Let’s go girls! The RITU RCT-study on breaking down menstrual barriers in Bangladesh. Cluster randomized controlled trial study on the impact and cost-effectiveness of improving Menstrual Health Management on wellbeing and academic achievement of girls in rural Bangladesh

Last registered on December 06, 2017

Pre-Trial

General Information

Title

RCT ID

AEARCTR-0002164

Initial registration date

December 04, 2017

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

December 06, 2017, 10:56 AM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country

Bangladesh

Region

South Asia

Primary Investigator

Name

Lidwien Sol

Affiliation

Maastricht University

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Vera Scholmerich

PI Affiliation

Erasmus University College

Contact Investigator

PI Name

Kellie Liket

PI Affiliation

Erasmus University Rotterdam

Contact Investigator

Additional Trial Information

Status

On going

Start date

2017-01-15

End date

2021-12-01

Keywords

Education, Health, Welfare

Additional Keywords

Gender, Menstrual Health, cost-effectiveness

JEL code(s)

Secondary IDs

IRB 2016-09

Abstract

Background: Around the time of menarche, the gap in academic achievement and psychosocial health between girls and boys in low and middle income countries substantially widens to the detriment of girls. This seems to be partially caused by girls’ poor ability to practice Menstrual Health Management (MHM). Poor MHM is also a challenge in Bangladesh, where 40% of girls reportedly miss three days of school during their menstrual period. We conduct a cluster randomized controlled trial of the impact of a complex intervention facilitating MHM in Bangladesh: the Ritu RCT study.

Methods: 150 Schools were randomized into three groups: i) receiving the basic school program ; ii) the basic school program and parent training; iii) a control group. The primary beneficiaries are schoolgirls. The program will last for 3 years, and the primary outcomes are academic attainment and psychosocial outcomes. Secondary outcomes include MHM knowledge, attitudes and practices, mobility, child marriage and teenage pregnancy. We will analyze both the short-term and long-term effects of both treatment arms on our primary and secondary outcomes. In addition, we will conduct cost-effectiveness evaluations of both treatment arms and a process evaluation of the entire intervention.

Discussion: Even though MHM programs are popular, there is very limited evidence on such programs. It is troubling that it is unclear what works, why, and at what cost. We aim to reduce these knowledge gaps by providing rigorous evidence. Different to most evaluations of public health programs, we evaluate a complex intervention and will include cost-effectiveness analysis for both treatment arms.

External Link(s)

Registration Citation

Citation

Liket, Kellie, Vera Scholmerich and Lidwien Sol. 2017. "Let’s go girls! The RITU RCT-study on breaking down menstrual barriers in Bangladesh. Cluster randomized controlled trial study on the impact and cost-effectiveness of improving Menstrual Health Management on wellbeing and academic achievement of girls in rural Bangladesh." AEA RCT Registry. December 06. https://doi.org/10.1257/rct.2164-1.0

Former Citation

Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Experimental Details

Interventions

Intervention(s)

We randomized schools into different groups: both groups receive the school level intervention, but treatment arm II also receives a household intervention.
The school-level intervention consists of three components:
1) Program launch at schools with a campaign to familiarize students, teachers and school management with the Ritu program.
2) MHM-friendly toilet facilities are realized at schools.
3) An MHM module is added to the school curriculum and teachers and headmasters receive training on how to integrate the module in their lesson plans.

Treatment arm II: school- and household-level intervention
Next to the school-level treatment, the parents of girls in this treatment receive a household level intervention consisting of two components:
1) Parent sessions are provided to fathers and mothers separately in their village. These parents session cover topics on MHM knowledge, attitudes, and practices and ways to support their daughters.
2) A module with visual information on basic MHM is given to girls to take home to increase household members’ knowledge, support and dialogue.

Intervention Start Date

2017-04-01

Intervention End Date

2019-12-31

Primary Outcomes

Primary Outcomes (end points)

School attendance, school performance, empowerment, happiness, physical outcomes, Menstrual Health Management

Primary Outcomes (explanation)

School attendance is constructed both by using administrative records as well as our own data collection, and survey questions.
School performance is constructed by administrative records, and survey questions.
Empowerment is constructed on survey questions (including the empowerment tool from OPHI - Oxford Poverty and Human Development Initiative.
Happiness is based on a survey question
Physical outcomes are constructed with survey data on eg. rash incidence, depressive feelings, irritations
MHM is contructed by using survey data on MHM knowledge, attitudes and practices

Secondary Outcomes

Secondary Outcomes (end points)

Demographics,
Household poverty rate (Poverty Probability Index)
Process evaluation measures

Secondary Outcomes (explanation)

Experimental Design

We conduct a cluster randomized control trial with 150 schools that were randomized into three groups: i) receiving the basic school program; ii) the basic school program and household program iii) a control group. All pupils in grades 6 to 8 (age 11-15) are automatically enrolled in the program and we collect school-attainment and survey data on a subset of girls. We conducted a baseline survey in January 2017, and continuously collect data on school attendance in two different ways: by primary data collection and administrative school records.

Experimental Design Details

We conduct a cluster randomized control trial with 149 schools that were randomized into three groups: i) receiving the basic school program; ii) the basic school program and household program iii) a control group. All pupils in grades 6 to 8 (age 11-15) are automatically enrolled in the program and we collect school-attainment and survey data on a subset of girls. The Ritu program started its implementation in April 2017 and is currently rolling out to all the 78 schools in treatment arm 1 and 2. We conducted a baseline survey in January 2017, and continuously collect data on school attendance in two different ways: by primary data collection and administrative school records.
The program is rolled out in Netrokona district in Bangladesh, all eligible co-education secondary schools in the district are part of the study.

Randomization Method

The randomization was done in office, on a computer with the STATA14 program in January 2017.
We randomized schools (i.e. the clusters) into one of the two treatment arms or the control group, stratified by 3 variables:
- Administrative attendance records, (low / high);
- Region (7 Upazillas); there are small difference in Upazilla-wise governance structures and size of schools.
- Quality-quantity of toilets (low / high); data collected by our implementing partner.
The first covariate reflects a primary outcome, and the latter two covariates are potentially important predictors of our primary outcomes.

Randomization Unit

Schools

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

149 schools

Sample size: planned number of observations

30,021 female students

Sample size (or number of clusters) by treatment arms

39 schools treatment arm 1, 39 schools treatment arm 2, 71 control schools

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

As the number of clusters (i.e. schools) we could use was limited to a maximum of 150 at the onset of the study, we performed calculations to estimate the number of schoolgirls we would need to sample per school for our survey. We did this using a mapping study conducted in 8 schools in the Netrokona district. We performed separate a sample size calculations for our primary outcomes using the open-source software program Optimal Design (3.01). We specified the following parameters: repeated measures with 3 data collection rounds, taking the 3-level nature of the data into account (Level 1: schoolgirls; Level 2: schoolgirls clustered into time (baseline, midline, endline); Level 3: schoolgirls clustered into time are clustered into schools), an alpha of 0.05, and ran different scenarios with the following options: an expected effect size of 0.2 or 0.3 and an ICC of 0.1 or 0.3. These calculations indicated that we would need to sample of at least 25 schoolgirls per school to be powered at 0.8 for all primary outcome measures included in the survey.

Supporting Documents and Materials

IRB