The current study is aimed at evaluating nutrition interventions to be implemented for improving health and nutritional status of female garments workers in 4 selected garment factories in Bangladesh. The interventions will attempt to increase either energy intake and or iron intake by female garments workers and expected to implement the interventions as planned:

1. Provision of a once weekly iron/folate supplement, and nutritionally improved, daily hot lunch including multi-nutrient fortified rice plus a nutrition and health behavior change communications (BCC) program in intervention factory compared with a control factory which will provide regular lunch meal plus a nutrition and health behavior change communications (BCC) program for 10 months

2. Provision of a twice weekly iron/folate supplement plus a nutrition and health BCC program in an intervention factory compared with a control factory with only nutrition and health BCC program for 10 months; on the following outcomes:

   • Prevalence of anemia
   • Nutrition knowledge and practices as per key messages of the BCC program
   • Worker satisfaction with the modified lunch and/or iron/folate supplementation, and the BCC program

Research Design and Methods Impact Survey Design To determine the impact of this program on the main outcomes (i.e., prevalence of anemia and dietary intake practices) we will use a quasi-experimental, controlled, pre-post design for both sets of interventions (Figure 1).

We are using a quasi-experimental design as it is not possible to identify a large pool of factories from which to randomly sample and assign factories to a control or intervention group. To minimize possible selection bias, we will:

i. To the extent possible, factories will be selected based on specified criteria, to minimize major differences in response to the intervention among them (as described in selection of factories section).

ii. Select individual women for pre-post studies based on matching or equivalence on key characteristics (eg, age); iii. Use difference-in-difference analysis to test impacts.

Intervention Description Two types of interventions will be tested against controls in a total of 4 factories. Among factories that already provide a hot lunch to workers, one intervention and one control factory will be selected, and among factories that do not provide hot lunches, one intervention and one control factory will be selected.

Intervention with hot lunch:

In a selected garment factory that already provides a hot lunch meal to workers, the intervention will consist of a nutritionally enhanced lunch meal program (increased dietary diversity plus inclusion of multi-nutrient fortified rice), provision of a weekly iron/folate supplement to female workers, and a BCC program conducted in the factory setting focused on improving health and nutrition for women and children, with a focus on anemia prevention. Change Associates will measure the nutritionally improved or enhanced diet and individual consumption. The BCC program builds on an existing program called 'HERhealth' that is currently being implemented by BSR and Change Associates in garment factories in Bangladesh. The HERhealth program will be enhanced to include modules on nutrition (balanced diets/dietary diversity, iron/folate rich foods), infant and young child nutrition (IYCN), and an overview module on anemia prevention, hand-washing with soap and dietary diversity at the very beginning of the intervention. Existing factory-run child day cares will also be enhanced by including additional training for the day care workers on Early Childhood Development methods and health and hygiene practices, and the improvement of facilities with regard to hygiene and sanitation. One control factory will also be included, where only the standard HERhealth modules will be implemented. After the completion of the endline impact survey, the additional BCC modules on nutrition and IYCF, and technical advice on improving the lunch program and day care centers, will be delivered as appropriate and desired.

Intervention without lunch program:

In a selected garment factory that does not have a lunch program, the intervention will consist of the twice weekly provision of iron/folate supplements to female workers, and the enhanced HERhealth program, as described above, including the same enhancements to the existing child day care program. One control factory will be selected where only the standard HERhealth modules will be implemented. As noted above, the additional intervention components of the enhanced HERhealth program will be implemented in the control factory after completing the endline impact survey, as appropriate and desired.

Dietary components of the intervention (lunch and supplements) Lunch meal Based on information compiled from menus from 7 factories currently serving hot lunches to factory workers the average frequency and estimated serving size of different food groups is summarized in Table 1. Rice and dal are served in large bowls to each table and workers can consume as much of these as they choose, and the serving size shown is an estimate. Other menu items are served individually in fixed portion sizes.

In addition to the dietary modifications, food service for the garment factories will procure multi-nutrient fortified , vitamin A fortified vegetable oil, and iodized salt for use in preparing the lunch meals, if they are not already doing so.

Behavior Change Communications Component This uses a peer-education model whereby one module is introduced each month to the peer educators selected from among the factory workers; the master trainers conduct each session in groups of 24-30 peer educators. For each module, there are 3 hour regular training sessions and a one hour refresher training in the subsequent month, i.e. a total of 4 hours over 2 months for every module. Peer educators are selected from among the workers based on leadership qualities.

For the outreach to factory workers, the factory management then schedule one weekly 15-minute session for the peer educators with approximately 15 of their respective colleagues. Additionally, factories will have visibility and awareness tools on the project board and relevant areas (i.e. hand washing chart in bathroom and wash areas) for continuous refreshers.

