Worldwide, malnutrition accounts for more than 45% (Black et al., 2013)of an estimated 5.6 children who die before their fifth birthday annually, 80% of deaths occurring in sub-Saharan Africa (SSA) and Asia  (Vakili et al., 2015).

Besides being a leading cause of child morbidity and mortality, malnutrition, especially undernutrition impairs child     normal     growth     and     development,    school performance and economic productivity in adulthood (Victora et al., 2008). Hence, ending all forms of malnutrition by 2030  in resonance with the Sustainable Development Goal (SDG) 2.2 target (United Nations, 2017), will significantly contribute to the achievement of other related child health goals. Malnutrition reduction strategies targeting evidence-based high-risk groups (Sultana et al., 2019)are fundamental in attaining  these goals.  

Among nutritionally vulnerable population groups are the world’s indigenous peoples, numbering more than 370 million in some 90 countries, and over 14.2 million in 35 countries in the African region (Department of Economic and Social Affairs, United Nations, 2015). Compared to benchmark populations, indigenous peoples experience higher disease burden due to persistent socio‐economic disadvantages, marginalization and limited access to health services (Department of Economic and Social Affairs, United Nations, 2015; Basnayake et al., 2017). Children in particular, tend to have worse health outcomes and die younger than their non-indigenous peers (Department of Economic and Social Affairs, United Nations, 2015; Anderson et al., 2016; Basnayake et al., 2017). Accordingly, improving the nutritional status of children, and overall health of the indigenous people is one of the global health priorities (Department of Economic and Social Affairs, United Nations, 2015).

Effective malnutrition prevention strategies among vulnerable populations hinge largely on reducing their exposure to contextual risk factors. According to the widely adopted (United Nations Children’s Fund, 1991)conceptual framework, the etiology of malnutrition is complex and multifactorial. In line with the framework, systematic reviews in sub-Saharan Africa (Akombi et al., 2017a)and Southern Asia (Akhtar, 2016)have adequately documented the principal factors that jointly explain malnutrition among under-five children. However, the magnitude and distribution of determinants vary across different geographical regions, populations, and various socio-cultural conditions, due to differences in patterns of food intake, cultural beliefs and access to health care services among others (Motedayen et al., 2019). Moreover, most reviews focused on non-indigenous populations, and related evidence for indigenous populations is scarce. Within this context, this review was undertaken to identify and map existing literature on factors that contribute to childhood anaemia, stunting, wasting and underweight among indigenous population groups in developing countries, and to identify knowledge   gaps.   This   paper    presents    a   narrative synthesis on the topic, drawn from 10 selected studies. To the best authors knowledge, this is the first and most current scoping review on the topic. The results will be relevant in informing undernutrition-reduction policies and interventions among various indigenous population groups. It is worth noting that this study is part of a wider PhD thesis titled “Assessing malnutrition and associated determinants among under-five Indigenous Mbororo children in two health districts of the West Region of Cameroon”.

Information source and search strategy

The study initially highlighted the medical subject headings (MeSH) and text terms related to the concept, context and population to guide an extensive on-line search from PubMed, Scopus and Medline databases. The MeSH and free text terms were used to construct search strings for each database. For example, using the Boolean operators (AND, OR) the following strategy was used for PubMed (Malnutrition OR Undernutr* OR Stunt* OR Wast* OR Underweight OR Anaemia [MeSH]) and (Determinant* OR Predictor OR Risk factor OR Correlate OR Cause [MeSH]) AND (Child* OR Under-five OR Infant OR Pre-school [MeSH]) and Indigenous OR Tribal OR Minority [MeSH]) AND English [lang]. We equally searched for unpublished literature from Google Scholar, World Health Organization (WHO) and UNICEF websites. A manual search for additional literature from the reference lists of selected studies was conducted.

