Elevated prevalence of malnutrition and malaria among school-aged children and adolescents in war-ravaged South Sudan

Abstract

Emerging as a sovereign state from decades of civil war, the Republic of South Sudan now faces poverty, a lack of health care infrastructure, a high burden of infectious diseases and a widespread food insecurity. School-aged children and youth, in particular, represent a high-risk demographic for malnutrition and infectious diseases. We screened 109 school-aged children and youth for nutritional status and malaria antigenaemia in Akuak Rak, South Sudan, and found a large proportion of underweight (77/109 = 73%) and prevalent malaria (44/109 = 40%). There was no significant association between malnutrition and malaria. This study represents one of the few published reports on child and youth nutritional status and malaria prevalence in South Sudan since its independence. The implementation of nutrition and malaria screening combined with evidence-based interventions in schools could help target this high burden vulnerable group.

Introduction

The Republic of South Sudan gained independence and was recognized as the 193rd United Nations member state on 9 July 2011. Born in crisis following decades of civil war, the country is characterized by a highly volatile political, economic and humanitarian situation.¹ In this unstable, war-ravaged context, the burden of infectious diseases,^2,3 food insecurity⁴ and child malnutrition⁵ are known to be elevated. Although there exist programmatic data reported from humanitarian organizations,⁵ published reports on child and youth nutrition and health from the region are scarce.

School-aged children and adolescents in the developing world are increasingly recognized as a high-burden demographic for nutritional deprivation and infectious disease, with adverse consequences for their developmental trajectory into adulthood.⁶ Beyond the preschool years, during which malnutrition and infectious diseases are the principal determinants of child mortality, school-aged children and youth are also at risk of poor health, cognition and educational achievement as a result of undernutrition and infection. Schools also represent a platform for public health surveillance for the community at large, as well as a convenient and cost-effective platform to deliver health interventions, such as antimalarials, anthelmintics and micronutrients.

A mutually amplifying relationship between malaria, a leading cause of hospitalization and death in sub-Saharan Africa, and malnutrition has been documented.^7,8 However, this interaction between malnutrition and malaria is complex and varies depending on the definitions and indicators used for malnutrition measurements.^9–11 More specifically, the literature has supported that stunting (low height-for-age), the indicator of chronic malnutrition, may increase the risk of malaria.¹² Whereas wasting (low weight-for-height), the indicator of acute malnutrition, appears to be associated with a decreased risk of malaria.¹³ Furthermore, a well-documented interaction is that re-feeding of famine victims increases clinical manifestations of malaria, with improved host immunity and/or nutrient replenishment that promotes parasite growth.^14,15

In the context of a humanitarian medical outreach in the remote community of Akuak Rak, Northern Bahr-el-Ghazal State, we screened school children (aged 5–18 years) for nutritional status and malaria. At the time of the study, January 2012, indigenous populations were returning to their ancestral lands and rebuilding, following a period of massive displacement due to violent conflict during the period of internecine civil war (1986–2005). Poverty (loss of cattle, crops, seeds and farming implements) among returnees may contribute to food insecurity.⁵ In addition, during displacement, they may have resided in areas with lower malaria transmission and lost partial immunity. Thus, this report provides rare health data in a post-war situation in a remote sub-Saharan African context characterized by near absence of infrastructure and human resources for health, as well as extreme poverty.

Methods

This was a cross-sectional study among children and youth voluntarily undergoing screening for malaria and malnutrition in the context of a medical outreach (Connecting Lives International Mission, Christian Reaching Mission) which included health screening at a local school (Hope Mission School). Oral informed consent was obtained from each participant prior to anthropometric measurements and malaria testing. Permission to conduct medical outreach and student health screening was obtained from local authorities. The study was conducted according to the guidelines laid down in the Declaration of Helsinki. The University of Alberta Research Ethics Committee approved the study.