The Enhanced HERhealth Curriculum will consist of the following modules;

1. Summary module
2. Health, Hygiene and Personal Health
3. Nutrition and Dietary Diversity
4. Maternal Nutrition and Hygiene (some focus on anemia)
5. Infant and Young Child Feeding (IYCF) + Breastfeeding (BF)
6. Reproductive Health and Family Planning
7. Sexually Transmitted Infections

Selection of factories

Given the large commitment required on the part of factory management to introduce hot meals, the selection of factories and assignment to study groups will not be done through random allocation. Randomization would help to minimize bias in sample selection (that is, to minimize the chance that differences in outcomes, or response to treatment, between the groups occurs due to existing differences between the groups). To minimize such differences, we will use a set of matching criteria to guide the selection, as summarized below:

1. Willingness to participate in the program and comply with research requirements over a 10-12 month period. Specifically, the requirements are:

   1. They would consider providing an improved hot lunch (Treatment)
   2. Willing to fully participate in the HERhealth program
   3. Willing to make data available on absenteeism
   4. Willing to allow data to be collected from selected workers in a baseline/endline survey and on a monthly basis for monitoring purposes throughout the pilot project period

   To the extent possible, for purposes of matching:

2. Factory has approximately - 1500- 6000 workers

3. Factory has similar ratio of women: men (e.g., 60:40 to 70:30)

4. Factory produces similar garment products (e.g., woven vs. knit clothing)

5. Factory has a day care facility

Sample Size Calculation and Outcome

The sample-size for the survey was calculated for anemia with a consideration of 10-15% absolute reduction in the prevalence from the baseline as a result of the impact of interventions compared to the control. Considering a 5% level of significance, 80% power and 20% non-response, the minimum sample-size was 328 female garment workers per factory.

Measurement of key outcomes Anemia: Hemoglobin will be measured in capillary blood samples using portable HemoCue machine at baseline after enrolment and at the end of 10 months to assess anemia in female garment workers. Single finger prick method will be used to assess haemoglobin status in field. The cut-off levels of haemoglobin that are considered to define anemia: <110 g/L for pregnant and <120 g/L for non-pregnant non-lactating or lactating women as the survey areas are at the sea level.

Absenteeism: Collect data from factory Human Resources and management on worker absenteeism. Data should be collected on a continuous basis (i.e., complete data throughout the intervention). Ideally data will be restricted to individual factory workers identified as being female, or to individuals in worker lines that are predominantly female. Ideally, data will be restricted to unscheduled time off, as opposed to vacation time or lay off time for non-permanent staff. Health related absenteeism can be monitored through collection of data from the factory doctor's office. However, the latter will only account for employees that request to leave work during the course of a work day, and does not include those that don't attend work. Health related absenteeism questions can also be applied directly to the sample of women included in the baseline/endline survey.

Worker retention: Human resources will be consulted for ongoing data on the number of female garment worker staff members who leave their employment positions throughout the intervention period. They may also be consulted on the number of female garment workers hired and trained to replace workers who have left.

Sampling procedure:

The required number of survey respondents will be randomly selected from the list of female workers available with the study factories. The place of interview inside the factory and prior appointment for interview will be noted by our field workers beforehand in consultation with the individual respondent and/or her supervisor. For hemoglobin estimation, the required number of female workers will be randomly selected from the list of survey participants.

For key informant interviews, around 2-3 respondents per factory will be tentatively selected resulting in a total of 8-12 key informant interviews. An iterative process will be followed for key informant interviews until saturation of information is reached. In addition to note taking by Research Assistants, audio recording of the interviews will be undertaken.

Data collection tools/measurements Questionnaire/ guideline/ checklist For the survey, a pre-tested semi-structured questionnaire will be used by the research staff to interview of selected female garments workers. Necessary guidelines highlighting areas of interests will be used for key informant interviews after necessary field test. Investigators and a trained Field Research Officer (FRO) will conduct the interviews. A checklist will be used by the Field Research Assistants (FRAs) for observation of the study factories.

Ethical Considerations Prior to implementation of the study, necessary approval of the Institutional Review Board (IRB) of icddr,b will be obtained for the protocol.

Data Analysis Data entry and cleaning Survey data will be double entered into computer using SPSS (version 17). Checks for errors of entered data will be done manually and errors detected will be corrected immediately. In addition, data will be validated by a series of logical and range checks, producing summary statistics and tables.

Data analysis The survey data will be analysed using SPSS (version 17) / STATA software (version 12). The summary statistics would be expressed as means with standard deviations, medians or percentages, with 95% confidence intervals (CI) where appropriate. To assess differences in mean among the factories, ANOVA and for categorical outcomes, chi-square statistical comparisons of proportions with 95% confidence intervals, will be calculated. A difference-of-difference (DID) analysis of quantitative outcome data will be done. To identify significant predictors associated with the main outcome indicators, bivariate and multiple regression analyses will be employed.

Thematic analysis will be followed for qualitative information collected through key informant interview, observation and record review. Following preparation of transcript, responses will be coded according to themes (a priori), sub themes and emergent issues, if any. Triangulation of data derived from different methods will be carried out to validate qualitative information.

Inclusion Criteria:

• Adult Non-pregnant female garment worker between 18-42 years of age from selected factories who are present and give consent for interview

Exclusion Criteria:

• Eligible participants who decline consent
• Eligible participants suffering from any known chronic illnesses

Overall study result will be disseminated without revealing individual participants data