Background: Burundi has one of the poorest child health outcomes in the world. With an acute malnutrition rate of 5% and a chronic malnutrition rate of 56%, under five death is 78 per 1000 live births and 47 children for every 1000 children will live until their first birthday. In response to this grim statistics, Village Health Works, a Burundian-American organisation has invested in an integrated clinical and community intervention model to improve child health outcomes. The aim of this study is to measure and report on child health indicator ahead of implementing this model. Methods: A cross sectional design was employed, adopting the Demographic Health Survey methodology. We reached out to a sample of 952 households comprising of 2675 birth, in our study area. Mortality data was analysed with R package for mortality computation and other outcomes using SPSS. Principal component analysis was used to classify households into wealth quintiles. Logistic regression was used to assess strength of associations and significance of association was considered at 95% confidence level. Results: The incidence of low birth weight (LBW) was 6.4% at the study area compared to 10% at the national level with the strongest predictor being malnourished women (OR 1.4 95%CI 1.2-7.2 p = 0.043). Fever incidence was higher in the study area (50.5%) in comparison to 39.5% nationally. Consumption of minimum acceptable diet was showed a significant protection against fever (OR 0.64 95%CI 0.41-0.94 p = 0.042). Global Acute Malnutrition rate was 7.6% and this significantly reduced with increasing age of child. Under-five mortality rate was 32.1 per 1000 live births and infant mortality was 25.7 per 1000 in the catchment with most deaths happening within the first 28 days of life (57.3%). Conclusion: Improving child health status is complex, therefore, investing into an integrated intervention for both mother and child could yield best results. Given that most under-five deaths occurred in the neonatal period, implementing integrated clinical and community newborn care interventions are critical.
This prospectively collected data collection was conducted at the Vyanda and Rumonge provinces located in the south of Burundi (Fig. 1). Predetermined collines (districts) of VHW’s study area constituted the program target area. However, to avoid outliers in the results, the capital towns of both provinces were excluded. Therefore, the sampling frame of this study constituted 18 collines with a total population of 142,953. Map of Study Area constructed using ArcGIS mapping software Programmatically, at present, VHW is the only organisation working on infant and child health in the area although this coincides with the implementation of the national free healthcare policy for pregnant women and children under-five. A two-stage sampling strategy comprising of cluster and systematic sampling was used for this evaluation. At the first stage, collines along with their respective populations were received from the administrative province authorities. This became the sampling frame from which a probability proportional to size (PPS) was applied to select a desired cluster size of 30. The second stage was conducted during field work. Enumerators received households list from the chief of collines (chef de la colline) and depending on the total number of households available in that colline, either a 2nd or 3rd consecutive household was systematically selected after a random ‘pen throw’ to select the first household. The main sampling units were households and selection was based on the following inclusion and exclusion criteria: The determination of an appropriate sample size was based on the methodology of the DHS [18], an international program that conducts national representative surveys on major maternal, infant and child health indicators. Following the statistically robust predetermined conditions of the survey, details of which, have been published elsewhere [18], 952 households were selected. These households then presented 2675 birth histories for computation of infant and child mortality indicators. The entire survey was managed by the operational research, monitoring and evaluation department of VHW. Standard questionnaires were adopted from French version of the standard DHS [20] and United Nations Children’s Fund (UNICEF) multiple indicator cluster survey [21]. Questionnaires were segregated into three targeting women (the caregiver in most instances), men and other members of household. Questions about child outcomes were collected from the caregiver and childhood mortality collected through birth histories of women of reproductive age. Data was collected by seven field teams comprising of a team lead (enumerator), measurer and a supervisor in each team from May 01, 2019 to June 28, 2019. Team leaders interviewed caregivers on child outcomes including taking birth histories and measurers were responsible for taking anthropometric measurements of children and women of reproductive age (height, weight and Mid-Upper Arm Circumference – MUAC). Supervisors ensured data quality and oversaw random selection of households. Measurements for resting systolic blood pressure (SBP) and diastolic blood pressure (DBP) were done using a digital portable