Patients and setting

The study was conducted at a combined boarding and day school, Hope Mission School (enrolment 160 students between the ages of 5 and 19), in Akuak Rak, Western Aweil County, Northern Bahr-el-Ghazal State (estimated State population 720,898), with a majority ethnically Dinka. The village is rural and remote, with no road access. The study took place during the dry season from 7 to 12 January 2012, 6 months after South Sudan became an independent state. Participants were included in the study if they attended the Hope Mission School and voluntarily took part in health screening; there were no exclusion criteria. A sample size calculation indicated that approximately 93 participants would need to be enrolled in order to estimate the prevalence of underweight and malaria to within ± 10%. Plasmodium falciparum is hyper-endemic in the area with peak incidence during the rainy season.³

Due to almost 50 years of protracted conflict, poverty, lack of infrastructure and disruption of social services, South Sudan is among the world’s least developed countries.¹ More than half of the country lives on less than $US 1.25 per day. Food insecurity is widespread, affecting an estimated 4.7 million people (37% of the country’s population) in 2012.⁴ Humanitarian agencies were involved in over 30 operations in the country at the time of this study, including food assistance and provision of basic medical services. Ongoing violent conflict and fluctuating levels of international donor aid have contributed to ongoing food insecurity despite an official end to the Sudanese civil war with the Naivasha Comprehensive Peace Agreement in 2005.

Anthropometric measurements

Height and weight were measured by professional nurses licensed in the USA, well trained and experienced in the measurement of paediatric weight and height. Instruments used were a digital scale and stadiometer. Weights and heights were plotted on growth curves at the time of data collection and were reviewed by a Royal College of Physicians of Canada certified paediatrician for plausibility and for clinical assessment at the time of data collection. We compared height-for-age, weight-for-age and BMI-for-age to US Centre for Diseases Control (CDC) and Prevention norms.¹⁶ Although alternative growth charts from the World Health Organization are available, the CDC norms provide a benchmark from a well-nourished population against which to compare our cohort. Of note, at the lower centiles, the WHO and CDC curves match closely such that the choice of reference norms is unlikely to substantively affect the findings.¹⁷

Malaria diagnosis

Trained nurses performed a fingerprick rapid diagnostic test (RDT) for malaria (ParaCheck-Pf device, Orchid Biomedical Systems). This test, based on detection of histidine-rich protein 2, is able to detect P. falciparum antigen in the peripheral circulation. Recent experience in 65 schools in Kenya showed that this RDT had a sensitivity of 99% and a specificity of 85% compared to light microscopic examination of Giemsa-stained peripheral blood films.¹⁸ Children found to be RDT-positive were treated with a six dose regimen of artemisinin combination therapy (ACT) according to WHO guidelines,¹⁹ irrespective of symptoms.

Results

A total of 109 students voluntarily participated in health screening, representing 68% of the total student population of the Hope Mission School. The self-reported median age was 13 years (range 5–19) and 38/109 (35%) were female (significant underrepresentation of females in cohort, p = 0.0016). Figure 1 shows the distribution of weight, height and body mass index (BMI) for the cohort. Noteworthy is the striking proportion of underweight children and youth (BMI < 5%ile), representing 73% (95%CI 64–81%) of the cohort. In particular, low weight-for-age was remarkably prevalent, with 56/106 (53%) of students below the fifth percentile and 44/106 (42%) of students below the third percentile (−2SD) for weight. Stature was closer to US norms (Figure 1(B)), although 16/105 (15%) were below the fifth percentile for height and all of these children also had BMI < 5%ile (“stunting”).

Figure 1.

Low weight-for-age and body mass index (BMI) in South Sudanese school-aged children and youth. Histograms illustrate the frequency distribution of weight (1a), stature (1b) and BMI (1c). The distributions of weight and BMI are both markedly left-skewed, reflecting high prevalence of underweight children and youth.

