Article Text

The downstream effects of COVID-19 on adolescent girls in the Peruvian Amazon: qualitative findings on how the pandemic affected education and reproductive health
  1. Lisa L Woodson1,
  2. Adriana Garcia Saldivar2,
  3. Heidi E Brown1,
  4. Priscilla A Magrath3,
  5. Nicolas Antunez de Mayolo4,
  6. Sydney Pettygrove1,
  7. Leslie V Farland1,
  8. Purnima Madhivanan3,
  9. Magaly M Blas5
  1. 1Department of Epidemiology and Biostatistics, The University of Arizona Mel and Enid Zuckerman College of Public Health, Tucson, Arizona, USA
  2. 2Universidad Peruana Cayetano Heredia, Lima, Peru
  3. 3Department of Health Promotion Sciences, The University of Arizona Mel and Enid Zuckerman College of Public Health, Tucson, Arizona, USA
  4. 4School of Medicine, Universidad Peruana Cayetano Heredia, Lima, Peru
  5. 5School of Public Health and Administration, Universidad Peruana Cayetano Heredia, Lima, Peru
  1. Correspondence to Lisa L Woodson; lisalabita{at}


Due to COVID-19, schools were closed to mitigate disease spread. Past studies have shown that disruptions in education have unintended consequences for adolescents, including increasing their risk of school dropout, exploitation, gender-based violence, pregnancy and early unions. In Peru, the government closed schools from March 2020 to March 2022, declaring a national emergency that affected an estimated 8 million children. These closures may have unintended consequences, including increased adolescent pregnancy, particularly in Peru’s rural, largely indigenous regions. Loreto, located in the Peruvian Amazon, has one of the highest adolescent pregnancy rates in the country and poor maternal and child health outcomes. The underlying causes may not be fully understood as data are limited, especially as we transition out of the pandemic. This qualitative study investigated the downstream effects of COVID-19 on adolescent education and reproductive health in Loreto’s districts of Nauta and Parinari. In-depth interviews (n=41) were conducted with adolescents and community leaders. These were held in June 2022, 3 months after the reinstitution of in-person classes throughout Peru. Focus group discussions (FGDs) were also completed with community health workers and educators from the same study area in October 2022 to supplement our findings (3 FGDs, n=15). We observed that the economic, educational and health effects of the COVID-19 pandemic contributed to reduced contraceptive use, and increased school abandonment, early unions and adolescent pregnancy. The interplay between adolescent pregnancy and both early unions and school abandonment was bidirectional, with each acting as both a cause and consequence of the other.

  • Qualitative study
  • COVID-19
  • Global Health

Data availability statement

Data not available due to ethical restrictions. Qualitative data collected for this study are not publicly available as they contain potentially identifying information of the research participants. Due to the participants residing in small and remote communities in Nauta and Parinari, the research team ensured that the data presented were de-identified in a manner that preserves their confidentiality, including the use of generic participant labels for quotes used in the report.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Disruptions to education have been shown to have adverse effects on adolescent girls. Longitudinal studies in Malawi and Kenya showed that COVID-19 led to lower school re-enrolment rates, higher dropout among older adolescent girls and increased pregnancy risk for girls aged 15–19 years. A systematic review across low-income and middle-income countries found varying increases in adolescent pregnancy during the pandemic due to factors like school closures, peer pressure and economic challenges. However, data from Brazil showed a decrease in unplanned adolescent pregnancy, likely due to social distancing measures.


  • We provide a better understanding of the effects of the COVID-19 pandemic on the education and reproductive health of adolescent indigenous girls in the Amazon basin of Peru. This study also highlights three key factors affecting adolescent pregnancy during COVID-19: economics, education and healthcare access.


  • The pandemic may have long-lasting sexual and reproductive health consequences for adolescent girls whose education was severely disrupted. The knowledge gained from this research can be used to inform interventions and policies that may be implemented in this region to reduce the risk of adolescent pregnancy and school dropout, especially targeted towards indigenous youth.


Adolescent pregnancy (AP) has adverse maternal and perinatal outcomes with increased risk of morbidity and mortality.1 Globally, pregnancy and childbirth complications are the leading cause of death among girls 15–19 years of age.2 3 To improve adolescent sexual and reproductive health (SRH) outcomes, increasing the educational attainment of girls is an effective strategy.4 5 However, school closures to mitigate COVID-19 spread had unintentional consequences. It was estimated that 11–20 million children globally dropped out of school due to economic shocks of the pandemic,6 7 a majority were secondary school-aged girls.8 9 Dropout was more profound among girls in vulnerable and marginalised communities.6 Girls who dropped out of school were at increased risk of exploitation, gender-based violence and child marriage.10 Decreased female schooling is associated with low contraceptive use11 and in turn, decreased contraceptive use has been shown to increase adolescent fertility.12 13 Conversely, higher education levels deter AP, especially in low-income and middle-income countries.14

