We used the SPIRIT reporting guidelines in writing this protocol paper (19).
Study setting {9}
The trial will be conducted in secondary schools in Wakiso and Kalungu districts in the Central region of Uganda. Wakiso district has a total area of 2,704 square kilometres and includes both urban and rural areas. It partly encircles Kampala (Uganda’s capital city). According to the census in 2014, the population was estimated as 2 million, of whom 339,137 (17.3%) were aged 10-17 years and 48.9% of the 13-18 year olds are estimated to attend secondary school (20). Kalungu district is largely rural with an estimated population of 183,232 in 2014, of whom an estimated 40,960 (23%) were aged 10-17 years and 36% of 13-18 year olds attending secondary school (21).
Eligibility criteria {10}
Schools
Inclusion criteria for schools: Mixed-sex secondary schools with S1-S4 classes; day or mixed day/boarding schools; at least minimal WASH facilities (including an improved water source and sex-specific sanitation facilities that are functional, usable, and accessible to female students at the time of the rapid assessment eligibility survey (22) estimated enrolment of approximately 50-125 female S1 students as of January 2020 (before COVID19 lockdown) in Wakiso, and 40-125 female students in S1 in Kalungu, based on the 2019 report on the Master List of Education Institutions in Uganda (23).
Exclusion criteria for schools: Schools that are currently participating in a menstrual health related programme; boarding schools with no day students; single-sex schools and schools exclusively for students with disabilities.
Trial participants
Inclusion criteria for trial participants: i) All female students in S2 in 2022 who are present at the time of the baseline survey, whose parent/guardian provided consent (if aged <18 years), and who have assented/consented to the trial and intervention procedures; ii) A simple random sample of 15 male students (in S2 in 2022) per school, present at time of the baseline survey who have consented to the trial procedures. Further, all female students present at the time of the endline survey in 2023, with consent/assent for the research will be eligible for the outcome assessment, as will the male students with baseline data.
Inclusion criteria for the nested sub-study: A simple random sample of 25 female students in each school will be randomly selected from those enrolled in the trial, to participate in a daily diary sub-study.
Who will take informed consent? {26a}
Informed consent procedures
Trained research team members will seek i) written informed school-level consent from head teachers or directors to cover school-level research and intervention activities; ii) written informed parent/guardian consent for female students obtained in person where possible, through information meetings held at each school. Parents/guardians who were unable to attend information meetings at school were reached through home follow-up visits or through phone calls to seek documented verbal informed consent. The consent forms include procedures for the nested sub-study (self-collection of two vaginal swabs, one of which will be stored and will be shipped overseas), and for all anonymised data collected to be shared with other researchers. A separate consent form was administered for the menstrual cup provision, so parents/girls could opt to participate in all trial activities except for receiving a cup. The ethics committees agreed a waiver for parent/guardian consent for boys, as their participation in the (knowledge-focused) boys’ survey was deemed to carry minimal risk.
For student assent, information videos explaining trial procedures will be available for girls whose parent/guardian consent to the trial either including or excluding the menstrual cup respectively. Male participants will be shown a video describing relevant procedures for boys. All three videos will be available in English and Luganda. Students will be asked by a trained research assistant to watch the relevant video on a tablet with headphones, they will also answer a few questions at the end of the video to assess their understanding of the trial and if they agree, they will provide secure electronic (e-) assent by signing on the tablet. Participants are given a paper copy of the assent form to retain.
Participants selected for qualitative interviews will be asked for parent/guardian consent and student assent prior to taking part in interviews (except for female students who already consented to the main trial that includes consent for qualitative interviews). Additional parent/guardian consent (for girls) and student assent (for all students) will occur prior to the endline survey for students who were not present at baseline.
Interventions
Explanation for the choice of comparators {6b}
The choice of comparator was chosen to align with Government MH guidelines, and is optimized usual care. Optimized usual care will include i) a copy of the Government Circular on Provision of MHM for primary and secondary schools (Circular No. 1/2015).1/2015) (13); ii) the 2018 National Sexuality Education Framework for each school (24), and iii) copies of the Government Menstruation Management Reader to distribute to each girl and boy in S2. The Reader was written for girls and boys in primary schools to help them understand what menstruation is, why it happens, how to manage it and who to talk to about it. Schools randomised to the comparator arm will be able to receive the MENISCUS intervention after the endline survey.