Overall, 44/109 (40%; 95%CI 32–50%) of participants tested positive for malaria. Characteristics of the cohort, disaggregated by RDT result, are given in Table 1. Infected children and youth were minimally symptomatic, with only mild non-specific complaints such as headache and/or myalgia. The median axillary temperature (T) was higher in the RDT-positive than RDT-negative group, and fever defined as T > 37.5 °C was more common among students with malaria (22% vs. 7%, p = 0.0044). Fever (T > 37.5 °C) had a sensitivity of 30% (95%CI 18–44%) and a specificity of 93% (95%CI 84–97%) to predict malaria in this cohort. However, a higher grade fever (T > 38.0 °C) was present in only 5% of students with malaria, which was not statistically significantly different from negative controls. Weight, stature and BMI were similar in both groups.

Table 1.

Characteristics of students screened for malnutrition and malaria in South Sudan.

	Malaria (n = 44)	No malaria (n = 65)	p-value
Age, median (range)	12 (7–18)	13 (5–19)	0.44
Female gender, n (%)	13 (30)	25 (38)	0.45
Stature (percentile)			0.81
<5	5 (11)	11 (18)
5–10	3 (7)	2 (3)
10–25	8 (18)	3 (5)
25–50	7 (16)	16 (26)
50–75	8 (18)	9 (15)
75–90	6 (14)	11 (18)
90–95	7 (16)	7 (11)
>95	0 (0)	2 (3)
Weight (percentile)			0.94
<3	17 (39)	27 (44)
<5	24 (55)	32 (52)
5–10	6 (14)	8 (13)
10–25	5 (11)	8 (13)
25–50	6 (14)	6 (10)
50–75	2 (5)	6 (10)
75–90	1 (2)	1 (2)
90–95	0 (0)	0 (0)
>95	0 (0)	1 (2)
BMI (percentile)			0.20
<3	33 (75)	38 (62)
<5	33 (75)	44 (72)
5–10	3 (7)	3 (5)
10–25	5 (11)	6 (10)
25–50	2 (5)	4 (7)
50–75	1 (2)	3 (5)
75–90	0 (0)	0 (0)
90–95	0 (0)	0 (0)
>95	0 (0)	1 (2)
Temperature (°C), median (range)	37.4 (36.2–40.2)	37.2 (36.0–39.6)	0.0036
Fever (T ax > 38.0 °C), n (%)	2 (5)	1 (2)	0.78
Fever (T ax > 37.5 °C), n (%)	13 (22)	4 (7)	0.0044

Discussion

The point prevalence of underweight (BMI < 5%ile) and malaria in our cohort of South Sudanese school children and youth was alarmingly high (73 and 40%, respectively). These findings reflect a high burden of undernutrition and parasitic disease in a severely resource-limited post-war context in remote sub-Saharan Africa. Evidence-based interventions, including school food programmes, antimalarials and long-lasting insecticidal nets (LLINs), might be efficiently distributed to these high-burden groups using schools as a platform for public health.

Our data corroborate several other nutritional surveys from South Sudan conducted in early 2012. The United Nations Nutrition Cluster of South Sudan, which aims to support and strengthen coordination of nutrition actors in the area, has compiled data from several Standardized Monitoring and Assessment of Relief and Transition (SMART) nutrition surveys.²⁰ Between January and June 2012, five independent surveys in the Northern Bahr-el-Ghazal State conducted by humanitarian organizations documented global acute malnutrition rates of 19–27% among children under 5. Our study, which focused on school-aged children rather than children under 5, likewise found high rates of underweight, consistent with widespread food insecurity across childhood age groups. Nutritional and genetic factors likely contributed to the striking degree of underweight in this cohort. Around the time of our study, a survey in Northern Bahr-el – Ghazal State showed that 100% of households were severely food access insecure and 88% of households reported sometimes or often going to bed hungry in the preceding month.⁴ Reasons for food scarcity in the area included crop failures resulting from poor 2011 rains, increasing reliance on market purchase as a primary source of food, near doubling of food prices since independence, vulnerable food supply chain and poor transportation infrastructure, and high demand for food from large numbers of returnees and internal displaced persons after quiescence of violent conflict in the area. The Dinka are a traditionally agro-pastoralist society; loss of cattle, crops, seeds, farming instruments during the period of civil war created a crisis of food security among returnees. We used US norms for computation of weight, height and BMI z-scores, allowing the possibility of genetic differences. The Dinka are among the world’s tallest races (mean height 176 and 170 cm among adult males and females, respectively) with BMI of 19.4 and 19.7 among adult males and females, respectively.²¹ Our study showed height distribution similar to US norms, but markedly reduced weight and BMI. The proportion and severity of underweight, together with documented food insecurity in the region, suggest that macronutrient deprivation is an important and modifiable factor contributing to underweight status.

