The objective of this study was to present the distribution of wasting according to socio-demographic circumstances with a focus on household food insecurity and to measure the risk of wasting as determined by those risk. In this study, the prevalence rate of wasting among assessed children was noted higher among children from food insecure than among children from food secure households since 60 percent of the children under five years of age belonged to food insecure households. Of that 60 percent, mildly food insecure household covered one-third (33.5%), moderately and severely covered 15.7 and 10 percent respectively. Additionally, the prevalence of wasting among children under five years increased with levels of food insecurity as the adjusted odds ratio (as well as prevalence) increased from 0.98 (95% CI 0.64–1.50) to 1.14 (95% CI 0.65–1.98) and to 1.36 (95% CI 0.71–2.57) from mild (prevalence 9.4%) to moderate (prevalence 10.8%) and severe (prevalence 11.3%) food insecurity respectively, which is in accordance with the study of John et al. reporting a dose-response relation between severity of food insecurity and children’s nutritional health [14]. In addition to this, moderate and severe food insecurities were also significantly associated with wasting in low and middle income countries such as Bangladesh and Vietnam in the adjusted analysis [15]. However, no significant association was found in the present study although the prevalence increased incrementally with food insecurity. This was surprising given the already established association of food insecurity and children nutritional status by UNICEF conceptual framework of undernutrition. This might have been due to the small intersection sample size of wasted children and children belonging to food insecure households. A larger study is recommended to further understand this relationship in Nepal.
The thematic report on food security and nutrition had shown greater prevalence of wasting in female (36.1%) than male children (33.8%), which contradicts the findings of the present study showing similar prevalence of wasting among male (9.5%) and female (9.7%) children. Harding et al., however, reported that sex was not significantly associated with wasting in Nepal aligning with the present study [16]. They also noted that the association was noted significant in data of other low income countries such as Afghanistan, India, Bangladesh, and Pakistan [16].
Less variation was noted in the prevalence of wasting among the wealth quintiles in 2016. The maximum prevalence was noted in richest quintile (7.2%) and minimum prevalence was noted in richer quintile (11.3%) compare to the variation of 7.4 percent (in richest quintile) as maximum to 12.8 percent (in middle quintile) as minimum in 2011. Similarly, while the under five years children belonging to the poorest quintile had prevalence of 8.7 percent, which is unexpectedly close to the prevalence of wasting among children belonging to the highest (richest) wealth quintile (7.2%) and also represents the second most lowest prevalence rate among the wealth quintiles. In 2011 NDHS reported the prevalence of wasting among children under five peaked at middle (12.8%) and then at the poorest (12.5%) wealth quintiles [6]. The lowest prevalence shifted from middle quintile in 2011 to the richer quintile in 2016. This might be due to a shift of overall population belonging to the poorest and poorer wealth quintile towards middle and richer wealth quintile as a part of a rise in economic condition as characterized by data on Gross Domestic Product from The World Bank [17]. The Gross Domestic Product has risen from 18.9 to 21.2 to 28.8 from 2011 to 2016 and to 2018 respectively [17]. Also, no significant relationship was found between wasting and wealth quintiles in this study as in other previous studies [18–23], which could have been affected by the above mentioned statement on less variation in prevalence and shift of wasting pattern in terms of wealth quintiles. This finding is supported by a study on trends in stunting with four NDHSs, which has also shown reduction from 25.7 to 20.5 percent (from 2011 to 2016) in proportion of stunted children under five years age among those belonging to a poorer wealth quintile whereas an increment was observed among those belonging to a richer wealth quintile from 16.9 to 21.7 percent from 2011 to 2016, [21] similar to the present study. However, a multinational cohort study conducted in Ethiopia, India, Peru and Vietnam found children belonging to the lowest wealth quintile households had significantly increased probabilities of being wasted in all four countries in comparison to children belonging to the highest wealth quintile households [24]. This is not consistent with the result of the present study having noted higher prevalence of wasting among middle and richer wealth quintiles children.
The highest prevalence of wasting was also noted among Province 2, which might be due to large volume of underprivileged people suffering from lack of basic facilities such as education and health [25]. Additionally, Nepal Multidimensional Poverty Index 2018 reports more than 47.9 percent of people living in Province 2 (second highest populated province holding 18.4%) are multidimensionally poor, which is greater by almost 20 percent than that of the national average (28.6%) [26].
Additionally, the Demographic and Health Surveys of 15 sub-Saharan African countries showed that urban–rural differentials are considerable in all countries, that they have narrowed in most countries primarily due to an increase in urban undernutrition [27]. An another study done in Nepal mentioned that a significantly higher prevalence of wasting was noted in rural areas as compared to urban areas [28]. Aligning with these studies, the prevalence of wasting in the present study was slightly higher among children living in rural (10.1%) than those living in urban (9.2%) and the highest proportion of wasted children were born to a mother without education (12.4%), which is similar to the finding of a thematic report on nutrition and food security stating the highest proportion (43.3%) of the undernourished children born to a mother without education [28]. A study done by Asfaw and team in Southern region of Ethiopia shows a significant association between mother's education and all three indicators of undernutrition (stunting, wasting and underweight) [29] again in agreement with the present study. A significant relationship was also found between wasting and provinces in the present study. A higher prevalence was found in Province 1 and 2. The odds ratio after the adjustment fell from COR 2.18 (95% CI 1.06–4.50) to AOR 2.06 (95% CI 1.01–4.19); however, it remained significant.
Due to cross-sectional nature of the study, the causal inference between wasting and study variables could not be estimated. However, this study has given provincial level information on wasting, which is not available in other studies of Nepal and is highly beneficial for formulating provincial level policies. The provincial level information on wasting would bring the focus on the current need of each province and this information would support the design of intervention within various provinces. The study warrant a high precision of the findings due to large sample size representing national population. Reliable and comparable standardized tools were used.