Article Text

Protocol
Protocol for the Adolescent Menstrual Experiences and Health Cohort (AMEHC) Study in Khulna, Bangladesh: A Prospective cohort to quantify the influence of menstrual health on adolescent girls’ health and education outcomes.
  1. Julie Hennegan1,2,3,
  2. Md Tanvir Hasan4,
  3. Abdul Jabbar4,
  4. Tasfiyah Jalil4,
  5. Elissa Kennedy1,5,6,
  6. Erin Hunter7,8,9,
  7. Adrita Kaiser4,
  8. Sabina Akter4,
  9. Afreen Zaman4,
  10. Mahfuj-ur Rahman10,
  11. Laura Dunstan1,6,11,
  12. Alexandra Head1,6,
  13. Nick Scott1,12,
  14. Helen Anne Weiss13,
  15. Thin Mar Win14,
  16. G J Melendez-Torres15,
  17. Kyu Kyu Than14,
  18. Chad L Hughes16,
  19. Sonia Grover6,17,
  20. Mahadi Hasan10,
  21. Sabina Faiz Rashid4,
  22. Peter Azzopardi1,6,17,18
  1. 1Maternal, Child and Adolescent Health Program, Burnet Institute, Melbourne, Victoria, Australia
  2. 2School of Population and Global Health, The University of Melbourne, Melbourne, Victoria, Australia
  3. 3Melbourne School of Population and Global Health, University of Melbourne, Melbourne, VIC, Australia
  4. 4BRAC James P Grant School of Public Health, BRAC University, Dhaka, Bangladesh
  5. 5School of Population Health and Preventive Medicine, Monash University, Clayton, Victoria, Australia
  6. 6Murdoch Children's Research Institute, Parkville, Victoria, Australia
  7. 7Department of Public Health Sciences, Clemson University College of Behavioral Social and Health Sciences, Clemson, South Carolina, USA
  8. 8Department of International Health, Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland, USA
  9. 9Sydney School of Public Health, The University of Sydney, Sydney, NSW, Australia
  10. 10WaterAid Bangladesh, Dhaka, Bangladesh
  11. 11School of Social and Political Sciences, University of Melboune, Melbourne, Victoria, Australia
  12. 12Department of Epidemiology and Preventive Medicine, Monash University, Clayton, Victoria, Australia
  13. 13MRC International Statistics and Epidemiology Group, London School of Hygiene and Tropical Medicine, London, UK
  14. 14Myanmar Country Program, Burnet Institute, Yangon, Myanmar
  15. 15Faculty of Health and Life Sciences, University of Exeter, Exeter, UK
  16. 16Disease Elimination Program, Burnet Institute, Melbourne, Victoria, Australia
  17. 17Department of Paediatrics, The University of Melbourne, Parkville, Victoria, Australia
  18. 18Telethon Kids Institute, Adelaide, South Australia, Australia
  1. Correspondence to Dr Julie Hennegan; julie.hennegan{at}burnet.edu.au

Abstract

Background Menstrual health is essential for gender equity and the well-being of women and girls. Qualitative research has described the burden of poor menstrual health on health and education; however, these impacts have not been quantified, curtailing investment. The Adolescent Menstrual Experiences and Health Cohort (AMEHC) Study aims to describe menstrual health and its trajectories across adolescence, and quantify the relationships between menstrual health and girls’ health and education in Khulna, Bangladesh.

Methods and analysis AMEHC is a prospective longitudinal cohort of 2016 adolescent girls recruited at the commencement of class 6 (secondary school, mean age=12) across 101 schools selected through a proportional random sampling approach. Each year, the cohort will be asked to complete a survey capturing (1) girls’ menstrual health and experiences, (2) support for menstrual health, and (3) health and education outcomes. Survey questions were refined through qualitative research, cognitive interviews and pilot survey in the year preceding the cohort. Girls’ guardians will be surveyed at baseline and wave 2 to capture their perspectives and household demographics. Annual assessments will capture schools’ water, sanitation and hygiene, and support for menstruation and collect data on participants’ education, including school attendance and performance (in maths, literacy). Cohort enrolment and baseline survey commenced in February 2023. Follow-up waves are scheduled for 2024, 2025 and 2026, with plans for extension. A nested subcohort will follow 406 post-menarche girls at 2-month intervals throughout 2023 (May, August, October) to describe changes across menstrual periods. This protocol outlines a priori hypotheses regarding the impacts of menstrual health to be tested through the cohort.

