Comparison of nutritional status of rural and urban school students receiving midday meals in schools of Bengaluru, India: A cross sectional studyCN Shalini, NS Murthy, S Shalini, R Dinesh, NS Shivaraj, SP Suryanarayana
Department of Community Medicine, M S Ramaiah Medical College, Bangalore, Karnataka, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.132309
Source of Support: None, Conflict of Interest: None
Background: The objective of the study was to assess the impact of the mid day meal program by assessing the nutritional status of school students aged 5-15 years receiving midday meals in rural schools and compare them with those in urban schools in Bengaluru, India. Materials and Methods: This cross sectional study involved a sample of 4378 students from government and aided schools. Weight and height were measured and compared with ''means'' and ''percentiles'' of expected standards as endorsed by the Indian Association of Pediatrics. Regression coefficients were also estimated to assess the rate of growth. Results: In all age groups and in both sexes, the observed mean weight and height were below the expected standards. The study findings showed that 13.8% and 13.1% of the studied students were underweight and stunted, respectively (below the third percentile for weight and height for age). A higher proportion of rural students were below the third percentile for both weight and height compared with urban students (weight: 16.3% and 11.5%; height: 17.0% and 10.0%; P < 0.05 for both weight and height). Only 2.4% and 3.1% were above 97 th percentile for weight and height. The rate of growth of height for weight showed a declining trend with increasing age in all the groups. Discussion: The authors believe that the magnitude of the burden of undernourished students as seen in this study would have been much greater in the absence of the midday meal program. Conclusion: Greater involvement of the private sector to assist the government would help augment nutrition in children and indirectly impact school performance, attendance and literacy.
Keywords: Anthropometric measurements, midday meals, nutritional status, rural-urban
In 1960, the Government of India constituted a school health committee to assess the standards of health and nutrition among school students and to suggest ways to improve them. Some of the health problems in school students in India today include malnutrition, infectious diseases, intestinal parasites, diseases of skin, eye and ear, and dental caries.  The school health program currently in place has the objectives of screening for malnutrition and health problems, including dental problems. The idea is to facilitate early diagnosis, referral of students to health facility and feedback, increasing awareness of and capacity building of teachers on common, and emerging health problems in childhood and to provide age appropriate health education to school students. 
The National Program of Nutritional Support to Primary Education was launched during 1995 as a centrally sponsored scheme to enhance enrollment and, attendance at school, and improving nutritional levels by providing a cooked mid day meal to students studying in classes I-V (in Government and aided schools).  The improvement in nutrition would then be assessed through annual monitoring of weight and height. Studies that have assessed the impact of midday meals have been rather few. ,,, The present communication reports the assessment and comparison of the nutritional status of rural and urban school students receiving the midday meal based on anthropometric measurements of weight and height according to age groups and sexes as assessed in 2010.
Setting, design and ethics
The Department of Community Medicine, MS Ramaiah Medical College, Bengaluru in collaboration with Sai Mandali Trust Malleswaram in Bengaluru city [a nonprofit organization (NGO) providing free mid day meals] has been providing school health services since 2007. The school health examination was carried out with the permission and approval of the Deputy Director of Public Instruction (DPPI), Government of Karnataka State, India. The DDPI informed the headmasters/headmistresses of the schools about the study objective and sought permission from them. Written, informed consent from parents was obtained. Forty one schools (25 rural and 16 urban) which were covered under the mid day meal scheme of Sri Sai Mandali Trust were included in this cross sectional study. These included both government and aided schools.
Assessment of nutritional status
Standards developed for the same were published first by the Indian Council of Medical Research (ICMR) several decades ago (1956-1965).  However, the studies by the ICMR were conducted among the lower socioeconomic group of students. It has been suggested by Indian Academy of Paediatrics that Agarwal standards for weight and height are better representative of the growth of normal Indian students than ICMR charts.  Hence Agarwal standards were used for the present study.
As per the records of the schools, there were a total of 4575 students of which 2506 were in urban schools and 2069 were in rural schools. Schools were classified into urban and rural schools by the Karnataka State government's Department of Education based on the boundaries of the Bengaluru City Corporation-North Zone. 
Sample size estimation
A sample size was estimated based on a previous unpublished analysis of urban and rural students where 55% of urban students and 64% of rural students were below 50 percentile of expected weight and a level of precision of 2.5% at 95% level of confidence. (data on file).  This worked out to be 1521 and 1416 students from urban and rural schools respectively. However, all the students in the school were included totalling 4378 students in all.
A health booklet is maintained for each child, which documents various aspects of health check up including demographic and anthropometric measurements. Actual measurements were taken by the investigators during the school health check up. This paper reports only the assessment of nutritional status based on anthropometric measurements of height and weight of the students.
