ABSTRACT

The aim of this article is to analyze school lunch programs and their role in the healthy nutrition of children in Córdoba, Argentina from 2013 to 2018. A descriptive, cross-sectional study was conducted that included a sample of 10 schools in 2013 and 10 different schools in 2018. A survey of school lunch programs and 24-hour nutrition reminders was carried out with 341 children. The nutritional value of school lunch programs decreased over the study period, particularly among schools in contexts of greater social vulnerability. Significant reductions in average calcium and energy intake were also observed. Indicators of chronic malnutrition, overweight, and excess calcium and vitamin A and C intakes worsened among children attending school lunch programs. Regarding the association between school lunch program attendance and low height for age/risk of low height for age, the odds ratio was not statistically significant. We argue for the necessity of promoting policies that seek urgent improvements in child nutritional indicators from a rights-based perspective.

KEYWORDS:
School Lunches; Child Nutrition; Food Services; Argentina

BACKGROUND

In Latin America and the Caribbean, overweight in children is an alarming health concern. Overweight continues to rise, affecting 7.3% of the population under 5 years of age, a figure that exceeds the world average of 5.6%.¹1. Organización de las Naciones Unidas para la Alientación y la Agricultura, Organización Panamericana de la Salud, Programa Mundial de Alimentos, UNICEF. Panorama de la seguridad alimentaria y nutricional en América Latina y el Caribe. Santiago: FAO, OPS, WFP y UNICEF; 2018. In Argentina, based on recent data provided by the Second National Health and Nutrition Survey (ENNyS 2, for its Spanish acronym), 41.1% of the population between 5 and 17 years of age was overweight.²2. Secretaría de Gobierno de Salud, Ministerio de Salud y Desarrollo Social. 2° Encuesta Nacional de Nutrición y Salud (ENNyS 2). Buenos Aires: Ministerio de Salud y Desarrollo Social; 2019.

Moreover, short stature is another health concern, being closely related to inequalities and poverty.¹1. Organización de las Naciones Unidas para la Alientación y la Agricultura, Organización Panamericana de la Salud, Programa Mundial de Alimentos, UNICEF. Panorama de la seguridad alimentaria y nutricional en América Latina y el Caribe. Santiago: FAO, OPS, WFP y UNICEF; 2018. In Argentina, in the period 2018-2019, short stature affected 7.9% of the population under 5 years of age.²2. Secretaría de Gobierno de Salud, Ministerio de Salud y Desarrollo Social. 2° Encuesta Nacional de Nutrición y Salud (ENNyS 2). Buenos Aires: Ministerio de Salud y Desarrollo Social; 2019. As several authors argued, malnutrition in its multiple forms is currently a public health challenge.³3. Kac G, García Alvear JL. Epidemiología de la desnutrición en Latinoamérica: situación actual. Nutrición Hospitalaria. 2010;25(Supl 3):S50-S56.

The Third Global School-Based Student Health Survey of 2018, which assesses the risk factors in adolescents between 13 and 15 years of age in Argentina, showed that 19% of them were overweight while 2.6% were obese, a figure that has been increasing since 2012.⁴4. Ministerio de Salud de la Nación. Encuesta Mundial de Salud Escolar 2018 [Internet]. Buenos Aires: Ministerio de Salud y Desarrollo Social; 2018 [citado 10 mar 2019]. Disponible en: Disponible Disponible en: Disponible https://tinyurl.com/y4gp2mle .
https://tinyurl.com/y4gp2mle...

In Argentina, in the last decades, there has been an important process of recognition of the rights of children and adolescents, such as the 1990 International Convention on the Rights of the Child, and the enactment in 2005 of Law 26061 on the Comprehensive Protection of the Rights of Children and Adolescents. With regard to the right to food, Argentina has the legal protection of Law 25724, which gave rise to the National Food and Nutrition Program and later on, to the National Food Security Plan (PNSA, for its Spanish acronym), implemented in 2003 with the aim of facilitating access of the socially vulnerable population to adequate, sufficient food that best suits the customs and particular characteristics of each region of Argentina. In Article 13 of the Law, reference is made to the express guarantees of regular execution of the funds destined to school feeding programs for each jurisdiction.

Furthermore, Law 26396, in its Article 8, focuses on the coordination between the Ministry of Health and the Ministry of Social Development to ensure that school feeding programs comply with nutritional requirements, placing special emphasis on the correction of nutrient deficiencies or excesses, and taking into account the particular features of the local food culture.

Today, the nutritional policies promoted by the State play a central role in achieving healthy eating in children. One of the oldest and far-reaching food policies in Argentina are the school feeding programs. Since the 1990 decentralization, the provinces and/or the municipalities have been in charge of their implementation, and based on evidence, there have been significant changes at their different levels of implementation.⁵5. Veleda C, Repetto F, Díaz Langou G, Bezem P, Sánchez B, Cano E. Comer en la escuela: nueve acciones para mejorar la gestión de los comedores escolares [Internet]. CIPPEC; 2014 [citado 10 mar 2019]. Disponible en: Disponible en: https://tinyurl.com/y287njol .
https://tinyurl.com/y287njol... ⁾ Although these policies have been in the limelight in the last decades, the oldest experiences date back to the beginning of the 20th century, as is the case of the School Feeding Program in Uruguay. Other programs were created in the 1950s and 1960s, such as the National School Feeding Program in Brazil, the School Feeding Program in Argentina and the School Feeding Program in Chile.⁶6. Cohen E, Franco R. Seguimiento y evaluación de impacto de los programas de protección social basados en alimentos en América latina y el Caribe [Internet]. Santiago; 2005 [citado 10 mar 2019]. Disponible en: Disponible en: https://tinyurl.com/y4897q7b .
https://tinyurl.com/y4897q7b...

