Promotion of Complimentary Feeding Practices in Bangladesh

Back ground:

Link between infant feeding practices and nutritional status among children has been well established. Incidence of malnutrition sharply rises during 6-9 months of age which coincide with the period of complimentary feeding (CF). It is reported that poor nutrition increases the risk of illness, and is responsible, directly or indirectly, for one third of the estimated 9.5 million deaths in children less than 5 years of age. Early nutritional deficits are also linked to long-term impairment in growth and health. There is evidence that adults who were malnourished in early childhood have impaired intellectual performance. They may also have reduced capacity for physical work. High proportion of childhood malnutrition has negative impact for national development. Based on evidence of the effectiveness of interventions, achievement of universal coverage of exclusive breastfeeding (EBF) could prevent 13% of deaths occurring in children less than 5 years of age globally, while appropriate complementary feeding practices would result in an additional 6% reduction in under five mortality. Although there is huge potential for EBF till 6 months of age, the CF practices is also important to maintain the increase demand and growth after the period of EBF to maintain the optimal growth and development during infancy and early child hood period. However, there are lacks of efforts observed to increase the knowledge of good complementary feeding practices at population level. Under five child mortality is still high for Bangladesh. Among these deaths prematurity of the newborn/LBW and malnutrition together contribute to about 45% of all deaths. Despite the progress achieved, child malnutrition in Bangladesh remains among the highest in the world, and more severe than that of most other developing countries. This suggests that children in Bangladesh suffer from short-term acute shortfall in food intake as well as longer-term under-nutrition. Exclusive breast feeding is quiet static (43%-46%) for last few years in Bangladesh and around one fourth infant from 6-9 months ages start complementary feeding either too early or too late. Inappropriate practices of CF reflect the knowledge gap at the population level. Mother and family members are usually unaware about the nutrition need immediately after the completion of exclusive breast feeding practices. It was reported that Bangladesh scored 91.5/150 for Young Infant and Child Feeding (YICF) practices which was done by International Baby Food Action Network (IBFAN), Asia Pacific where community outreach and information support warranted much improvement. To reduce malnutrition: Comprehensive programme of integrated actions on many fronts has been addressed which includes "Promoting improved infant feeding practices, including breast feeding practices". Mothers and families need support to initiate and sustain appropriate infant and young child feeding practices through support person or peer group is recommended at the community level. Educational intervention on CF practices shows improvement in several studies in India, china and Brazil. Indian study observed feasibility of CF practices promotion through existing system and found limited improvement of infant length (difference in means 0.32 cm, 95% CI, 0.03, 0.61) between groups. Brazilian study assessed impact on child growth of the nutrition counseling component of the Integrated Management of Childhood Illnesses strategy and found children 12 month of age or older had improved weight gain (changes of +.25 and -.06 Z-score, respectively) and a positive but non-significant improvement in length (0.40 vs. 0.12). In this study medical doctors were used for counseling which is not applicable in poor resource settings. The China study was to improve infant growth by improving infant feeding practices. The education group infants were significantly heavier and longer, but only at 12 month (weight-for-age -1.17 vs. -1.93; P=5 0.004; height-for-age -1.32 vs. 1.96; P =0.022). But this study differ from exclusive breast feeding (EBF) duration and starting of complimentary feeding practices as recommended by WHO (EBF is practice 4-6 month and weaning starts at 4 months). None of the above studies evaluated feeding practices through CF index. In early 90's Bangladesh Rural Advancement Committee evaluated educational messages on CF practices other than WHO recommendation, Population level study on CF practice promotion is poor in Bangladesh. Given the above background, with lack of knowledge on CF practices in the community level and high burden of childhood malnutrition, we plan to improve complementary feeding practices in Matlab.

Hypothesis to be tested:

The proposed study hypothesized that an appropriate delivery of CF messages at the community level will improve the CF practices and subsequently decrease the burden of malnutrition during infancy.