A longitudinal study in Malawi of pre-COVID-19 and post-COVID-19 data found that extended school closures resulted in lower school re-enrolment rates and higher dropout rates, especially among older adolescent girls.15 In Kenya, there were observed significant differences in secondary school-aged girls between the pre-COVID-19 and COVID-19 cohorts, with the latter group reporting a twofold to threefold higher risk of pregnancy and dropout.16 A systematic review examining the effect of COVID-19 on adolescent health, education, and other social and economic factors in low-income and middle-income countries found that increased AP rates were influenced by school closures, peer pressure, lack of family planning and economic pressures.17 Conversely, Latin America, particularly Brazil, experienced a decrease in unplanned adolescent pregnancies during the first year of the pandemic likely due to enforced social distancing measures.18 These data, however, may not have fully captured the long-term risks of AP due to COVID-19 mitigation efforts, did not differentiate between high-risk groups and were not representative of all Latin American countries.19

In Peru, there was a slight increase in AP in 2021, especially in the selva or jungle region, although it is unclear if this was a result of the pandemic as the study used live birth certificate data.20 The study furthermore does not take into account Peru’s COVID-19 migration efforts put into effect on 12 March 2020. The government imposed some of the earliest and strictest lockdowns in the world and mandated school closures nationwide21 affecting over 8 million children from pre-primary to secondary school.22 23

Latin America and the Caribbean have the second-highest rate of AP in the world despite reductions made in the past two decades.6 24 In Peru, the fertility rate among adolescents 15–19 years was 58.0 in 2020 and remains considerably high when compared with other middle-income countries.24 Moreover, these rates are considerably higher among adolescent girls residing in rural areas, the selva (jungle) regions and within the lowest wealth quintiles of Peru.25 Thus, it is not surprising that Loreto, located in the Amazonian basin of northeastern Peru with a predominately rural population and high rates of poverty, has one of the highest rates of AP in the country.25 Loreto also has some of the country’s poorest maternal and child health outcomes.26

Peru’s COVID-19 containment measures may have unintentionally affected adolescent girls in the Loreto region. However, available data are limited, making it challenging to fully understand the extent of these effects. The purpose of the study was to gain greater insight into the downstream effects of the COVID-19 pandemic on the educational and reproductive health outcomes of adolescents within the Amazon basin—specifically focusing on school dropout and AP.


Study sites

The study sites were located in Nauta and Parinari districts of Loreto, Peru along 210 km of the Marañón River, a major tributary of the upper Amazon River. We conducted research in 11 communities. The study sites ranged in size from 120 to over 400 people and were largely designated by the Peruvian government as indigenous population centres or communities. Most were only accessible by boat.

Less than 15 000 people reside in the rural communities of Nauta and Parinari of our study area, a majority identifying as either Catholic or Evangelical and indigenously as Kukama.27–29 The predominant language spoken is Spanish. The main livelihood activities are fishing and farming; communities closer to the capital city of Iquitos also make and sell handicrafts.

Communities had limited access to health centres along the Marañón River depending on location and population density. In Peru, primary health centres are divided into four types from lowest (I-1) to highest (I-4) service provision. The urban centre of Nauta, which has the only paved road to Iquitos, services a larger population with I-4 and I-3 health centres. The population centre of Parinari has the other I-3 health centre. In our study area, there are 11 I-1, 1 I-2, 2 I-3 and 1 I-4 health centres (see figure 1).

Figure 1

Map of communities in the districts of Nauta and Parinari, Loreto, Peru embedded with community health workers (CHWs) through Mamás del Río service area. Primary health centres are divided into four types from the lowest (I-1) and highest (I-4). I-1 refers to health posts which usually include one nurse’s aide, nurse or nurse midwife. I-2 refers to a health post with a medical doctor which includes one of the personnel/services from an I-1 establishment and a general physician or surgeon. I-3 refers to a health centre and includes personnel/services from I-1 and I-2, a dentist, laboratory technician, pharmacy technician and a statistical technician. I-4 refers to a health centre with an inpatient facility and includes personnel/services from I-1 to I-3, a specialist physician and professional pharmacist.

Not all communities have a secondary school. There are 38 secondary schools in our study area including 10 located within and directly around the port city of Nauta. Secondary school consists of 5 years from grades 7 to 11, with students typically aged 13–18 years.30

Study design

We conducted both in-depth interviews (IDIs) and focus group discussions (FGDs). For the IDIs, we collected data from adolescents defined as either a pregnant girl, non-pregnant girl, or boy, and community leaders. Inclusion criteria for adolescent IDIs were Spanish-speaking youth between the ages of 15 and 17 years who attended school full-time before March 2020. Pregnant adolescents included those who were primigravida or primiparous (having become pregnant after August 2020 or 6 months after the start of the pandemic). Non-pregnant adolescent girls were nulliparous. Adolescent boys had no paternity exclusion criteria. We included male participants to enhance our comprehension of peer and relationship dynamics, aiming for a more comprehensive understanding of the experience of the pandemic for adolescents. Community leaders were identified as those who hold a place or occupation of distinction in our study sites, including certified medical professionals, and indigenous or youth leaders. Participants were required to be Spanish speaking and either worked or resided in Nauta or Parinari.

