Potentially modifiable micro-environmental and co-morbid factors associated with severe wasting and stunting in children below three years of age in Aligarh district
S Sachdeva, A Amir, S Alam, Z Khan, N Khalique
microenvironment, stunting, wasting, z scores
S Sachdeva, A Amir, S Alam, Z Khan, N Khalique. Potentially modifiable micro-environmental and co-morbid factors associated with severe wasting and stunting in children below three years of age in Aligarh district. The Internet Journal of Epidemiology. 2008 Volume 7 Number 2.
Undernutrition among children is a major public health problem in developing countries like India. Poor nutrition during childhood affects growth potential and risk of morbidity and mortality in later years of life. Malnourished children are more likely to grow into malnourished adults who face heightened risks of disease and death 1 . A number of factors affect child nutrition, either directly or indirectly. The most commonly cited factors are food availability and dietary intake, breastfeeding, prevalence of infectious and parasitic diseases, access to health care, immunization against major childhood diseases, vitamin A supplementation, maternal care during pregnancy, water supply and sanitation, socioeconomic status, and health-seeking behavior 1 . Children, especially the infants and toddlers constitute the most disadvantaged group. Severe acute and chronic malnutrition setting in at this stage could permanently impair the psychosocial development of the children 2 . The present study therefore attempts to investigate the potentially modifiable distal and proximal factors that cause severe malnutrition in children under three years of age and suggest ways to mould them to their advantage.
Study areas and subjects
This community based cross sectional study was conducted (during the period of August 2007 to June 2008) in the field practice areas of the Department of Community Medicine, J N Medical College, Aligarh. The urban health training center (U.H.T.C) has four registered peri-urban localities with 1670 households and a registered population of 10,250. The rural health training center (R.H.T.C.) has seven registered villages having 2400 households with a registered population of 14,600. A community based household survey was conducted in the registered areas.
Sample size and Sampling Method
The estimated sample size was calculated according to the formula: N=4pq/d^2 where p is the prevalence of malnutrition, q=1-p, and d is relative error. Taking the prevalence of malnutrition in children under three years of age as p=45.9% (NFHS-3) 3 and relative precision d as 10 % of p, the sample size was calculated to be 468. The sample was taken from children under three years of age present in the household and whose parents gave consent for the interview. The response rate was 96% and a total of 486 households (146 urban and 340 rural) were visited to attain the required sample. Systematic random sampling was done in the respective areas such that one eligible child from every tenth household (common sampling interval) was included randomly. For simplification, the lowest
A preformed proforma (questionairre) was used for the study. The proforma was divided into two parts. First part of the proforma contained the identification data, socio demographic and environmental profile of the child’s family. The social class of the child’s family was determined using the Modified Prasad Scale 5 . A precise history of dietary intake of the child was elicited from the mother (recall of food items consumed in last 24 hours). Second part of the proforma included anthropometry and general physical examination of the child including signs of protein-calorie malnutrition and micronutrient deficiencies.
Height and weight measurements were recorded following the standard techniques. The spring balance with an infant tray was used for weighing children up to 10 kg and one with a floor dial for older children (more than 10 kg). The scales could easily be ‘zeroed’ without any weight on them. Further, daily calibration was enabled before survey. Weights were taken with the subjects wearing minimal clothing to the nearest 0.1 kg and 0.5 kg, in the infant and toddler scales, respectively. Supine length was recorded to the nearest of 0.1 cm using an infantometer for children up to two years of age .A portable stadiometer with a head piece was used to measure heights of older children. The exact age of the child was computed from the child’s date of birth. When data on the exact date of birth was not available, the age as told by the mother was used, corrected to the nearest month. A regional local-events calendar was used to assist the mothers for better recall. Wasting and stunting were used to evaluate the nutritional status of the subjects as per Centre for disease 2000 norms. Age and sex specific – 2 z-scores were followed to define wasting and stunting.
Social classes I, II and III of the modified Prasad’s classification were categorized as upper class and IV and V as lower class. Appropriate breast feeding was exclusive breast feeding for six months of age and continued up to two years along with semi solids. Appropriate feeding was as per the Integrated Management Neonatal Childhood Illness guidelines. The definitions for appropriate and inappropriate care utilized for the study were appropriate care sought from qualified medical professionals in government health facilities and private hospitals/clinics. Purchasing medicines from pharmacy, home remedies, visiting pharmacies, temples and traditional healers was defined as inappropriate care
Analysis was performed using SPSS version 10.0 (SPSS, Chicago, IL). Continuous variables were expressed as mean standard deviation (Gaussian distribution) or range and qualitative data was expressed as percentage. Unpaired t test for independent samples was used in comparing continuous data between sexes. Chi square test and Fisher’s exact test were used for univariate analysis. All p values were two tailed and values of <0.05 were considered to indicate statistical significance All confidence intervals were calculated at 95% level. Binary logistic regression was used to do the multivariate analysis.
The majority (282/ 60.25%) of children were seen in the 12-36 months age group and the least (101/18.16%) in the 0-6 months age group (Fig 1).
Female children constituted 48.08% of the study group. Almost half (51.46%) of the 486 children belonged to Muslim families and rest were from Hindu families. Sixty percent were living in nuclear families. Almost 83% were living in overcrowded dwellings and 76.5% belonged to the lower socio economic class. About 69.1% were practising open air defecation.
Mean (SD) z scores for height for age and weight for height of the study population were -2.93(1.07) and -1.46(1.34) respectively. The z scores for height for age in males (-2.52±1.02) and females (-2.25±1.12) were comparable (95% CI -0.13 to 0.46). Respectively the weight for height (-1.43±1.46 Vs -1.49±1.20, 95% CI-0.29 to 0.18) were comparable between the two genders in the study population (Fig 2 & 3).