In recent years, the analysis of the literature on school feeding at the global, regional and national levels has evidenced the barriers that exist when trying to implement healthy school meals in line with the state policy.⁷7. Reilly K, Nathan N, Grady A, Wu J, Wiggers J, Yoong SL, Wolfenden L. Barriers to implementation of a healthy canteen policy: A survey using the theoretical domains framework. Health Promotion Journal of Australia. 2019;(Suppl 1):S9-S14. doi: 10.1002/hpja.218. The difficulties in complying with the regulations⁸8. Varman S, Bullen C, Tayler-Smith K, Van Den Bergh R, Khogali M. Primary school compliance with school canteen guidelines in Fiji and its association with student obesity. Public Health Action. 2013;3(1):81-84. doi: 10.5588/pha.12.0063.^,99. Gabriel CG, Vasconcelos FA, Andrade DF, Schmitz BA. First law regulating school canteens in Brazil: evaluation after seven years of implementation. Archivos Latinoamericanos de Nutrición. 2009;59(2):128-138. and the factors that affect the menu planning are highlighted.¹⁰10. Drummond C, Sheppard L. Examining primary and secondary school canteens and their place within the school system: a South Australian study. Health Education Research. 2011;26(4):739-749. doi: 10.1093/her/cyr036. The studies found consistently show a less than optimal quality in the food services provided, evidencing remarkable deficiencies in the provision of micronutrients and energy.⁹9. Gabriel CG, Vasconcelos FA, Andrade DF, Schmitz BA. First law regulating school canteens in Brazil: evaluation after seven years of implementation. Archivos Latinoamericanos de Nutrición. 2009;59(2):128-138.^,1111. Ministerio de Sanidad y Política Social. Programa piloto escolar de referencia para la salud y el ejercicio contra la obesidad - Programa PERSEO [Internet]. s/f. [citado 10 mar 2019]. Disponible en: Disponible en: https://tinyurl.com/yxzhsyx8 .
https://tinyurl.com/yxzhsyx8... ^,1212. Moyano D, Perovic NR. Nutritional contribution the school canteens program to the children of ten municipal schools in the city of Cordoba, Argentina. Revista de la Facultad de Ciencias Medicas. 2018;75(3):194-202.^,1313. Buamden S, Graciano A, Manzano G, Zummer E. Proyecto “Encuesta a los Servicios Alimentarios de Comedores Escolares Estatales” (PESCE): alcance de las metas nutricionales de las prestaciones alimentarias de los comedores escolares de Gran Buenos Aires, Argentina. Diaeta (Buenos Aires). 2010;28(130):21-30.^,1414. Bracamonte S, Carranza C, Olivero I, Laquis M. Calidad nutricional y consumo de los menúes del comedor escolar y estado nutricional de los alumnos del nivel primario de la escuela Jorge Peyrano, Ciudad de La Calera, Provincia de Córdoba, 2012. Córdoba: Escuela de Nutrición, Universidad Nacional de Córdoba; 2012. Several authors suggest that in Latin America, the nutritional scenarios with a double burden of malnutrition are reflected in school meals.¹⁵15. Aguirre P. Pobres gordos, ricos flacos: la alimentación en crisis. Buenos Aires: Capital Intelectual; 2004.^,1616. Aguirre P. Reflexiones sobre las nuevas formas del hambre en el siglo XXI: La obesidad de la escasez. Boletín Científico Sapiens Research. 2011;1(2):60-64.^,1717. Del Castillo SE. La situación nutricional de la niñez en Latinoamérica: entre la deficiencia y el exceso, de brecha nutricional a deuda social. Biomédica. 2012;32(4):471-473.

However, there is background information that shows that healthy interventions in the context of the school feeding programs could positively impact on food selection,¹⁸18. Clinton-McHarg T, Janssen L, Delaney T, Reilly K, Regan T, Nathan N, et al. Availability of food and beverage items on school canteen menus and association with items purchased by children of primary-school age. Public Health Nutrition. 2018;21(15):2907-2914. would be effective in reducing the risk of obesity,⁸8. Varman S, Bullen C, Tayler-Smith K, Van Den Bergh R, Khogali M. Primary school compliance with school canteen guidelines in Fiji and its association with student obesity. Public Health Action. 2013;3(1):81-84. doi: 10.5588/pha.12.0063.^,1919. Ministerio de Sanidad y Política Social. Informes técnicos de evaluación del programa PERSEO de promoción de la alimentación y la actividad física saludables en el ámbito escolar. España: AESAN; 2009. and favorably influence in decreasing unhealthy food consumption and increasing fruit consumption¹⁹19. Ministerio de Sanidad y Política Social. Informes técnicos de evaluación del programa PERSEO de promoción de la alimentación y la actividad física saludables en el ámbito escolar. España: AESAN; 2009. that can be sustainable over time.²⁰20. Reilly KL, Nathan N, Wiggers J, Yoong SL, Wolfenden L. Scale up of a multi-strategic intervention to increase implementation of a school healthy canteen policy: findings of an intervention trial. BMC Public Health. 2018;18(1):860. Nevertheless, in the local context, during 2013 and 2018, when this work was developed, there were no apparent government initiatives regarding the school feeding programs beyond the provision of food.

Earlier international and national research works show that the meals served to children attending school feeding programs do not meet the nutritional goals required to satisfy the nutritional needs in childhood,¹²12. Moyano D, Perovic NR. Nutritional contribution the school canteens program to the children of ten municipal schools in the city of Cordoba, Argentina. Revista de la Facultad de Ciencias Medicas. 2018;75(3):194-202.^,2121. Garbayo Solana J, Tomás Herrero N, Navarro Elizondo M, Sánchez Hernández B, Murillo Zardoya R, Lorente Simon M, et al. C-22. Influencia de los comedores escolares en la nutrición de niños de 4 a 6 años. Pediatría Atención Primaria. 2011;13(Supl 20):S40-S41.^,2222. Ongan D, Inanc N, Cicek B. Comparing school lunch and canteen foods consumption of children in kayseri, Turkey. Pakistan Journal of Medical Sciences. 2014;30(3):549-553. and exhibit a deficit in vegetable, fruit and legume intake ⁽²⁴24. Castro M, Ríos-Reina R, Ubeda C, Callejón RM. Evaluación de menús ofertados em comedores escolares: comparación entre colegios públicos, privados y concertados. Revista de Nutrição. 2016;29:97-108., although the theoretical food offer of the school feeding programs may be adequate.²³23. Campos Díaz J, Rodríguez Álvarez C, Calvo Pacheco M, Arévalo Morales MP, Sierra López A, Arias Rodríguez A. Valoración nutricional de los menús escolares de los colegios públicos de la isla de Tenerife. Nutrición Hospitalaria. 2008;23:41-45.

