The Effect of a School-based Nutrition Intervention to Promote Healthy Eating Behaviour Among Adolescents in Indonesia - a Cluster Randomized Pilot Study

A program which is called "Nutrition Goes to School" (NGTS) has been initiated by the Southeast Asian Ministers of Education Organization Regional Centre for Food and Nutrition (SEAMEO RECFON) in 2016 to improve the health of the children through implementation of nutrition good practices at schools.The program was started by providing training to school teachers about nutrition and continuously improved until it currently includes multiple components consisting of school readiness survey, healthy school canteen, school garden for nutrition literacy, nutrition education, school hygiene and sanitation promotion and promotion of physical activities at school.

Since its initiation in mid-2016, some schools from primary to secondary level in several provinces in Indonesia have been trained in the implementation of some components of the program. However, since the program was developed gradually and schools' participation is on voluntary basis, no schools participated in the training currently implement the whole components, no standard protocol for the activities was implemented and an impact assessment is yet to been conducted. SEAMEO RECFON is one of the regional organizations under the Ministry of Education in Association of South-east Asian Nations (ASEAN) member countries which effort focused on the human resources development through education, research and community development activities in food and nutrition area. Evidence on the effectiveness of the NGTS as the flagship program of the institution is needed to advocate the implementation of the program nationally and even regionally among the ASEAN member countries.

A cluster randomized controlled pilot trial among secondary school in Jakarta, Indonesia. The school-based nutrition intervention which consisted of multi-components could be regarded as a complex health intervention. Therefore, evaluation of its effectiveness follows the steps suggested by the UK Medical Research Council (MRC) whom devised a four-phase framework for developing and evaluating RCT's of complex interventions. This pilot trial is the 2nd phase of the framework, which will be guided by the result of the program evaluation, and aimed to assess the adherence to the whole NGTS package, determine intra-cluster correlation coefficients and effect estimates. This stage is important to inform sample size calculations for future definitive trials designed to assess the effectiveness and cost-effectiveness of a full-scale intervention as currently high-quality controlled trial on school-based health intervention in Indonesia is still lacking.

Study population is adolescents (grade 10-11; 15-17 years old) in Jakarta from ten schools (five intervention and five control);each school consisted of at least 100 students. To accommodate the Indonesian school system in urban areas, we will intentionally select schools from Jakarta Provinces, which cover 5 different municipalities, North Jakarta, East Jakarta, South Jakarta, Central Jakarta, and West Jakarta. First, recruitment will be based on the discussion with the local education office at the provincial level and ask for a list of schools considering eligibility criteria, such as having a school canteen, implementing the School Health Program (Usaha Kesehatan Sekolah/UKS), and having potential land to be utilized as a school garden. Two schools in each municipality with comparable characteristics will be selected, and then each pair will be randomized to become either the intervention or the control group. Second, selected schools will be invited to participate in the study. The intervention group will receive baseline educational intervention (Nutrition Education Capacity Building for Teachers and Management) and all components of the NGTS program. While the control group will receive only the baseline educational intervention. The baseline educational intervention will be conducted in 5 days through face-to-face delivery training mode.

The sample size for this study was determined using the formula from Hayes R and Bennett S (1999). Determination of the calculation of the main parameter assessed was based on a prior study in Ecuador of a school-based intervention to improve dietary intake outcomes, the proportion of breakfast intake. Randomization was based on the number of clusters with a total of 10 schools, considering the significant level (α) of 0.05, the coefficient of variation between groups (Intra Class Correlation/ICC) of 0.15, the power of study (β) of 80%, the ratio of cases and controls (k = 1), and the anticipated dropout of 10%. Based on the sample calculation, the number of samples per intervention group and control group was 486 people each, which was increased to 500 people. So that the total sample needed is 1,000 people, with each school consisting of 100 students.