Schools were given prior intimation of the date of visit of the health team. A one-day sensitization workshop for school teachers was conducted in order to increase awareness of the need for health promotion and detection of health problems in school students. The health care team consisted of the team from the Department of Community Medicine viz. physicians, medico social workers, and a coordinator from the NGO. Anthropometric measurements of students were recorded with the use of standardized instruments with appropriate quality control checks by the physicians. Height was measured with a nonstretchable standardized measuring scale with an accuracy of up to 0.5 cm and without footwear. Weight was measured with a standard ''Indian Standards Organization'' certified weighing scale with minimum clothes and without footwear with an accuracy of up to 100 gm. This was followed by a medical checkup for all students and treatment as necessary.
Analysis was done by SPSS version 18.3 All the results were tabulated by age, sex, and rural and urban schools. The age of the students was noted from the school records which in turn were based on the date of birth certificates provided by the parents at the time of admission of the student to the schools. Age of students were classified into 5-9, 10-14, and 15 years based on the Census of India classification of age class intervals for general purposes.  Nutritional status was assessed for weight and height by comparing the weights and heights of the students with the standards provided by Agarwal et al.,  and endorsed by the Indian Academy of Paediatrics. 
Reporting of outcome measures
Weight for age and height for age were used as indices to measure nutritional status of the students. Weight and height for age were analyzed by tabulating them as percentiles- ≤3 rd percentile, 4-50 th percentile, 51-97 th percentile, and >97 th percentile based on the Agarwal's standards.  The students below the third percentile were classified as undernourished and over 97 th percentile as obese. Demographic data was analyzed using descriptive statistics. The chi-square test was used for assessing association of proportions in various nutritional categories between rural and urban schools. The correlation coefficient was used to derive a value of quantitative relationship between weight and height. Regression coefficients between weight and height were estimated according to age, sex, and area in order to understand the extent of relationship as well as rate of change of weight for unit change in height.
A total of 1942/2069 students in rural schools (93.8%) students were present on the day of the study. . In the 16 urban schools corresponding figure was 2436/2506 (97.2%) The total evaluable sample size thus was 4378 children. Female students outnumbered male students in both urban and rural schools. Nearly a third of the students (32.9%) were aged 5-9 years and over half of them (53.4 %) were in the age range 10-14 years [Table 1]. In rural schools, nearly all males (95.2%) and females (93.4%) were in the 5-14 year age group, whereas in urban schools all males and 74.7% of females were in the 5-14 age groups. In the urban area, the high schools (8 th standard -10 th standard) included for midday meal and the study, had provision for girl students only (15 years of age) which accounted for 25% of girls. Hence, there were no boys of 15 years of age in the urban area.
The mean weights and heights of students by age groups, sex, and area are presented in [Table 2]. Both of these increased with increasing age. In the 5-9 year age group, the mean weights and heights of urban male students were found to be slightly higher relative to rural students (19.2 kg, 117.4 cm versus 19.0 kgs, 115.8 cms) In 5-9 year age group of female students, both mean weights and heights of urban students were found to be slightly lower relative to rural students (19.1 kg, 117.2 cm versus 18.6 kgs, 115.0 cms). However, only the difference in height was statistically significant (P = 0.04). In the age groups of 10-14 years and 15 years, lower weights and heights were found in rural males relative to urban males [rural-10-14 years: 28.2 kg, 136.8 cm; 15 years: 40.9 kg, 156.2 cm; urban-10-14 years 29.8 kg 138.5 cm; 15 years - no male students in this age group. The differences were significant (P = 0.002 and 0.02 respectively). Similarly in 10-14 years and 15 years females, lower weights and heights were found among rural students (10-14 years: 30.5 kg, 137.9 cm; 15 years: 41.8 kg, 151.5 cm, P < 0.001 for both). Except for students in the 5-9 year age group, weights and heights of female students were greater than male students in urban and rural areas. However, statistically significant differences between rural and urban students were found in all age groups (P < 0.05).
A comparison of the means of weights and heights of all study participants with the expected standard means of Agarwal et al.,  without subclassification ],[ revealed that except for weight and height of females in the 10-14 year age group, the mean weight and height of study students were lower than expected standards [Table 3].
Sub group analysis-weight for age
Overall, one in seven students, 13.8% (95% CI : 12.8 - 14.8) (n = 607) were underweight (below 3 rd percentile of weight for age). In all age groups in males and females in both urban and rural areas, the largest proportion of students was found between 3 rd and 50 th percentile which ranged from 44.2% to 65.8%. Students who were below the 3 rd percentile (undernourished) ranged between 4.5% (rural females ≥15 years) to 38% (urban males 5-9 years). The differences between undernourished rural and urban students was found to be statistically significant among 5-9 year males (P = 0.008) and females (P < 0.001) as well as 10-14 year old females (P < 0.0001). Only a small proportion of students (2.4%) were in the obese group (>97 th percentile) in all age groups and in both rural and urban areas [Table 4].