FGDs were conducted with educators and community health workers (CHWs) who volunteered to participate in the study. Educators eligible for inclusion had to be Spanish speaking and had current or prior experience working as an educator, staff, or administrator in a local secondary school in Nauta or Parinari. CHWs were required to be Spanish speaking, residents of Nauta or Parinari, had previous experience as a CHW before the pandemic and be current active members of Mamás del Río, a project through Universidad Peruana Cayetano Heredia. Mamás del Río is a community-based health intervention project focused on improving maternal and neonatal health outcomes in rural Amazonian communities of Peru.31–36 We sampled CHWs and educators in key access points along the Marañón River. The FGDs were held in larger population centres because of the availability of secure and private spaces. Gender-specific FGDs were held with the CHWs. However, owing to the small number of educators, we did not separate these participants by gender for the FGDs.

We used purposive sampling using the snowball principle to identify interview participants for the study. We first identified pregnant adolescent girls in Nauta and Parinari with the help of local obstetricians and CHWs. In study sites with identified pregnant adolescents, we first introduced our research to the indigenous community leader or apu when available. Once permitted to conduct the study in the community, we would interview the apu. Next, we would interview the pregnant adolescent and ask her to refer an adolescent boy and a never-pregnant adolescent girl for an interview. In circumstances where they could not make a referral, we relied on the CHW or local obstetrician for the recruitment of participants in the same community who met our inclusion criteria.


Guides for the IDIs and FGDs were developed to gain a contextual understanding of factors that directly and indirectly influenced adolescent educational attainment and reproductive health during the pandemic. Additionally, our semistructured interview guides included sensitive SRH topics, which participants may under-report because of confidentiality concerns and reluctance to disclose potentially embarrassing personal information.37 To address this, SRH questions were phrased to capture the experiences of adolescents in their community rather than of the individual. This technique was adapted from Westgard et al.38

The numbers of IDIs and FGDs were determined by thematic saturation,39 conducted in Spanish, and lasted approximately 45–90 min. Demographic data were collected from each participant at the start of the interview or FGD. IDIs and FGDs were recorded using a voice recording device. If the participant declined to be recorded, notes were handwritten.

Study participants

Table 1 shows that the mean age of adolescent participants was 15.9 years. Over half (55.2%) self-identified as indigenous. 44.8% did not specify an ethnic identity despite living in a recognised indigenous population centre or community. All adolescent participants were enrolled in secondary school before the pandemic, of which 58.6% were currently enrolled and 27.6% had since graduated. Four adolescents (13.8%) permanently dropped out of school between March 2020 and March 2022.

Table 1

Demographic characteristics of study participants in the study sites of Nauta and Parinari by type, n=56

IDIs with community leaders included seven apus, four medical professionals and one youth leader. Not all apus were available in the study sites for an interview; however, none had declined our request to conduct the study. Only one study site did not have an apu. All male community leaders in this sample were apus. Female community leaders consisted of obstetricians, a medical technician and a youth leader. The mean age of community leaders was 44.5 years with two-thirds (66.7%) having completed secondary school or higher. Most community leaders were indigenous (75.0%).

The mean age of the FGD participants was 45.7 years with 33.3% having completed primary school and 53.3% having completed secondary school or higher. All CHW participants identified as indigenous while educators identified as non-indigenous.

Data collection

Field observations were conducted in June and October 2022. Field notes were compiled and organised by date. This included notes on adolescent interactions and SRH education. Brochures and other health materials provided by the local health clinics of the ‘Decidamos Ya! Reducir el Embarazo Adolescente en Loreto’ campaign were also collected and reviewed.40 41 We used these data to triangulate the findings from IDIs and FGDs on how and in what context SRH information was provided to adolescents.

IDIs were conducted in June 2022, 3 months after the reinstitution of in-person classes throughout Peru. Empirically, to reach thematic saturation, we had aimed to complete 12 IDIs among each homogeneous group42; however, we had difficulty recruiting adolescents for the study due to the sensitive nature of discussing SRH, the unfamiliarity of participating in qualitative studies and time constraints. In total, we completed 41 IDIs consisting of 11 pregnant adolescents, 9 never-pregnant adolescent girls, 9 adolescent boys and 12 community leaders in Nauta and Parinari. FGDs with educators and CHWs were conducted in October 2022 to supplement interview findings.

Data analysis

Notes and recordings were transcribed and uploaded to ATLAS.ti Windows (V. Using a thematic approach, we derived patterns from our qualitative data. Themes with subthemes were identified through an iterative process using notes gathered from field observations, field debriefings and IDIs across participant groups. Codes were developed both inductively and deductively. Broader code categories included: social norms, SRH access, SRH outcomes, contraceptive use, confidence/aspiration, COVID-19 effects, education, mobility, social interactions and technology use. For this paper, we only present data that were expressively linked to youth’s COVID-19 experience. A more in-depth examination of social norms is referenced in a compendium paper.44 No new codes emerged from the FGDs. Once the codebook was finalised in Excel, codes were uploaded to ATLAS.ti and used by the research team to code each interview and FGD. Six IDIs were selected to test intercoder reliability with a Krippendorff alpha-binary of 0.80 or greater across code groups. Identified discrepancies were reviewed and discussed between coders, then recoded through a consensus process.