In Argentina, the provinces and/or municipalities²⁵25. Centro de Implementación de Políticas Públicas para la Equidad y el Crecimiento (CIPPEC). Comer en la escuela: nueve acciones para mejorar la gestión de los comedores escolares. Buenos Aires: CIPPEC; 2014.⁾ are in charge of the implementation of the school feeding programs, and specific data indicate that approximately 4.5 million children, most of them from vulnerable sectors, eat at school.²⁵25. Centro de Implementación de Políticas Públicas para la Equidad y el Crecimiento (CIPPEC). Comer en la escuela: nueve acciones para mejorar la gestión de los comedores escolares. Buenos Aires: CIPPEC; 2014. The few evaluated experiences regarding the school feeding programs at the regional level, and especially in South America, make it difficult to assess the actual impact of this policy on the nutrition of the target population. In Argentina, there are few published studies on the topic; one of them is an analysis conducted in 1990²⁶26. Organización Panamericana de la Salud. Evaluación de un programa de alimentación escolar: el caso argentino. Washington: OPS/OMS; 1990., almost three decades ago. Without evaluative research, there is little chance of achieving sustained improvement in managing this situation and of verifying the compliance of objectives and goals.²⁷27. Abeyá Gilardon EO. Una evaluación crítica de los programas alimentarios en Argentina. Salud Colectiva. 2016;12(4):589-604.

This study is framed in the doctoral thesis of the first author of this work, entitled “La política pública alimentaria y nutricional destinada a la población infantil: un análisis de la implementación y del impacto de los Comedores Escolares de Córdoba, Argentina” [“Food and nutrition public policy directed to children: an analysis of the implementation and impact of school feeding programs in Córdoba, Argentina”]. The aim of this study was to analyze the intervention of the school lunch program policy and its role in the healthy nutrition of children of Córdoba, Argentina in the periods 2013 and 2018.

METHODS

Study design

A descriptive and cross-sectional study was carried out, based on a primary data source collected during the year 2018, and a secondary database with data collected in 2013.²⁸28. Britos S, Virgolini M, Saraví A, Chichizola N, Moyano D, Perez L, Arroyo Parisi S. Análisis de la alimentación en el ámbito escolar. Anuario Becas de Investigación “Ramón Carrillo-Arturo Oñativia”. 2013;(3):241. A comparison of the nutritional quality variables of school feeding programs and the food intake of children was sought during two survey periods over a 5 year-period. During the year 2018, the comparison between school lunch program attendees and non-attendees was deepened with the incorporation of nutritional status indicators.

The data obtained at two different times enabled to make comparisons in a period in which important changes were observed in the Argentine social and political situation, being a useful and important analysis for public health monitoring and surveillance as well as for State interventions.

Sampling and units of analysis

The universe consisted of 37 municipal schools in Córdoba, Argentina, their respective school feeding programs and the services provided on two non-consecutive days during the same survey year. A convenience sample of 20 schools was selected from this universe: ten of these schools were surveyed in 2013,²⁸28. Britos S, Virgolini M, Saraví A, Chichizola N, Moyano D, Perez L, Arroyo Parisi S. Análisis de la alimentación en el ámbito escolar. Anuario Becas de Investigación “Ramón Carrillo-Arturo Oñativia”. 2013;(3):241. and ten different units in 2018. In order to ensure the feasibility of the study during the two survey periods, non-probability sampling was implemented, in which school quotas were established according to socioeconomic level, taking household crowding as a proxy variable, in accordance with the National Population, Households and Dwellings Census, 2010.²⁹29. Instituto Nacional de Estadística y Censos. Censo Nacional de Población, Hogares y Viviendas 2010 [Internet]. Buenos Aires: INDEC; 2010 [citado 10 may 2019]. Disponible en: Disponible en: https://tinyurl.com/y383xk6w .
https://tinyurl.com/y383xk6w... Considering the feasibility of the study and the possibility of obtaining a minimum sample size that would enable to conduct the proposed analyses, children from each school were randomly selected, with the aim of reaching minimum quotas composed of at least 15 schoolchildren of both sexes. During the 2013 sample collection²⁸28. Britos S, Virgolini M, Saraví A, Chichizola N, Moyano D, Perez L, Arroyo Parisi S. Análisis de la alimentación en el ámbito escolar. Anuario Becas de Investigación “Ramón Carrillo-Arturo Oñativia”. 2013;(3):241. only those schoolchildren who attended the school feeding programs were selected, whereas in 2018, in view of the social, economic and political context in Argentina, both school feeding program attendees and non-attendees were included with the purpose of being able to compare both scenarios.

Inclusion Criteria

Schools, their school feeding programs and urban location services that had provided at least two of the frequent feeding services (breakfast, lunch and/or snacks) during the survey days were included in the study. Children of both sexes aged 8 years or older in 4th, 5th and 6th grade of the schools under study, attending the morning or afternoon school shift were recruited.

To determine the inclusion of the children who attended the school feeding program, the self-report of attendance, the day before the survey, to at least one of the meals provided by the feeding program was defined as an inclusion criterion. Exclusion criteria included children of both sexes exhibiting a degree of cognitive dysfunction, language barriers or any other situation that would not enable to implement the 24-hour dietary recall. However, during the field stage, no situations were detected that hindered the progress of the interviews. All participating children had their informed consent forms signed by their father, mother or responsible person.

Main Variables

School feeding program

Dietary intake in children

Nutritional status: Categories and indicators

Covariates

Data collection techniques and instruments: Food and nutritional quality of school feeding programs

A survey of primary and secondary data was conducted.²⁸28. Britos S, Virgolini M, Saraví A, Chichizola N, Moyano D, Perez L, Arroyo Parisi S. Análisis de la alimentación en el ámbito escolar. Anuario Becas de Investigación “Ramón Carrillo-Arturo Oñativia”. 2013;(3):241. The same methodology was used for the variables collected in both periods. This instrument consisted of a direct observation guide and a record of all food and beverages in all the food services provided on two non-consecutive days. The weights of randomly selected food rations were included, using a portable electronic scale with a capacity of up to 3 kg; accuracy ±1 g. Based on the food data obtained, estimates of the amount of nutrients were calculated using the food composition databases of Latinfoods, Argenfoods, the Food Analysis and Registration System (SARA, for its Spanish acronym), nutritional labeling of food industries in Argentina published on web pages and the United States Department of Agriculture (Release 28).