Intervention description {11a}
The schools randomised to the intervention arm will receive the optimised usual care plus the MENISCUS intervention which includes puberty education, a drama skit, provision of a menstrual health kit and training, pain management, WASH improvement and creation of a Menstrual Health Action Group. Details of the MENISCUS intervention are given in Table 1, reported according to the TIDieR Framework.
The MENISCUS intervention is grounded in social cognitive theory which suggests that individuals learn by observing others (25). The theory suggests that the following factors reinforce behaviour change: increased social support for and instilling expectations of behaviour change; increased self-efficacy; and using observational learning. During our formative work, we developed a logic model of change with stakeholders. Aligning this logic model with the core constructs of social cognitive theory, we developed a theoretical framework for the intervention which asserts that the intervention (i) increases girls’ self-efficacy to manage their menstruation (e.g. through provision of an MH kit and pain management options), (ii) positively reinforces girls’ learning (and also that of boys, teachers and parents) to create a more supportive MH environment (e.g. through drama) and (iii) reinforces behaviour change through positive reinforcement and expectations (e.g. through improving WASH facilities and reduced teasing) (18).
Outcomes {12}
The primary outcomes are educational attainment and mental health problems at endline among girls included in the endline survey (Table 2). Secondary outcomes are shown in Table 2.
Educational attainment will be assessed through bespoke pre-tested assessments administered and marked by the Uganda National Examination Examinations Board (UNEB). The baseline assessment will test Mathematics and Biology material taught to S1 students pre-intervention (2020-2021), and the endline assessment will assess recently-taught Mathematics, Biology and English material. UNEB assessors will be masked to trial arm. The outcome will be the composite examination score at endline. The rationale for the subject choice is that the Biology curriculum taught in S2 includes puberty knowledge in the reproduction topic and is directly related to our intervention, and that English and Mathematics are core subjects recognised by policymakers as influential for scale-up.
Mental health problems will be assessed for all girls at baseline and endline using the Total Difficulties Scale of the Strengths and Difficulties Questionnaire (SDQ-25). This includes emotional symptoms, attention and peer-relationship problems. The outcome will be total SDQ-25 score, analysed as the examination data above. The SDQ-25 has been validated among adolescents in sub-Saharan Africa including in Uganda (26).
Participant timeline {13}
Different items will be measured according to the point in time when the data are collected. The enrollment into the study is three months and assessments for the primary and secondary outcomes will take place at different time points with majority of the assessments at both baseline and endline.
Sample size {14}
The sample size of 60 schools is based on an estimated harmonic mean of 60 female students per school at endline. This sample size provides 84% power to detect a standardised mean difference of 0.2 for continuous outcomes, assuming an intra-cluster correlation (ρ) of 0.05 and Type 1 error of 0.05. The effect size of 0.2 is that observed for the primary outcome of SDQ score for girls in the pilot (17), and is considered moderate for educational outcomes and hence of policy relevance. This sample size provides 90% power to detect 12% vs 20% participants answering all knowledge questions correctly in the control vs intervention arms for girls (assuming ρ=0.05), and 85% power to detect 4% vs 12% for boys (as in the pilot), assuming 10 boys per school are seen at endline.
The sub-sample of 1500 girls selected to provide daily diary data on school/class attendance and their menstrual cycle over 12 weeks (60 school days) at endline provides >98% power to detect a 20% reduction in period-related absence at endline based on pilot data (17), and ~89% power to detect this difference by district or other equally-sized sub-groups. All girls at endline will be asked about urogenital symptoms and genital hygiene, and to provide self-collected vaginal swabs that will be tested for bacterial vaginosis (BV) and vaginal yeast.
We have not adjusted the Type 1 error because these two outcomes are independent, and an improvement in one without the other will be informative. From our pilot, we expect >95% of girls to be menstruating at endline, and this is accounted for in the power calculation where relevant.
Recruitment {15}
Schools were selected and assessed for eligibility in two stages. First, an initial desk review was conducted using the 2019 Master List of Education Institutions. Of 631 secondary schools listed in the two Districts (583 in Wakiso and 48 in Kalungu), 160 schools (138 in Wakiso, 22 in Kalungu) were deemed potentially eligible based on the listed information (i.e. mixed-sex, not exclusively boarding, and 40-125 (Kalungu) or 50-125 (Wakiso) female students registered in S1). The remaining schools were stratified by district and Government/private ownership and randomly ordered.
