, and loss of confidentiality among adolescents seeking contraception [37].

Our data elicited little diversity in participants’ current pregnancy preferences: they reported that an unintended pregnancy would be unacceptable. Adolescent pregnancy is common in Kenya: among women aged 20–24 in 2014, nearly 25% of women had given birth by age 18, and 46.8% by the end of the 19^th year. Among 15–19 year-olds, 18% were pregnant or had already give birth [38]. Given that context, we anticipated a wider spectrum of positive, ambivalent, or indifferent perspectives on pregnancy “now.” Studies from the United States among young women advocate that unintended pregnancy is not a universally poor outcome; many unplanned pregnancies are welcomed and acceptable [39], and social context, including limited financial resources and social status, may make pregnancy planning less relevant for some [40]. The present study’s narrow scope of pregnancy preferences likely reflects its young, 15–19 year-old population, and the exclusion of married and parous adolescents; it is also important to note that participants viewed future pregnancy extremely favorably. Interestingly, our findings are consistent with a recent quantitative study of ambivalence in pregnancy intentions in Kenya, where the youngest age group of women (age 15–24) were the least likely to express ambivalence [41]. Social desirability bias may have influenced some adolescents’ responses, especially in the FGDs, though many adolescents freely shared other behaviors and experiences that would be considered contrary to social norms. Additionally, as parental consent was required for participants under age 18, adolescents who did not feel comfortable involving their parents, perhaps due to parents’ more conservative attitudes towards sexuality or contraception, may have avoided sampling. However, only 3 adolescents who were otherwise eligible for the study declined due to lack of parental availability, and no parent declined to provide permission. Quotes from IDI participants are primarily used in the results, as they tended to be more concise than the longer FGD conversations, but thematic elements were similar between data sources. Our study was designed to take advantage of the social nature of FGDs and anticipated that that the way peers conversed about the topics may uncover new insights. However, we did not find that the FGDs promoted interaction that elicited different perspectives than the IDIs. While our team thought carefully about word choice when translating interview questions from English into Luo and Kiswahili, there may have been challenges with managing the connotations and translation of various descriptors such as “unplanned” and “unintended.” Despite translation quality control measures, it is possible that some meaning and nuance around pregnancy preferences were lost in translation.

This study’s conceptual framing distinguishes it from other studies assessing adolescent perspectives on contraceptive use. We explored norms and preferences around adolescent sexuality, pregnancy, contraception, and abortion to contextualize adolescent contraceptive behaviors and gain insight into their contraceptive decision-making. This approach allowed us to closely examine the incongruence between adolescent reproductive desires and contraceptive behavior through the lens of “spoiled” girls’ narratives, beliefs, and life experiences. With this incongruence in mind, individual- and community-level interventions to prevent undesired pregnancy should focus on supporting adolescent decision-making agency rather than increasing contraceptive prevalence, with the goal of better aligning reproductive preferences and health behaviors to achieve those preferences. Such efforts might support adolescents in navigating various “layers” of social norms, such as providing adolescent-centered contraceptive counseling and decisional support, addressing misperceptions, and allowing adolescents to access contraception with peers or in locations not associated solely with contraception such as pharmacies [42, 43] and other community-based venues. Clinical interventions must acknowledge and address pervasive social messaging linking contraceptive use with future infertility. Furthermore, efforts to engage adolescents and young adults, including boys and men, in shifting harmful social and gender norms at the level of communities [44, 45] are also critical to enabling social environments for adolescent SRH care-seeking and healthy behaviors.

The adolescent voices represented in our study support the perspective that contraceptive prevalence rate and unmet need for contraception are useful indicators to compare over time, but are not person-centered outcome measures [46] for programs and interventions aiming to improve adolescent SRH. Indeed, our findings and conceptual model identify domains that should be considered in such adolescent-specific measures of reproductive desires and preferences. Senderowicz calls attention to the mismatch between stated rights-based global family planning goals and the metrics currently used to evaluate family planning programs [21]. Contraceptive autonomy, defined as “the factors that need to be in place in order for a person to decide for themselves what they want in regards to contraceptive use, and then to realize that decision,” is proposed as a novel family planning indicator [21]. Measures such as contraceptive autonomy, reproductive autonomy [47], and sexual and reproductive empowerment [48] should be tailored for adolescents and used to guide programs and funding priorities. Finally, contraceptive access and the ability to prevent pregnancy with contraception when desired are clearly critical to efforts to improve adolescent SRH outcomes. However, as was poignantly expressed in our participants’ stories, a focus on prevention of undesired pregnancy alone is inadequate and unrealistic: along with contraceptive services, quality maternity care, safe abortion services, and post-abortion harm reduction interventions are urgently needed to reduce maternal morbidity and mortality and improve quality of life among adolescents living in Sub-Saharan Africa.

Supporting information

Acknowledgments

This study received administrative support from the Fellowship in Complex Family Planning. The authors thank community advisory board members and the County Health Management Teams in Kisumu and Siaya counties, Kenya. This work would not have been possible without the contributions of the local study team, including Pherez Inda, Lydia Okumu, Ruth Awinja Okumu, Winnie Owade, Mercy Awuor, and Daniel Matemo. We thank the UW-Kenya staff and the University of Washington Global WACh Center for logistical and scientific support.