For food and nutritional intake and the achievement of healthy nutritional goals, the 24-hour food recall method³¹31. Ferrari M. Estimación de la ingesta por recordatorio de 24 horas. Diaeta (Buenos Aires). 2013;31(143):20-25.^,3232. Gibson RS. Principles of nutritional assessment. 2nd ed. New York: Oxford University Press; 2005. was used, validated for children older than 8-years of age. In order to minimize possible biases, the Gibson’s four-step technique³²32. Gibson RS. Principles of nutritional assessment. 2nd ed. New York: Oxford University Press; 2005. as well as validated visual food models, standardized recipes and household utensils (spoons, cups and others) were used. Two non-consecutive observations to a 15% subsample per school were conducted, using the “multiple source method” to statistically adjust the intake.

The amount of energy and critical nutrients due to deficiency and excess was estimated, and the abovementioned nutritional composition databases were used. The achievement of healthy nutritional goals was determined taking as reference the dietary reference intakes recommended by the United States Institute of Medicine³³33. Institute of Medicine FaNB. Dietary reference intakes (DRIs). Washington DC: National Academy Press; 2010., and the WHO/FAO 2001³⁴34. World Health Organization, Food and Agriculture Organization. Expert consultation report on human energy requirements: Interin report. Rome: WHO, FAO; 2001.^,3535. World Health Organization, Food and Agriculture Organization. Consultation report on human vitamin and mineral requirements. Bangkok: WHO, FAO; 2001. and 2003.³⁶36. World Health Organization, Food and Agriculture Organization. Diet, nutrition and the prevention of chronic diseases. Geneva: WHO, FAO; 2003. The age ranges of energy proposed by the WHO/FAO 2001³⁴34. World Health Organization, Food and Agriculture Organization. Expert consultation report on human energy requirements: Interin report. Rome: WHO, FAO; 2001. and, for its application, the 50th percentile by age and sex was taken as reference weight for the children, according to the Guide for the evaluation of growth, published by the Argentine Society of Pediatrics in 2013³⁷37. Sociedad Argentina de Pediatría, Comité Nacional de Crecimiento y Desarrollo. Guías para la evaluación del crecimiento físico 2013. 3a ed. Buenos Aires: SAP; 2013.. The activity factor considered for the calculation was that for moderate physical activity. In the case of the micronutrients for which there were no available data on recommendations, the adequate intake values were used.

To define the nutritional status of the children, the guidelines of the WHO and the Guide for the evaluation of physical growth, of the Argentine Society of Pediatrics³⁷37. Sociedad Argentina de Pediatría, Comité Nacional de Crecimiento y Desarrollo. Guías para la evaluación del crecimiento físico 2013. 3a ed. Buenos Aires: SAP; 2013., were followed to implement an instrument for the collection of anthropometric measurements. Prior to the implementation of the survey, the team of researchers were trained in order to standardize the measurements. To estimate the weight, a platform scale with 100 mg accuracy was used as well as a stadiometer to measure the height. The WHO Anthro software was used to calculate the children’s nutritional status indicators.

Data analysis

A descriptive analysis of the main variables and covariates of interest was performed, using measures of central tendency and dispersion (mean); standard deviation, median and 25th and 75th percentiles for the continuous variables and the absolute and relative frequencies for categorical data.

The distribution of continuous data (nutritional) was evaluated by using graphic methods (histogram) and the Shapiro-Wilk test to contrast the normality of the distribution of data; the Bartlett test was used for the analysis of variances. In all the anthropometric indicators, standardized data through the Z-score were always used,

To explore possible associations between dichotomous variables of achievement of healthy nutritional goals and the school feeding program attendance, a multivariate analysis through logistic regression methods (melogit), taking the schools as variable effect and the variables from the children as fixed effects. The outcomes are expressed in crude odds ratios and adjusted for potential confounders. The multicollinearity of the independent variables was verified using the variance inflation factor. Possible interactions that were not statistically significant (p>0.05) were explored. For the goodness- of-fit, the ROC curve and the Hosmer-Lemeshow test were used as discrimination and calibration parameters of the model. In all cases with 95% confidence intervals and a statistical significance value of p<0.05. Statistical software Stata version 14.0 was used for all the analyses.

Ethical aspects

All ethical requirements at the national level were followed in this study. The informed consent of the mother/father/guardian of the participating children was requested in accordance with the current national regulations of the 2016 Argentine Civil and Commercial Code - with respect to the age scale for signing the informed consent - and the requirements of the local ethics committees. The secondary base protocol (2013) was evaluated by the Research Ethics Committee of Hospital Nacional de Clínicas, Universidad Buenos Aires and, in 2018, by the Institutional Ethics Committee of Health Research of Hospital Nacional de Clínicas, Universidad Nacional de Córdoba (RHNC No. 3575/2018).

RESULTS

The food service sample of the school feeding programs in the first survey was n=58 and in the second survey was n=57. The total sample of children of both sexes during the two data collection periods was =341, consisting of 150 children of both sexes in the first survey and 191 in the second survey. As shown in Table 1, the sociodemographic indicators in the two periods under analysis, followed a similar structure, with a predominance of the girl subgroup, with ages equal to and/or older than 10 years, attending grades below 6^th grade and the morning shift.

In the global sample, and taking as a reference the food groups proposed by the 2016 Dietary Guidelines for the Argentine population, it was found that, of the total sample, the most consumed foods were those in the group of optional foods (33%) and to the group of legumes, cereals, potatoes, bread and pasta (21%); 17% of the foods under analysis belonged to the group of vegetables and fruits, 8% to the group of milk, yogurt and cheeses, 8% to meats and eggs, 6% was water, 7% belonged to the group of oils, dried fruits and seeds.