The second stage was a rapid assessment which consisted of stakeholder workshop, phone calls, school visits and interviews to confirm eligibility and assess the schools’ physical and social environments for MH, in preparation for the trial. Of 110 head-teachers of schools contacted, 84 were deemed potentially eligible for the trial from an initial screening phone call and 75 (89% of those potentially eligible) were invited to participate in the rapid assessment. Following the rapid assessment, 67 schools were confirmed eligible and 58 (87%) agreed to participate. Head teachers of these schools and district officials were invited to a second stakeholder workshop in each District in October 2021, where results of the rapid assessment were discussed and schools were asked to sign a Memorandum of Understanding (MOU) and provide a list of female students due to start S2 in 2022 if they agreed to participate in the trial. We assessed a further two schools (to replace two that declined to participate) from the original list for eligibility in February 2022 to achieve the final sample size of 60 trial schools (16 in Kalungu District and 44 in Wakiso District).
Assignment of interventions
Sequence generation {16a}
The 60 trial schools will be allocated using 1:1 restricted randomisation within each district (strata). The allocations will be restricted to minimise imbalance with respect to potential confounding factors (e.g. mean baseline examination score, past school examination scores, mean baseline SDQ score, school type (Government vs. private), mean score on the Menstrual Practice Needs Scale, and number of S2 female students). We will minimise biased recruitment by conducting the randomisation after the baseline survey (27).
Concealment mechanism {16b}
The primary outcome of the educational assessment will be masked as it will be independently administered by UNEB with no knowledge of arm allocation. We will minimise bias in other outcome assessments by having separate teams for implementation (WoMena Uganda) and outcome assessment (MRC/UVRI and LSHTM Uganda Research Unit). Masking of participants, school staff and implementers is not possible as it will be clear whether the intervention is being implemented. Laboratory technicians testing self-collected swabs for bacterial vaginosis, vaginal yeast and UTI will be masked. All outcome data will be anonymised, analysed and interpreted masked to trial arm. The clinical officer will not be masked to arm as they will be training staff in the intervention schools, and will be responsible for assessing causality of adverse events.
Implementation {16c}
Eligible allocations meeting restriction criteria will be sent to the data manager by the trial statistician. One allocation will be randomly selected at randomisation ceremonies held in each district following completion of the baseline survey, in the presence of representatives from schools, the District Education Officer (DEO), the MoES and MoH. We will use a random allocation method to select which schools are in the intervention arm and which are in the control arm using a randomisation method.
Plans to promote participant retention and complete follow-up {18b}
The trial is powered to compare outcomes between participants in the two randomised groups at endline, adjusting for school-level baseline measures of outcome variables. There may be differential loss-to-follow-up by arm, if, for example, the intervention reduces school dropout. We will collect data on school dropout in girls and adjust for the proportion dropout in endline analyses.
Data collection and management {19}
Baseline and endline survey data will be self-collected by participants on tablets using ODK Collect software, and uploaded to the MRC/UVRI and LSHTM Uganda Research Unit secure ODK Central server. Other quantitative data (daily diaries, laboratory forms and analgesic vouchers) will be collected on paper Case Report Forms (CRFs) and entered into a REDCap database. The data manager will regularly monitor data quality and any data queries will be resolved in collaboration with the relevant field team member or data entrant. In-depth interviews and focus group discussions will be audio-recorded and transcribed verbatim, including relevant non-verbal communications, and translated into English by the study interviewers for those conducted in Luganda. Cost data will be inputted into an Excel-based costing tool.
In the endline survey, female participants will be asked a set of additional questions on current genital symptoms. After completing the survey, trained study staff will explain and demonstrate the procedure for obtaining the vaginal swab, using an animated video that will be co-developed with students. After the student has watched the video and asked any questions, the study staff will pass the student the 2 swabs for self-collection in a private setting. The swabs will be returned to the staff member. Participants reporting urogenital symptoms will be referred to an appropriate clinic for syndromic management and treatment as per Ugandan Clinical Guidelines. Those with symptoms of a UTI (e.g. urinary burning and frequency) will have their urine tested by Multistix 8 dipstick to look for signs (e.g. nitrite and leucocyte esterase positivity) to aid diagnosis of a UTI as per Ugandan Clinical Guidelines. The test result will be sent with the participant to the local clinic for management.
