 |
|
Original Articles Indian Pediatrics 2000;37: 1060-1071 |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Changes
in nutritional status and morbidity over time among pre-school
children |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
RAPID industrialization has resulted into phenomenal growth of urban slum settlements in many big cities of India in the recent past. Slum populations living in adverse conditions represent nutritionally vulnerable groups that need immediate attention. Information regarding socio-economic factors, dietary intakes and morbidity in relation to nutritional status of children living in slums is scarce and changes with time in nutritional status is an important area in this context. Earlier studies from India have usually examined malnutrition in slum children in a cross-sectional manner. Prasada Rao et al.(1) observed overt signs of vitamin deficiencies and growth retardation among children from Hyderabad slums. The National Nutrition Monitoring Bureau reports(2,3) too revealed that diets of urban slum dwellers are poor and that undernutrition is visible in these communities. Recent studies from Delhi(4) and Maharashtra(5) also indicate a high prevalence of nutritional defiency disorders and larger proportions of wasting and stunting among slum pre-school children. There is thus a paucity of longitudinal data in this context. The present manuscript therefore, examines changes in nutritional status among pre-school children from urban slums in Pune in a follow-up study.
The study was carried out during the period 1992-1995. Initially five to six big slums in Pune city were visited. Consideration was given to those slums which had separate place for common activities such as running a kindergarten as such a place was thought to be useful for recording anthropometric measurements of children at one place rather than carrying the equipments from house to house. The final selection of three slums was done only after prior visits and meetings with the slum leaders for ensuring sustained co-operation. Once the slum was selected, all the children below 5 years of age were enrolled in the study. Complete coverage of the slum areas was done which included 877 households with a population of 4479 individuals. Slum Community: Detailed socio-economic information was collected in the beginning, using structured questionnaire for each household regarding housing condition, family information, literacy and occupation of parents, income, place from where migrated, hygienic conditions etc. Most houses were poorly constructed and about 40% had area less than 100 sq ft. Majority of the houses had only a door and half of them had a small window often not placed rightly to have cross ventilation. No direct sunlight could reach inside of the house. Average family size was 5.77 and overcrowding was obvious. Considerable proportion (96%) of houses had unclean surroundings and only 19% were clean from inside. Socio-economic information revealed that 44.6% mothers were illiterate and 15% had only primary education. One quarter of fathers were also illiterate and most were engaged as daily wage workers or laborers in low levels of jobs. Average monthly per capita income was between Rupees 100 to 300 in case of 60% families and only 8.9% had per capita income above Rupees 500/-. In each household, we enquired about the age of the mother at first delivery and also about the death of pre-school child, if any, in the preceding calendar year and noted down related details about age, sex and cause, wherever available, to have some idea about the child mortality rates. About 32% of mothers had their first delivery before 18 years of age. Mortality rates were also very high. The crude estimates of infant and child mortality were 87 per thousand and 70 per thousand, respectively. Measurements: It took about four to six weeks to complete measurements of children in one slum. Thus children from all the three slums could be covered in four month’s time. The sequence of slums covered in a season was kept same in the subsequent season so that the time between the two successive measurements in a given slum remained around four months. The average duration between two successive measurements was thus 4 mo (± 15 days). Since the natural seasons also change by about four month's period (viz., June to September - rainy, October to January - winter and February to May - summer) we were able to comment on seasonal changes in the variables measured. Age assessment was done critically and only few mothers required help to recall the birth date with the help of local calendar describing important events and festivals in each month. Weight was recorded (up to 20 g.) using a digital weighing balance (ATCO, India) while height was measured up to 0.1 cm. using standard equipment (UNA, India). Supine lengths were measured for children below 2 years age using Infantometer. These measure-ments were recorded by trained research staff and the same investigator