What to expect when girls are expecting: psychosocial support challenges and opportunities in the context and aftermath of teenage pregnancy in Kenya

Background An understanding of the psychosocial support challenges and opportunities in the context of teenage pregnancy is important for developing appropriate interventions for pregnant and parenting girls. This qualitative study was conducted in Homa Bay County, Kenya, to examine the experience of teenage pregnancy and the resultant psychosocial support needs from the perspectives of both pregnant/parenting girls and their own parents, who are typically expected to provide various forms of support. Methods The study used a descriptive case study design, drawing on counseling case notes documented by psychologists who held counseling sessions with 20 pregnant or parenting girls and 6 of their parents as part of a psychosocial support intervention. The counseling case notes formed a qualitative data set, which was analyzed thematically. Results Emerging master themes were related to pregnant and parenting girls’ experience of sexual violence and adverse childbirth outcomes; psychological trauma confronted by girls and their parents alike; parental need for support in communicating with pregnant/parenting girls about sex and sexuality; and the availability of family support as a resource for teenage mothers. Conclusion Pregnant and parenting teenage girls require a range of psychosocial support responses that recognize the realities of sexual violence and other challenges in the lives of the girls themselves, as well as in the lives of their parents and caregivers. While parents and other caregivers can serve as an important resource for supporting affected girls, they often need assistance as well, in order to support pregnant/parenting girls effectively. These realities need to be taken into account to maximize the effectiveness of health and development programs for pregnant and parenting girls. Furthermore, emerging themes from actual counseling sessions with affected girls and parents can provide important insights into the potential psychosocial support needs of the broader population of pregnant and parenting girls.


Plain language summary
This study looked into the lives of pregnant and parenting girls in Kenya by examining case notes that were written up by psychologists during the counseling sessions they held with 20 different teenage girls who were pregnant, or already had children, and 6 of their parents. The case notes provided an understanding of the kind of support pregnant and parenting girls need, including mental health care to deal with the trauma of sexual violence (and other issues) that some had experienced. It also helped put a spotlight on the traumas that parents of affected girls face after a teenage pregnancy occurs. These issues need to be addressed in order to ensure that health and development professionals who work with pregnant and parenting girls are as helpful to them as possible.

Background
Despite the worrisome prevalence of teenage pregnancy in Africa-with nearly 1 in 5 adolescent/teenage girls becoming pregnant in this region [1]-the evidence base on the psychosocial support concerns of affected girls within this context is surprisingly thin. The current literature on the experiences of teenage mothers (including psychosocial support/mental health issues) is characterized by 'few studies' [2], with evidence on best practices around psychosocial support for pregnant and parenting girls described as being 'limited' [3]. Yet, early pregnancy is associated with tremendous psychosocial effects [2,4,5], which are likely to be exacerbated in the low-resource settings that many African countries represent.
Recent review studies support these findings, highlighting the dearth of evidence on this subject in Africa, in particular. For instance, an integrative review of adolescent girls' experiences of transitioning to motherhood [6] included only one Africa-based study. More recently, a systematic review of psychosocial interventions designed for pregnant and parenting girls [3] found no eligible studies in low-and middle-income countries, and called for a focus on the psychosocial needs of this population to better target their mental health outcomes.
A number of studies have begun to respond to this call, noting that the psychosocial support needs of pregnant and parenting teenage girls in Kenya, for instance, revolve around issues such as economic support [7], mental health concerns [8][9][10] and stigma [10,11]. These important psychosocial support issues emerged from conventional datasets focusing on pregnant/parenting girls and other related populations. However, data derived from actual psychosocial support contexts are yet to inform the emerging body of literature. Yet, such data have the potential to deepen our understanding of a complex issue requiring urgent solutions.
To address this gap, this paper draws on the voices and experiences of pregnant/parenting girls (and, by extension, their parents) in the context of home-based counseling sessions held in Homa Bay County, Kenya. Examining affected girls' lives in this unique context provides an opportunity to gain deeper insight into their worlds, and to pick up on critical psychosocial support challenges and needs that have yet to make their way into the literature. Such evidence is critical for informing the design of effective health and development programs and policies for pregnant and parenting girls.