The design, data collection and analysis phases of the research were informed by the interdisciplinary expertise and experiential backgrounds of our research team in the fields of public health, medicine, sociology, epidemiology and anthropology. See the online supplemental appendix for reflexivity statement.

Supplemental material


Effects of COVID-19-related restrictions

Adolescents clearly remembered the day when they received news of the emergency declaration by the Peruvian government of a country-wide shut down due to the COVID-19 pandemic. The initial 15-day lockdown was extended for 2 years, and community members were severely restricted in their movement outside their homes.

At the time of the pandemic, it was different… For example, you couldn’t go out to play, you couldn’t go out to have fun in the street, everything was controlled, sometimes security took care of us, and you couldn’t go out anywhere. (adolescent boy, IDI)

With consecutive lockdown periods under the state of emergency, adolescents grew more hopeless. They also described this period as sad, fearful, lonely and boring.

That’s how we went and until I think the month of June when there was an indefinite lockdown. There we lost hope of everything. (non-pregnant adolescent, IDI)

I gained a lot of weight because, what were we doing? In the house, we made our breakfast, our lunch, our dinner. We ate and laid down… everything was boring. We didn’t go out, we just looked out the window. We closed it because we were afraid of getting infected. (non-pregnant adolescent, IDI)

Drinking and smoking also increased during this time as a coping mechanism.

Well, here was a time that… they were a lot into alcohol. Yes, alcohol, and they [adolescent boys] smoked a lot… Yes, they were like that for a while. (non-pregnant adolescent, IDI)

The COVID-19 pandemic had negatively affected communities along the Amazon basin. Due to a growing fear of contagion, rural communities along the Marañón River closed their ports and followed strict social distancing precautions starting in early 2020 as the rapid spread of disease, and a large number of deaths were reported from Iquitos. In addition, adolescents feared leaving their homes. Despite these mitigation efforts, ‘in the end, we all ended up getting the disease’ (pregnant adolescent, IDI). Some shared personal experiences of deaths in their families; others recalled hearing about people dying of COVID-19 at the health posts. Yet, these restrictions may have been hard to enforce in rural areas, and over time, communities eased their COVID-19 restrictions as it became economically unviable to continue strict isolation measures.


Households reported a loss of income during the pandemic. Fishing, as the primary livelihood activity and part of the cultural identity of the Kukama people in this region, was affected by mobility restrictions along the Marañón River and port closures. The trade linkages were severely disrupted as fish and products could not be sold in the urban markets of Nauta and Iquitos, and family businesses were unable to get supplies.

It was pure state law. The other communities too, are a bit difficult to enter another community. In other words, they locked us up. (male community leader, IDI)

There has been a port blockade here, right? No exit or entry, right? But that led to chaos. There was no rice, there was no chicken, there wasn’t much. (male community leader, IDI)

Households experienced financial hardships which were further exacerbated by increased food prices during COVID-19. Households faced food insecurity and resorted to negative coping strategies such as reduced food intake, selling non-essential household items, relocating, and/or subsistence fishing and farming.

Here are times, like they say, we don’t have the money. Sometimes, uhum (laughs) we eat little. Or sometimes my partner goes to look for work… on the farm; [what] he produces, he sells. Sometimes we have [money], we just buy it… sometimes you have little and you can’t buy so many things [like] what you want to eat sometimes… I know that a lot has gone up [in price]. What you could eat before… with little money and now it’s no longer possible. (pregnant adolescent, IDI)

Additionally, many fathers were forced to find work outside of the community leaving adolescents with less parental supervision. There was also a growing demand for adolescents to work in the chacras or farm/fields, family gardens or fish to supplement household food.

[We] sold everything. I didn’t know how; all the money had gone down the drain. We were left without money and so we went to the farm. (pregnant adolescent, IDI)

Other youth went ‘to look for work in the city’ (pregnant adolescent, IDI), which usually consisted of informal or labour-intensive employment such as construction. Adolescent girls hired as domestic workers reported unpaid and exploitative working conditions:

During the pandemic, I could not go out because, because I did not live with my mother. I lived with another lady, that is, she made us work a lot… What did I not do!? I didn’t rest. I [worked] mostly the fields, cultivated [all day] in the sun, sometimes washed clothes, and had no rest. (pregnant adolescent, IDI)

Young girls forced to work outside of their homes in larger urban areas due to economic insecurity during the pandemic were more vulnerable to early unions and AP; some girls had sexual relations with men in the home or extended family of their employer. However, from the accounts of the participants, it is unclear how these relationships were initiated.

Access to healthcare

Health centres, sporadically spread along the Marañón River, were not accessible as movement was severely restricted, and many facilities were closed during the pandemic. If open, they were understaffed and diverted their limited resources toward the COVID-19 response.

Additionally, health staff did not travel to the communities, and medicine was scarce, further disrupting regularly provided services, including contraceptive access, during this time.

Many continued to take care of themselves, and many did not because, as I say, they did not have as much access since the medicines did not arrive at the clinics. (female community leader, IDI)

Yes, it has changed with the pandemic. The nurses are in charge of giving contraceptive methods but in the pandemic, they were not here [in the community], and there was no one to turn to. (adolescent boy, IDI)

Botiquines, or community medicine cabinets, were quickly depleted of their medical supplies and were not replenished, forcing participants to rely more heavily on CHWs and locally available medicinal plants or remedio vegetal for the treatment of COVID-19, other ailments and family planning. The use of medicinal plants was not uncommon to prevent unwanted pregnancies and may have had increased use during the pandemic as formal SRH services and modern contraceptives were severely limited.

