Study area
The study area was Mbita Sub-County, Homabay County, on the shores and islands of Lake Victoria in western Kenya. The study area is, under a health and demographic surveillance system (HDSS), Rusinga east and west on the island and Gembe east and west on the mainland. During the 2011 survey, the population of the Mbita HDSS was 55,929. Parasitic infections, especially schistosomiasis, are an urgent public health concern [26, 27]. Mbita is dominated by fishing communities near the lake. Temperatures range from 15 to 30 ℃ with a bimodal rainy season. The short rainy season occurs between October and December, whereas the long rainy season occurs from March to May. The average annual rainfall is 800–1200 mm in the western part of the study area on Rusinga Island, whereas Gembe receives 800–1900 mm.
Study design and population
This was a secondary analysis of a cross-sectional study that explored malnutrition risk factors in the study population. The study was conducted between September and November 2011, at the end of the dry season, among fourth-grade primary school children. According to the education office in Mbita Sub-County, the primary school enrolment rate was 91.6% at the time of the survey. The schools included were full-grade primary schools with no history of mass chemotherapy 1 year before the study. Of the 64 public primary schools in the study area, 39 met the inclusion criteria. From these, eight schools were randomly selected. None of the selected schools was implementing school feeding programs. At the time of the survey, medications designed for treatment of schistosomiasis were not available at the study sites, save for their utilization in mass chemotherapy, and thus were not considered within the exclusion criteria. We enrolled 310 children, representing 98% of the children who met the inclusion criteria in the eight selected schools. We included children with complete anthropometric, parasitic infection and dietary data who were unlikely to have significant under- or over-reporting errors in daily energy intake.
Anthropometric measurements
Anthropometric measurements were used to assess nutritional outcomes. Weight and height were measured at school using a Seca scale with a height rod (Seca 786; Seca GmbH & Co. KG, Hamburg, Germany). Body mass index (BMI) was calculated as the weight (kg) divided by the square of the height (m). The height-for-age Z score (HAZ) and the BMI-for-age Z score (BAZ) were calculated using AnthroPlus software from the World Health Organization [28]. Based on the WHO definitions, children with HAZ and BAZ less than − 2 were categorized as stunted and underweight, respectively.
Dietary evaluation
Dietary intake was assessed using a food frequency questionnaire (FFQ) tailored to capture individual-level relative dietary patterns within this study population. The FFQ’s development involved preliminary 24-h diet recalls among ten local staff members to ensure its relevance. The FFQ included 41 food items asking about the frequency of consumption (nine options: never, 1–3 times a month, once a week, 2–4 times a week, 5–6 times a week, once a day, 2–3 times a day, 4–5 times a day, and > 6 times a day) and portion size. The enumerators conducted inquiries in Luo or Swahili, asking parents/guardians of the target children about the frequency and quantity of food items consumed in the past month. The amount was estimated as an approximate number of portions per meal, based on one portion using utensils such as bowls, tablespoons, teaspoons, plates, and cups. According to the answers in the FFQ, the daily intake of each item (g) was calculated as follows: number of portions × 1 portion amount × frequency. Daily energy and macronutrient intakes (energy ratios) were calculated using the Kenya food composition tables 2018 [29].
We evaluated diet quality based on the adherence level to the Kenyan food pyramid recommendations [22,23,24] called the FP score. In brief, the score was based on the recommended serving range for five food groups (staple foods, protein-rich foods, dairy products, vegetables, and fruits), as shown in Additional file 1: Table S1. Ten points were assigned if the number of servings consumed was within the recommended range. Fewer points were assigned for serving numbers outside the recommended range. The degree of adherence to the recommended serving range in the food guide pyramid was expressed as the sum of the scores of the five groups, each with a maximum of 10 points (maximum of 50 points). According to the Kenya food pyramid, we categorized the 41 food items in the FFQ into five food groups. The number of servings consumed for each food group was also calculated. For better interpretation, general starches were divided into two subgroups: (i) cereals and grains and (ii) potatoes, tubers, and starches. Protein-rich foods were also divided into two subgroups: (i) pulses and (ii) animal-source foods. The conversion of gram amounts to servings has been described previously [24].
Parasitic tests for malaria and helminthic infections and hemoglobin assessment
Malarial and helminthic infections were examined in a previous report on this study population [26]. Stool samples were examined for the presence of Schistosomiasis (S.) mansoni and STH eggs using the Kato–Katz fecal thickness smear technique. S. mansoni infection was defined as at least one egg count on either day of examination. S. mansoni infection intensity was expressed as eggs per gram of feces (epg). Based on the WHO categorization, S. mansoni infections were categorized as light (1–99 epg), moderate (100–399 epg), or heavy (≥ 400 epg) [30]. The STHs examined were Ascaris lumbricoides, Trichuris trichiura, and hookworms. Thick blood smears were examined under a light microscope to detect malaria infections. Positive cases were defined as the detection of at least one malaria parasite in a microscopic field of 200 white blood cells on thick films or 2000 red blood cells on thin films [31].
For hemoglobin (Hb) assessment, we conducted compete blood count (CBC) using a Hemolyzer® 3NG machine (Analyticon Biotechnologies GmbH, Lichtenfels, Germany). After Hb determination, participants were categorized as being anemic based on the WHO’s threshold of < 11.5 g/dL Hb for children 5–11 years, < 12 g/dL Hb for children 12–14 years and nonpregnant females above 15 years, and < 13 g/dL Hb for males above 15 years [32].
Questionnaire survey
Trained enumerators gathered information on children’s age, sex, mother/female guardian’s educational attainment, and household possessions from parents/guardians in home settings using a pretested questionnaire. Each child’s age was confirmed using official birth certificates or church baptism cards during household visits. In addition, an observation checklist was used to collect information on the house structure, latrines, and electricity availability in each household. Principal component analysis was conducted with variables including land ownership, electricity availability, latrines, and housing structure to create the household wealth index variable, which was categorized into tertiles of poorest, poorer, and least poor.
Statistical analysis
The participants were classified into tertiles according to their FP scores (T1, T2, and T3). Food group intake was energy-adjusted using density methods, shown as servings per 1000 kcal. We examined the trend associations between FP score tertiles and energy-adjusted food group intake using the Jonckheere–Terpstra test to confirm the FP score in this specific study population. The trend associations of dietary intake by wealth index, which is thought to influence dietary intake, were also examined in Additional file 1.
Logistic regression analyses were used to examine the association between stunting and diet quality (FP score). The outcome (dependent variable) was stunting, and the FP score tertile was the explanatory (independent) variable. Age-adjusted and multivariate-adjusted odds ratio (OR) and 95% confidence interval (CI) of stunting were determined for each FP-score tertile relative to the first tertile (T1). We adjusted for the child’s age and sex, total energy intake, household wealth index, mother/female guardian educational attainment, S. mansoni infection status, coinfection of S. mansoni infection and malaria (Plasmodium falciparum), and anemia status, as these variables might affect both nutritional status and dietary intake [11, 25, 33]. We did not include STH infection status as a covariate because they were rare in this population. To delve deeper into the interlinkage between diet, parasitic infection, and risk of stunting, we conducted stratified analyses-based S. mansoni infection intensity (negative/light and moderate/heavy) [12]. In addition, trend associations were examined by treating the median FP score for each tertile as a continuous variable. Sex-stratified analysis was also performed as in Additional file 1. All statistical analyses were conducted using IBM SPSS Statistics for Windows, version 28.0 (IBM Corp., Armonk, NY, USA). The significance level was set at 5% for two-tailed tests.
Add Comment