Study context, sample, design
This study draws on a qualitative, secondary data set composed of counseling case notes recorded by trained psychologists from November to December 2019. The psychologists were engaged to support a data collection team carrying out a household survey with girls and their parents conducted in three sub-counties within Homa Bay County, Kenya, namely, Homa Bay, Ndhiwa, and Rangwe [12]. Nationally in Kenya, Homa Bay County has the second highest childbearing levels among girls ages 15-19 [13]: A third of girls in this age range are pregnant and/or parenting, compared to 18% of their peers, nationwide.
Under the household survey, pregnant and parenting girls were offered home-based counseling and referrals for psychosocial support, in recognition of the response needs associated with these populations. A field team conducted the household survey, and psychologists doubled up as interviewers when they were not engaged in counseling sessions. A total of 20 pregnant/parenting girls requested counseling by a psychologist during or after their interviews. Although counseling support was intended solely for girls, parental demand for similar support resulted in about a third of the girls' parents (6 out of the 20 cases) being counseled as well. Pregnant/parenting girls who responded in the affirmative when asked whether they would like to speak to someone further regarding any of their psychosocial issues were included in the current study, along with any parents of such girls who proactively asked to speak to someone. Pregnant/ parenting girls and parents who requested to speak to someone were not eligible for inclusion in the study if they could not speak to the psychologist privately. Three psychologists attached to the household survey team documented counseling case notes for all pregnant/parenting girls (along with the parents concerned) that asked for counseling. The present study is qualitative in nature and based on the narratives contained in these case notes.
Standard validation techniques employed in qualitative research were applied to gauge the alignment of this study's findings with reality. These techniques involved devoting attention to the qualitative constructs of confirmability (comparable to the notion of objectivity in quantitative research), dependability (comparable to 'reliability' in quantitative methods), and credibility (comparable to internal validity in quantitative methods) [14]. The confirmability of the data in this study was enhanced by preserving an audit trail of the data collection, processing, and reduction that led to the study's conclusions, providing evidence of the grounding of these conclusions in the actual data. The dependability of the data was enhanced by minimizing the number of psychologists recording the case notes (one per sub-county), and through holding check-in meetings with the psychologists to discuss emerging counseling and case note recording issues. To assess the credibility of the data, areas of uncertainty and negative evidence were deliberately sought out within the data set, and alternative explanations were considered and weighed against the existing data prior to arriving at conclusions.
The study used a descriptive case study design. This type of case study design is employed 'to describe an intervention or phenomenon and the real-life context in which it occurred' [15]. Homa Bay County served as a single, holistic case within which the phenomenon of teenage pregnancy was explored, along with the psychosocial support issues associated with it, and the context in which these occurred. The descriptive case study drew on the case notes emanating from counseling sessions with pregnant/ parenting girls at the household level. Where a parent of a pregnant/parenting girl also requested counseling at the household level, the resultant case notes were merged with those of the index pregnant/parenting girl to gain a deeper understanding of her overall context. This led to a total of 20 case notes (representing 20 pregnant/parenting girls). Of these, 6 case notes included descriptions of counseling sessions with the parent of the girl concerned.

Data collection and analysis
The psychologists on the data collection team handrecorded their 'counseling case notes' (i.e., a record of their management of each girl that requested counseling, along with descriptions of the counseling encounter and the context in which it occurred). Their documentation of the cases involved a two-pronged process reminiscent of fieldnote documentation: initial 'jottings' about the counseling encounter (in order not to forget important issues raised by the client, and to be able to follow up on them during the counseling session), followed by the expansion of the notes later in the day [16,17]. Once the case notes were fully written up, they were processed in Microsoft Word and anonymized before being shared electronically with the corresponding author.
Data from the case notes were analyzed manually and thematically -a process that entailed reading through a first case note in its entirety and documenting a list of tentative themes identified at 'descriptive' and 'conceptual' levels [18]: straightforward descriptions of specific experiences were noted, in addition to more complex concepts that required deeper reflection and piecing together.
Themes were either confirmed or disconfirmed by the first author after reading subsequent case notes in their entirety and finding that they either re-emerged or did not. New tentative themes were also added to the list as the reading of subsequent case notes ensued, and they were subjected to the same process of confirming/disconfirmation by comparing them to other case notes in the data set. The final list of themes was produced via multiple readings of all case notes to be sure nothing was overlooked or erroneously labeled as a theme without evidence of multiple tellings across counseling cases.