Between the two cut-off points, most of the means of the critical nutrients due to deficiency, especially zinc, vitamin C, iron and folic acid, exhibited a reduction in the contribution of food services of the school feeding programs. For example, the mean of iron of the portion offered by the school feeding programs went from 3.06 [95%CI (1.96-4.16)] in 2013 to 2.16 in 2018 [95% CI (1.57-2.76)]; of folic acid from de 142.09 [95%CI (91.97-192.21)] to 117.17 [95%CI (95.13-139.21)] and of vitamin C from 20.08 [95% CI (4.92-35.24)] to 8.03 [95% CI (2.48-13.57)] in the high crowding stratum (greater vulnerability). This was similarly reflected in the means of nutrients considered to be critical due to excess such as simple sugars and energy. In the case of energy, it went from a mean of 371.5 [95%CI (264.02-479.05)] in 2013 to a mean of 305.12 [95%CI (241.68-368.55)] in the stratum of highest crowding (Table 2).

Another interesting fact is that in the school feeding program subgroup in the context of high level of crowding, both in 2013 and 2018, the median of calcium and vitamin D was lower compared to the subgroup with a medium and/or low level of crowding (Table 2).

Table 3 shows a comparison analysis between the intakes of the children that participated in the first survey (n=150) and those in the second survey (n=191) (Table 3). It was observed that almost all the means of the critical nutrients under analysis were higher in the subgroup that participated in the 2013 survey compared to 2018; however, the statistically significant differences were found in the mean total calcium intake, where the mean in 2013 was 573.49 [95% CI (524.61-622.38)] and in 2018 was 493.99 [ 95%CI (463.81-524.17)] and in total energy with a mean of 2480.59 [95%CI (2333.38-2627.79)] and 2243.68 [95%CI (2146.99-2340.38)], respectively. Only in the case of fiber, a slightly higher intake was observed in the second survey [mean 9.83; 95% CI (9.19-10.46)] with respect to the first survey [mean 11.39; 95% CI (10.62-12.15)] (Table 3). A relevant fact is that taking the median of nutrients provided by the school feeding programs, this value would be covering 23% or less of the median global intake (depending on the nutrient), with the exception of vitamin D, which reached 48.3% of coverage (Table 3).

From the analysis of the coverage of the healthy nutritional goals recommended by international organizations, it was observed that an important part of the nutrients under study in both periods of analysis did not achieve the proposed goal (taking the median value), this being more evident in 2018. The most critical cases of insufficient achievement of the goal were observed in relation to calcium, vitamin A, fiber and vitamin D, as well as to simple sugars (Table 3).

When conducting the subgroup analysis comparing the nutritional intake of children attending and non-attending the school feeding programs (Table 4) during the 2018 survey, it could be observed that part of the nutritional indicators were lower in the group covered by school feeding programs compared to those who were not covered by the programs, especially regarding calcium, in which the median in the group of attendees was 438.8 and in that of non-attendees was 512.4, while in relation to vitamin C a median of 32.2 and 37.0 was observed and regarding vitamin D the median was 2.0 y 2.2 between attendees and non-attendees, respectively. In the critical nutrients due to excess, an intake according to the highest median of total sodium was observed in the subgroup of attendees compared to the non-attendee subgroup. However, total energy was lower in the group of attendees compared to the group of non-attendees (Table 4).

With regard to the achievement of healthy nutritional goals, it was observed that an important part of the nutrients under analysis, especially calcium, vitamin C and vitamin D, both in the group of attendees and non-attendees was inadequate. Nevertheless, this latter scenario was more evident in the subgroup that attended the school feeding programs.

In Table 5, a column is included that represents the proportion of children with “adequate diagnosis” both in the subgroup of school feeding program attendees and non-attendees in each of the Z-scores under analysis (BMI-for-age, height-for-age and weight-for-age) following the WHO guidelines. With regard to the nutritional status, differences were observed in the percentiles in the subgroup of school feeding program attendees compared to those in non-attendees, for example in the case of the weight-for- age Z-score. The first group had a median of 1.0 and the second a median of 0.4. The height-for-age percentile in the group of attendees had a sample proportion with “adequate diagnosis” of 90.8% and 95.8% in the non-attendee group, while the BMI-for-age percentile was higher in the first group compared to the second group (Table 5).

As shown in Table 6, in the simple and multiple logistic regression models adjusted for potential confounders on the relationship between the category of failure to achieving the healthy nutritional goals and attendance to school feeding programs, an OR of 1.37 [95%CI (0.47-3.98); p=0.566] in saturated fats and an OR of 1.42 [95% CI (0.62-3.28); p=0.406] in total energy, although these associations were not statistically significant. Increased ORs were observed, although not significant, in the category of failure to achieve vitamin C and folic acid goals in those attending school feeding programs.

When the odds ratio between the indicators of the nutritional status of children and school feeding program attendance were analyzed, both in crude and adjusted models, no statistically significant differences were observed in any scenario. Although an OR of 2.59 [95%CI (0.53-12.80)] was found in the category of short stature/ short stature risk in children attending the school feeding programs in the adjusted model, which was not statistically significant (p=0.241) (Table 6).

DISCUSSION

Based on the results obtained, it is concluded that it would be useful to promote a policy of school feeding programs urgently seeking improvements in the quality indicators of food services, as well as in the intervention indicators in child food and nutrition. Although the school feeding programs provide part of the critical nutrients (both due to deficiency and excess) of the global average intake of children attending the school feeding programs, more aggravated indicators of excess weight and short stature were found in this group with respect to those not attending the programs, with a deterioration in the nutritional contributions of the school feeding programs, as well as in the intakes between 2013 and 2018.

The generalized micronutrient deficiency and, consequently, a low presence of food sources such as fruits and vegetables, was alarmingly reflected both in this work and in a previous study.³⁸38. Diez González S, Rodríguez Bernardo C, Alonso Alonso O, Gutiérrez Diez MC, Allande Díaz R. Evaluación de la variedad y calidad en los menús escolares de Asturias. Curso 2015/2016. Revista Española de Salud Pública. 2018;92:e1-e12.