The process evaluation will use quantitative data from the endline survey supplemented with data from log-books, structured observations, focus group discussions and in-depth interviews (Tables 3 and 4). This will enable us to understand how the intervention was implemented and received and the social, structural, and logistical factors that impede or facilitate this, as well as how staff and students were able to engage with it. We will also explore implementation processes and mechanisms of impact, and how these vary across the schools, using qualitative data. Data will be collected from all intervention schools via: phone interviews with one senior staff member at baseline and endline; semi-structured interviews with one member of the school’s MH Action Group member purposively sampled to ensure overall variation across schools in gender and type of member (e.g. staff, student, other stakeholder) at mid-line and endline.
The economic evaluation will evaluate the costs of setting up and running the intervention package, the unit cost per female student reached, and the incremental cost-effectiveness of the intervention per unit increase in selected policy-relevant outcomes, relative to optimized usual care. The cost-effectiveness measures will be compared to similar school-based interventions in the region to inform scalability and financial sustainability. The policy analysis will assess the policy environment and financial sustainability of our intervention in line with Government policies and frameworks for MH and sexual and reproductive health.
Confidentiality {27}
Identifiable data collected will be stored securely and their confidentiality protected in accordance with the Data Protection Act 1998. The risk of disclosure will be minimised by training staff on the need for confidentiality (including training on Research Ethics), secure storage of documents and use of linked data with a unique participant ID instead of names or other personal identifiers. All information regarding the participants will remain confidential to the extent allowed by law. Screening forms and case report forms will be kept in a secured location with access limited to authorised study staff. Consent and assent forms will be stored separately in a locked cabinet as will all other forms, logbooks and appointment books that link participant ID numbers to other identifying information.
Plans for collection, laboratory evaluation and storage of biological specimens {33}
The self-collected vaginal swabs will be transported to the MRC/UVRI laboratory in Entebbe for processing and storage. Upon receipt, one swab will be Gram stained and Nugent scored (28).The second swab will be stored immediately in a cool box before being transferred to a -80C freezer within 24 hours for storage and potential future testing related to vaginal microbiota.
Statistical methods
Statistical methods for primary and secondary outcomes {20a}
We will follow CONSORT guidelines for analysis of cluster-randomised trials (29). A Statistical Analysis Plan will be finalized prior to data unblinding and analysis. A participant flow diagram will be completed, following CONSORT guidelines for CRTs. Descriptive statistics for school-level and individual-level baseline characteristics will be shown by arm and will include mean and SD (or median and range if non-Normal) (for continuous variables) and counts and proportions (for categorical variables). Factors showing substantial baseline imbalance by arm (agreed in advance of unmasking by the Co-Investigators) will be noted for inclusion in effectiveness analysis. The primary and secondary outcome analyses (Table 2) will be carried out for all schools as randomised (intention-to-treat).. Intervention effects will be estimated using individual-level regression analysis, adjusted for clustering by school and for baseline measures of outcome variables, restriction variables and factors imbalanced by arm at baseline. Results will be presented as standardised mean difference (continuous) or odds/prevalence ratio (binary) with a 95% confidence interval. For the analyses of daily diary data, there are repeated measures per girl, and analyses will additionally adjust for within-girl clustering using random-effects. Sensitivity analyses will be conducted, including randomisation inference.
Methods for additional analyses (e.g. subgroup analyses) {20b}
Effect modification will be assessed for a-priori defined subgroups (e.g. District; Government vs private school; age group; baseline examination score; baseline WASH facilities). The cost-effectiveness analysis will include cost–utility analysis comparing costs and utilities of the two groups with costs considered from a provider perspective and health outcomes measured by QALYs based on the Child Health Utility–9D (CHU9D); cost-effectiveness analyses of the intervention effect on SDQ score and cost-wellbeing analysis, with costs considered from a provider perspective and outcomes measured by improvement in happiness and life satisfaction scores. To assess bias due to attrition, we will compare reasons for students leaving school, and baseline characteristics associated with dropout, between arms.