Study selection

Inclusion and exclusion criteria

Full articles were considered for review if they reported the determinants of anaemia, stunting, wasting and underweight; focused on indigenous children under five years old were conducted in developing countries; were qualitative, qualitative or mixed-methods studies; and were published in English from 2009 to 2019. Records were excluded for the following reasons: Repeated publications; lack of access to full text of articles; lack of reference to the determinants of malnutrition; non-indigenous children; children above five years; published in languages other than English; books, policy briefs and conference papers; and studies conducted in affluent countries notably: Canada, United States of America and Australia. As presented in the PRISMA flow diagram (Figure 1), 10 of the 136 records which were retrieved met the priori eligibility criteria. Two review authors independently screened the abstracts and full articles using the eligibility criteria.

Data extraction

Data extraction was done using a data extraction template designed  by the team and pilot tested on a representative sample of articles. The first author’s surname, year of publication, country, study design, sample size, age of study population, and determinants of anaemia, stunting, wasting and underweight were extracted from included studies. Only risk factors with reported p-values, odd ratios (OR), adjusted odd ratios (AOR) and prevalence ratios (PR) were considered. Data extraction was undertaken independently and in duplicate by two authors with disagreements resolved via consensus. A narrative approach was followed to synthesize the evidences from the reviewed studies.

Characteristics of selected studies

Table 1 presents the characteristics of 10 studies that were selected for the review. Of these, 4 were  conducted in Brazil, 3 in India, 2 in Guatemala and 1 in Malaysia. Seven were cross-sectional studies, 2 were population-based studies, and 1 was a mixed method study. All studies focused on children under five years of age, ranging in sample size from 74 to 6075. As concerns anthropometric indicators, 1 study focused on anaemia, while, 2, 3, and 8 studies, respectively, reported on wasting, underweight and stunting.

Maternal characteristics

Maternal educational status

Low maternal educational status was  significantly associated with underweight and stunting in India (Meshram et al., 2012)and Brazil (Horta et al., 2013). The  odds of underweight were 1.7 times  higher among  children whose mothers were illiterates (Meshram et al., 2012), in agreement with a similar study in Ecuador (Walrod et al., 2018), and a systematic review in sub-Saharan Africa (Akombi et al., 2017). The association between maternal formal schooling and child nutritional status finds an explanation in the presumptions that, high maternal education attainment translates into greater health care utilization, adoption of modern medical practices, and child-centered caring practices that improve child nutritional outcomes (Abuya et al., 2012; Duah and Adisah-Atta, 2017). Additionally, maternal education enhances efficient management of limited household resources to provide adequate nutrition for children, and prevent childhood infectious diseases (Walrod et al., 2018). Otherwise, mothers with little or no formal  education    are  more  likely  to  have  inadequate child feeding practices, and poor sanitation and hygiene practices (Adhikari et al., 2017). Hence, improvements in access to indigenous women’s education are likely to contribute to better child nutrition outcomes.

Maternal health care seeking behaviour

Maternal health care seeking behaviour which includes utilization of maternal and child health services was reported as a significant determinant of nutritional status of indigenous children. Inadequate pre-natal care significantly predicted stunting  in Brazil (Araújo et al., 2016). Meanwhile, child growth monitoring and being up-to-date with the vaccination schedules were inversely associated with stunting (Araújo et al., 2016). In another study in India (Singh et al., 2015),  immunization against measles reduced the risk of wasting among under-five children.  These relationships  are not unexpected, as access to primary health care  has proven fundamental to the prevention of maternal and child deaths (Leite et al., 2013; Chomat et al., 2014; Basnayake et al., 2017; Dagnew et al., 2018). Regrettably, accessing health care services is a  major  problem  for  indigenous  populations worldwide (Department of Economic and Social Affairs, United Nations, 2015).  In Africa and Asia in particular, access to healthcare services is limited by several structural, social, economic, linguistic and cultural barriers (Chomat et al., 2014; Department of Economic and Social Affairs,  Nations, 2015). Accordingly, mothers may tend to have low levels of health seeking behaviour which contribute to higher rates of child under nutrition, morbidity and mortality. This finding highlights the need to ensure maternal and child health care services that are accessible, affordable, acceptable, adaptable and adequate to meet the needs of the indigenous populations at large. 