The context of inequalities in which the school feeding programs are inserted, would be apparently conditioning the nutritional contributions. Incorporating this approach into the analysis is relevant when thinking about child nutrition from a comprehensive perspective and within a human rights-based perspective. In this sense, it was observed that in the subgroup of school feeding programs inserted in contexts with a level of critical crowding - both in the surveys conducted in 2013 and 2018 - most of the critical nutrients and total energy provided were lower. This finding calls attention to the importance of protecting the right to healthy food from a public policy perspective in all social groups, especially those with greater social vulnerability and even more so in a scenario of crisis and increased poverty indicators, leaving an open door for future studies addressing inequality variables, especially taking into account that the most vulnerable groups are the direct targets of these policies. In line with our findings, and based on 2019 data from the Childhood Social Debt Barometer of the Argentine Catholic University, extreme poverty due to financial indigence and/or severe deprivation affected 24.7% of children between 0 and 17 years of age in 2018.³⁹39. Tuñón I, Poy S. Pobreza, derechos e infancias en la Argentina (2010-2018). Ciudad Autónoma de Buenos Aires: Educa; 2019.

The abovementioned differences regarding the impoverishment of the school feeding programs between 2013 and 2018 may be related to the economic and social context of Argentina, with the increase in poverty indicators in 2018.³⁹39. Tuñón I, Poy S. Pobreza, derechos e infancias en la Argentina (2010-2018). Ciudad Autónoma de Buenos Aires: Educa; 2019. Although the menu offered by the school feeding programs surveyed in both mentioned periods, apparently remained unchanged in the period under study, the decrease in the global provision of the analyzed nutrients could have been due to the policy decision to reduce food portion sizes, replace ingredients or increase the number of attendees at food service time, which impacted on global nutrient provision. Understanding the public intervention from the perspective of the territory and the local reality in a national context and the potential role within the nutritional food processes of children in a context of deprivation and from a multidimensional approach will be key in fully achieving the right to food.

Child malnutrition in frequent users of school feeding programs should be a priority in the public agenda of the countries in Latin America and the Caribbean, and at the country level in a current context of alarming malnutrition indicators.⁴⁰40. Organización Mundial de la Salud. Malnutrición [Internet]. 2017 [citado 10 mar 2019]. Disponible en: Disponible en: https://tinyurl.com/y6dxfrtn .
https://tinyurl.com/y6dxfrtn...

The findings of this research would be indicating that the foods most consumed by schoolchildren were, in more than one third, those in the group of optional foods (according to the GAPA),³⁰30. Ministerio de Salud. Guías Alimentarias para la Población Argentina 2016 [Internet]. 2016 [citado 10 may 2019]. Disponible en: Disponible en: https://tinyurl.com/y5k9h3ku .
https://tinyurl.com/y5k9h3ku... ⁾ being those that most provide critical nutrients due to excess and which need to be reduced within the framework of healthy nutrition policies. These results are similar to those reported by ENNyS, 2018-2019²2. Secretaría de Gobierno de Salud, Ministerio de Salud y Desarrollo Social. 2° Encuesta Nacional de Nutrición y Salud (ENNyS 2). Buenos Aires: Ministerio de Salud y Desarrollo Social; 2019., as well as to those recently published by Dunford and Popkin⁴¹41. Dunford EK, Popkin BM. 37 year snacking trends for US children 1977-2014. Pediatric Obesity. 2018;13(4):247-255. in children aged 2 to 18 years in the United States. When specifically analyzing the nutrient intake in Dunford and Popkin’s work, multiple nutritional inadequacies were observed in the children’s intake, with calcium, fiber, vitamin A and simple sugars being the most alarming in both survey periods. Similarly, these inadequacies were reported in national data.²2. Secretaría de Gobierno de Salud, Ministerio de Salud y Desarrollo Social. 2° Encuesta Nacional de Nutrición y Salud (ENNyS 2). Buenos Aires: Ministerio de Salud y Desarrollo Social; 2019.

A worsening of the indicators was observed in an important part of the nutrients under analysis between 2013 and 2018. However, statistically significant differences were found in the mean intake of calcium and total energy, which were higher in the first survey compared to the second survey. As mentioned in the section on the analysis of the nutritional quality of food services in school feeding programs, these results may be due to the context of increased child poverty indicators³⁹39. Tuñón I, Poy S. Pobreza, derechos e infancias en la Argentina (2010-2018). Ciudad Autónoma de Buenos Aires: Educa; 2019. that directly affects the access to healthy and sufficient nutrition.

Relating situations of deprivation to the social context is essential to understand nutritional outcomes, as was shown by Monteiro et al.’s study in Brazil⁴²42. Monteiro CA, Benicio MH, Conde WL, Konno S, Lovadino AL, Barros AJ, et al. Narrowing socioeconomic inequality in child stunting: the Brazilian experience, 1974-2007. Bulletin of the World Health Organization. 2010;88(4):305-311. during the period 1974-2007, the socioeconomic development along with equity-oriented public policies have been accompanied by remarkable improvements in living conditions and a substantial decrease in child malnutrition, as well as by a reduction in the gap in the nutritional status of children in the highest and lowest socioeconomic quintiles.

Along this same line, a study that analyzed malnutrition and nutrition disparities showed that overweight is more prevalent in low and middle-income countries, where children are among the most affected⁴³43. Perez-Escamilla R, Bermudez O, Buccini GS, Kumanyika S, Lutter CK, Monsivais P, et al. Nutrition disparities and the global burden of malnutrition. BMJ. 2018;361:k2252. and also the weight of social and economic inequalities is blamed for growth stunting over time.⁴⁴44. Restrepo-Mendez MC, Barros AJ, Black RE, Victora CG. Time trends in socio-economic inequalities in stunting prevalence: analyses of repeated national surveys. Public Health Nutrition. 2015;18(12):2097-2104. Furthermore, according to a longitudinal study conducted in 2009, growth stunting in children was associated with poor school attendance, lower school performance and an increased likelihood of living in poverty in adult life.⁴⁵45. Maluccio JA HJ, Behrman JR, Martorell R, Quisumbing AR, Stein AD. The impact of improving nutrition during early childhood on education among Guatemalan adults. The Economic Journal. 2009;119(537):734-763.⁾ This indicates how child malnutrition may be the cause and effect of situations of vulnerability, in which public policies are crucial to intervene in these complex scenarios.