Subgroup analysis-height for age
Similar to weight for age, the total proportion of students below the 3 rd percentile of height for age irrespective of gender or place of school was 13.1% (95% CI: 12.1 - 14.0) (n = 573). In all age groups in males and females in both urban and rural areas, the largest proportion of students was found between 3 rd and 50 th percentile which ranged from 44.7% to 65.7%. Students who were below the third percentile (stunted) ranged between 4.9% (urban females ≥15 years) to 25% (urban females 5-9 years). The differences between stunted rural and urban students (both sexes combined) was found to be statistically significant (P = 0.001) in all age groups in both sexes. Only a small proportion of students were in the tall category (>97 th percentile) in all age groups and in rural and urban areas combined (3.1%) [Table 5].
A larger correlation coefficient between weight and height was observed among urban students aged 5-9 years as compared with rural students for both males and females, whereas a higher correlation between weight and height was observed in rural area in the 10-14 year age group among both males and females r = 0.796 vs 0.727; 0.801 vs 0.701. In rural males of 15 year age group, a correlation of 0.641 was found; however, among females of the same age, the extent of correlation was lower (0.410 and 0.358 respectively). A higher rate of growth as revealed through regression was found in the 5-9 year students as compared with 10-14 year or 15 year students in both sexes and in urban and rural schools. The rate of growth showed a declining trend with increasing age in all the groups and was the least in urban and rural females of 15 years [Table 6].
The mid day meal program is the popular name for the school meal program in India and involves provision of lunch/snacks/meal free of cost to school students on all school working days. The age group of 5-15 years is a period of dynamic physical and mental growth. Mid day meals are known to reduce school dropout rates and increase literacy in the country.  Inspite of several efforts to improve the health of school students, the services provided in India tend to fall short largely due to shortage of resources and facilities.  Philanthropic organizations and nongovernmental organizations lend support to the government machinery in the form of provision of midday meals in many areas in Bengaluru. The schools in which this study was done received midday meals from an NGO (Sri Sai Mandali Trust, Malleswaram). This scheme benefits all students for 6 days of the week (except in the summer holidays of 2 months duration).
The well-being of a child can be assessed indirectly by measuring the weight and height of the child. Although NCHS standard  were employed as reference values to categorize students as malnourished or otherwise, they were found to be unsuitable for Indian students, due to ethnic differences in growth between countries and within India. Further, the WHO Reference (2007) is a reconstruction of the 1977 National Center for Health Statistics (NCHS). It uses the original NCHS data set supplemented with data from the WHO child growth standards sample for under-fives. Therefore, standards for Indian students were published first by the ICMR several decades ago (1956-1965).  The studies by ICMR were conducted among the lower socioeconomic group of students and as such may not represent the entire population. Moreover, the improvement in the economic status and resultant better health and nutrition has resulted in an upward trend in growth in India.  Considering the need for new reference values, Agarwal et al.,, published the results of a large multi-centric survey of growth and development of Indian students. These standards were thus used for the present study.
In a study carried out in urban primary schools in Meerut, among 5-11 year boys and girls to assess nutritional status by using the Agarwal's standards, it was found that 49.5% of students were found to be undernourished (<3 rd percentile). The findings of this study revealed that a lower number of 32.2% of 5-9 year old urban students were undernourished (below the 3 rd percentile of Agarwal's standards). The possible reason for better nutritional status seen in our study is likely due to the effect of the midday meal. In India, the coverage of more than 12 crore students from rural and urban areas under the midday school meal is one of the largest nutrition support schemes in the world. The findings of this study are similar to that by Saluja et al.,  who found that a large number of students were found to be undernourished (below the 3 rd percentile for age for weight and height). These findings are also similar to those of a study carried out in rural Haryana among school students of 15 years, in which 14.5% boys and 20.8% girls were found to be undernourished (below the 3 rd percentile for height).  The difference between rural and urban students in the 10-14 and 15 year age group is likely due to the economic constraints faced by rural parents and the fact that these children contribute significantly to farming and household activities. The study is however limited by being restricted to a single centre, and non rating of sexual maturity.
Levinger  reported that school food programs have a marked affect on attendance and school performance. The authors believe that the magnitude of the burden of undernourished students in this study would have been much greater in the absence of the midday meal program. In order to augment the efforts of the government, private organizations like the Sri Sai Mandali Trust, Malleswaram need to come forward. This in the long run will improve nutritional status as well as school attendance, performance and literacy.
We are grateful to the Trustees and members of the Sri Sai Mandali Trust Malleswaram for their support in providing midday meals to school students. We would like to thank the Management of M S Ramaiah Medical College and Hospitals and M S Ramaiah Dental College for their support of all the school health related activities of the Department of Community Medicine. We sincerely thank Mr Dinesh H K, Mr Chethan and Mr D K Shivaram, medico-social workers, in the Department of Community Medicine, M S Ramaiah Medical College for assistance in the data collection and entry.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]