Maternal height

Low maternal stature (height <145 cm)  was highlighted as a significant predictive factor for child stunting in  Brazil (Araújo et al., 2016), India (Orellana et al., 2019)and Guatemala (Solomons et al., 2015). Children whose mothers were below 145 cm in height, were twice as likely to be severely stunted compared with  their counterparts with  taller mothers (Orellana et al., 2019). This finding is consistent with other studies among non-indigenous population groups (Khatun et al., 2018; Khan et al., 2019; Nshimyiryo et al., 2019). The positive association between maternal short stature and child stunting reflects an inter-generational problem (Orellana et al., 2019). Existing evidence suggest that, repeated height deficits for several generations, results  in premature births, low birth weights and small-for-gestational age babies (Black et al., 2013; Orellana et al., 2019). The incidence of low birth weight and being small-for-gestational age babies is common among indigenous populations (Gracey and King, 2009). Thus, improving maternal nutritional status before and during pregnancy among the indigenous populations may significantly reduce child linear growth faltering.

Child characteristics

Child age

Child age was invariably associated with anaemia, stunting and underweight (Meshram et al., 2012;Orellana et al., 2019; Ferreira et al., 2017). An inverse relationship was shown between increasing child age and anaemia, as  children less than 2 years were at greater risk of moderate and severe anaemia than their older counterparts (Ferreira et al., 2017). Whereas, children above 3 years were 0.66 times less likely to be stunted compared with those below 3 years of age (Meshram et al., 2012), similar to other studies in  Malaysia (Anto Cordelia et al.,  2019;  Orellana  et  al.,  2019),  and  India (Saxton et al., 2016).The positive correlation between stunting and increasing child age could possibly be attributed to an imbalance between the demand and supply of essential nutrients as children transit from infancy to childhood. This period is characterized by intense growth and physical activity, with correspondingly high needs for nutrients. Unmet nutritional needs during this critical period of growth could lead to poor nutrition outcomes. Additionally, young children unlike infants are  more vulnerable to diarrhoea and other gastro-intestinal enteritis due to exposure to microbial contaminants in food and the immediate household environment  (Budge et al., 2019). While child age is not a modifiable factor, ensuring adequate age-appropriate feeding, as well as food hygiene practices are crucial measures that could reduce under nutrition burden among indigenous children. 

Child sex

Child sex (male) was highlighted as a significant  risk factor for anemia in India (Meshram et al., 2012)and Brazil (Ferreira et al., 2017)underweight in Brazil (Horta et al., 2013)and Guatemala (Tumilowicz et al., 2015). Whereas, a study in India (Saxton et al., 2016)showed that girls had higher height-for-age scores than boys, male sex was a predictive factor for stunting in India (Meshram et al., 2012), Brazil (Horta et al., 2013)and Malaysia (Anto et al., 2019). These findings are consistent with  a meta-analysis of 17 demographic health surveys in sub-Saharan Africa (Wamani et al., 2007). While this relationship may not be adequately explained, some authors (Akombi et al., 2017b)suggest that females are culturally expected to be less active and stay at home with their mothers near food preparation, while males engage in high physical activities that expend high amounts of energy that would have been channeled into increasing growth. Another school of thought (Singh et al., 2015)posits that, the sex of a child is significant only in settings where there is gender discrimination regarding access to food and other intra-household resources, hence the need for interventions to address such gender inequalities.

Child morbidity status

Fever, diarrhoea and acute respiratory infections (ARI) during the preceding fortnight were significantly associated with under nutrition (Meshram et al., 2012, Horta et al., 2013). Children with ARI were 1.85 times more likely to be underweight compared with their counterparts (Meshram et al., 2012). Similarly, children with fever and diarrhoea were twice at risk of wasting than their peers without any of the conditions. The role  of  infectious diseases on child  nutritional status is well documented (Guerrant et al., 2008; Humphrey, 2009; UNICEF, 2013; Papier et al., 2014; Debnath et al., 2018). Persistent or repeated diarrhoea and infections cause loss of appetite, vomiting intestinal damage, decrease absorption of micronutrients, thereby, leading to under nutrition, which in turn impairs the immune system, thus, worsening the intensity of infections (Ibrahim et al., 2017). Moreover, the presence of infections may cause a shift in energy expenditure to combat the infectious process to the detriment of child growth.