I believe that there are, uh, enough plant medicines to take care of ourselves, as you say, ‘toronja’ [grapefruit] is one, for me, it is a very effective remedy. I took care of myself just like that… I cannot get pregnant. (female CHW, FGD)

Adolescent girls refrained from seeking reproductive healthcare beyond their community because of COVID-19-related restrictions. When intercommunity travel was permitted, girls were required to travel to the health post by boat chaperoned by a parent or older adult male relative. Unaccompanied travel was not permitted. This prompted concerns from girls about accessing confidential and private SRH services.

SRH information

Before the pandemic, adolescents accessed SRH education from visiting organisations or in school. Health clinics also provided SRH information to adolescents primarily during on-site appointments or health checks. When financed, the obstetrician from the health clinic visited neighbouring communities through campaigns to provide abstinence-based SRH education in schools (as noted in our field observations), at church or in other open community spaces. However, the subject matter varied depending on the instructor. Despite formal efforts to provide adolescents with SRH information, it was considered inefficient, inconsistent and at times, incorrect.

That is, they gave us talks but… we couldn’t prevent ourselves. In other words, even though they gave us instructions on how to take care of ourselves with the methods… most of my classmates have gone out… like this… that is, they are all moms. (pregnant adolescent, IDI)

The teacher told them that they can also be infected [with HIV] when they use the same bathroom: ‘You have to wash the bathroom with bleach, do that… when you sneeze the virus spreads, just like COVID, you have to take care of yourself.’ (adolescent boy, IDI)

Regardless of how effective formal SRH education was perceived among adolescents, during the pandemic, it was largely absent or unavailable despite calls from Peru’s Ministry of Education to teach contraceptive methods to youth.

In the pandemic, in the first year they didn’t talk…, and now it [is said] that the Ministry of Education itself tells teachers to come and talk about contraceptive methods. (non-pregnant adolescent, IDI)

This can be attributed to lockdown measures that restricted adolescents from accessing education at the health posts. Also, virtual SRH education was considered less optimal and largely halted as schools and obstetricians lacked the resources to effectively teach the subject.

In the absence of formal SRH education during the pandemic, adolescents relied on informal networks for information and advice. For many, parents were identified as primary source of SRH information. However, the extent of information shared is unclear as one adolescent boy described it as ‘not enough’. The conversation between parents and daughters centred around ‘[taking] care not to get pregnant’, while conversations with sons had greater emphasis on ‘[taking] care of themselves’ (pregnant adolescent, IDI).

Thus, adolescents depended on friends for more explicit information on sexual intercourse and contraceptive methods, sharing what they learnt from personal experience, health posts, previous SRH workshops at school, parents, older siblings, partners or the internet. Due to increased access to technology during COVID-19, adolescents turned to this resource to obtain SRH information.


Prior to the pandemic, secondary schools located in the Marañón River employed teachers from urban areas who stayed for short periods with a schedule that allowed them to travel back and forth. The quality of teaching was considered low due to the poor recruitment of qualified teachers. Many also did not ethnically identify as indigenous. Not all communities had secondary schools and students had to travel by peke-peke or gasoline-powered wooden boats to attend in person. This presented challenges in terms of increased transportation costs and the dangers of traversing hazardous river conditions and inclement weather. In cases where students had to attend school in another community, they needed to rent accommodations in a dormitory, home or a designated hostel nearby, incurring additional educational costs.

During the pandemic, schools were closed, teachers did not travel and students were socially isolated in their homes. To meet the educational needs, teachers would make photocopies of the homework to be sent by boat to the communities each month. When available, the Peruvian government distributed tablets to promote virtual education. Households were required to pay for photocopy charges and to recharge mobile data plans for the tablets; however, many could not afford these additional costs as they were already experiencing economic hardships because of the pandemic. Furthermore, several communities in the Amazon basin do not have reliable access to electricity or internet via cellular towers. Government-issued tablets came with data plans from a single telecom company that were not uniformly available across communities. This rendered online education as an impractical option for many students. To help offset the educational costs during the pandemic, parents gathered small groups of students to share resources.

It was something…not as normal as it was before. You didn’t see your classmates anymore, it’s not like being at school with everyone together…it was something separate, we only met as a group [with] some of our classmates, that’s how it was. (adolescent boy, IDI)

Adolescents were less motivated to attend classes online. Some expressed frustration in adapting to this new format of learning and felt that it was less engaging, of lower quality, inconsistent and less productive. Virtual classes were offered less frequently and in shorter duration than in-person classes had been before the pandemic. Students, during this period, temporarily abandonado or abandoned virtual classes and stopped submitting homework; however, many did not consider this dropping out or leaving school permanently. In the end, students passed or graduated regardless of meeting academic proficiency or performance for their grade level by government order (Vice-Ministerial Resolution No. 193-2020-Minedu)45:

Yes, there were some changes as everything was virtual. All the students passed [their] grade, just like that, without knowing anything… that we have learned almost nothing. (adolescent boy, IDI)

Teachers recounted that when schools reopened for in-person classes in March 2022, many students could not be accounted for. It was assumed that these students dropped out.