blood pressure monitor (Panasonic, Germany) which were taken for women of reproductive ages at the households. For each study participant, two measurements were taken, 10 min apart and the average was taken as the final reading. Main outcomes for assessment in this study were: low birth weight, childhood fever and malnutrition and childhood mortality. To allow comparability of our results with national figures, we defined these outcomes according to standard definitions by the DHS [18]. Low birth weight (LBW) was assessed from birth card of the last pregnancy of women within the reproductive age. We classified low birth weight as children who had a weight below 2500 g at the first measurement after birth. Fever incidence was defined as child with elevated temperature any level any day within the 2 weeks preceding the survey. This definition is line with the DHS definition and the recall method for data collection has been validated in different settings [22, 23]. Malnutrition was classified under two main measures: acute malnutrition (wasting) and chronic malnutrition (stunting). Global Acute Malnutrition, a combination of moderate and severe wasting was defined as children between 6 and 59 months with weight-for-height (WfH) z-score less than − 2 according to the WHO growth standards and global Chronic Malnutrition on the other hand was defined as children with height-for-age z-score less than − 2 according to the growth standards. A woman was classified as malnourished if she had MUAC ≤23.0 cm to ≤25.5 cm. From birth and death histories acquired from women and using a synthetic cohort life table approach [24, 25], childhood mortality rates classified as neonatal, post-neonatal, infant, child and under-5 mortality were calculated as: Minimum Acceptable Diet, Blood Pressure status of caregiver, wealth status of households, Nutritional status of caregiver and Household Hunger Scale of households were included as exposure variables for this study. Minimum Acceptable Diet (MAD) consumed was defined as the proportion of children who consumed four out of seven food types the day preceding the survey. The food types were; grains, root and tubers, legumes and nuts, dairy products (milk, yoghurt, cheese etc.), flesh food (meat, fish, poultry etc.), eggs, Vitamin A-rich fruits and vegetables and other fruits and vegetables. This information was obtain through recall of food consumed 24 h before the survey. Blood Pressure (BP) classified was into normal or abnormal (high or low). Normal blood pressure classified as SBP/DBP of 90/60 mmHg and 120/80 mmHg and abnormal blood pressure, < 90/60 mmHg or ≥ 140/90 mmHg. Household hunger scale, a global indicator for assessing household access and frequency to food was defined as households who reported ‘yes’ to one or more of the following events: 1. no food at all in the house; 2. went to bed hungry, 3. went all day and night without eating. For households that who confirmed ever experiencing any of these events, further questions on frequency were asked which were classified into never (value = 0), rarely or sometimes (value = 1), often (value = 2), summing to a total score of 6. Households with a score between 0 and 1 were classified as ‘No hunger detected in households’, those with score between 2 and 3, ‘Moderate hunger detected in household’ and between scores of 4 and 6, ‘Severe hunger detected in household’. In assessing if a child had received Vitamin A supplement, a sample was presented to the caregiver and asked if child had received it 6 months preceding the study. When confirmed, the child was considered as having received Vitamin A supplementation. Wealth quintiles were constructed from principle component analysis of 15 household items, consisting of household possessions, a state of housing and access to essential services. From the component coefficients generated, rank analysis was applied to classify households into five levels with lowest being the poorest and highest being the richest. Finally, to determine the influence of malnutrition on some outcomes, it was used as an exposure and the definition is same as stated above in outcomes section. Mortality rates were computed using version 0.7.0 of a predeveloped R package [24] which was originally developed for calculation of childhood mortality using the DHS methodology [25]. Childhood mortality rates were calculated from birth and death histories acquired from women of reproductive age 60 months (5 years) preceding the survey. All other indicators were calculated using IBM software – SPSS Statistic version 20 [26]. Chi-square test was used to assess relationships between outcomes and exposure variables (those variables disaggregated by the outcome variables). All outcomes were binary, as such a binary logistic regression was used to assess strength of association. However, when independent variables had more than two variables, a multinomial logistic regression was used. Significance of association was considered at 95% confidence level p < 0.05 (two-tailed). Nutrition data was analysed with Emergency Nutrition Assessment (ENA) software [27] for determining the individual malnutrition level of every child (using the WHO defined z-score parameters) and results exported to SPSS for further analysis.
N/A