Ethics and dissemination AMEHC has ethical approval from the Alfred Hospital Ethics Committee (369/22) and BRAC James P Grant School of Public Health Institutional Review Board (IRB-06 July 22-024). Study materials and outputs will be available open access through peer-reviewed publication and study web pages.

  • Adolescent
  • Dysmenorrhea
  • EPIDEMIOLOGY
  • Health Equity
  • Health policy
  • PUBLIC HEALTH
http://creativecommons.org/licenses/by-nc/4.0/

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: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

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.

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • Hypothesis-driven cohort with a priori specification of core analyses informed by past research, with prospective design to capture the impact of menstrual health on broader outcomes.

  • Capturing menstrual experiences and needs over time generates unique data on the variability of experience and measures of sustained unmet needs.

  • Proportional random sampling of schools through full subdistrict listings will provide a representative sample of girls attending class 6 in 2023 in Khulna, Bangladesh.

  • The experiences of girls who have dropped out of school prior to secondary school (class 6) are not captured.

  • Findings may be compromised by loss to follow-up.

Introduction

Menstrual health (MH) is essential for gender equity and the well-being of women, adolescent girls and other people who menstruate. Defined as a ‘state of complete physical, mental and social well-being in relation to the menstrual cycle’,1 MH requires the individual to be able to meet a range of needs including access to information and education, materials, facilities and services to care for the body, timely care for discomforts and disorders related to the menstrual cycle, and a positive environment free from stigma and discrimination. The past decade has seen broad recognition that millions of people globally are unable to meet these needs and experience poor MH.2–7

A large body of qualitative research has reported linkages between poor MH and consequences for education and employment,8–10 social participation,11 12 physical and mental health,8 11 13–15 along with hypothesised linkages with broader sexual and reproductive health (SRH).16–19 However, links between MH and broader health, education and social outcomes have rarely been quantified.20–22 While there is an established relationship between early menarche and subsequent poorer mental health and SRH,23 the contribution of MH to this association is unexplored.24 Although qualitative research and advocacy have galvanised attention to MH in policy and practice,2 5 it remains difficult to secure substantial and sustainable financial investment without estimates of the impacts of MH on broader health and social outcomes.

Cross-sectional studies have tested relationships between menstrual hygiene practices and reproductive tract infection and irritation,24–27 and a small number of studies have investigated associations between menstrual experiences and mental health or school engagement.28–32 While these have provided useful estimates, cross-sectional studies are unable to infer directionality. Moreover, cross-sectional studies often capture MH experiences at a single time point. Menstruation is a frequent, repeated experience and the hypothesised impacts of unmet needs are linked to sustained exposure over time, rather than experiences during a single menstrual period. The small number of trials of MH interventions provides some causal indications of the impacts of poor MH on SRH knowledge, urogenital infection, well-being and education33–37; however, these estimates primarily reflect the effectiveness and quality of the tested interventions. Prospective cohort studies are needed to deliver rigorous quantitative estimates of the impacts of MH on health and social outcomes over time.

Cohort studies are also positioned to test the pathways through which different exposures contribute to MH.8 11 The sociocultural context—including gender norms and stigma surrounding menstruation—limits menstrual education, constrains social support including from healthcare providers and imposes restrictive behavioural expectations related to menstruation resulting in less effective self-care practices, poorer confidence and feelings of shame and distress.8 Resource limitations including inadequate water, sanitation and hygiene (WASH) infrastructure and inaccessible or unaffordable menstrual materials impede effective and comfortable menstrual blood management and cause feelings of shame throughout the menstrual period. These diverse challenges contribute to MH and they are then hypothesised to contribute to broader social and health outcomes. This diversity of exposures complicates intervention development and evaluation. It is plausible that different deficits, such as inadequate WASH or education, impact different aspects of menstrual experience and different health and social outcomes. Yet, policies, programmes and evaluations often assume that addressing any MH need (eg, WASH or education) will result in improvements on the same set of broader outcomes (eg, education, urogenital infection). This may lead to erroneous conclusions that MH interventions are ineffective if the wrong primary outcomes are selected. By following adolescents over time, a cohort study is positioned to understand changing MH needs over this important developmental window and provide a clearer theory tracing the pathways to impacts on girls’ lives. Findings can enable more suitable targeting of interventions at appropriate ages and for desired outcomes.

