Title: The Impact of Maternal Obesity on Pediatric Malnutrition in Rural Uganda – a Household Survey



  1. Background




Chronic pediatric malnutrition is a serious problem affecting low and middle income countries across the world. 1-3 Children who are affected by this condition can suffer from stunting, cognitive delay, and increased risk of overall mortality. 1 Additionally, being chronically malnourished can lead to less education, decreased ability to be productive in work, and eventual lower earnings as adults. 2 A WHO estimate states that one quarter of all children – approximately 178 million – suffer from chronic malnutrition. 1 The prevalence of chronic malnutrition is highest in sub-Saharan Africa, where almost 40% of children are affected. Within sub-Saharan Africa, Uganda in particular has an estimated prevalence of 33% of children under five years of age stunted, six percent wasted, and 14% underweight. 2 . In spite of initiatives such as the 2004 Uganda Food and Nutrition Policy Reform and the 2010-2015 National Development Plan these numbers have remained largely unchanged. 2,4 Other interventional trials that promote adequate complementary feeding, including specially formulated foods and education on optimal feeding practices, have been shown to reduce childhood stunting but adherence is low, with only six percent of children meeting the overall recommendations for a minimally adequate diet. 5

Rural children in Uganda are twice as likely to be stunted compared to children in urban areas. 6 Some theories put forward for this disparity have looked at livestock ownership, birth order of the child, age of the mother and child, marital status, and occupation (farmer vs. pastoralist). From a broader public health standpoint, child feeding practices seem to be influenced by household food security as well as the health environment, including hygiene, and caring practices. 6

While under-nutrition and infectious diseases continue to persist in Uganda, overweight, obesity, and associated health concerns are emerging. 7 This rise in those who are overweight and obese has lead to an increase in chronic illnesses and a significant burden of disease. 7 The nutrition transition, a shift from an active lifestyle with the consumption of fewer processed foods to a sedentary lifestyle with the consumption of high-calorie foods, is occurring in Uganda. 7 This change in lifestyle coupled with the poor pediatric nutrition described above creates a dual-burden of malnutrition with underweight and overweight individuals in the same communities or even the same households. 8 This transition has been described in some middle income nations and in urban areas compared to rural areas of Uganda. 7

The prevalence of obesity among adults in Uganda is 3.9 percent. 9 There is a greater burden of overweight and obesity in females, with the prevalence between ages 15-49 being 19%. 7 In a study concerning overweight and obesity in young adults in Uganda, females were 4.81 times more likely to be overweight than men. 10 This has been seen in various studies and is related to differences in metabolism and hormonal balance between genders. Another factor that may contribute to this is the susceptibility of females to gain weight because of cultural factors. Females were found to be less likely to exercise regularly. 10

The prevalence is higher in urban areas, with 53.4% of women in Kampala being overweight or obese. 7 Some data is conflicted regarding this urban-rural divide, and seems to indicate that consuming large portions of carbohydrates and level of activity may play a role in different regions of the nation. 10 In one study in particular, there was not a significant difference in overweight prevalence between the rural and urban areas, but there was a difference in behaviors that led to overweight and obesity. 10

As of yet, there have been no studies demonstrating significant relationships between parental obesity and pediatric malnutrition in a low income nation, although parental education may play a role. Therefore, we decided to conduct a house-hold survey of sub-counties in rural southwestern Uganda. Specifically, we will conduct a cross-sectional analysis to determine the prevalence of malnutrition in children under five years of age, as well as prevalence of obesity and level of education of the parents of these children. We will explore a possible relationship between these two factors, and collect additional data regarding infectious diseases and micronutrient deficiencies to help assess potential confounding variables.


  1. Hypothesis / Objectives of the study



Given the abundance of food in the Kabale region and the juxtaposed high prevalence of pediatric malnutrition factors other than access to food may play a role. In particular, factors such as parental obesity, cultural beliefs and knowledge of nutritional content of food may be associated with higher rates of pediatric malnutrition.




  1. To determine whether children with at least one obese parent are more likely to be malnourished
    1. And if parental lack of nutrition education is associated with higher rates of pediatric malnutrition


  1. To obtain background data on prevalence of pediatric malnutrition and parental obesity in a difficult to access rural population


  1. Design and Methodology

We intend to conduct a randomized cluster-based house-hold survey of children under five years of age in sub-counties of Kabale district, located in southwestern rural Uganda. We will assess nutrition status and other health indicators of children in addition to surveying parents.


