Malnutrition among children in developing countries is a major public health concern since it places a heavy burden on already disadvantaged communities. Poor physical growth, an indicator of poor nutritional status, is high in sub-Saharan countries, where approximately 21.9% of children are underweight and 40.1% are stunted [1]. The most vulnerable group of children are those under 5 years of age [2,3]. Anaemia, another indicator of poor nutritional status, is also widespread, with estimates indicating prevalence rates of 40–70% in Sub-Saharan countries. Both growth restriction and anemia in the early years of life increase the risk of mortality [4,5,6] and morbidity [7,8,9,10], and are associated with developmental and cognitive impairment [11,12,13,14,15].

The link between poverty and poor nutritional status among children has been widely reported [16,17]. Varying indicators of social economic status (SES) such as maternal and paternal educational level [18,19], parental income [20], and family assets such as the ownership of land, quality of housing, and foods harvested [21,22] among many SES indicators have all been associated with children’s nutritional status. Regardless of the method by which SES was estimated, its influence on child’s nutritional status was significant and consistent. It was observed that children from less advantaged families were more likely to experience growth restriction (i.e., stunting and being underweight) compared to their peers from more advantaged backgrounds [13]. Despite this link several factors give compelling reasons to carry out further investigation into the relationship between SES and growth restriction. One of the most salient reasons for this is the fact that the prevalence of stunting and being underweight has been found to show both between- and within-country variation in sub-Saharan Africa [1,23,24]. There is thus a need examine prevalence and risk factors for poor growth outcomes in a variety of contexts so as to be able to plan targeted interventions. Compared to stunting and being underweight, relatively few studies have looked at the link between anaemia and social economic and demographic indicators, with most of the studies focusing on the relationship between anaemia and ill-health [25,26].

In this paper we report the prevalence and socioeconomic risk factors of stunting and being underweight among children aged 1–35 months in Same district, Kilimanjaro region, thus attempting to replicate earlier findings from other regions in Tanzania. Going beyond earlier findings, we evaluate the association between socioeconomic, demographic and health risk factors at the same time; a novel approach to studying the correlates of nutritional status in this region. Evaluating the different risk factors at the same time is significant since these variables are strongly correlated and have a synergic effect. Thus situating them together in the same model allows for the evaluation of the relative contribution of each in the growth status of the children. Moreover, we investigate the correlates to being anaemic and evaluate the relationship between anaemia and anthropometric status.

Our study indicates that there is a high prevalence of anaemia, stunting and being underweight among children in Kilimanjaro region of Tanzania. The high rates observed are similar to those observed elsewhere is East Africa, where up to 40% of the children have been observed to experience stunting [13]. The mean anthropometric status of this population is significantly below the WHO references populations which is generally consistent with findings from other parts of East Africa [13,33]. We observed that the risk factors for each of these nutritional deficiencies varied. For instance, stunting was closely associated with several sociodemographic factors in contrast to being underweight which was much more closely related to child’s health. These findings are consistent with the different aetiology of these two nutritional status indices and are consistent with earlier findings. HAZ is usually compromised following a period of chronic malnutrition, while being underweight is an indicator largely of short term nutritional compromise [2,31]. These results indicate the need to monitor different indicators of a child’s nutritional status as part of the health monitoring programmes to be able to provide adequate and timely intervention.

Consistent with earlier reports we observed that age was a significant risk factor for poor physical growth [33]. These results emphasize the need to implement early intervention to reduce the number of children experiencing growth restriction. Moreover, the results further emphasized the need to investigate factors that contribute to continued growth faltering during toddlerhood. Regional based studies of factors contributing to early growth faltering may be salient since growth faltering has been associated with socio-cultural and feeding practices during the weaning period factors that are likely to vary on a regional basis.

We observed high rates of anaemia in our sample which is consistent with findings from other developing countries [34]. Moreover, the significant relationship between anaemia and maternal education has also been reported in earlier study from setting similar to ours [35,36]. Further investigation on the exact mechanism by which maternal education influence the child’s anaemic status is warranted. We hypothesize that maternal education may simply be a proxy for other factors such as childrearing practices, health seeking behaviour or feeding practices which significantly affect the children’s health. For instance, in Korea [36], it was observed that children of mothers who were better educated were less likely to be anaemic largely because better educated mothers were more likely to adopt healthier feeding habits.

Our results indicate that parental concerns of child’s growth and development was a strong predictor of child anthropometric status. These results have clear practical implications. The association between parental concerns and child anthropometric status provides validation for parental concerns and raise the question: to what extent can health workers utilize parental concerns as an indicator of child health? Can parental concerns be used as a screening instrument to identify children in need of closer monitoring, surveillance and intervention? In other fields dealing with child well-being, e.g., child development parental reports of concerns with child growth and development have been used to identify children who may be at-risk of poor developmental outcomes [37,38]. Based on these finding it can be recommended that as part of the post natal health services, health workers need to discuss and address parental concerns. This approach has two major benefits: a) it may potentially provide a quick and relatively cheap screening tool for health workers and b) allows the parent to become an active participant in monitoring the child’s health status.

The most consistent indicator of child nutritional status was the maternal educational level; it was the only factor that was observed to significantly predict all the three nutritional indicators. The influence of maternal education could arise from various sources key amongst them is the impact it has on economic means (higher educated mothers have better jobs and more money) or more directly its influence on childrearing practices since mothers with more education have been observed to provide optimal care for their children. Regardless of the source of this influence our results further emphasize the need to invest in girls’/mothers’ education as a way to enhance child wellbeing in developing countries [39,40].

The current study adds to the body of knowledge linking social, economic and demographic factors to the nutritional status of children in different settings. However, this study is cross-sectional in nature therefore limits our ability to make inferences on causation. Additionally, we did study other potentially salient predictors of nutritional outcomes. For instance we did not evaluate the impact of ill-health on nutritional status. Yet, anaemia has been strongly linked to ill health and especially helminthic infections [41] and malaria particularly in malaria endemic areas [42]. This forms another limitation of our study.

In conclusion, we observed that high rates of nutritional deficiency and these deficiencies were related to maternal, individual and household characteristics. However, the most consistent risk factor was observed to be maternal education further emphasizing the need to invest in women education as an intervention for enhancing child wellbeing. The current study further emphasizes the need to implement context relevant interventions to combat malnutrition in this region of Tanzania and other similar settings.