Study objectives

The Adolescent Menstrual Experiences and Health Cohort (AMEHC) Study will follow a cohort of adolescent girls in Khulna, Bangladesh to address a range of gaps in the MH evidence base.

The AMEHC Study aims to:

  1. Describe girls’ MH needs and experiences throughout adolescence.

  2. Quantify the impact of sustained exposure to met and unmet MH needs on girls’ education (participation, performance, attainment), mental health (anxiety, depression), physical health (urogenital symptoms, healthcare seeking) and SRH (body image, empowerment, early marriage, contraceptive uptake and early pregnancy) over time.

  3. Advance understanding of the pathways to MH and to health and social outcomes, including the role of contextual factors.

Methods and analysis

Study design

AMEHC is a closed, prospective longitudinal cohort study. It includes (a) annual surveys with adolescent girls, (b) surveys with girls’ guardians at baseline and wave 2, (c) annual data collection in schools capturing WASH infrastructure, education provision, and recording girls’ school attendance and performance data. Data collection activities are presented in more detail below. The cohort is supplemented by a nested subcohort of post-menarche girls followed at three time points in 2023. The subcohort provides unique data capturing the variability in menstrual experience and needs within a single year.

Throughout the study, engagement with stakeholders through a national advisory committee and dissemination with communities supports the translation of findings to recommendations for policy and practice.

In preparation for the cohort, the research programme included a preliminary phase of community engagement, exploratory qualitative research, and a pilot survey to inform and refine the cohort measures. This foundational phase ensured cohort survey tools were informed by nuanced understanding of menstrual experiences in the study context. Figure 1 presents an overview of the AMEHC Study activities and timeline.

Figure 1

AMEHC Study methods and timeline. AMEHC.

Study setting

The AMEHC Study is situated in Bangladesh, a country with world-leading prioritisation of adolescent MH national strategy, and interventions delivered through national and international non-governmental organisations (NGOs).38 Thus, findings from the AMEHC Study can be positioned to influence policy and programming nationally, while addressing global research questions about the relationships between MH exposures and outcomes. Our study brings together research institutions and NGO partners to support stakeholder engagement and rapid translation of the findings relevant to policy and practice. The existing body of MH research in Bangladesh39–41 has highlighted that adolescent girls throughout the country experience a wide range of sociocultural and resource-related MH challenges. The Bangladesh National Hygiene Survey in 2018 reported a mean age at menarche among school girls as 11.8 years.42 Findings are likely to provide generalisable insights for similar settings across South Asia and can be considered in the context of country and regional MH needs elsewhere.

The AMEHC Study is being undertaken in Khulna district, Bangladesh. The study district and subdistrict areas were selected based on the following criteria: (1) the study NGO partner (WaterAid Bangladesh) had strong relationships with local government but was not currently implementing or aware of other large-scale NGO implementation of programmes focused on MH; (2) an area in which WaterAid is likely to implement MH programming in the future that could be informed by study findings; and (3) cost and feasibility of transport for study teams. In 2015, Khulna district had a population of 2 million, with 66% residing in rural areas. According to the Poverty Maps of Bangladesh (2016), Khulna district has a poverty head count ratio of 30.8%.43 77% of the population is Muslim and 23% Hindu.44 Representative data from the Multiple Indicator Cluster Survey undertaken in 2018 found that most of the people in Khulna use improved sanitation infrastructure (95%). Khulna also represents a climate-vulnerable setting experiencing upheavals in WASH conditions, including high salinity in soil and water.45 The 2022 Demographic and Health Survey reports that in Khulna, 32% of women are married before age 16 years (56% prior to age 18 years), and 31% of those 15–19 years old have ever been pregnant.46 Khulna district is comprised of nine upazilas (subdistricts) and five metropolitan thanas (under a city corporation). For the AMEHC Study, we selected one rural upazila, Dumuria, and the urban area Khulna City Corporation (including its five metropolitan thanas).

