Developing a model for prevention of malnutrition among children under 5 years old

Background Serious consequences leading to the increase of infectious diseases and mortality of children justifies the importance of interventions for eradication of malnutrition. Thus, this study aimed to provide a model for the prevention of malnutrition among children under 5 years old (CU5) in Iran. Methods This paper is part of a bigger study, conducted in 2017 using a mixed methods approach. A model for prevention of malnutrition in CU5 was proposed based on the earlier steps, with the cooperation and consultation of experts and specialists. In the final step, a Delphi method was used to determine the validity of the proposed model. Results The main dimensions of the prevention model of malnutrition for CU5 in Iran included four level: basic causes, interventions, outcomes and impact. The proposed interventions are presented based on twelve areas: structural, intersectoral, political, economic, sanitary, health-oriented, research, educational/cultural, evaluation related, production, infrastructures and legal. Based on these areas, 118 solutions were finally selected for the final model. This model is designed based on the current conditions in different regions of Iran, the factors related to child malnutrition, affective context on policy making, the content of previous policies, the process of policy making in Iran, and key stakeholders and actors in policy making. Conclusion In order to prevent malnutrition, the causing factors should be identified and resolved. The adopted policies should be, more seriously, based on the presence of key stakeholders and actors. Most of the existing nutritional problems among children are because of inappropriate consumerism culture and habits in families and its transfer to children.

Under nutrition exposures children to a higher risk of dying from common infections and slower recovery when infected [9]. Malnutrition among children also affects their cognitive-sensory function and consequently disrupts their ability to have a productive and efficient life [10].
This type of malnutrition is caused by many factors such as inadequate care in pregnancy period, low literacy of family members, inadequate community malfunctions (injustice, war, natural disasters, etc.), polluted environments, poor nutrition of the household, frequent and severe infections, inadequate supply of food, and poverty, especially [11]. Despite the importance of malnutrition and its relevance to the cause of important diseases, this issue continues to be underestimated [12] .
Considering the fact that stunning, underweight and wasting are among the most important indicators of malnutrition, the current study aimed to use these indicators to investigate malnutrition among children. In 2018, globally 149 million CU5 (22%) were stunted [13] and 49.48 (7.3%) million wasted [14]. The prevalence of underweight, wasting and stunning in Iran, according to the latest national study were 3.8, 4 and 4.6%, respectively [15]. Although malnutrition prevalence has been declining in recent years, Iran is still far from a World Without Malnutrition [9].
Using the indigenous models, specific for each region of world, needs to be considered to reduce malnutrition and improve the situation. Hence, given the importance of malnutrition in children, the aim of this study was to develop a model for the prevention of malnutrition among CU5 in Iran.

Methods
This paper is a part of a bigger study that was conducted in 2017 using a mixed methods (quantitative and qualitative) approach in four steps (Fig. 1). The first and second steps were a systematic review and a policy analysis conducted and published earlier [2,5,6]. In the third step of the study, based on the findings of the first two steps, a malnutrition prevention model was developed for the CU5, enjoying the cooperation and consultation of experts and specialists. In the final step of the study, a Delphi method was used to determine the validity of the model. The Delphi technique is an approach for gathering data used to gain consensus among respondents within their domain of expertise. This is achieved through a series of rounds using questionnaires (Table 1), where information is fed back to panel members [16,17]. The initially proposed model was formulated in the form of a questionnaire. For this step, the study sample consisted of 38 key policy makers, senior health system managers, specialists and faculty members of nutrition departments of universities. The inclusion criteria were as follows: the individuals with education and research history on nutrition, especially CU5, senior managers with the history of conducting research (at least 5 years), individuals with the experience of executive jobs (at least 5 years), and finally, having effective experience related with policy making in this area. Snowballing method was also used to detect the eligible samples for entering the study.
Data collection tool comprised of three sections: demographic information of respondents, policy options, and also a section for recording the participants' responses about the proposed options. The validity of the questionnaire was evaluated through the opinions of experts in this field. To assess the reliability, a pilot study was done on five individuals. A summary of the objectives along with a proposed model was sent to the experts in two rounds. The results of the first phase were analyzed and a revised model was developed based on the comments. It was then sent back to the experts and to collect their final comments. To prepare the final model, the Delphi step findings were analyzed by Fig. 1 Research steps. The final model is the result of a mixed methods approach in four step. In the final step, a Delphi method was used to determine the validity of the proposed model descriptive method. If the agreement was over 75%, the policy option was approved and if the agreement was between 50 and 75%, the policy option entered the next round. Moreover, the agreement below 50% would lead to the rejection of options [18]. At the end, the final model was formulated based on the approved dimensions and aspects.