Household socio-economic characteristics

Household income and wealth index

Two studies in India (Meshram et al., 2012; Saxton  et al., 2016)revealed that stunting rates were lower among children from  middle-income groups compared with those in the lower-income groups. Similarly, a study in Brazil (Horta et al., 2013)reported that children belonging to households with no regular  source of income were more likely to be underweight and stunted. A plausible explanation for this relationship could be attributed to the fact that, adequate income increases the ability of households to purchase and access good quality food to meet children’s nutrition needs (Titaley et al., 2019). Thus, poor households are at higher risk of food insecurity and related child growth deficits (Bain et al., 2013). Furthermore, poverty drives families to reside in resource-limited rural and urban slum areas (Goudet et al., 2017)where childhood morbidities are higher (Singh et al., 2011). Globally, indigenous populations are more likely to reside under poor living conditions (Basnayake et al., 2017). Hence, policies and interventions that seek to improve the socio-economic status of poor indigenous households in a sustainable manner are likely to improve the nutritional well-being of their children.

Inadequate water, hygiene and sanitation conditions

Household environmental factors notably the unavailability of  piped drinking water, absence of in-door sanitation facilities, and limited access to trash collection services correlated positively with child under nutrition in Brazil (Horta et al., 2013)and India (Singh et al., 2015). This could be explained by the fact that, most endemic diarrheal diseases, intestinal parasite infections, and environmental enteropathies that harm children’s nutritional status are transmitted under conditions of inadequate water, sanitation and hygiene (WASH) (Horta et al., 2013; World Health Organization, 2015; Adhikari et al., 2017; Fatihah et al., 2018). Although access to improved     household     WASH    facilities    remains    a worldwide problem (World Health Organization, 2015), the issue is more pronounced among indigenous populations (Basnayake et al., 2017; Rajoo et al., 2017; Adhikari et al., 2017). Hence, they tend to have much higher rates of infections than do their non-indigenous counterparts (Gracey and King, 2009). Ensuring adequate access and use of safe drinking-water, personal and environmental hygiene services among indigenous populations are therefore likely to improve child nutritional status.

The results of this study need to be considered in the light of a few limitations. First, the search identified only 10 eligible articles, and none from Africa. This was probably due to bias introduced by examining studies published in the English language only, and for a ten years period. Second, some other factors known to significantly predict child under nutrition such as maternal nutritional status, cultural child feeding practices, and household food security status were not included in the analysis because they were not reported in the reviewed articles.

This study reveals that indigenous populations groups share a wide range of interrelated maternal, child, socio-economic and environmental factors that contribute to child under nutrition.  The most significant predictors are low maternal educational status, low maternal height, inadequate health seeking behaviour, child sex, child age, child morbidity status, poverty and inadequate water, sanitation and hygiene practices. These findings represent important prevention points, and support holistic multi-sectorial policies and interventions to combat child malnutrition among various indigenous populations groups. Such interventions should be aimed at addressing barriers to formal educational opportunities for the girl child and women, facilitating access to mother and child care health services, promoting adequate maternal nutrition, reducing childhood infectious disease, improving WASH conditions and access to income generating opportunities.  The lack of related studies from the African region revealed by this study is equally of great concern. There is therefore need to address such knowledge gaps in order to foster child development and ensure sustainable socio-economic development particularly among the growing numbers of indigenous people in the region.

The authors have not declared any conflict of interests.

This is part of a Ph.D. thesis by Florence Titu Manjong under the supervision of Associate Professors Nsagha Dickson Shey and Thomas Obinchemti Egbe, and Dr. Vincent Siysi Verla, of the Faculty of Health Sciences, University of Buea.