Preliminary activities: qualitative research and measure development

In 2022, prior to the cohort, we undertook a phase of community engagement, focused qualitative investigation, cognitive interviewing and pilot survey to develop and refine survey measures along with our data collection protocols and enumerator training. Our qualitative investigation took a phased approach to identify and develop measures to capture social norms surrounding menstruation in the Khulna context, and to explore girls’ knowledge about menstruation, MH literacy and information seeking experiences. This included qualitative activities with girls, mothers, fathers, adolescent boys and teachers. Cognitive interviews with adolescent girls further served to adapt translations, ensure relevance and support revalidation of the Menstrual Practice Needs Scale (MPNS)47 for use in Bangladesh. Cognitive interviews also served to explore the suitability and adapt measures of body image48 and social norms.

The pilot survey collected data from 360 schoolgirls in class levels 6–9 (ages 12–16 years were eligible) from 10 schools, including single sex, co-education and madrassas (religious education system schools), to test questions across ages. Our pilot survey included questions that we anticipate using throughout the cohort, and was used to refine Bangla translations, check question performance and undertake psychometric assessments on scales for use. For example, the pilot served to support the development of a short form of the MPNS,47 select items for capturing body image, and to remove knowledge-quiz items that performed poorly or did not distinguish between participants. Questions capturing menstrual practices49 were refined to ensure we accurately reflected the contextual realities in Khulna. We found in the pilot that girls appreciated having a printed copy of the questions to follow along and enjoyed self-completing small sections of the survey, so we retained these activities in the cohort. Finally, through the pilot survey, we compared the performance of candidate mental health measures in the study context to select the most acceptable tool for the cohort. Detailed findings from this work will be reported elsewhere.

AMEHC population and recruitment

We aim to recruit a sample of 2000 adolescent girls enrolled in class 6 (the first year of secondary school) at baseline, including 1000 participants from each of the urban and rural study areas. Class 6 was selected to capture many girls prior to menarche (with an expected age of 12 years in class 6), while avoiding loss to follow-up from girls transitioning from primary to secondary schools. The sample size was selected based on cost and feasibility to maximise the data available for testing our research questions and to allow for loss to follow-up. The sample size is comparable with that of the country samples within the Global Early Adolescence Study (sample sizes range from 614 to 2809), which investigates broader gender norms over time among young people.50 We anticipate a loss to follow-up of 6–9% per year, noting the recent RITU MH trial in Bangladesh was unable to follow up 9% of the baseline sample at 2-year endline.51 A loss of 8% per year would result in a sample of 1569 in wave 3 (78% retention).

Schools are the entry point for sampling. Participants will be recruited based on a full listing of schools in the two study areas. In 2022, the NGO partner WaterAid undertook a full listing of co-educational, single-sex and madrassa education system schools in Khulna City Corporation and Dumuria subdistricts and collected information on the number of girls enrolled. We calculated the proportion of girls enrolled in each school type in Khulna City Corporation and Dumuria based on school listings to inform a proportional random sampling approach. Schools of each type were ranked using a computer-generated random number and will be recruited according to rank until the target sample size of girls attending each school type is reached. Schools declining participation will be replaced by the next school on the list. To minimise disruption to schools, we planned to recruit girls from a single class level (class 6) and include all girls in the class to avoid individual girls feeling targeted or excluded from the study activities.

This approach resulted in a total sample of 2016 to avoid excluding girls in final selected schools. A total of 101 schools were included.

Data collection

Data collection activities are summarised in figure 1.

Girls’ survey

Adolescent female participants will be surveyed annually through an enumerator-administered survey. Girls will be surveyed at school or at home dependent on school scheduling and girls’ availability. The timing of surveys undertaken in schools will be arranged with teachers and school administrators to minimise disruption such as during a health class or break. Girls who have dropped out of school will be surveyed at home. We will endeavour to reach girls who move out of Khulna district for in-person or telephone-based survey. Surveys at baseline will last for approximately 30 min for menstruating girls, while subsequent surveys may last up to an average of 50 min. Participant surveys are administered by trained female enumerators. The survey is loaded on a tablet with the enumerator reading questions aloud and entering participant responses. For girls’ surveys, participants are provided with a printed copy of the survey which they can also use to indicate responses and follow along with questions. For a small subset of scale-based questions, participants are asked to complete their own paper copy of the survey and responses are entered immediately into the tablet application (app). This provides variation in the survey experience, and in our pilot survey was received well and helped to maintain attention. It also enables the use of scale questions using ‘I’ statements. The girls’ baseline survey is presented in online supplemental materials 1.