Results
In order to design the primary model (Fig. 2), the findings of the first phase of the study [5], the factors related to malnutrition in Iranian CU5, were classified into six categories: the social, economic, biological, environmental, child-related and family-related causes. The findings of the meta-analysis study [2] showed that the prevalence of malnutrition was higher in the deprived regions of the country. Therefore, different regions of the country need different interventions; some regions need urgent and short-term interventions while some preventive and supportive interventions, due to their low economic situation and higher prevalence compared with the standards assigned by World Health Organization. These issues are considered in the present model. Given the cultural, ethnic and regional differences, the level of education and different food styles in different regions of the country, we need individual and social policies which are included in the final model. The findings from the interviews [6] are also used in more details for the regions in need of intervention.
The first and second phases of the Delphi attended by 25 and 20 individuals respectively. In Table 2, the proposed interventions are presented within the following twelve areas: structural, intersectoral, political, economic, sanitary, health-oriented, research, educational/cultural, evaluation related, production, infrastructures and legal. Based on these domains, 118 policy options were finally selected for preventing malnutrition among CU5.

Discussion
The prevalence of malnutrition among Iranian CU5 is much lower than the global average and is in relatively good condition. About contextual causes, mother education level, father education level, child gender, birth weight, and age group were mentioned as the most important factors in the literature. Based on evidence, the main factors in malnutrition are low socioeconomic status of family, parental education level, household health index, health literacy, nutrition culture, maternal characteristics such as BMI, nutrition during pregnancy, and the number of childbirths, child characteristics such as age, gender, birth weight and common infectious diseases [19][20][21][22][23][24][25].
Considering that nutritional problems are multifactorial, different creating causes need to be resolved to prevent them [26]. The problem of food insecurity or malnutrition in a region would be solved only if  Developing and enforcing the laws which protect of high-quality food producers all its contextual, and mediating and immediate factors are addressed [27]. The interventions are suggested to be classify based on the area, scale, type, or time to reduce the children malnutrition. Considering the contextual causes, the most important intervention could be maternal measures. Maternal interventions, particularly during pregnancy, may have intergenerational effects and on birth weight and child growth [28]. Other interventions in the studies include providing nutritional aids and paying cash, micronutrient supplements such as vitamin A, fertility and children health (immunization, prenatal care, the presence of a skilled person at childbirth, and the treatment of childhood common diseases), hand washing program, supplement foods enriched with micronutrient [29][30][31][32][33]. Another powerful potential strategy to prevent malnutrition in poor and sick children is consuming nutritional supplements [34]. Studies have also shown that the prevalence of malnutrition in lower social classes is higher [35] and by improving the economic status and reducing poverty, the chance of a better nutrition at higher income levels is higher. In this regard, some studies have shown a relationship between income and a reduction in the risk of stunting [36][37][38][39].
Low cost and time are the two points mentioned in previous studies regarding the implementation of interventions. Results of the study by Peru et al. showed that nutritional interventions are low cost and may improve growth in the short or midterm [40]. Clombati et al. also showed that in countries with a high prevalence of malnutrition, low cost and short-term interventions are easily applicable and effective [41]. Furthermore, Shimpton et al. showed that these interventions should start from the early pregnancy or at birth [42].
Nevertheless, results of some studies showed that some interventions had low impact. The results of three meta-analysis studies showed that the use of micronutrient as an intervention was effective in improving the development of children, but iron and vitamin supplements did not have a significant effect in the improvement of children's growth [43]. A study by Bandari et al. in rural areas of India revealed that educational interventions aimed at increasing energy consumption and improving nutritional methods for infants from 6 to 18 months could not improve their weight [44]. Some double-blind controlled studies also showed that zinc intake could not improve the growth of malnourished children [45].
In general, it seems that screening in short intervals, the status of children can be monitored regularly and the effectiveness of interventions can be controlled in order to modify them. It should also be noted that health policies, including nutrition, differ greatly from other policies, and interests and objectives of the governors have a direct impact on nutrition policies [46]. Some studies showed that strong political support programs were of the main strengths of nutritional policies; and such support could improve the nutritional status of the children [47,48].
The findings of policy analysis in the current study showed that using evidence, documents, policies, previous experiences, and experiences of other countries, a successful model for preventing child malnutrition can be developed for the country. A model that, in addition to the children's health and nutrition needs, involves all individuals, organizations and stakeholders, considers the health of mother and child before the childbirth and has the ability to align the activities of all stakeholders. Furthermore, a nationwide program and a comprehensive document in the field of the nutrition for CU5 is needed; to do so, strengthening of the political process is crucial [49].
Some studies presented a model for preventing and reducing malnutrition in children. Using appropriate models to improve the quality of services in different organizations can be considered to develop the final model [50][51][52]. Different countries around the world may design specific models based on their epidemiological situation or use a combination of several models. The models of UNICEF and WHO are among most important models and frameworks for prevention and control of malnutrition. In the UNICEF model consists of three levels of causes: basic, underlying, and immediate. Each of these levels contains components that need to be considered to prevent and control malnutrition [26]. The WHO model addresses the context and causes of malnutrition. Some other models, in addition to context causes, consider children and mothers as two main pillars, and necessitate a deeper understanding of cultural models in each region to formulate the needed programs [53,54].
Policy options of the final model for malnutrition prevention in CU5 in Iran are presented based on educational, research, structural, economic, health-oriented, hygiene, political and inter-sectorial dimensions. This model was developed based on the current conditions in different regions of the country, factors related to child malnutrition, context effecting policy making, the content of previous policies, the process of policy making in Iran, and stakeholders and actors effective in policy making.