was responsible for measuring a particular measurement throughout the study to reduce the personal error. Morbidity Assessment: Morbidity was also assessed along with anthropometric measure-ments, for recent illness in last 7 days prior to the visit using a proforma describing symptoms of common childhood illnesses. The duration of illness was also recorded. The illnesses were grouped according to the system. Thus, vomiting, diarrhea, dysentery, stomach-ache, etc. were considered as gastrointestinal (GI) ailment. Similarly, cold, cough, otitis media, breathlessness, etc. were considered as respiratory (RTI) ailment. Fever was considered separately and other illnesses such as skin infection were categorized as others. Proportion of ill children in each age group offered the estimates of prevalences and mean days of illness were computed. Clinical Assessment: A detailed proforma was prepared by a pediatrician for clinical assessment of individual child at the time of anthropometric measurements. This included examination of head, eyes, oral cavity, skin, edema, signs for active rickets (rickety rosary, widened wrist and hypotonia) or old rickets (Harrison sulci and deformity), anemia and systemic examination. Vitamin A deficiency was assessed by presence of conjunctival xeroxis/Bitot spot/corneal xerosis/corneal ulcer and presence of follicular hyperkeratosis/dermatosis. Similarly, vitamin D deficiency was assessed by presence of frontal bossing/wide fontanelle/active rickets/old rickets. PEM was confirmed by presence of edema while anemia was assessed in three categories as mild, moderate or severe pallor by examination of eyes and nails by the pediatrician. Assessment of Nutrional Status: Nutritional status was assessed using conventional indicators such as weight for age (W/A), height for age (H/A) and standard deviation scores (Z score). Cross classification of children by their nutritional grade in the initial round against their grade in the final round was done to assess the changes in the nutritional status of children over time(6). A child shifting from one grade of malnutrition to the other grade in the upper direction was considered to show the ‘improvement’. For example, shift from ‘moderate’ grade mal-nutrition to ‘mild’ grade malnutrition was considered as an improvement while the shift in the reverse direction was considered as ‘deterioration’. Statistical Methods: Two sample ‘t’ test was used to compare mean duration of illness in two age or sex groups while independent proportion tests (Chi-square) was used for comparing proportion/prevalences between groups. Mean observed velocities in each age group were compared with standard (NCHS) velocities for respective age groups using one sample ‘t’ test.
The number of children observed in each round are given in Table 1. It can be observed that 85 to 90% children could be observed in various follow-up rounds. Children were not available when they went out with parents or out of station or when the household shifted to other slum area. Table I - Number of Children Observed in Successive Rounds
The computed mean heights and weights in initial round for different age groups are depicted in Table II and compared with NCHS reference in Fig. 1. Attained heights and weights of boys and girls from urban slums were satisfactory only upto first four months, but fell below the standard significantly by six months of age. By two year’s age they were even below 3rd percentile of NCHS. Mean Z-scores for weight and height by age in fact show (Fig. 2) that lowest values occurred around 18 months age in both sexes. It indicates that peak prevalence of stunting and underwight occurs at an earlier age. On the basis of weight for age it was observed that, during infancy, 22% children were underwight (W/A <75) and by five years age it was over 40%. Similar proportions considering height for age showed that only 7% infants were stunted (H/A <90) and by five years age it was 35%. Stunting increased considerably beyond two years age. Table II - Mean (± SD) Weights and Heights of Slum Children by Age and Sex
The results of clinical assessment for major nutritional deficiency disorders are given in Table III. Prevalences were compared for two major broad age groups, viz., 0-2 yr and 2-5 yr as reported studies indicate peak prevalence of malnutrition around two year's age in Indian children. Prevalence of vitamin D deficiency was significantly (p <0.05) higher in younger boys (<2 yr) than older (2-5 yr), while Vitamin A deficiency was significantly (p <0.05) higher among older girls (2-5 yr) than younger. Prevalences for anemia were also higher in older girls than younger, though difference was not statistically significant. These prevalences were significantly higher (p <0.01) in rainy and summer seasons as compared to winter. Prevalence of PEM was about 6-10% among slum children and was maximum in summer. Table III - Clinical Assessment (% Prevalence) for Slum Children in Different Seasons