Guardians’ survey

One of girls’ guardians will complete a short, enumerator-administered survey at baseline and wave 2 to capture household demographics such as parental education, employment and wealth index items (assessed through asset ownership), and their perspectives on menstruation. The guardian baseline survey is presented in online supplemental materials 2.

Guardians’ and girls’ surveys are administered using the KoboCollect app and uploaded to the BRAC James P Grant School of Public Health Kobo Toolbox server. Participant contact information was collected using a separate survey administered using the REDCap survey app and uploaded to the Burnet Institute REDCap server.

School audit

School information is collected by a trained enumerator which includes direct observation of WASH infrastructure (for example, sanitation facility type, functionality and cleanliness at the time of observation) and questions answered by a school representative (for example, if the school provides education on menstruation and at what class level). From wave 1 onwards, school administrative data including participating girls’ attendance records and performance in core subjects will be collected from participating schools. If girls transfer to schools within the study areas (Dumuria and Khulna City Corporation), these will be added to school checklist and administrative data collection. This will not be possible for schools outside of the study area. The baseline school checklist is presented in online supplemental materials 3.

Outcomes, exposures and covariates

Primary outcomes

Through the cohort study, we will estimate the effects of MH on girls’ broader health and social outcomes. These outcomes have been identified through past review of qualitative research,8 and drawn on recent efforts to define linkages between MH and the Sustainable Development Goals.2 Our outcome selection is also informed by recent efforts to co-define priority outcomes of interest across sectors, including health (SRH, psychological health, physical health), WASH and gender, through a workshop among global stakeholders (including researchers, practitioners, United Nations (UN) agency and national government representatives).52

Education

MH is hypothesised to impact school attendance and participation in class due to difficulties enacting menstrual hygiene behaviours, feelings of shame surrounding menstruation, social expectations for girls to remain at home during menstruation and unaddressed menstrual pain. This in turn is hypothesised to contribute to poorer school performance and ultimately dropout/lower level of educational attainment. Current estimates suggest only 64% of girls in Bangladesh complete secondary-level education.53 Throughout the AMEHC Study, we seek to test the hypothesised contribution of the progressive effects of MH on education. At each survey wave, participants will self-report school attendance during their most recent menstrual period and participation in class during menstruation. Girls will also report current class level which will assess grade progression and time to dropout. From schools, we will capture attendance register data for the 2 months preceding each survey. Finally, we will collect girls’ annual school-based examination scores on literacy (writing) and numeracy (mathematics). End-of-year examinations undertaken in class 8 are set by the education board (encompassing multiple districts, including all of Khulna district), providing comparability across schools.

Social participation

Beyond education, girls are hypothesised to miss other social activities due to menstruation. Through yearly surveys, girls will self-report difficulties participating in social activities, cooking, eating with others and bathing in their regular location due to menstruation. These measures are aligned with the Multiple Indicator Cluster Surveys and Joint Monitoring Programme indicators for MH.4

Mental health

Poor MH is hypothesised to impact on girls’ mental health, particularly anxiety and depression, along with more generalised well-being through multiple pathways. Stress surrounding menstrual self-care, unaddressed pain, shame associated with menstruation, withdrawal from social settings and teasing from others are all experiences which may exacerbate existing anxieties or feelings of depression and contribute to social isolation and depression during menstruation. Poor MH literacy including understanding normal and abnormal bleeding or symptoms and the timing of each menstrual period may also exacerbate feelings of anxiety or disempowerment. Based on findings from the pilot survey, we will use the depression and anxiety subscales from the youth-Depression Anxiety Stress Scale (DASS),54 55 with translation drawing on validated Bangla translations of the DASS-21.56

Physical health

Physical health outcomes will include symptoms of urogenital infection or irritation, and healthcare seeking for menstrual pain or disorders. Menstrual hygiene practices and genital care have been linked to reproductive tract irritation and infection.57 Girls will be asked to self-report symptoms of urogenital infection including itching and burning, pain on urination, frequent urination, genital odour and abnormal vaginal discharge. Girls will also be asked to self-report care seeking for genital concerns and the treatment received. Throughout follow-ups, girls will be asked to report interactions with healthcare services and seeking care for menstrual pain or concerns.

