Study setting

A Working Group comprised of focal points from UNFPA and UNICEF country offices, two youth advisors (females aged 18–24), and two government representatives from the Ministry of Planning and Investment was established to review methodology and study materials, provide guidance on site selection, and support interpretation and dissemination of findings. Site selection was done in consultation with the Working Group, and informed by UNFPA internal analysis of adolescent fertility and estimated premarital conception using DHS data for each province, along with cost and feasibility constraints.

Vientiane Capital and Vientiane Province were selected to provide urban and rural sites with median levels of adolescent fertility and premarital conception (i.e., close to national average rates), while urban and rural sites in Luang Namtha were selected to represent a high level of adolescent fertility and premarital conception. Within each province, site areas were selected with input from the Working Group with the aim of representing the demographics of the district. Vientiane Capital (Prefecture) is the second most populous administrative region in the country and is home to almost one million people (population density of 247 people per km²) [21]. Vientiane Province is the fifth most populous administrative region in Lao PDR, with a population of 468,000 and occupies 15,610 km² (population density of 30 people per km²) [21]. Vientiane Province contributes about 4% of the national gross domestic product largely from the service (including tourism) and industry sectors; however, in rural areas, agriculture remains the main source of livelihood. Luang Namtha, a predominantly agricultural province in the country’s northwest, is home to 202,000 people and occupies a total land area of 9,325 km² (population density of 5 people per km²) [21].

Ethical approvals

Ethical approval was provided by the Alfred Hospital Research Ethics Committee (ID: 14/21) and the Lao People’s Democratic Republic Ministry of Health National Ethics Committee for Health Research (NECHR) (ID: 2020.87.MC).

All participants provided written informed consent, including for audio recording, prior to timeline interview. Participants provided verbal consent for follow-up telephone interviews and were sent an electronic copy of the participant information sheet and notified that support organisations had hard copies of the information sheet available. Lao family law recognises young people aged 15–17 years as able to provide informed consent, and girls invited to participate in our study were pregnant or parenting at the time of interview and had already made decisions regarding their healthcare. Thus, girls recruited for the study were considered as having the evolving capacity and maturity to understand the study and provide informed consent to participate in research without the need for parental/guardian’s consent [22].

Two participants were 15 years old at interview. These were reported to the Alfred Hospital Ethics Committee and the in-country research team received immediate refresher training to reemphasise the need to clearly establish participant age prior to obtaining consent. The research team and ethics committee determined that the most ethical course of action was to retain the participants’ data in the study. The ethics committee recommended no further action.

Recruitment

Adolescent girls aged between 16 and 20 years who had become pregnant at age 18 or younger were identified through partnership with social support organisations in each study area. Most rural participants were invited (in person) a day before interview, while urban participants were invited via phone call to participate in the study. We undertook rolling recruitment and recorded participant details against a sampling matrix, recruiting purposively to achieve diversity in rural-urban residence, ethnic background, and pre- and post-marital conception.

Data collection 1: In-depth timeline interviews

Two female interviewers (KR, SP) received five days of training, including a pilot interview. The study youth advisors supported two days of training as paid consultants, participating in activities to inform the language used in the interview and interviewer practice engaging adolescents.

Interviews were conducted between March and August 2021. Interviews were done at a time and location convenient for participants (often in a private location on household grounds with auditory, and where possible, visual privacy) and lasted between 60 and 90 minutes. Participants were informed of their right not to discuss topics and to decline to answer any questions or end the interview. Discussions were paused if needed for childcare or other interruptions.

