Child undernutrition is largely attributed to inadequate nutrition including micronutrient deficiency. Undernutrition is prevalent among indigenous children as compared to the general population. This cluster randomized controlled trial aimed to determine the effect of multiple-micronutrients supplement (MMS) on growth and iron status of Orang Asli (indigenous group in Peninsular Malaysia) young children (6 to 24 months) in Selangor. MMS is a blend of 15 micronutrients in powder form that can be used for home fortification of foods for young children. A total of 98 children recruited in this study with 49 children randomly assigned for each intervention (IG) and control group (CG). At baseline, all children were normal in weight-for-age (WAZ>-2SD), length-for-age (LAZ>-2SD), weight-for-length (WLZ>-2SD) and blood haemoglobin (>11g/dL). IG was supplemented with three sachets of MMS each week i.e. every other day for 12 months and received health and nutrition advice. CG only received health and nutrition advice. Both groups were assessed for body weight, length and dietary intake at baseline, month 3, 6, 9, 12 of intervention, and 3 months post intervention. Blood haemoglobin was assessed at baseline, month-12 of intervention and 3 months post intervention. Compliance to MMS was measured in IG. The primary outcomes were weight, length, WAZ, LAZ, WLZ and haemoglobin, while the secondary outcome was dietary intake. The independence sample t-test and the chi-square test were used to determine the difference in the baseline variables between the groups. ANOVA using General Linear Model (GLM) for repeated measures was performed to determine the difference in the growth measures z scores, haemoglobin, energy, nutrients and food group within and between the groups over the period of the study. Per protocol analysis was performed. This study hypothesized that there were significant differences in the changes (before and after MMS intervention) related to growth [(weight-for-age (WAZ), length-for-age (LAZ) and weight-for-length (WLZ)]; iron status (haemoglobin concentration); and dietary intake (energy, nutrients and food group intakes) between intervention and control groups.

1. Study Objectives This study was undertaken to determine the effect of Multiple micronutrients supplement (MMS) on the nutritional status of Orang Asli (indigenous people in Peninsular Malaysia) children (6 to 24 months) in Selangor. This study assessed growth (weight, length, WAZ, LAZ, WLZ) and iron status (haemoglobin concentration) of children as primary outcomes, while the secondary outcome included dietary intake (energy, nutrients and food groups) of children.

2. Study Location This study was conducted in Orang Asli (OA) communities in Selangor, Peninsular Malaysia. The specific communities were identified with the assistance of Department of Orang Asli Development Selangor (JAKOA). The majority of residents are Temuan (78%), followed by Mah Meri (15%), Semai (2%) and mixed (5%). This study included only the Temuan tribe as study population.

3. Study Design This was a cluster randomized controlled trial. Villages as clusters were randomly allocated into either intervention or control group. All households in each cluster were included in the trial.

4. Sample Size A formula for mean difference between two groups was used for sample size calculation. The sample size was calculated to have 80% chance of detecting a mean difference in length of 0.56 cm or more with an estimated standard deviation of 0.31. An additional 50% was applied to the calculation of sample size to account for withdrawal and non-compliance. Hence, the total number of respondents per group was 49.

5. Sampling Procedure There were 74 OA clusters (villages) in Selangor. Screening of clusters were conducted according to inclusion and exclusion criteria of clusters. Upon screening, 17 non-Temuan clusters and 6 Temuan clusters were excluded from this study. Temuan cluster is defined as having at least 80% of the households in the village are of Temuan ethnicity. The remaining 51 Temuan clusters were then randomly assigned either to the intervention or control group by considering the distance and geographical locality of the clusters. A total of 22 clusters were randomly allocated to the intervention group, while 29 clusters randomly allocated to the control group. There were 7 small clusters in the control group which grouped into 2 clusters as the small clusters have a small number of young children.

   Screening assessment of children then were conducted in clusters (that obtained consent from head of villages) based on the lists of children aged between 6 to 24 months given by the head of villages. All listed children were screened according to inclusion and exclusion criteria of children. Upon screening, 68 children in four clusters of intervention group, and 102 children in seven clusters of control group were excluded as not meeting the inclusion criteria of children. Based on the screening assessment, the total number of children recruited initially was 98 children, with 49 children in each of the two groups.

6. Selection of Clusters (Villages) and Children Approval by JAKOA was obtained prior to the selection of clusters (villages). OA villages were visited by investigator and a research assistant to conduct the selection of clusters based on inclusion and exclusion criteria. Head of villages were interviewed to obtain information on majority of households' ethnicity and availability of child/children aged between 6 to 24 months. For Temuan cluster that has at least one child aged between 6 to 24 months old, the head of villages were given an invitation letter, letter of study approval by JAKOA and information of the study. The head of villages were briefed about the purpose, benefits and potential risks of the study to the children. They were given the opportunity to ask for further details if they were not clear about the given explanation. Consent from the head of villages were obtained after the briefing. List of households with children aged between 6 to 24 months was then given to the investigator by the head of villages for the selection of children.

7. Treatments The intervention group was supplemented with three sachets of MMS in powder form each week for every other day for a period of 12 months. Each sachet containing 1 g consisting of ten vitamins and five minerals. The amount of each micronutrient provided by MMS was mostly ≥ 42% of the Malaysian Recommended Nutrient Intake, (RNI). This was adequate to meet the daily nutrient requirement of children in addition to nutrients provided by daily complementary foods. The MMS was manufactured by DSM Singapore Industrial Pte. Ltd. It was in powder form and to be sprinkled over a cooked meal or dissolved in a drink for the child. The intervention group was visited weekly in first 2 months and two-weekly in the following months to distribute and monitor the supplementation of MMS. Written instruction for using and storing the MMS in simple language with visuals was given prior to the supplementation. The intervention group also received health and nutrition advice delivered by investigator at 3, 6 and 9 months after supplementation begins. The control group received health and nutrition advice that were similar to intervention group and delivered at 3, 6 and 9 months after the study began.