Sexual and reproductive health

MH has been hypothesised to serve as an entry point for improving adolescent SRH outcomes.16 58 MH is hypothesised to affect adolescent SRH through multiple pathways. First, inadequate knowledge about menstruation and the menstrual cycle initiates a trajectory of misinformation, sociocultural taboos, and a poor foundation for SRH and contraceptive education.16 17 Second, positive experiences of control, and positive body image and autonomy in relation to MH and hygiene behaviours have been hypothesised to contribute to improved SRH empowerment and outcomes.16 There is limited evidence, mostly from high-income settings,59 that negative menstrual experiences are associated with lower sexual decision-making power, mediated by body image. In early years of the cohort (ages 12–14 years), we will collect girls’ self-reported body image using an adaptation of items from the Body Appreciation Scale48 and empowerment using adapted items from the Sexual and Reproductive Empowerment Scale for Adolescents and Young Adults.60 Girls will also self-report if they are married or have children. We aim to extend the cohort to older ages (15–18 years) and at these ages will seek to measure contraceptive awareness and uptake, age at sexual debut, along with continued data on girls’ age at marriage and childbearing.

Exposures and covariates

Specific primary exposures, secondary (more distal) exposures and covariates will differ for each outcome of interest based on hypothesised pathways, expanded on further below (with figures depicting the relationship between proximal and more distal exposures). Table 1 presents an overview of the most common primary exposures, distal exposures and covariates, how they will be measured and the data collection waves in which they will be collected.

Table 1

Summary of primary and distal exposures, and covariates

Analysis plan

We anticipate using generalised linear mixed-effects models to test the relationships between exposures and outcomes, with choice of link function to follow from distribution of outcome values. However, specific analyses will vary according to each research question. Primary exposures will be investigated to determine the most appropriate form of predictor; for example, we will explore whether aggregated measures to capture experience over time or the identification of trajectories is most appropriate. Girls will be clustered within schools for analysis, and analyses may be stratified by rural and urban residence or by type of school where relevant. Theoretically relevant confounders will be included in each analysis.

Exploratory analyses and greater understanding of the ways girls’ menstrual experiences and needs change over time may revise specified hypothesis. Similarly, extension of the cohort to include SRH outcomes will shape the research questions we are able to address. Below, we outline a priori hypotheses. Core hypotheses are provided alongside conceptual diagrams (illustrative directed acyclic graphs).

  • Education. Sustained exposure to poor MH including (a) negative menstrual hygiene experience, (b) shame, (c) menstrual pain that is not able to be reduced and (d) parental restrictions on school attendance during menstruation predicts poorer educational performance and attainment, mediated by poorer school attendance and participation. Less supportive school WASH infrastructure, household WASH infrastructure, parental support for MH, menstrual product affordability, menstrual knowledge and social support including positive support from friends and household members and less teasing will contribute to MH for education. Hypothesised pathways are depicted in figure 2.

  • Social participation. Girls experiencing (a) more negative menstrual hygiene experience, (b) greater shame, (c) menstrual pain that is not reduced, (d) poorer social support and greater parental restriction will experience higher levels of reduced participation in social activities due to menstruation.

  • Mental health. Sustained exposure to poor MH including (a) negative menstrual hygiene experience, (b) shame, (c) menstrual pain that is not reduced, (d) greater personal endorsement of norms of privacy surrounding menstruation and (e) poorer MH literacy will lead to higher levels of depression and anxiety. Hypothesised pathways are depicted in figure 3.

  • Urogenital symptoms. Girls reporting sustained exposure to poorer menstrual and genital hygiene practices including (a) use of improvised menstrual materials, (b) infrequently changing menstrual materials, and (c) inadequate washing and drying (non-use of soap, materials not dried in the sun and not completely dried) will predict greater risk of urogenital symptoms.

  • Healthcare seeking. (a) Lower MH literacy and (b) poorer social support for menstruation at school and at home will predict lower rates of seeking support from a healthcare provider for girls experiencing menstrual pain that impacts daily activities or other concerns about their menstruation.