From initial sample size of 435 boys and 410 girls, children above age 5 yr were excluded. Additionally some children were not available for clinical assessment. Sample size was thus: Winter - 418 boys and 398 girls, total 816 (exclusion of 14 and 15 were not available). Summer - 347 boys and 310 girls, total 657 (exclusion of 64 and 60 were not available). Rainy - 326 boys and 308 girls, total 634 (exclusion of 96 and 42 were not available). Prevalence and mean duration of illness for different morbidities in various seasons is described in Table IV for the above two broad age groups. Fever, gastrointestinal and respiratory illness were the major illnesses. Prevalence of total morbidity was also higher among younger boys and girls compared to older ones, but statistically significant (p <0.05) only in case of boys. Total morbidity was highest in rainy season followed by that in summer season. Prevalence of gastrointestinal and fever morbidity was also higher in younger children as compared to older but the differences were significant (p <0.05) for rainy season in boys and during winter season for girls. Prevalence of RTI was however, higher in rainy and winter season and was also higher (p <0.01) among younger boys than older boys. Table
IV - Prevalence
(%) and Mean (±
SD) Days of Illness for Different Morbidities
* p <0.05; ** p <0.01 for comparison of prevalence / mean days of illness between the two age groups. From initial sample size of 435 boys and 410 girls, children above age 5 yr were excluded. Sample size observed is thus: Winter - 430 boys and 401 girls, total 831 (exclusion of 14). Summer - 382 boys and 335 girls, total 717 (exclusion of 64). Rainy - 354 boys and 322 girls, total 676 (exclusion of 96). The mean days of illness in each age-sex group are also given in Table IV. Mean days of illness due to RTI were higher, though not statistically significant in younger children during rainy season. However, days of illness due to gastrointestinal and fever were significantly higher (p <0.05) among younger boys and girls during rainy season. Thus younger children had higher prevalence as well as higher duration of illness as compared to older children. The association of nutritional status and morbidity - a major determining factor for growth retardation - was therefore examined (Table V). Morbidity due to gastrointestinal illnesses and fever especially during rainy season was considered for this purpose. Prevalence of gastrointestinal illnesses and fever as well as mean days of illness were higher among underweight boys and girls in both the age groups compared to their normal counterpart but the differences were not statistically significant. Stunting didn’t affect morbidity prevalence or duration. Table
V - Prevalence
and Mean Days Lost Due to Gastrointestinal Illness and Fever
Sample size observed in the rainy season was 354 boys and 322 girls, total 676.
The observation that malnutrition sets at an early age and that younger children experience higher morbidity, has obvious implications for growth velocity. The mean annual increment in weight and height by age are plotted in Fig. 3. Weight increments were significantly lower as compared to NCHS values, throughout pre-school age. The height velocity was also significantly (p <0.01) lower upto three years age and were similar thereafter. The smaller velocities in early life thus, result in shorter attained heights by almost 10 cm. by the age of 5 years. There was no scope for catch-up growth as velocities beyond 3 years age remain similar. This highlights the importance of the early life undernutrition which may be responsible for shorter adult size observed in poor communities of India. The changes in nutritional status over time were also examined using the method of Rao and Kanade(6). Children were classified by their grade of malnutrition in the initial round against their grade at the end of two year’s period, to compute proportions of children showing improvement, deterioration or no change in their nutritional grade over time. It was observed that the extent of deterioration in weight as well as height status was significantly higher (p <0.01) than the extent of improvement in the younger group (Table VI). Deterioration in weight for age status was observed in 34% children, while that in stunting was observed in 49% children. In contrast, the proportion showing improvement in weight and height status was only 13% and 10%, respectively. Beyond two years age however, shifting from one grade to another was not prominent as most children remained in ‘no change’ category. The analysis thus, highlights that slum children not only experience malnutrition at an early age, but deterioration in nutritional status starts in the younger age and was higher, probably due to higher morbidity. Table
VI - Proportion of Children in Broad Age Groups Showing Changes in
Nutritional
+ Sample size (n=608) and consists of children present both in the initial and final round. Figures in parentheses are percentages.