Interviews used a participatory, timeline approach and were supported by a semi-structured topic guide. Interviews commenced by asking participants to draw a timeline of their life story and the way that pregnancy and marriage fit into their story. In cases where participant literacy was low, the interviewer drew the timeline for the participant with their permission. The semi-structured topic guide was then used to explore girls’ timing of pregnancy and feelings about becoming pregnant; pregnancy intentions; relationship with the partner at the time of pregnancy; timing of union; experiences of learning about sex and reproduction, sex, and romantic relationships, and using contraception; other important family and community relationships; timeline and experience of schooling; and hopes and plans for the future. In moving through discussion topics, participants were asked if they wanted to place other experiences onto their timeline and timelines were used to clarify the temporality of events. Interviews were audio-recorded and transcribed into Lao by the interviewers (KR, SP), and timelines were photographed. Participants were assigned unique participant identification (ID) numbers (e.g., LA0101) and any personally identifiable information in the interview transcripts were replaced with generic alternatives (e.g., replaced names with ‘boyfriend/partner’, ‘mother’, ‘regional healthcare facility’). Only authorised members of the Lao PDR research team had access to participant contact information during data collection and solely for the purpose of recontacting participants who were willing to participate in follow-up interviews. Participant contact information was stored separately from study data and could not be linked to the participant ID.

Analysis

We undertook framework analysis to describe, compare, and contrast the pathways to first adolescent pregnancy shared by participants [23]. Framework analysis is a thematic analysis approach which involves five steps: 1) familiarisation, 2) developing an initial analytical framework, 3) indexing (during which we summarised interviews according to the framework), 4) charting and summarising through development of a complete framework matrix, and 5) interpretation and mapping (during which cross-case analysis was undertaken to develop preliminary and final pathway typologies) [24].

For familiarisation, the study team undertook regular debriefing to share initial impressions from interviews. Lao-speaking team members developed initial themes. The analysis team (JH, MH, RK, KR, SP) then developed a comprehensive interview summary template capturing preliminary drivers and themes identified along with other temporal details. Team members fluent in Lao (RK, KR, SP) produced in-depth summaries of interviews in English and recorded key quotations for a diverse selection of 30 interviews. Through discussion, we developed a framework matrix to capture key temporal events and themes. Each participant’s data was entered into the matrix from English summaries by a Melbourne-based team member (JH) and by Lao team members directly from the remaining 27 transcripts, with the framework modified through discussion if needed to capture girls’ experiences.

A distinct component of framework analysis is the development of a framework matrix which enables data summary and display [24], and where each row corresponds to a case (a participant), and each column to a code (i.e., a theme or sub-theme, such as topics or issues observed across the dataset) [23, 25]. Consistent with the framework analysis approach, we used the framework matrix to undertake cross-case analysis [26] and identify pathway typologies. Preliminary pathways were discussed among the study team and presented to the study Working Group for feedback. We identified topic areas for validation and clarification with follow-up participants.

Data collection 2: Follow-up telephone interviews

After preliminary analysis of the timeline interviews, we undertook follow-up interviews. These aimed to provide participant validation, clarification to preliminary findings, and to gather participant recommendations for policy and practice. Interviewers received a further three days of training on the follow-up content and to develop the videos, which was again supported by the youth advisors.

A subset of girls who provided permission and a phone number to be recontacted were called and invited to participate. Interviews were undertaken at a time nominated by the participant through a phone call. Four short and engaging videos and corresponding ‘radio-play’ audio recordings were developed to summarise key preliminary findings along topics of knowledge of sex and reproduction; contraceptive access and use; relationships and negotiating sex and pregnancy; and pregnancy and marriage. A semi-structured topic guide complemented the videos/audio recordings and explored girls’ impressions of the study findings, points of clarification on each topic, and recommendations for action around these topic areas. Videos were sent to girls with smartphones and audio recordings played back to girls with non-smartphones before interview. Interviews lasted no longer than 45 minutes and were paused if required. Interviewers made written notes during follow-up interviews and developed a written summary in English, after which all call and text history was deleted.

Insights from follow-up interviews were used to revise preliminary findings and pathways, and incorporated into written and presentation materials used during dissemination of study findings to stakeholders. Participant recommendations for policy and practice were analysed thematically and are reported elsewhere.