  • SRH. The relationship between early menarche and SRH outcomes (earlier sexual debut, early pregnancy)23 is partially mediated by (a) distress at menarche, (b) early negative menstrual hygiene experience, (c) early experiences of menstrual shame and (d) poorer puberty knowledge at menarche.

  • SRH. Sustained exposure to poor MH, including (a) negative menstrual hygiene experience, (b) shame, (c) poorer MH literacy and (d) unaddressed menstrual pain will predict lower body appreciation and SRH empowerment. Poorer menstrual knowledge scores will also predict lower SRH empowerment. Hypothesised pathways are depicted in figure 4.

  • SRH. Sustained exposure to poor MH, including (a) negative menstrual hygiene experience, (b) shame, (c) poorer MH literacy and (d) unaddressed menstrual pain will predict an increased risk of poor SRH outcomes including (1) lower contraceptive knowledge and intended use, (2) early marriage and (3) pregnancy, mediated by body image and SRH empowerment. In addition, poorer early menstrual knowledge will predict poorer contraceptive knowledge. Hypothesised pathways are depicted in figure 4.

Figure 2

Hypothesised proximal and distal exposures and their association with education outcomes.

Figure 3

Hypothesised proximal and distal exposures and their association with mental health.

Figure 4

Hypothesised proximal and distal exposures and their association with sexual and reproductive health outcomes.

Missing data

We will endeavour to minimise loss to follow-up through the collection of multiple sets of contact information to reach participants in successive data collection waves. Contact details will be updated every 6 months, alongside sharing findings back with participants, to support cohort retention. While participants will be recruited through schools, girls who move or drop out of school will be followed up at home using contact information for girls (where available) and for their guardians. Educational outcomes that can be collected from schools will be collected for girls enrolled in schooling and will be used to confirm school dropout, alongside self-report.

Scope for exploratory research

Alongside core a priori hypothesised pathways of impact, the AMEHC datasets provide extensive opportunities for exploratory quantitative research. The collection of school-level indicators, alongside girls’ self-report data and baseline survey among guardians, provides a rich data source.

Patient and public involvement

Adolescent girls and the public were not involved in the initial design of the study. We sought feedback on the study design, methods and measures from a National Advisory Committee comprised of stakeholders from government, UN agencies, community service and NGOs and researchers. Further, prior to the cohort, we undertook community workshops in Khulna City Corporation and Dumuria to meet with principals, community leaders and students to provide information about the study and seek feedback on community engagement, tracking mechanisms, and the collection of school and education data. The National Advisory Committee will continue to provide feedback on the study methods and interpretation of findings for the life of the cohort. In 2023, following baseline recruitment, we will establish a Girls Advisory Group comprised of up to 24 girls volunteering from the cohort. This group will provide guidance on study implementation, dissemination activities and co-develop recommendations and health promotion strategies for policymakers, schools and NGOs. Yearly reflection workshops in Khulna City Coporation and Dumuria with will provide opportunities for further feedback from communities on study methods, interpretation of findings and recommendations for policy and practice.

Ethics and dissemination

Ethics

The study received ethical approval from the Alfred Hospital Ethics Committee (369/22) and the Institutional Review Board of BRAC James P Grant School of Public Health, BRAC University (IRB-06 July 22-024). National and district-level education offices provided endorsement for the study. Schools provided administrative approval to participate. At baseline, guardian consent will be sought in person or over the phone if in-person meeting is not possible. Guardians will be provided with a 4-minute video explaining the cohort study and data collection. This is supplemented by a written consent form and explanation from enumerators. Following guardian consent, adolescent participant assent will be sought through a similar procedure using the study informational video, assent form and interaction with a trained female enumerator. Guardian consent and participant assent will be sought for each wave of the cohort. Participant contact information is collected through a separate survey system (REDCap) and stored securely and separately from study survey data (collected through the KoboCollect app). A numerical participant ID is the only link between participant-identifying information and study responses. Participants can withdraw from the cohort at any time.

Dissemination

Findings from the AMEHC Study will be reported in peer-reviewed publications and presented at appropriate research conferences. Further, findings will be summarised for broad audiences and presented through online webinars to a wide variety of stakeholders. Each year, participating schools and girls will receive an age-appropriate summary of survey findings. In addition, we will hold reflection workshops at the upazila level with participation from invited schools, communities, parents and girls to share more detailed study findings, and seek community input in interpretation and implications for government policy and NGO practice. Policy briefings will be developed and shared with stakeholders and government in Bangladesh.