Ethical considerations
Several levels of ethical approval were obtained prior to conducting the overall study from which this paper's data were derived, including from the Ethics and Research Committee of the Kenyatta National Hospital-University of Nairobi in Nairobi, Kenya, and from the New York-based Institutional Review Board of the Population Council. In addition to the consent and assent processes required when interviewing minors, the study also took into account the potential trauma that interview questions on early pregnancy could pose for pregnant and parenting girls. Three psychologists were therefore incorporated into the data collection team to provide first line psychosocial support (in the form of immediate counseling) to girls who asked to speak to a professional. In addition to providing first line counseling, the psychologists were trained to provide referrals for further care to those who had received counseling. The project established a relationship in advance with the Kenya Association of Professional Counselors in the neighboring County of Kisumu to attend to any acute cases referred.

Results
Five overarching themes emerged from the analysis, revolving around sexual violence, adverse childbirth outcomes, psychological trauma (of girls and their parents alike), parents' need for support, and the availability of family support for pregnant/parenting girls. These themes are unpacked in the sub-sections below.

Sexual violence
Sexual violence turned out to be the most prominent theme across the majority of the case notes, with many girls indicating that they had gotten pregnant as a result of rape. Indeed, only 5 out of the 20 pregnant/parenting girls who received home-based counseling became pregnant as a result of consensual sex. Oftentimes, incidents of sexual violence occurred in the context of girls' 'boyfriend-girlfriend' relationships. Other perpetrators that emerged from the case notes included relatives, friends, and strangers.
Pregnant girls also struggled with disclosure, due to the notion that their experience of rape was their fault, and that support and sympathy would not be readily available.
I was interviewing this girl, and when we got to the section on sexual experiences, she started crying. I stopped the interview and asked her if she would like to talk to me as I assured her of total confidentiality. She trusted me and she opened up that she is pregnant, and she has not told anyone, including her mom. She is fearing her mom will kick her out of the house and she has no idea what to do. She said her boyfriend, who is in [

Psychological trauma Girls' experiences
During their counseling sessions, pregnant/parenting girls showed evidence of psychological trauma as a result of their pregnancies (and other associated concerns), whether the latter were rape-related or not. All psychologists noted the deep traumatization of the girls concerned, with one specifying that, while the first line of counseling provided was important and helpful, many of the girls actually required long-term psychotherapy.
The sexual debut of many girls occurred in the context of sexual violence. Several girls found it challenging to disclose their horrifying experiences to anyone, and had not spoken about them until they were asked if they wanted to speak to someone during data collection for the wider study.
The circumstances described above resulted in a wide range of psychological problems and mental health issues described by pregnant and parenting girls, including depression, fear, suicidal ideation, insomnia, reduced self-esteem, and hopelessness, as outlined below: (The girl started crying

Parents' experiences
The case notes analysis also surfaced the psychological trauma confronted by the parents of pregnant and parenting girls. This trauma manifested in the forms of sadness, despair, bitterness, and anger, for instance, and led some parents to act out against their pregnant/parenting daughters, including by being harsh (verbally and physically), belittling pregnant/parenting girls before their siblings, lashing out, and simply ignoring or ostracizing their daughters. This often led to situations of 'ping-pong' trauma within affected girls' households, as such girls inadvertently inflicted trauma on their parents as a consequence of their pregnancy/parenthood status, while parents, in turn, unknowingly triggered further trauma in their daughters when they left their own trauma unattended to. The quotes below illustrate the complex psychological terrain navigated by parents of pregnant and parenting girls: Her mom was so bitter after realizing she was pregnant that she didn't say a word to her. … She told me she is a single mother, and her daughter has really disappointed her: "Can't they see how I am struggling?'' she said, as she broke into tears. She