Specific objectives:

The objectives of the project are to evaluate the effect of CF practice messages through community health workers to mothers and family members in comparison to mothers and family member who do not receive CF practice messages on the following outcomes:

Reducing the burden of stunting, underweight and wasting Evaluate complementary feeding index

Research Design and Methods:

Study site:

The study will be implemented in a rural community of Matlab under a sub-district of Bangladesh. Matlab is located 56 kilometer southeast of the capital Dhaka, where ICDDR,B has been running a health and demographic surveillance system (HDSS) in a population of about 220,000 since 1966. In HDSS, the community health research workers (CHRWs) collect all the demographic events and selected morbidities in the area by monthly household visits. The study area is divided into two parts. One is the ICDDR,B service area, where an extensive Maternal, Child Health and Family Planning Programme has been operating to strengthen the government's services since 1978. The other is the Government service area, where the population receives services from government facilities as in areas all over Bangladesh. The ICDDR,B service area is again divided by four blocks and each one is comprised of an approximate 27,000 population. Each block has a sub-centre that provides 24-hour services by midwifery staff. All the sub-centre and field activities are supported by the hospital at Matlab Township. Sub-centres are comparable with the union level health structure of government facilities. The study will be implemented in the ICDDR,B service area.

Currently at Matlab a comprehensive maternal, neonatal \& child health (MNCH) programme has been ongoing since March 2007. Under this programme pregnancies are identified by CHRWs during their routine bimonthly household visits. Once pregnancy is identified by urine test, CHRWs visit each woman at their home twice during pregnancy (12-14 weeks \& 33-34 weeks of pregnancy) and counsel them about facility-based delivery, antenatal care, and birth preparedness. After delivery, the CHRWs visit women on Days 0, 3, 7, and 28, 45, 75,105 and 180 to provide additional counseling, immediate newborn care, anthropometry measurements, \& collect morbidity information (Diarrhoea, Respiratory Tract infections \& other illness as well as also breast feeding \& other feeding status). The women also visit CHRWs houses for immunization, family planning and also for Home based Life Saving Skill (HBLSS) training along with their child and family members.

Study Design:

This randomized community intervention will be conducted on mother-infant pairs of two randomly selected areas from four blocks: intervention and control.

Implementation:

The study participants (the infant who has completed his/her 6 month of age from the date of birth along with their mother) will be enrolled from MNCH database according to criteria for the intervention group and same number of control will be selected as above mentioned procedure for each selected CHRW. At the beginning, baseline information will be collected for each participant after informed consent. All the mothers of the selected infants along with family member/s will receive training on standard complementary feeding practices at respective CHRWs houses. A Training manual will be developed according to WHO and Bangladesh Breast Feeding Foundation guideline. Refresher training will be arranged 3 months after the first training at the CHRWs houses with the same participants. The CHRW will visit each infant house till he/she reaches at 12 month of age at every two month interval.

Data collection:

During each visit CHRWs will record anthropometrical measurement, details feeding information both intervention and comparison group. CHRWs and Field Research Assistant (FRA) will also receive training on study questionnaire. Field Research Supervisors (FRS) will be included also in the training programme to include them in supervision process. The socio-demographic information will be collected from HDSS database. The recruited FRA will work with FRS and will be responsible for all trainings, supervision and monitoring the data collection. Monitoring will be done by a standard check list. CHRWs will bring the questionnaires in each subcenter meeting and FRA will receive and will send it to the computer room at Matlab after reviewing each questionnaire. Any inconsistency or quarries will be checked by FRA with the help of respective subcenter supervisor.

Sample Size:

The number of woman-infant pairs required was calculated based on the studies conducted earlier with peer counselors. We determined the sample size based on 20% difference (based on previously conducted same study) in the prevalence of height-for-age, and weight-for-age (40%) between intervention and control groups, with 5% significance level and 80% power. With a maximum of 30% lost to follow up a total of 120 pairs will be needed for each study arm (intervention and control). With design effect of 1.5, we need 180 woman-infant pairs from each arm.

Data Analysis:

The total observations will be categorized into two groups by using baseline status and after. Data will be presented as mean ± SD for continuous variables and proportion for categorical variables. Unadjusted odds ratio between exposure variables and complementary feeding will be determined by using univariate analysis. Multivariate logistic regression analysis will be done to evaluate the simultaneous effects of various exposure variables after adjusting for any confounding variables. A p-value <0.05 will be considered for significance.