Materials developed for the AMEHC Study, including surveys and school checklists, will be publicly available through a study Open Science Framework page and collaborating institution web pages. Publications will be published open access to ensure accessibility.

Discussion

To the best of our knowledge, the AMEHC Study is the first dedicated prospective cohort to investigate changes in a broad set of menstrual needs and experience throughout adolescence and the first study that is positioned to estimate the impact of sustained exposure to poor MH on health and education outcomes. The study addresses priority gaps in the MH evidence base which have tempered investment and presented barriers to intervention development and evaluation. By attending to exposures across the breadth of MH challenges, we seek to offer comprehensive estimates of the impacts of poor MH and identify pathways through which these can be addressed.

The health and education outcomes measured in our study have established methods for translation into economic outcomes: the DASS can be mapped to quality-adjusted life years,61 years of school attainment can be modelled in lifetime earnings and societal economic costs,62 as can SRH outcomes including child marriage63 and unintended pregnancies.64 Thus, the estimates generated by the AMEHC Study have further utility in comparing the impacts of MH against other health priorities and populating an investment case.

Strengths of our study methods include the a priori specification of hypotheses and comprehensive data collection including information collected from guardians about the household and their views, school-level data and girls’ own self-reported experiences. Investment in preliminary research to pilot and validate measures, refine translations and enumerator training prior to commencing the cohort strengthens the quality of self-reported data. Six-day in-person training with enumerators and the repeated employment of consistent enumerators throughout the cohort will further support positive participant experiences of the study and data quality. Our hypotheses and survey measures are informed by rigorous systematic review and synthesis of past qualitative evidence,8 audits of measurement in MH research,65 66 together with qualitative interviews in the study context and primary measure development activities including cognitive interviews and pilot survey.40 47 67 Collaboration between research partners and NGO/practitioner partners supports the translation of findings with relevance for practice and stronger engagement with national, regional and local stakeholders.

Challenges for implementation will include cohort mobility and tracking. Few adolescent girls have their own phone or contact information, and thus contact information for parents and guardians will be required to reach girls for follow-up. Girls who move schools, residence or get married may be difficult to find in subsequent cohort waves. Six-month contact with the cohort seeks to maintain updated contact details. Participatory upazila reflection workshops and sharing adolescent-friendly dissemination materials with participants will also support positive engagement with the cohort participants and their communities.

Ethics statements

Patient consent for publication

Acknowledgments

We thank the many field research assistants who administered the AMEHC surveys, without whom the study would not have been possible. Our deepest thanks to the participating schools, families and girls. We are grateful for the input from our National Advisory Committee members, and the Directorate of Secondary and Higher Education.

References

Supplementary materials

Footnotes

  • MTH, SFR and PA are joint senior authors.

  • X @julie_hennegan, @JabbarTopu, @ErinHunterPH

  • Contributors Conceptualisation—JH and PA. Methodology—JH, PA, MTH, M-urR, EH, AJ, TJ, AK, EK, HAW, NS, GJM-T, CLH and SG. Investigation—MTH, AJ, TJ, EH, LD, AK, SA, AZ, AH and JH. Writing (original draft)—JH. Writing (review and editing)—PA, MTH, SFR, AJ, TJ, EK, EH, LD, AK, M-urR, NS, SA, AH, AZ, HAW, KKT, TMW, CLH, SG and MH. Visualisation—JH, LD and AH. Supervision—JH, MTH, PA and SFR. Project administration—JH, MTH, AJ, AK and TJ. Funding acquisition—JH, PA, EK, EH, NS, HAW, KKT, TMW, CLH and SG.

  • Funding The AMEHC Study is supported by funding from the National Health and Medical Research Council of Australia (NHMRC) (GNT2004222 and GNT2008600) and the Reckitt Global Hygiene Institute (RGHI). The authors gratefully acknowledge the contribution to this work of the Victorian Operational Infrastructure Support Program received by the Burnet Institute. JH is supported by an NHMRC Investigator Grant GNT2008600 and RGHI Fellowship. PA is supported by an NHMRC Investigator Grant GNT2008574.

  • Disclaimer The views expressed are those of the authors and not necessarily those of RGHI.

  • 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; externally 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.