The malaria point prevalence in our study (40%) was high relative to other African reports. In another school-based screening study of 2400 children from 51 schools in coastal Kenya, the overall prevalence of malaria was 13%, but ranged markedly between different schools (0–75%), with a prevalence exceeding 40% in only three schools.²² Given the timing of our survey (dry season) and the relative health of our cohort, P. falciparum prevalence rates may be even higher in wet seasons, in clinic- or hospital-based cohorts, and in children under 5. Infected children were minimally symptomatic, likely reflecting acquired premunition from repeated past exposure to P. falciparum. Low-grade fever was associated with malaria positivity (specific but insensitive sign) but T > 38 °C was present in only 5% of students with malaria. Similarly, only 3.5% of infected children had documented fever in another school-based study from Kenya.²² We did not observe an association between underweight and malaria in our cross-sectional survey, similar to a recent study in Ethiopia¹¹ but unlike other studies from Uganda.^7,8 A prospective study examining malaria incidence in underweight school children and healthy-weight controls would help to address this discrepancy. Given the potential for exacerbation of malaria signs and symptoms upon re-feeding, caution should be exercised with nutritional interventions in this group, considering the high prevalence of malaria.

School-based malaria screening, in use for over a century, is an inexpensive framework for malaria surveillance that can complement community-based surveys.¹⁸ For example, large-scale school parasite surveys documented the decline of malaria in the USA during the 1920–1940s, and continue to be used by the Ministries of Health of modern African states to monitor the impact of malaria control efforts.¹⁸ The representativeness of a school survey for the community at large depends on the catchment area and may be influenced by gender, wealth and health differentials in school enrolment.²³ Unfortunately the school enrolment for South Sudan may be very low as children and youth are not able to attend school in the aftermath of war that decimated the education infrastructure. Our study involved the only mixed day and boarding school in the large geographic area around Akuak Rak. The school provided free education under foreign sponsorship, such that students from poor households may be adequately represented in our study, although we did not measure household wealth. Underrepresentation of female students in our cohort may reflect a predilection for education of males.¹⁸ Thus, high malaria prevalence in the school likely implies high transmission in the community of Akuak Rak, although the degree to which the community prevalence may be over- or underestimated is difficult to quantify.

School-based screening offers a natural and convenient platform for delivery of public health interventions to a high-burden target group. Previous studies have shown positive impact of school-based intermittent preventive treatment (IPT) for malaria on health and cognitive function.^24,25 On the other hand, a recent school-based cluster-randomized trial in Kenya did not demonstrate any impact of malaria intermittent screening and treatment (IST) on prevalence of anaemia, P. falciparum infection or scores of classroom attention after 1–2 years of follow-up.²⁶ Finally, infected school children may act as a reservoir for onward transmission of malaria to others in the community, and treatment may block parasite spread.¹⁸ This may be an operationally efficient mechanism to target a high-prevalence group for malaria control.¹⁸ On the other hand, a recent trial in Burkina Faso indicated no impact of IST on community-wide malaria transmission.²⁷ Further studies are ongoing to test whether school children can be used to detect and monitor malaria “hotpots” within a community, where localized transmission and re-infection rates are high. This could, in turn, trigger intensive hotspot-focused community interventions to interrupt malaria transmission.

Conclusions

These rare data of underweight and malaria among school children in war-ravaged South Sudan call for implementation of known inexpensive and evidence-based interventions in this high-burden vulnerable group.