Nine of 22 sub-counties in Kabale region are eligible for random selection using probability proportional to size calculations. The total population available for sampling across the nine sub-counties is 217,723. Using a random number generator in MS Excel a starting point of 24,725 was calculated. From there, a sampling interval of 54,431 was calculated by dividing the total population size by the number four, which was the pre-determined number of clusters in this stage. The four sub-counties of Kyanamira, Rwamucucu, Bubare and Rubaya were selected. These four sub-counties contain 147 villages. The villages were arrayed similarly in a probability proportional to size method, and 30 clusters – each cluster representing one village – were randomly selected across the four sub counties. Teams of two will go door to door to each of the 30 village clusters and will survey seven households, providing a final sample size of 210 households.


Two teams each consisting of one Ugandan clinical health officer and one final year medical student will conduct the survey using an electronic data collection tool Magpi. Data will be obtained for anthropometric measurements of children and mothers. Additionally, the research team will conduct a questionnaire regarding nutrition and overall health of mothers and children, utilizing Ugandan nationally validated guidelines for diagnosis and treatment of malnutrition. Additional data will be collected regarding infectious diseases in children, income, and education level of mothers. In order to do so, in the household survey mothers of children under five years old who consent to engage in the study will be asked about their nutritional knowledge, foods they commonly eat, and any relevant medical conditions. A relevant medical condition is defined as a disease process that is known to lead to malnutrition. These include diseases such as HIV, gastroenteritis, malaria, pneumonia, parasitic helminthic infections and otherwise.


Each team of researchers will visit approximately seven households per day in each cluster. Data will be collected over the course of a three-week period in early 2016, with teams working over fifeteen clinic days. A mapping study will be completed in advance of the research team to indicate locations of households in the selected villages that contain children under five years of age.


Based on previous data it is estimated that each household will contain 2-3 children under five. We therefore expect to be able to survey approximately 525 children and 210 mothers. We will be able to use this data to calculate the prevalence rates of obesity and malnutrition in children under five in the district, as well as the primary outcome, which is a comparission of households that contain obese mothers and one or more malnourished children with households that contain obese mothers and no malnourished children. We will be looking at parental obesity as an exposure for malnutrition in this cross-sectional analysis. Based on past studies we anticipate a prevalence of 30% for maternal obesity, and 40% for pediatric malnutrition. We estimate a prevalence of 20% for malnourished children in the same home as a mother with a normal BMI, and a 45% prevalence for malnourished children in the same home as an obese mother.


Based on these estimations, using Fleiss, Statistical Methods for Rates and Proportions, formulas 3.18 &3.19 with an alpha of 5%, a power of 80%, an estimated ratio of 2.3 for unexposed to exposed children (ie. for every 7 non-obese mothers there will be 3 obese mothers) we will require a sample size of 43 exposed households and 99 non-exposed households. This sample size is well within the reach of the current study design.


All diagnoses of malnutrition will be based upon the 2006 WHO growth standards, as adopted by Ugandan clinical guidelines. Any mother with a severely malnourished child (three standard deviations below the mean) will have the option to send her child to the KIHEFO re-feeding center for treatment. Treatment received at KIHEFO will not be recorded for study purposes; all study data will be collected during the survey and health status exam.


  1. Justification for Use of Deception


Deception will not be used in this study.


  1. Inclusion/Exclusion Criteria


Inclusion criteria include:

The respondant must reside in the household being sampled, be over 18 years of age, and be the mother or guardian of the children. As well, the respondant must be able to understand a verbal consent agreement as translated from English to Ruchiga and provide their written consent for themselves and their eligible children to participate. Children must be above 6 months and less than 5 years of age.


*Exclusion criteria include:

Any male guardians. Any mothers below age 18 or unable to understand verbal consent and provide written consent.



*All participants excluded from the study will still be offered appropriate nutritional supplementation.


  1. Age Range


Participants enrolled in this study will be between age 6 months to five years for children, and greater than 18 years for parents.


  1. Statistical Design


The study design is a randomized cluster-based house-hold survey.


  1. Recruitment Process


The study will be conducted by going to patients homes as described above, with participants being recruited from communities in the Kabale Region. In order to ensure that participants are at home when the survey team arrives, the specific cluster that is randomly selected for that particular day will be informed that a team from KIHEFO will be arriving on the day in questions. They will be informed that any household visited by the team will be asked to participate in a research study, and will be receiving de-worming treatment, irrespective of their choice to participate in the research study.