I have students who have not been on my attendance list since March [2022] … and there are at least five in each classroom. We have not heard from them since March to this day [October 2022]. I don’t even know them either because they didn’t attend [school]. (male educator, FGD)

Adolescents were more likely to drop out in their final years of secondary school, usually reflecting changing life priorities related to work, relationships or having children.

I think now just to raise my child. Because I’m not thinking about going back to school… Everything is different – I mean that’s your reality, and besides that, the jobs that come don’t give you time for anything. (pregnant adolescent, IDI)

Falling in love along the way, so after that, many do not finish secondary school. (male community leader, IDI)

During the pandemic, many adolescents were also pressured to discontinue their education due to the economic stress of the pandemic, choosing instead to support their families and, in some cases, work outside of their communities. Many did not re-enrol when schools reopened in their communities.

Interviewer: In the pandemic years, who has left school the most?

Participant: The girls… Sometimes some young people, for work reasons, money…yes, mostly for that reason. (adolescent boy, IDI)

[My friends/classmates] they didn’t want to go [to school] because maybe… well, it was boring, they worked, they saw the money and they didn’t want to study anymore. (pregnant adolescent, IDI)

Technology use

Communities along the Marañón River have limited access to the internet, and electricity was largely restricted to nightly use. However, our findings suggested that adolescents in this region used technology in a greater capacity than before the pandemic. Adolescents increasingly relied on technology to access virtual education through government-distributed tablets or personal cellphones, and as a means for social engagement when strict COVID-19 mitigation efforts limited individual movement and required youth to self-isolate. Adolescents also sought SRH information online because personal devices provided greater privacy. Yet, the use of technology for this purpose is not well supervised and community leaders were concerned about the growing technology use among adolescents and increased access to pornography.

They get into some programs, uh, let’s say that does a lot of harm to the child, to the adolescent… they use it [the internet] for bad things… and that is the issue for us; it is really worrying to see the child go crazy, right? (male community leader, IDI)

Early unions/AP

Many participants believed that the pandemic led to more unions and partnerships in their communities despite having limited social interactions during the pandemic.

No, the pandemic has made even more couples join together, and there have been even more couples here in the community because they did not go out, they did not go out to any town, and no town was allowed to enter… And that has meant that everyone… with that, the number of young people who have their partners within the community has increased. (male community leader, IDI)

Sexual and romantic relations were maintained in ‘secret’ (male community leader, IDI), using online platforms to connect and coordinate meetings in person despite strict lockdown restrictions.

Yes, sometimes when we go out there alone, we find ourselves, sometimes a chair, a bench, in a dark place as they say. (pregnant adolescent, IDI)

Many of the partnerships were observed between underage girls (<18 years) and older adult men 8–15 years their senior, with unions/cohabitation becoming obligatory upon pregnancy. The pandemic may have pushed more adolescent girls into relationships with older partners for economic stability and pressure to find a suitable partner. For example, the adult male partner of one pregnant adolescent was identified by his prominent position in the community and reported by the study participant and others as being ‘a good person’. The subsequent quote from a non-pregnant adolescent provided valuable insight into the dynamics of these relationships:

He is supporting me a lot, well, I tell you that from a very young age. Many girls always say: ‘I like older people’, right? (laughs) Because I think that an older person can help you, I say, or can guide you. (non-pregnant adolescent, IDI)

However, not all pregnancies result in a union. Some pregnant adolescents experienced abandonment during the pandemic.

During the pandemic, participants believed that there was an increase in APs in their communities. A pregnant adolescent shared that before the pandemic, many young girls played outside in the afternoons but afterward, she noticed most now have children. Others believed that AP was most common among girls aged 14–15 years but also included those as young as 12 years:

Well, in the pandemic it did affect a lot because people, students ah, about 15, 12 years old, 13, 14, 17, not much 17; more were 15 to 14 years old; there were quite a lot that they were gestating. (non-pregnant adolescent, IDI)

Causal diagram

Based on our research, we developed a causal diagram illustrating how COVID-19 affected education, health access and economics/household financial stability, and in turn, AP and school dropout (see figure 2). COVID-19 also inadvertently affected early unions and contraceptive use. The relationship between AP, school dropout/abandonment and early unions is bidirectional, indicating that AP is both a consequence and a cause.

Figure 2

Causal diagram of the effects of the COVID-19 pandemic on education and sexual and reproductive health of adolescents in the Amazon basin, Peru. Arrows indicate directionality, dashed lines indicate bidirectionality; red lines represent a negative relationship; black lines represent a positive relationship; outcomes are in grey boxes; the primary outcome of interest (adolescent pregnancy) is in the red diamond.

We observed that girls dropped out of school during the pandemic because they became pregnant or were forced to work due to economic insecurity. They migrated for work, often under exploitative conditions. They were vulnerable to advances made by adult men in the households they were employed and may have been subject to coercive sexual relations. Many returned to their communities pregnant; however, this would need further exploration as it was unclear from participant accounts how these relationships were formed. Likewise, girls may have also sought these relationships for greater economic stability.