Parents' need for support
A recurrent theme among parents of pregnant and parenting girls was their need for external support in talking to their daughters about sex and sexuality. After participating in the quantitative survey under the wider study, parents of pregnant/parenting girls commonly asked for someone to speak to their affected daughters about these issues, admitting their difficulty in doing so themselves.
The following narratives depict this communication challenge faced by parents of pregnant/parenting girls: [O]ne of the Head Teachers … had two teenage girls, and after interviewing the girl who was sampled in the study, he requested me to counsel his daughters: "You know, these things affect us as well. We are also parents, despite being leaders at school. " He said that this is one of the reasons he was advocating for girls to remain in school as long as they want ( 16-year-old girl A mother … requested the data collectors to help counsel her … daughter who gave birth and went back to school, [along with] her eldest daughter, who was equally intelligent, but whom she feared would be influenced. This mother of six and widow feared that her young daughter might have not learned from her previous experience, and may be taken advantage of again, [even though her mother] was struggling to provide for them with her meager income (she sold roasted groundnuts for a living).

Counseling case notes, 13-year-old girl
The fact that all parents who received counseling responded positively to this service (even when they had not requested for it, but were drawn into counseling by the proactive psychologists) is a further indication of the sheer need that parents have for support in order to properly support pregnant/parenting girls in turn.

Availability of family support
Despite the displeasure of most parents about their daughters' teenage pregnancies, the vast majority expressed willingness to support their daughters' return to school, or were already doing so. Parental support was demonstrated in other ways, as well, including by showing kindness and unconditional love toward their daughters, despite their disappointment, or by providing emotional support.
"I wanted to commit suicide, but my mom encouraged me and walked the entire journey with me. I would have been dead by now. Later on, I gave birth, and this was very traumatizing, and I felt the world had come to an end. I thank God for the support I got from my parents. And after a few months, I told them I wanted to go back to school. They were ready to take me back to school, and that is why I am in Class  Her paternal grandmother loves her and her baby, so she will possibly take care of the baby while she will be at school. Counseling case notes, 16-year-old girl