8. Measurements 8.1 Demographic and Socioeconomic Information Demographic and socioeconomic information were obtained through face-to-face interview with parents/ caregivers using a pre-tested questionnaire and assistance from a local translator. Age of children was calculated from childbirth date that obtained from birth certificate or child health record. Data were collected prior to the intervention as baseline data.

   8.2 Anthropometry Birth weight and length of children were obtained from child health record. Body weight and length of children were measured at baseline, 3, 6, 9, 12 and 15 months after supplementation begins. Weight and length of children were measured using Tanita Digital Weighing Scale HD 325 (to the nearest 0.1 kg) and Recumbent Measuring Board (to the nearest 0.1cm), respectively using standard procedures. All measurements were taken twice, and the average value were used as final data. Age, weight and length of the children were translated into three indices of weight-for-age, length-for-age and weight-for-length. WHO Anthro program version 3.2.2 was used to estimate the Z scores for weight-for-age (WAZ), length-for-age (LAZ) and weight-for-length (WLZ). Z score was used to distinguish between normal and underweight (WAZ <-2), stunted (LAZ <-2) and wasted (WLZ <-2) children according to WHO classifications.

   8.3 Haemoglobin Haemoglobin level was assessed using finger-prick blood method. A staff nurse conducted the assessment using a portable HemoCue®201+ Analyzer. Haemoglobin assessment was performed at baseline, 12 and 15 months of study.

   8.4 Dietary Intake Dietary intake was assessed using a standard 24-hour diet recall questionnaire at baseline, 3, 6, 9, 12 and 15 months. This assessment involved face-to-face interview with parents/ caregivers to obtain information on time, types and amount of foods and beverages consumed by children during the 24 hours prior to the interview. Household measurements (e.g. cups, teaspoons, plates, bowls) were used to assist parents/ caregivers in estimation of portion sizes of consumed foods and beverages. Dietary intake data was analysed using a computerized dietary analysis program, Nutritionist Pro. The USDA Food Database was used in this software for dietary analysis.

   8.5 Child Vaccination, Infection and Household Sanitation and Hygiene History of child vaccination, infection and household sanitation and hygiene were assessed using a standard questionnaire at 12 months of study (end of the intervention). History of child vaccination was obtained from child health record. Assessments of child infection history involved face-to-face interview with parents/ caregivers to obtain information on common infections in children in the preceding two weeks. Interview with parents/ caregivers were conducted as well to obtain information on household sanitation and hygiene.

   8.6 Compliance to MMS Every week, parents/ caregivers in intervention group were given a pack which contains 3 sachets of MMS to be taken by their children in a week. After a week, the parents/ caregivers were required to return the empty sachets (including unused or half used) to the investigator. Compliance to MMS was measured by dividing the total number of empty sachets with a total number of distributed sachets for the 3, 6, 9 and 12-months periods.

   8.7 Adverse effect of MMS Any adverse effect was assessed during a weekly visit. The information was obtained through a weekly face-to-face interview with the parents/ caregivers.

9. Study Withdrawal Children who withdrew the study were visited several times and attempts were made to persuade the parents/ caregivers to continue with the study. Withdrawals were recorded in Case Report Form to determine percentage of the participants who withdrew for reasons related to the intervention or those who missed at random.

10. Ethics Approval and Funding The study protocol was approved by the Ethics Committee for Research Involving Human Subjects (JKEUPM) of Universiti Putra Malaysia. Permission to conduct the study in Orang Asli communities was obtained from the Department of Orang Asli Development (JAKOA). Written informed consent was obtained from parents /caregivers who agree to participate in the study. This study was funded by a research grant from Universiti Putra Malaysia (GP-IPS/2016/9514200).

11. Data Analysis Data were analyzed using 'Statistical Package for Social Sciences' programme for Windows version 22.0. Categorical data were described using count and percentage. All numerical data were tested for normality of distribution and were presented as mean and standard deviation. Non-parametric tests were used whenever the assumptions of normality and/ or homogeneity of variances were not met. Logarithmic conversion was performed for data not normally distributed. Per protocol analysis was performed, and children who withdrew from the study, noncompliant, and has missing outcomes were excluded from analysis.

The independence sample t-test was used to determine the difference in the mean of continuous variables between the intervention and control groups, while for categorical variables the Pearson's chi-square test was used. Paired t-test was conducted to assess changes in outcomes variables within each group (intervention and control) between baseline and 12 months as well as 15 months follow-up. ANCOVA using General Linear Model for repeated measures was performed to determine the effects of intervention between the groups on main and specific outcomes. Demographic and socioeconomic characteristics of both groups such as gender of children, age of children at selected time points, education of mothers and baseline measures were controlled in the analysis. Differences between groups were considered statistically significant at p<0.05 using two-tailed tests.

Inclusion Criteria:

• Aged between 6 to 24 months old; and
• Normal weight-for-age (WAZ > -2SD), length-for-age (LAZ > -2SD) and weight-for-length (WLZ > -2SD); and
• Normal haemoglobin level (> 11.0g/dL).

Exclusion Criteria:

• Has/ had history of chronic diseases e.g. failure to thrive and/or metabolic or endocrine disorders; and
• Has history of congenital and/or acquired neurological condition; and
• Under treatment for communicable diseases such as Tuberculosis, HIV.