Our research aimed to understand how adolescent SRH and education were affected during the COVID-19 pandemic. We were able to access remote indigenous communities in the Amazon basin of Peru and learn from vulnerable youth in this region. We found from our thematic analysis that the economic, educational and health effects of COVID-19 contributed to school dropout/abandonment, decreased contraceptive use, early unions and AP. However, the relationship between AP was not unidirectional between school dropout and early unions. In a longitudinal study in Peru, cohabitation and early unions were found to be intrinsically related to and often a result of pregnancy.46

Economically during the pandemic, Peru experienced the greatest drop in gross domestic product among all Latin American countries, an estimated contraction of 11%. This decline can be attributed to the government’s COVID-19 mitigation efforts, and its effect on education, health and mobility.47 Additionally, the pandemic pushed more families into poverty as unemployment increased. The informal job sector was particularly affected.47 The financial repercussions of COVID-19 may have elevated the risk of AP, given that both low socioeconomic status and lack of employment opportunities in low-income and middle-income countries are associated with an increased likelihood of AP.48

Likewise, the COVID-19 policies lacked cultural appropriateness and consideration for Peru’s indigenous people. The emergency declaration of compulsory social isolation by the Peruvian government gave little advance notice and left many indigenous groups without supplies and food.49 When products were available locally, many households were priced out due to the high demand for these products when available.49 The economic pressure of the pandemic not only pushed girls to drop out of school to seek employment outside of their communities, increasing their risk of AP, but changed nuclear family structures. Fathers, in particular, were forced to find work outside of the community resulting in a single-parent household, which may help drive AP.46 50

COVID-19 mitigation efforts had further detrimental SRH consequences for adolescents. The pandemic made it difficult for youth to access SRH services.51 This was affected by factors such as fear of COVID-19 contagion, access restrictions and cost.17 52 Concerns around accessing confidential and private healthcare may have also reduced the utilisation of SRH services and contraceptives among adolescents.52 Indigenous girls in our study sites were required to travel to the health post accompanied by a parent or adult man, potentially creating impediments to accessing these services. Such factors increased their susceptibility to AP as they face greater geographical, economic and cultural/linguistic barriers to accessing appropriate education and healthcare compared with their non-indigenous counterparts.53

Education serves as a buffer against AP.14 Favara et al found that in Peru, attendance and better school performance were associated with increased age of sexual initiation and a lower risk of early childbearing.46 Conversely, girls with lower education levels were less likely to use contraceptives and were at a higher risk of pregnancy.13 School retention is a key AP prevention strategy for secondary school-aged girls.54 However, COVID-19 school closures affected how youth were able to access education.17 Disadvantaged children in Peru (those whose parents did not complete secondary school) lost an average of 34% instructional time during the pandemic.55 Ultimately, this may lead to wider educational inequalities and greater dropout, especially among poorer households like those in our study sites.56 57

SRH information was largely absent during the pandemic and is unlikely to be compensated for in the future.52 Even so, formal health education may have limited effectiveness in reducing AP among this population even before COVID-19 due to inconsistent programming and incomplete information. It has also been shown that SRH interventions based solely on school curricula have limited evidence of changing the sexual practices and behaviours of adolescents.58 Apart from formal SRH education, adolescents also relied on parents, peers and the internet to acquire more knowledge. The reliability and accuracy of this information remain unclear and need further investigation as cultural taboos may restrict what is shared.59 Notably, previous research in Peru’s central jungle region, as demonstrated in a case–control study, highlighted the protective effect of adolescent SRH communication with parents against AP.60 While our study sample was of a different indigenous background, this finding underscores the significance of parent–adolescent communication and parent-sourced SRH education in AP prevention.

Furthermore, early unions between pregnant adolescent girls and older adult men were observed in our study and reportedly increased during the pandemic. Adolescent girls may be motivated to seek out relationships with older men for financial security or out of fear of not finding a suitable partner.61 Studies have shown that adolescent girls were at higher risk of engaging in exploitative work during COVID-19 and were subject to unwanted sexual advances,51 a known determinant of unintended pregnancies.54 It is important to also consider the unequal gender power relationships between adolescent girls and their adult partners in the Amazon basin. Girls lacked agency to make decisions around their reproductive health.62 The COVID-19 pandemic likely intensified these concerns. However, our data were limited as older male partners were not included as participants in our current research, restricting us from gaining deeper insight into this phenomenon within the specific cultural context of our study site.

Other limitations of our study included possible recall and social desirability bias. Participants may have had difficulty recalling events over the last 2 years since the national school closures in March 2020. To reduce this bias, we conducted IDIs with adolescents and community leaders within 3 months of school reopening in Peru.

We also recognise that social desirability may have influenced responses among adolescent participants when discussing SRH and education. In the demographic survey, only four adolescent study participants (13.8%) had declared that they had permanently dropped out of school during the pandemic; however, in interviews, we found that a larger number may have temporarily ‘abandoned’ school during this same period. Using other terminologies or descriptions in the demographic survey and interviews may help elicit this information better.