Profile of participants

Fifty-seven interviews were undertaken between March and August 2021 with young women aged 15–20 years; the median age was 18 years. Eight participants were from Vientiane Capital (all urban), 16 from Vientiane Province (6 urban, 10 peri-urban), and 31 from Luang Namtha (16 urban, 15 rural). Definitions or ‘urban’ and ‘rural’ villages were based on the 2015 Lao Population and Housing Census [27], while the definition of a ‘peri-urban’ village is that used by the research partner organisation.

Twenty participants were married prior to becoming pregnant, 14 became pregnant after they started cohabitating with their partner but were not yet married, and 23 became pregnant prior to cohabitation or marriage. All participants had conceived their first pregnancy prior to age 18; the youngest at age 13, with a median age of 16 years. In Vientiane Capital and Province, almost all participants were of Lao Loum (‘lowland Lao’) ethnic background. In Luang Namtha, a diverse set of ethnicities were represented including Khamu, Hmong, Lantan, Lamet, Tai Dam, Lue, Phounoi, Sila, and Yung. Most participants had completed lower secondary school, while one-third had attended some lower secondary school but not completed. Four participants were secondary school graduates.

Outside-union pregnancy pathways

For 23 girls, their pathway to pregnancy unfolded outside of union (cohabitation or marriage). Across the sample, girls began engaging in romantic relationships between ages 10 and 16. Many girls shared a similar story to that shared by LA0213: ‘The boy asked for my number from my friends. … He started chatting with me via WhatsApp. After six months of chatting, I felt that I like him. He then asked me out for Pai Len (party-going, usually involving alcohol consumption). I said I would go if my friends would come along. So, we went out drinking beer with a group of friends.’ Most girls who got pregnant outside of union had other boyfriends before the partner who got them pregnant. Prior relationships usually involved text messaging and did not involve physical intimacy. Girls typically met the partner who got them pregnant through common spaces (e.g., school, work) and chatted through social media platforms before spending time together in person.

Within-union pregnancy pathways

For 34 girls, adolescent pregnancy occurred in the context of child marriage and early union. Their pathways to pregnancy varied according to their relationship context prior to union, who initiated the union, and pregnancy intention.

Barriers to SRH information and services, including contraception

Many girls in our study lacked accurate information about sex, reproduction, and contraception before sexual debut, as also observed in other research [29]. Some girls received information about sex and reproduction in school, but the most named friends and peers as their main sources of information. Currently, school-based comprehensive sexuality education (CSE)-life skills education (LSE) in Lao PDR is integrated into other subjects at both primary (e.g., moral education, science and environment) and secondary (e.g., biology, civic education, and safe use of social media) school levels [30]. CSE is mandatory in primary school but optional in secondary school, and coverage varies according to the type of school (e.g., public, private, faith-based) [30]. Some have noted gaps in the quality of teacher training and supervision, as well as learning materials (e.g., books, other materials), and that teachers did not feel confident delivering lessons on sexuality education [31].

At follow-up, participants elaborated that girls’ lack of knowledge about sex, reproduction, and contraception was sometimes because they did not think that the information they received in school was relevant to them (they felt these were ‘only for married people’). Girls often received more information from family members and healthcare providers about SRH (particularly contraception) after marriage or birth, suggesting that girls’ views that SRH information was ‘only for married people’ were being reinforced by social norms and structures. Prior studies in Lao PDR have found that parent-adolescent communication about SRH, relationships, and sex is low, with only one in five young people reporting ever discussing these topics with their parents [15, 17]. Our follow-up interview participants confirmed that some parents do not want to discuss sex with their adolescent daughters because they believe that they are ‘too young’. Negative cultural attitudes regarding sexual activity outside of marriage were also key contributors to young people’s reluctance to access SRH services and discuss SRH topics with healthcare providers [17]. Though there are no laws that explicitly prohibit young people from accessing SRH services [32], existing policies (e.g., 2019 National Sexual and Reproductive, Maternal, Newborn, Child and Adolescent Health Policy) lack specific guidance on parental consent requirements for minors to access contraceptives [33]. In the absence of laws or specific guidance, parental consent requirements are subject to healthcare providers’ discretion [33], and recent research suggests that healthcare providers could be limiting adolescents’ access to family planning services [34], especially since sex outside of marriage is socially unacceptable in Lao culture [12].