  1. Data Collection


Patient care data will be collected by two research associates with tablet computers. This data will be password protected and only accessible by members of the research team.


Additionally, study data will be anonymized and replaced by a study number then kept on a password protected electronic database. This database will only be made available to the immediate study staff


The data will be kept for 7 years. Following this time period the protected electronic database will be deleted.


  1. Consent Process

In the attached consent form you will find that we have explained to the parents that they are agreeing to participate in a research study regarding pediatric malnutrition in rural Uganda. We explain that they will still be able to receive all medical services free of charge even if they decide to not participate in the research study. Following the completion of the consent form the parents will be asked a number of questions in our survey form. Following completion of the parental survey they will have their body measurements taken. They will then be asked a survey specific for their children, and children will have anthropometric data collected.


The consent form will be delivered by certified Ugandan medical professionals translated from English using a printed document.


  1. Risk and Benefits


Possible Risks:

(i) Physical risks (e.g., any bodily contact or administration of any substance:   Yes      No x


(ii) Psychological/emotional risks (e.g., feeling uncomfortable, embarrassed, or upset): Yes x     No


(iii) Social risks (e.g., loss of status, privacy and/or reputation):                       Yes      No x


(iv) Legal risks (e.g., apprehension or arrest, subpoena):                                  Yes      No x


To eliminate any emotional risks of feeling uncomfortable or embarrassed while having measurements taken and the survey conducted the parents will be interviewed privately in their home. If the research associate is male and the participant is female she will have the right to request a female chaperone. No additional personnel will be able to observe the survey or measurements unless they have received specific verbal consent from the patient.


Voluntary enrollment in a refeeding centre will be offered for children who are severely malnourished. In addition, nutritional supplementation and deworming will be provided for all children and parents. These initiatives are strongly advocated by all consensus national and international guidelines. All care will be provided as per Ugandan national guidelines.


Possible Benefits:


The patients will directly benefit from being involved in this project as they will receive free medical care including nutritional supplementation. By engaging in this study participants will be providing data on the prevalence of malnutrition and obesity. It will provide the ability to demonstrate a relationship between parental obesity and pediatric malnutrition.


  1. Feedback to Communities


Data regarding the prevalence of pediatric malnutrition and parental obesity will be disseminated to local healthcare organizations and governmental agencies. Additionally, the data will be published in recognized public forums to ensure that novel programs may be developed to provide additional care to the population.

The study results will also be presented at an international meeting and will be submitted for publication in a peer reviewed journal.



  1. References
  2. World Health Organization. Essential nutrition actions: Improving maternal, newborn. Infant and Young Child Health and Nutrition.Part II: Effectiveness of large-scale nutrition programmes: evidence and implications.Geneva: WHO. 2013.
  3. Habaasa G. An investigation on factors associated with malnutrition among underfive children in nakaseke and nakasongola districts, uganda. BMC pediatrics. 2015;15(1):134.
  4. UNICEF. Strategy for improved nutrition of children and women in developing countries. 1990. New York, UNICEF.
  5. Mosites EM, Rabinowitz PM, Thumbi SM, et al. The relationship between livestock ownership and child stunting in three countries in eastern africa using national survey data. PloS one. 2015;10(9):e0136686.
  6. Uganda Bureau of Statistics (UBOS) and ICF International Inc. Uganda demographic and health survey 2011. . 2012.
  7. Ickes SB, Hurst TE, Flax VL. Maternal literacy, facility birth, and education are positively associated with better infant and young child feeding practices and nutritional status among ugandan children. J Nutr. 2015.
  8. Turi KN, Christoph MJ, Grigsby-Toussaint DS. Spatial distribution of underweight, overweight and obesity among women and children: Results from the 2011 uganda demographic and health survey. International journal of environmental research and public health. 2013;10(10):4967-4981.
  9. Black RE, Victora CG, Walker SP, et al. Maternal and child undernutrition and overweight in low-income and middle-income countries. The Lancet. 2013;382(9890):427-451.
  10. Central Intelligence Agency. The world factbook: Uganda. Accessed October 27, 2015.
  11. Baalwa J, Byarugaba B, Kabagambe E, Otim A. Prevalence of overweight and obesity in young adults in uganda. African health sciences. 2010;10(4).