Conclusion
The findings showed that in order to prevent malnutrition, first, the context causes should be identified and resolved. Context causes, solutions and interventions may differ in different regions of the country. The interventions can be supportive and health care related. The adopted policies should be strongly based on key stakeholders and actors. One of the most important needs in children nutrition, especially CU5, is formulating a comprehensive national document designed for this age group. The issue of targeted subsidies, helping poor people, job creation and production are also so important and need to be considered. Economic empowerment will ultimately lead to an increase in welfare of the family and the improvement of the nutritional status of children. One of the important strategies for improving economic conditions of the community is the balanced distribution of resources. Also, the correct reinforcing of laws will improve many indicators and ultimately improve the nutritional status of children. In many cases, there are proper and comprehensive rules for resolving issues but they are not correctly implemented. Finally, nutritional culture and literacy need to be considered. Low nutritional literacy will result in inadequate or inappropriate nutrition, malnutrition and other complications in children. Many of the current nutritional problems are due to wrong consumption culture in family that transfers to children. Solving the abovementioned problems and meeting the desired goals require strong cooperation between the organizations involved in the nutrition of children; and it can be achieved with sufficient expertise and being commitment to the goals.

Consent to publication Not Applicable.
Authors' contributions MM and AA contributed to the concept, design, analysis and interpretation of the data. MM contributed to the writing of the manuscript. All authors have read and approved the final manuscript.

Funding
This study was part of a PhD thesis supported by Iran University of Medical Sciences (grant No: IUMS/SHMIS_1394/9221557201). The funder did not have role in any process of the study.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate
This study was approved by the Ethics Committee of Iran University of Medical Sciences (IR.IUMS. REC1394.9221557201). Written informed consent was obtained from all of participants, all adult, for interviews. The interviews were confidential.