The age-wise mean z scores for weight for height and height for age in both genders did not reveal any significant difference except in the age group 7-12 months wherein the mean height z scores was significantly ( 95% CI 0.33 to 1.08) lower among boys than girls. Males were more severely stunted than females in all age groups. Males were more wasted in the two younger age groups. All these differences did not reach the level of significance (Table 1).
The overall age and gender combined rates of wasting (37.60%) and stunting (59.18%) were high. The overall prevalence of severe wasting and severe stunting were 8.33% and 30% respectively. Age combined rates of severe malnutrition (< -3 z scores) for height (28.8 % Vs 26.2 %) and weight for height (9.01% Vs 7.5%) were more common among males than for females.Social class (OR 2.3, 95% CI 1.0 to 5.1) and feeding practices (OR 2.6, 95% CI 1.3 to 5.1) were significant risk factors associated with wasting. However, only rural children were more likely to be stunted (OR 2.1, 95% CI 1.3 to 3.1) and none of the other socio demographic factors had significant association with stunting (Table 2).
Among the childhood morbidities presence of measles (Odd’s ratio 3.9, 95% CI 1.4 to 11.3), vitamin A deficiency (OR 3, 95% CI 1.2 to 7.8), vitamin D deficiency (OR 6.7, 95% CI 2.5 to 17.8) and worm infestation (OR 3.3, 95% CI 1 to 10.3) were significantly associated with wasting. Pallor (OR 129.7, 95% CI 54.2 to 310.8) was found to be the most significantly associated with stunting. Another factor that had a borderline association with stunting was the presence of vitamin D deficiency (OR 3.01, 95% CI 1.3 to 7.2) (Table 3).
On binary logistic regression analysis social class (Adjusted OR 1.7, 95% CI 1.1 to 2.6), feeding practices (Adjusted OR 2.8, 95%CI 1.4 to 5.9), Measles (Adjusted OR 5.2, 95%CI 1.4 to 19.3), vitamin D deficiency (Adjusted OR 5.3, 95%CI 1.6 to 17.9) were the factors found significantly associated with wasting. On applying logistics regression analysis only pallor (Adjusted OR 1.7, 95%CI 1.1 to 2.6) was found significantly associated with stunting.
The two nutritional indices (wasting 37.6 %; stunting 59.2 %) were staggeringly higher in the present study as compared to the national (19 %; 38. %) (http://www.nfhsindia.org/pdf/IN.pdf) and state figures (14 %; 46 %) (http://www.nfhsindia.org/pdf/UP.pdf)
An earlier study done in south Indian children revealed 8% and 21% severe wasting and severe stunting 6 . These figures are lower than that found in the present study specially in case of severe stunting. In some other better developed states of India (Haryana and Kerala) the surveys have not found children with severe malnutrition 10,11 . The pooled prevalence of severe wasting and severe stunting in Punjab districts were 2.8% and 38% 7 . The mean z scores for height among both the sexes declined as the child progressed from infancy through becoming a toddler. A largely similar trend was observed with the mean z scores for the weight for height. The better nutritional status of infants is probably due to their nutritional needs being met through breast milk and some complementary foods. However, after the first year of life, when breast feeding no longer meets their nutrient needs and complementary food is inadequate, there was an increase in the prevalence of under-nutrition. The findings are consistent with those of Benjamin and Zachariah 12 of Punjab who found that the highest proportion of malnourished children was in the second year of life. Better mean z scores during infancy further confirm the nutritional superiority of breastmilk and propose a case for exclusive breast feeding. The mean age and gender combined z scores for height and weight for height in the 0-6 months age group were -1.51± 0.96 and -1.12 ±0.98 respectively. Undernutrition and failure to thrive in this age group could be due to intra mural and neonatal events. However these factors have not been studied in the present work.
Children belonging to the lower social classes and where feeding was inappropriate were significantly more wasted than the rest. These results were confirmed by other workers 13 indicating that unavailability of food, insufficient purchasing power, inappropriate distribution and inadequate utilization might make the children vulnerable to malnutrition in a deprived community. Long duration of breastfeeding without introduction of appropriate complementary feeds may be associated with higher malnutrition because it reflects lack of resources to provide children with adequate nutrition 14 . It is also possible that children who are breastfed for a long time are more reluctant to eat other foods, as was found in a study on a cohort of Ghanaian children 15 . Moreover the present study found measles an important co-morbidity associated with severe wasting which has been seen in a case control study of severely malnourished children with diarrhea in Bangladesh 16 . Whereas a Ugandan study found presence of fever in the preceding two week period as a risk factor for wasting 17 . Surprisingly we also found vitamin D deficiency significantly associated with wasting. This could be due to wasting and stunting present together in some children. Contrast to the present study which only found pallor as a risk factor for severe stunting, some workers from the developing world 18 have found family size, parental education 17,19 , poor breastfeeding pattern and inadequate complimentary feeding associated with chronic malnutrition. In a study done in Kerala anemia was seen to be significantly associated with mild and moderate malnutrition 11 .
There is a close positive link between the nutritional status of pre-school children and the stages of development of the states. Mothers' education and household conditions are important influences on children's health status irrespective of the stage of development 20 . Uttar Pradesh, where the present study was carried out, is one of the worst affected states in India. Focus, therefore should be on domiciliary management of severely malnourished children (without medical complications). The same can be achieved at primary health level with appropriate training of the anganwadis and auxillary nurse midwives. Intervention programmes for the most vulnerable groups should be planned to mitigate childhood morbidity and mortality. Micronutrient supplementation and health education of the caregivers through simple health packages would go a long way in alleviating the co-morbidities.