For many girls in our study, marriage was accompanied by expectations regarding contraceptive use and pregnancy. Hormonal contraceptives were accessed by some girls once they were in a union. However, side effects associated with these, or fears of side effects (particularly infertility), resulted in discontinuation of these methods. Couples underestimating their risk of pregnancy was also a salient theme across pathways resulting in unplanned pregnancies. These couples seemed less likely to use any contraception because they did not think they would get pregnant. Prior experiences of unprotected sex that did not result in pregnancy led to false confidence that unprotected sex would not result in a pregnancy, reflective of poor SRH knowledge. This was observed among participants who conceived before and after union.

Many girls in our study who had unplanned and partner-led pregnancies considered abortion and a few girls successfully undertook a termination. Even though abortion is illegal in Lao PDR except to save a woman’s life or to protect a woman’s physical health [28], recent data indicate that a drop in contraceptive use during the COVID-19 pandemic was associated with an increase in the unsafe abortion rate [35]. Our findings suggest that despite its illegality, abortion remains common in Lao PDR and women–including adolescent girls–need access to safe abortion services.

Partner’s control over reproductive decision-making

Partner’s control over reproductive decision making (including contraceptive use and desire for a pregnancy) was another key driver across pathways. For pathways in which the couple jointly did not plan pregnancy, condom use was usually discontinued when the boyfriend decided to stop using and instead, promised to take responsibility for any pregnancy. Some couples, but particularly males, reportedly viewed condoms as inappropriate for a couple in union. During follow-ups, participants elaborated that married couples found to be using condoms may invite suspicion, because condom use was associated with sexually transmitted infection or marital infidelity. This aligns with other research where male participants expressed that they did not like condoms and felt that condoms were for men who were cheating on their wives [36]. Follow-up participants added that girls often felt they had no choice but to accept the partner’s preference. This is consistent with the findings of other studies in Lao PDR, that unmarried young women usually deferred to their male partner when it came deciding whether or not to buy and use contraceptives [18, 19], even when the young woman’s preference differed from that of their partner [17].

Some participants experienced a partner-led pregnancy despite being aware of contraceptive methods and, in some cases, having attempted to use them. In these cases, the partner/husband wanted to have a baby, but in some cases, family members also encouraged pregnancy soon after union. A study in Savannakhet Province found that some women accepted their partner’s decision on contraception (including not using any), although some women from urban areas reported that they would try to negotiate or covertly use contraceptives if they disagreed with their partner’s decision [36]. Some of our follow-up interview participants believed that married women had more say in timing of subsequent pregnancies and contraceptive use after they had already had one baby; more so if the baby was a boy. Yet, we noted a few cases of married girls attempting to use contraceptives covertly, indicative of unequal power in their relationships [37].

Male sexual entitlement and alcohol use driving and pressured/forced sex

Experiences of pressured and forced sex were observed across pathways, usually involving alcohol, and some of which were soon followed by pregnancy. Alcohol consumption is common among Lao young people, especially in rural settings [38]. In our study, alcohol consumption combined with notions of male sexual entitlement contributed to girls’ risk of pressured and forced sex as some boys/men took advantage of girls under the influence of alcohol. However, in contrast to findings from related research in Indonesia [39], among the subset of girls who experienced forced sex outside of union, it was more common for girls to not marry their assailant. Girls becoming pregnant due to forced sex outside of union represented their own pathway to adolescent pregnancy, and these participants either refused to continue a relationship with the partner who got her pregnant, or her parents/partner decided against pursuing or continuing a union. Alongside interventions seeking to engage men and boys in eliminating violence against women and girls, there is an urgent need for gender transformative programmes that promote respect in relationships and address harmful use of alcohol and other drugs.