Discussion
This study is associated with some limitations: In particular, its core findings emanated from the narratives of a small, select group of pregnant and parenting girls who requested counseling support when it was offered, and from the accounts of an even smaller group composed of affected girls' parents. We therefore craft our discussion of the findings with great care, and with the important caveat that our conclusions are not driven by a concern for 'representativeness' in the quantitative sense. They are, rather, propelled by the conceptual issues emerging from the data. As Ulin and colleagues [19] remind us, 'Our goal [with qualitative inquiry] is to produce data that are conceptually, not statistically, representative of people in a specific context' . It is noteworthy, nonetheless, that the key findings from this study are in concurrence with results from several population-based studies, which are also discussed in this section. Findings from this study provide a portrait of existing psychosocial support challenges and opportunities in the context of teenage pregnancy, based on the lived experiences of some affected girls and parents in Kenya. This portraiture could serve as a useful starting point for developing psychosocial support interventions for pregnant and parenting girls in low-and middle-income countries where such interventions are rare [3].
Sexual violence, for instance, turned out to be the most prominent theme emerging across the majority of the case notes, with 15 out of the 20 girls in the sample indicating that they had gotten pregnant as a result of rape. The high prevalence of sexual violence among pregnant/parenting girls is reflective of findings from an earlier, population-based study in the same context, which showed that over a third (35%) of ever pregnant 12-to 16-year-olds reported getting pregnant as a result of rape or coercion [20].
Oftentimes, sexual violence occurred in the context of girls' 'boyfriend-girlfriend' relationships, supporting findings from a population-based study in the same sites which indicated that about 90% of teenage mothers reported getting pregnant by their boyfriends, with a considerable proportion of these boyfriends (37%) being their fellow students at the time [21]. Other perpetrators that emerged from the case notes included relatives, friends, and strangers. Pregnant girls also struggled with disclosing their experience of sexual violence.
The realities of girls in this study who experienced sexual violence mirror those of their peers nationwide: A nationally-representative survey in Kenya [22] demonstrated that 43% of 13-to 17-year-old girls who experienced an incident of sexual violence in the past 12 months did not tell anyone about their experience.
Screening interventions to support sexual violence disclosure have been observed to be feasible and effective in Kenyan settings [23] and could potentially be embedded into psychosocial response efforts for pregnant and parenting girls. Additionally, perpetrators were often the girls' schoolmate boyfriends or peers, and pregnant/ parenting survivors were reticent about disclosing sexual violence. These findings draw our attention to the need to incorporate boys into the portrait of teenage pregnancy. Related to this is the critical need to embed sexuality and relationships education (which would include sensitization to certain concepts such as gender-based violence, consent, self-management, kindness, and relationship skills [24], for example) in the discourse and programming around teenage pregnancy. Such education is clearly required early on in the lives of girls and boys within and outside of school contexts. While the Kenyan government has demonstrated support for sexuality education, it has privileged abstinence-only approaches [25], which, in addition to being incomprehensive, overlook the realities of sexual violence (and other grave, sexualityrelated issues) occurring as early as primary school age.
The need for sexuality and relationships education is all the more important, given parents' recurrent request in the present study for external support in communicating with their children about sex and sexuality. This parentchild communication challenge plausibly stems from cultural prescriptions in some Kenyan sub-cultures, which hold that conversations about sex and sexuality contravene notions of 'respect' within parent-child relationships [26]. Consequently, parent-child communication around these subjects is often only prompted by negatively-perceived events, including early pregnancy [27]. Sexuality and relationships education efforts could, therefore, be designed to support parents in this regard, and to complement related endeavors within and outside schools.
The study's findings demonstrate that mental health concerns are not the preserve of pregnant/parenting girls alone. Parents of girls who had experienced pregnancy sometimes faced psychological trauma of their ownan issue that has received much less attention in the literature. Intersections between adolescent and parental trauma also requires a response and further investigation in the context of teenage pregnancy, given evidence of their co-occurrence in this study. While this intersection could potentially have a cyclical effect, prolonging recovery for both girls and their parents, it also holds potential for whole-family healing, if properly addressed. Although evidence of this intersection emerged only once in the present study (in a case where a pregnant girl and her mother alike were physically abused by her father in response to her teenage pregnancy), intersections between violence against girls and violence against women in the context of teenage pregnancy are also worth exploring further, given the likelihood that this is occurring much more often than this qualitative study was able to capture.
Adolescent mothers face higher risks of adverse birth outcomes than older mothers [28]. As seen in this study, this reality can place extra pressure on girls whose babies face health challenges (or whose babies do not survive), coupled with the stigma of being a young mother of school-going age, and the prevalent experience of raperelated pregnancy. Adverse birth outcomes and their psychosocial effect on teenage mothers are rarely considered, and should become a part of response interventions for this population.
It is also important to highlight existing family support for pregnant and parenting girls as a major theme that emerged in this study. Childcare is a major barrier to parenting girls' advancement [21]. Nonetheless, as the study findings suggest, parents, grandparents, and other relatives can serve as an important childcare resource to support the future endeavors of pregnant and parenting girls. Grandmothers, in particular, emerged from the case notes analysis as relatives who offered a safe space and 'guaranteed' support for pregnant/parenting girls. These family resources should be part and parcel of psychosocial responses in low-resource settings.
Overall, the findings from this study are quite similar to results from research around the psychosocial issues of teenage mothers in non-African contexts. Interpersonal violence, mental health challenges, and economic issues are commonly experienced among pregnant and parenting girls in other settings as well [4,5,29], and incorporating the parents and extended family of teenage mothers into support interventions is considered important and effective [4]. Parent-teen communication about sexuality issues is also noted as being challenging in other settings [30]. However, conduct disorders among pregnant/parenting teens did not emerge in the current study, but were highlighted as relevant in the broader literature [5]. Its absence from the present study could be due to the small sample size (which may not have allowed for certain issues to be captured), or could be associated with sociocultural factors in Kenya that might moderate conduct problems (such as oppositional defiant disorder) in households with pregnant/parenting teenage girls. The broader literature also extends the framework of psychosocial support, going beyond pregnant/parenting girls and their parents alone, to include the children of teenage mothers, who are noted to require psychosocial support as well [4,5]. Future psychosocial studies and interventions that give attention to the children of teenage mothers would be valuable for strengthening needed responses.