As was observed in this study, a significant number of children, both in the group of school feeding program attendees and non-attendees, did not achieve the healthy nutritional goals for their age. Although in the subgroup of school feeding program attendees it was found that the average intakes of some critical nutrients such as vitamin A and calcium and total energy were lower than in the group of non-attendees. A deterioration in nutritional status indicators was also observed, especially those of overweight and short stature/short stature risk, although, broadly speaking, the nutritional profiles in both groups were less than optimal. This finding enables to account for and verify that both in the school feeding programs and in the households children eat the same kind of food; therefore, the policy that intends to be “compensatory” becomes “conservative”, in terms of quantity and not quality, as several authors have already argued in a similar manner,¹⁵15. Aguirre P. Pobres gordos, ricos flacos: la alimentación en crisis. Buenos Aires: Capital Intelectual; 2004.^,1717. Del Castillo SE. La situación nutricional de la niñez en Latinoamérica: entre la deficiencia y el exceso, de brecha nutricional a deuda social. Biomédica. 2012;32(4):471-473.^,2727. Abeyá Gilardon EO. Una evaluación crítica de los programas alimentarios en Argentina. Salud Colectiva. 2016;12(4):589-604. highlighting the urgent need to improve the food policy in school settings.

With regard to the relationship explored between the children’s intake and school feeding program attendance (models adjusted for potential confounders), despite the fact that no statistically significant associations were found, it was observed that in the critical nutrients due to excess, in the group of school feeding program attendees, the OR for saturated fats was 1.37 and for total energy was 1.42. Moreover, an increased OR was also observed in the category of failure to achieve the goals of other critical nutrients due to deficiency, such as vitamin C and folic acid in school feeding program attendees.

This outcome is relevant as it leaves the door open to use longitudinal studies to go deeper into the weight that the school feeding program policy has in the excess of critical nutrients in the intake. As the WHO suggests, these nutrients have been associated with various non-communicable diseases,⁴⁶46. Organización Mundial de la Salud. Dieta, nutrición y prevención de enfermedades crónicas: informe de una consulta mixta de expertos OMS/FAO [Internet]. 2003 [citado 10 may 2019]. Disponible en Disponible en https://tinyurl.com/y2pukjfp .
https://tinyurl.com/y2pukjfp... leading to greater risks for the future development of chronic diseases in children who have had part of their diet covered by the school feeding programs. There are scant track records published in this line of research. An earlier study stated that the intake of energy, carbohydrates and fats in the group of school feeding program attendees was significantly higher than in those that did not attend the school feeding programs,¹⁹19. Ministerio de Sanidad y Política Social. Informes técnicos de evaluación del programa PERSEO de promoción de la alimentación y la actividad física saludables en el ámbito escolar. España: AESAN; 2009. although it was based on a bivariate analysis.

A study conducted by Ongan et al. found no statistically significant differences in body weight, height and body mass index in the group of school feeding program attendees compared to non-attendees,²²22. Ongan D, Inanc N, Cicek B. Comparing school lunch and canteen foods consumption of children in kayseri, Turkey. Pakistan Journal of Medical Sciences. 2014;30(3):549-553. although no confounding variables were reported in this study.

In Argentina, we found some isolated studies analyzing aspects of school feeding programs in public schools and describe the characteristics of the services,¹²12. Moyano D, Perovic NR. Nutritional contribution the school canteens program to the children of ten municipal schools in the city of Cordoba, Argentina. Revista de la Facultad de Ciencias Medicas. 2018;75(3):194-202.^,1313. Buamden S, Graciano A, Manzano G, Zummer E. Proyecto “Encuesta a los Servicios Alimentarios de Comedores Escolares Estatales” (PESCE): alcance de las metas nutricionales de las prestaciones alimentarias de los comedores escolares de Gran Buenos Aires, Argentina. Diaeta (Buenos Aires). 2010;28(130):21-30.^,1414. Bracamonte S, Carranza C, Olivero I, Laquis M. Calidad nutricional y consumo de los menúes del comedor escolar y estado nutricional de los alumnos del nivel primario de la escuela Jorge Peyrano, Ciudad de La Calera, Provincia de Córdoba, 2012. Córdoba: Escuela de Nutrición, Universidad Nacional de Córdoba; 2012.^,2828. Britos S, Virgolini M, Saraví A, Chichizola N, Moyano D, Perez L, Arroyo Parisi S. Análisis de la alimentación en el ámbito escolar. Anuario Becas de Investigación “Ramón Carrillo-Arturo Oñativia”. 2013;(3):241.⁾ although more robust analyses on child nutrition intervention are limited. Unfortunately, this limitation also extends to several Latin American countries.⁴⁷47. Herscovic CR, Kovalskys I. Obesidad Infantil: Una revisión de las intervenciones preventivas en escuelas. Revista Mexicana de Trastornos Alimentarios. 2015;6(2):143-151.

The study published by Rausch Herscovici et al.,⁴⁸48. Rausch Herscovici C, Kovalskys I, De Gregorio MJ. Gender differences and a school-based obesity prevention program in Argentina: a randomized trial. Revista Panamericana de Salud Pública. 2013;34(2):75-82. could be an exception to this situation. This study, among several interventions, focused on providing food with better nutritional value to be offered in the school feeding programs, within the framework of other educational and physical activity interventions. Although the authors found significant changes in some indicators of healthy food consumption, the limitation of the study was that the intervention was short-term (six months), as mentioned by the authors themselves,⁴⁸48. Rausch Herscovici C, Kovalskys I, De Gregorio MJ. Gender differences and a school-based obesity prevention program in Argentina: a randomized trial. Revista Panamericana de Salud Pública. 2013;34(2):75-82.⁾ and no aspects related to school feeding program exposure in outcome variables, such as the achievement of nutritional goals in children, were explored.

Another outcome that should be explored deeper in future studies is the association between the exposure to school feeding programs and nutritional status indicators. Along this line, we found an OR of 2.59 in the category of short stature/ short stature risk in the subgroup that was exposed to school feeding programs compared to those that were not exposed, although this was not statistically significant. Failure to find significant differences in the multivariate analysis could have been due to the fact that the sample size of the subgroup of school feeding program non-attendees was small with large confidence intervals. In addition, as stated by Galván et al.⁴⁹49. Galván M, Amigo H. Programas destinados a disminuir la desnutrición crónica: Una revisión en América Latina. Archivos Latinoamericanos de Nutrición. 2007;57(4):316-326. the impacts on chronic malnutrition indicators need to be analyzed in the long term.