Cultural acceptance of child marriage and early union

Consistent with prior research [18], community acceptance of child marriage and early union was an important driver of adolescent pregnancy in this study. Our findings indicate that child marriage and early union is viewed as the most acceptable response to adolescent pregnancy outside union. This was consistent with other research in rural communities in Lao PDR that found that young people, parents, and community elders had favourable attitudes and norms regarding child marriage and adolescent pregnancy as it was the norm in these settings [18]. Most of our participants experienced sexual debut before cohabitation or marriage, often following assurance from the boyfriend that they would take responsibility for any pregnancy through ‘Pai Khor’. When an unmarried girl in a romantic relationship became pregnant, parents expected the couple to marry. Betrothal through ‘Pai Khor’ was crucial as it served to lessen the negative social consequences for the pregnant girl and her family. Notably, for a few girls who had unplanned pregnancies, there was less urgency for marriage to happen before the birth of the baby–some parents accepted cohabitation with the promise of marriage later. This is consistent with another study that found that among women aged 20–24 years who have ever been in a union and conceived before union, only 58% of women from Laos were married by the time they gave birth [14].

In some cases, girls acknowledged that it was common in their community or village for adolescent girls to marry and begin childbearing, and some stopped attending school by choice once they entered a union. Similar findings were noted in Indonesia, where adolescents viewed becoming a wife and mother as a valued and realistic life goal, particularly when there were constraints to education and work [40]. Cohabitation or marriage both served as an indication of a couple’s long-term commitment to each other. However, during follow-up interviews, participants clarified that though cohabitation was common, girls in cohabiting unions still aspired to be formally married. Other research has found that it is common for Lao girls to aspire for marriage and childbearing in adolescence, with some young women expressing apprehension that they may not be able to attract a suitable partner if they wait until age 20 to marry [18].

For a smaller number of participants, particularly in Luang Namtha, girls were pressured into marriage by parents for economic reasons, and this was soon followed by pregnancy. Some parents expected or pushed for their daughter to cohabitate or marry to ensure financial security for the girl by marrying her into a family with a better economic position. In other cases, parents (and some girls) saw marriage as a way to ease the financial burden on the family, or to gain an additional labourer to help the farming household. For these participants, parents’ expectations of marriage remained a primary driver of adolescent pregnancy.

Follow-up interview participants elaborated that if a married couple does not have a baby soon after marriage, they may be mocked, ridiculed, or suspected of ‘something wrong’–such as the man having ‘weak sperm’ or the woman having ‘too much sex’ or using contraceptives ‘too much/for too long.’ This was indicative of the social pressure experienced by both females and males to prove their fertility once in a union, and illustrates how child marriage and early union norms within the community are contributing to adolescent pregnancy.

Attitudes and norms regarding sex and pregnancy outside of union

In almost half of our sample, premarital sex and avoiding the risk or shame of a pregnancy outside of union served as drivers of child marriage and early union, which then facilitated adolescent pregnancy. For some girls, the discovery by family that they had engaged in premarital sex prompted some boyfriends to propose or parents to arrange (or consent to) their daughter’s engagement to avoid embarrassment or having to pay village fines. In our study, about half of participants in the post-union pathways and most of the participants in the pre-union pathways experienced sexual debut before union, indicating discordance between social ideals disapproving of premarital sex and young people’s lived realities. This aligns with similar research from the Philippines noting that even amid social disapproval of premarital sex, many young people engage in sexual relations before cohabitation or marriage [41, 42]. Others have noted that while premarital sex is stigmatised and not socially accepted in Lao culture [12, 17, 34], in certain areas (i.e., rural, and rural-off-road) and ethnic minority communities, traditional values, beliefs, and practices are more permissive regarding adolescent premarital sexual relations [10, 18, 43].

Strengths and limitations

We note some limitations of our study. First, we used a combination of purposive and snowball sampling in two provinces in Lao PDR to recruit young women who were best placed to address our research topic. As such, our findings are not necessarily representative of adolescents’ pathways to pregnancy in other provinces, regions, or the country. However, we captured diverse contexts and crosscutting factors that will be important to explore with girls from other geographical areas, and socioeconomic and ethnolinguistic backgrounds.