Due to the nature of how participants for the FGDs and IDIs were recruited, they may not have been representative of the population. CHWs who volunteered to be in our study could easily travel to the location of the FGDs; however, this inadvertently may have biased our sample as more isolated communities along the Marañón River may not have been represented. Likewise, educators for our FGD were only recruited from one secondary school in our study site, and medically trained community leaders did not identify as indigenous although they lived in these communities. Adolescents who agreed to participate in interviews may also be inherently different from those who declined. For future research, we would include adult male partners of adolescent girls, a group not originally part of the study design. It was only through IDIs and FGDs that we identified this knowledge gap.

Other effects of COVID-19 on adolescent health were not explicitly explored in our research. These include mental health, gender-based violence and poverty. However, we found that participants often alluded to these topics when relating their experiences of the pandemic in terms of their relationships, AP and school. These factors would need further investigation as they are important to uncovering the gaps in current knowledge of the pandemic’s effect on our study population.


As we transition out of the COVID-19 pandemic, we recommend concentrated efforts to counteract the effect COVID-19 had on adolescent girls in the Amazon who are the most vulnerable in Peru. This includes specific strategies aimed at reducing their risk of AP. The insight gained from our research, especially around how youth access and use information, health resources and technology, can help inform and develop post-pandemic interventions that aim to reduce the incidence of AP in these communities. Overall, our research underscores the growing importance of investing in tailored, comprehensive and consistent SRH education in indigenous communities that involves a broader network of formal and informal educators. Furthermore, we propose leveraging the increased use of technology by integrating it into SRH education. This approach must be paired with more accessible, adolescent-friendly, confidential and private SRH services. It is crucial to ensure that SRH services and education are culturally appropriate and readily accessible for indigenous populations, particularly during future pandemics, to effectively address their unique needs and geographical isolation.

Data availability statement

Data not available due to ethical restrictions. Qualitative data collected for this study are not publicly available as they contain potentially identifying information of the research participants. Due to the participants residing in small and remote communities in Nauta and Parinari, the research team ensured that the data presented were de-identified in a manner that preserves their confidentiality, including the use of generic participant labels for quotes used in the report.

Ethics statements

Patient consent for publication

Ethics approval

Ethical approval for the research was obtained from Universidad Peruana Cayetano Heredia as the IRB of record (SIDISI Code: 2071919) approved on 13 May 2022, and with amendments on 6 June 2022 and 3 October 2022. The University of Arizona served as the external IRB (STUDY00001109). Before data collection, we obtained written informed participant consent. For participants under the age of 18 years, we obtained both assent from minors and parental consent. Informed consent was obtained prior to community leader interviews and FGDs. Adolescent participants received a compensation of S/10 PEN (equivalent to US$2.50) per interview, whereas community leaders, CHWs and educators were given S/15 PEN (equivalent to US$3.75). For CHWs who incurred travel costs to attend the FGDs, we provided reimbursement for the cost of gasoline used (~S/20 PEN per gallon).


We would like to thank the Agentes Comuntarios de Salud of Mamás del Río, the community leaders, local health staff, teachers and adolescents of the communities of Nauta and Parinari for their contribution to our research. We would also like to acknowledge the support and guidance of Mamás del Río, Universidad Peruana Cayetano Heredia, Gerencia Regional de Salud de Loreto and the University of Arizona. Special thanks also go to Elizabeth Jacobs and Cory Mabel for their contributions.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • PM and MMB are joint senior authors.

  • Handling editor Manasee Mishra

  • PM and MMB contributed equally.

  • Presented at The research was previously presented as a poster presentation at the Ninth Annual Stanford Global Health Research Convening at Stanford University, California on 25 January 2023.

  • Contributors LW, PM and MMB led the conception and design of the study with input from PAM, LVF, SP and HEB. LW, PM, HEB and MMB helped to secure funding for the research. LW, PM and MMB coordinated efforts for ethics approval through the University of Arizona and Universidad Peruana Cayetano Heredia. Study site logistics were coordinated with MMB, NAdM and AGS. LW and AGS conducted data collection, codebook development and analysis. LW and AGS interpreted the data with guidance from MMB. LW drafted the manuscript. PAM, MMB, HEB, PM, AGS and NAdM provided revisions to the manuscript. All authors reviewed the final manuscript for publication. LW, as the guarantor, accepts full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish.

  • Funding This work was supported by the Fulbright-Fogarty and Global Health Equity Scholar’s Programme (grant #: D43TW010540). Additional funding was provided by the University of Arizona’s College of Public Health Dean’s Fund (grant #: N/A) and the Graduate and Professional Student Council Research and Project grant (grant #: N/A).

  • Map disclaimer The inclusion of any map (including the depiction of any boundaries therein), or of any geographic or locational reference, does not imply the expression of any opinion whatsoever on the part of BMJ concerning the legal status of any country, territory, jurisdiction or area or of its authorities. Any such expression remains solely that of the relevant source and is not endorsed by BMJ. Maps are provided without any warranty of any kind, either express or implied.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.

  • Provenance and peer review Not commissioned; internally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.