The full protocol will be made available on the study website (https://www.lshtm.ac.uk/research/centres-projects-groups/meniscus), and participant-level data and statistical code will be available through the LSHTM Data Repository (https://datacompass.lshtm.ac.uk/)
Oversight and monitoring
Composition of the coordinating centre and trial steering committee {21a}
The trial sponsor, the London School of Hygiene and Tropical Medicine (LSHTM), will organise monitors to review the source documents as needed, to determine whether the data reported are complete and accurate. The Trial Steering Committee (TSC) is chaired by an independent scientist, and consists of 6 further independent members (including Government representatives), 3 observers (including a funding representative) and 7 non-independent members. The Trial Management Group (TMG), chaired by the PI and attended by Co-Investigators, meet every 1-2 months to oversee trial progress and report to the TSC. The Independent Data Monitoring and Ethics Committee (IDMEC) consists of 3 independent members who meet 1-2 times per year to monitor the recruitment, overall conduct of the trial, and to review data to ensure the safety, rights and well-being of trial participants. The trial steering group and the independent ethics committee will meet regularly for review throughout the trial.
Adverse event reporting and harms {22}
Prior to randomisation, the trial Clinical Officer will map and visit Health Centres III, IV, district hospitals and referral hospitals in each District and meet with District Health Officers to confirm referral procedures. Health Centres III are sub-county level clinics led by a senior clinical officer, with a general outpatient clinic, maternity ward and functioning laboratory. Health Centres IV are county or parliamentary constituency level mini-hospitals, led by a senior medical officer, with in-patient wards for men, women, and children respectively, and a theatre for emergency operations. Hospitals provide comprehensive medical services, receive referrals from lower health facilities, are relatively equipped with advanced diagnostics and have advanced capabilities for resuscitation and life support. As part of the intervention, girls will be provided with the toll-free number for Public Health Ambassadors Uganda (PHAU) toll free number 0800-300-600 and the Marie Stopes Uganda (MSU) contact centre (phone: 0800 220 333; WhatsApp: 754001 503), which operates 7 days per week and is managed by a team of 12 qualified counsellors and 2 nurses to provide confidential advice on sexual and reproductive health and referral to services in Luganda and English.
School nurses or designated person in both arms will be trained to monitor and report all suspected serious adverse events (SAEs) (i.e. death, life-threatening event, hospitalization, persistent or significant disability or incapacity). These will be assessed by the trial Clinical Officer and noted in the SAE reporting form, including determination of a possible relationship with the intervention (analgesic use or the menstrual cup). Participants with SAEs will be rapidly referred to previously mapped Health Centre IVs by the MENISCUS Clinical Officer, or the school or Health Facility III as appropriate. Transportation to the hospital will be provided by the study if needed. The trial Clinical Officer will be responsible for initially assigning causality and SAEs will be reported to the trial manager and PI within 24 hours, using an SAE form. SAEs that are at least possibly related to the study interventions will be reported to the IDMEC within 24 hours of the PI becoming aware of it. If awaiting further details, a follow-up SAE report will be submitted promptly upon receipt of any outstanding information.
Non-serious adverse events will be reported in the intervention arm schools only, following training by the trial Clinical Officer. Each participant experiencing a non-serious adverse event in intervention arm schools will have a record of their adverse events archived confidentially at the MRC MENISCUS office.
Girls who self-report urogenital symptoms during the endline survey will be referred to a local clinic (both private and government) for syndromic management as per Ugandan clinical guidelines (30), together with the Multistix 8 test result for those with UTI symptoms. The trial Clinical Officer will provide training to the school nurses on symptom identification and referral.
Plans for communicating important protocol amendments to relevant parties (e.g. trial participants, ethical committees) {25}
Protocol modifications which may impact on the conduct of the study, potential benefit of the patient or may affect patient safety, will be approved by the ethics committees as a formal amendment prior to implementation. Administrative changes of the protocol are minor corrections and/or clarifications that have no effect on the way the study is to be conducted. These administrative changes will be agreed upon by the TMG and TSC, and will be documented in a memorandum.
Dissemination plans 31a}
Results will be disseminated among schools, and local and national stakeholders to discuss policy implications of the findings. Policy makers and partners will be engaged through the Technical Working Groups (TWGs) for Ministry of Education and Ministry of Health. Communication with the wider academic community will include publication of the trial protocol and results as well as the economic, qualitative, and process evaluations in peer-reviewed academic journals with open access, guided by a dissemination plan.
Plans to give access to the full protocol, participant-level data, and statistical code {31c}
Access to full protocol and datasets will be made available on reasonable request.