Second, and related to our sampling and study areas, we were able to collect rich qualitative data through our in-depth timeline interviews. While we were able to gather the depth of information that we needed to address our research objectives, given our focus on strands within individual narratives [44], it is possible that other pathways featuring novel contextual and crosscutting factors may still eventually emerge if more data were to be continuously collected [45]. Nevertheless, our study was able to identify eight diverse pathways based on key life events, motivations, and decision-making related to sexual relations, contraceptive use, non-use and discontinuation, pregnancy, and union formation, all of which would likely still be important in any alternative pathways to adolescent pregnancy.

We conducted follow-up interviews to validate and clarify our study findings with a selected number of participants from the individual interviews. We opted to conduct follow-up interviews over the phone due to prevailing COVID-19 restrictions and to minimise exposure risk for our participants and Lao PDR research team. It is possible that conducting the follow-up activity through in-person, participatory group discussions may have generated different (e.g., more in-depth, detailed) feedback from our participants, as participants would have had the opportunity to engage with each other and elaborate on their ideas, which was not possible through one-on-one follow-up interviews over the phone. Yet, the follow-up interviews validated and clarified our findings and interpretations, particularly regarding social pressure for couples to begin childbearing soon after child marriage and early union, and the common reasons for male unwillingness to use condoms. These will be important considerations for future research as well as programmes and policy.

Implications for research and practice

Our findings contribute valuable insights regarding the different ways that adolescent girls in Lao PDR navigate romantic relationships, sex, and union in relation to their experiences of adolescent pregnancy.

Our study found that many adolescent pregnancies in Lao PDR occur outside of formal marriage despite social disapproval of premarital sex and pregnancy. Adolescent girls’ experiences of sex and pregnancy were affected by a lack of knowledge about SRH, power imbalances with their partners, and girls’ lack of agency over if and when to have sex, use contraceptives, and begin childbearing. These were also occurring within sociocultural contexts where parents, community members, and adolescent girls themselves viewed child marriage and early union as an acceptable alternative to education (especially when the girl’s family faced financial hardship), and as the most acceptable response to a pregnancy outside of union.

Moving forward, it will be imperative to address adolescent girls’ barriers to SRH information and services, including contraception. This requires strengthening comprehensive sexuality education (CSE)/life skills education (LSE) both within and outside school settings by increasing young people’s access to CSE/LSE, but also by improving the content covered and the quality of delivery. CSE/LSE content should include a strong focus on addressing myths about contraception, equipping girls with the skills (e.g., communication, negotiation, consent) they need to make informed decisions and assert themselves in their relationships, teaching young people to engage in respectful relationships, and engaging girls and boys to critically analyse harmful gender norms. Further implementation research is needed to understand how to effectively deliver CSE curriculum at scale, and how to effectively address harmful gender norms.

Alongside this, we recommend ensuring that girls have access to non-judgemental, adolescent-responsive healthcare, and increasing availability of and improving access to affordable (or free) contraception. SRH service providers will need to support all girls (whether in a union or not) to access and use their chosen contraceptive method. This would need to be complemented by initiatives to transform parents’, community members’, and healthcare workers’ perceptions that SRH information and services are not appropriate for unmarried adolescents, including addressing individual and social attitudes that contribute to male resistance to condom use.

The acceptance of child marriage and early union–as well as adolescent pregnancy–is deeply rooted in Lao customs, traditions, values, and beliefs, including the view that associates womanhood with being a wife and mother [18]. Culturally responsive approaches that foster adolescent girls’ sexual and reproductive agency will be required to shift these, and it will be important to ensure girls have the resources they need to pursue their educational and work goals. At the community and societal levels, this requires addressing sociocultural and financial drivers of child marriage/early union and adolescent pregnancy, by working toward shifting notions of child marriage and early union as a ‘solution’ for premarital pregnancy, and helping families build economic resilience as a preventive measure against child marriage/early union and adolescent pregnancy.