Within the framework of the region’s progress in promoting efforts in school setting interventions and that these are evaluated, different international organizations have been establishing the issue of healthy school settings in the global, regional and national agenda, as a starting point in the fight against childhood overweight, obesity and malnutrition.⁵⁰50. Fondo de las Naciones Unidas para la Infancia. Situación de la Salud de las y los adolescentes de Argentina [Internet]. 2016 [citado 10 may 2019]. Disponible en: Disponible en: https://tinyurl.com/y38dxmqx .
https://tinyurl.com/y38dxmqx... However, there are no apparent and updated data on actions at the provincial and municipal levels in Argentina in relation to this topic. Moreover, there is a lack of impact assessments that can give an account of the efforts made in this connection. All this has already been highlighted in a previous study.²⁷27. Abeyá Gilardon EO. Una evaluación crítica de los programas alimentarios en Argentina. Salud Colectiva. 2016;12(4):589-604.

The world literature is highlighting the potential of school feeding programs in improving nutritional indicators in the infant-juvenile population. A review published in 2016 by Meiklejohn et al., which evaluated the impact of nutritional interventions in adolescents, showed that the changes in school feeding programs were associated with significant changes in food intake.⁵¹51. Meiklejohn S, Ryan L, Palermo C. A systematic review of the impact of multi-strategy nutrition education programs on health and nutrition of adolescents. Journal of Nutrition Education and Behavior. 2016;48(9):631-646. Furthermore, the contributions to school permanence, school performance, the implications of social protection and individual and family food and nutritional security, the positive effects on healthy habits and on children’s socialization processes are also underlined.⁵²52. Food and Agriculture Organization. School feeding and the possibilities for direct purchases from family farming - Case studies in eight countries. Santiago de Chile: FAO; 2013.^,5353. Food and Agriculture Organization, Brazilian Cooperation Agency. Strengthening school meal programmes under the Hunger Free Latin America and the Caribbean 2025 Initiative. Brazil: FAO, ABC; 2008.

However, there is scant published information in the region regarding the actual contribution of school feeding programs and all the more, from healthy interventions. In recent years, there are other alternative proposals for school feeding programs in Latin America.⁵²52. Food and Agriculture Organization. School feeding and the possibilities for direct purchases from family farming - Case studies in eight countries. Santiago de Chile: FAO; 2013.^,5353. Food and Agriculture Organization, Brazilian Cooperation Agency. Strengthening school meal programmes under the Hunger Free Latin America and the Caribbean 2025 Initiative. Brazil: FAO, ABC; 2008. Brazil is one of the most remarkable cases, given that despite having a long-standing school feeding program, several actions are currently being performed that transcend the welfare nature that these programs historically had.⁵⁴54. Fundo Nacional de Desenvolvimiento da Educação. Programa Nacional de Alimentação Escolar (PNAE) [Internet]. 2012 [citado 10 may 2019]. Disponible en: Disponible en: https://tinyurl.com/yyymm2bd .
https://tinyurl.com/yyymm2bd...

Today it has its respective legal framework and seeks to favorably contribute to children’s growth, biopsychosocial development, nutritional status, learning and school performance, as well as to promote healthy eating habits, act on adverse environmental factors, incorporate food nutritional educational instances (EAN, for its acronym in Spanish) and strengthen the participation of the local community in the control and management of the program.⁵⁴54. Fundo Nacional de Desenvolvimiento da Educação. Programa Nacional de Alimentação Escolar (PNAE) [Internet]. 2012 [citado 10 may 2019]. Disponible en: Disponible en: https://tinyurl.com/yyymm2bd .
https://tinyurl.com/yyymm2bd...

Among the limitations of this work, the biases from cross-sectional studies can be mentioned, such as the selection that could have occurred at the time of excluding schoolchildren with a cognitive and/or intellectual impairment, although the official data⁵⁵55. Ministerio de Educación. Información y Estadística educativa [Internet]. s/f [citado 10 mar 2019]. Disponible en: Disponible en: https://tinyurl.com/y3qjgaff .
https://tinyurl.com/y3qjgaff... may be indicating that the percentage of children with disabilities currently included in mainstream schools is low. It was also possible to report the presence of information bias and social complacency (information) that is frequently observed in population studies that survey food data; nevertheless, the data were collected through a questionnaire with open-ended questions, guided by trained researchers, which could reduce the influence of this bias.

In the evaluated relationships, a common confounding and reverse causality bias could have existed in epidemiological cross-sectional studies, although we had a comparison group that enabled to make comparisons, and models adjusted for potential confounders were performed. In addition, the small sample size could have been a limitation at the time of analyzing the relationships; therefore, future more robust studies need to be conducted, with larger samples that will enable to improve the estimates. However, the fact of having data from two time periods (2013 y 2018), and different socio-political contexts of Argentina, and including a comparative analysis from an approach of inequalities, is key when defining the public intervention aimed at achieving the right to food and other fundamental rights, such as health and education

This study has multiple and important strengths in the field of public health and public policies. The sample is representative of the municipal schools under analysis with a sufficient number of children that enabled to make population estimates. In addition, this proposal provides evidence for an issue that is now controversial and with multiple knowledge gaps in our region.

Below are several recommendations related to the issue addressed in this study and to future lines of work in the national and local context.

At the level of management and implementation of the school feeding program policy

At the level of the impact of the school feeding policy

At the level of regulations, local, subnational and national public policies of school feeding programs

ACKNOWLEDGMENTS

We would like to thank Universidad Nacional de la Matanza through the “Formando UNLaM” Program, 2019 for the funds contributed for the analysis stage of this work and to the Health Research Directorate, Salud Investiga scholarships, 2013 and 2018 for the provision of funds for conducting the two field stages. We are grateful to Dr. Jorge Chaui and Dra. Paula Vázquez for reviewing the analyses

References

Publication Dates

History