Urban slums in all big cities of India are growing at an alarming pace and yet, sufficient attention has not been paid to understand nutritional problems of slum populations. Gopalan(7) in fact, pointed out that problems of urban slums in India would be of greater dimensions and that they would need far more attention than that in the past. Nutritional status of slum children is even poorer than rural children(8-10). However, unlike rural areas, urban slums are characterized by adverse living conditions, higher population densities and poor hygienic and sanitary conditions and therefore need special attention. Growth of infants is influenced initially by adequacy of breast milk and subsequently by infective morbidities they experience. Children in our study had comparable body weights and heights upto first four months, indicating adequacy of breast milk despite the poor nutritional status of slum mothers. However, subsequent rapid fall in their nutritional status could be attributed to two major factors. Firstly, duration of exclusive breastfeeding (3.5 mo) was significantly shorter in slum children than that observed in rural communities(11-12), thereby depriving them of protection from morbidities(13). In addition, lack of appropriate supplementary weaning foods was responsible for early undernutrition. Secondly, the poor nutritional status beyond six months age increases risk for morbidity due to poor environmental conditions in urban slums. Similar observations have been reported from other Indian studies(14-18). Peak prevalence of malnutrition was thus observed around 18 mo in our study and is much earlier than that reported for rural children. Growth studies carried out in developing countries report peak prevalence of malnutrition and growth faltering between the age of 2 and 3 years(19). Saxena et al.(20) too report that peak prevalence of wasting was observed in 12-23 mo age while that for stunting was 48-59 mo age in children from Delhi slums. Promotion of breastfeeding is, therefore, highly essential among slum communities. Clinical assessment indicated higher prevalence of anemia. Symptoms of vitamin A deficiency such as conjunctival xerosis, corneal xerosis or folicular hyperkeratosis which are believed to be rare, were also seen in the slum children and indicate the severity of the problems. Severe nutritional disorders in infants(21) and pre-school children(5) have been reported from slums of Calcutta and Maharashtra, respecitvely. Anemia continues to be a significant public health problem(22) among children from slum communities. PEM too was higher in our study children than usually reported in rural children. Clinical Vitamin D deficiency was 10 to 20% in different seasons in our study but has rarely been reported earlier. The mothers mostly working as housemaids, and the elder siblings often playing outside, left the young children (<2 yr) inside the house, often depriving them of sunlight due to adjacent huts from the three sides. Thus young children were hardly exposed to sunlight and high prevalence of vitamin D deficiency actually appeared characteristics of slum children. Prevalence of gastrointestinal illnesses and fever was higher in younger children than older and contributed more than 50% of total days lost in illness. Morbidity was higher in two of the three seasons in a year. Kapil et al.(4) observed similar prevalence of diarrhea in children below 3 years from Delhi slums. We compared our finding with the recent information available from NFHS data(23). Since it does not give separate figures for slum communities, our findings are compared with rural data. Even for rural data, the estimates of morbidity and mortality are not available for age and sex. Nevertheless, the comparable figures for mortality show higher infant mortality (81 per 1000) among slum children than respective figures for rural children (61 per 1000). Duration for exclusive breastfeeding for slum children observed in our study was considerably smaller (3.5 mo) compared to that for rural children reported in NFHS data (6.2 mo). The NFHS estimate for prevalence of fever and diarrhea among pre-school children are 22.2% and 10.2%, respectively indicating that fever was major illness. In slum children too, we found that the prevalence for fever was higher (about 22%) than that for diarrhea (about 15%). However, prevalence of diarrhea was higher among slum children than that for rural children. The NFHS data also show that the morbidity prevalences were higher among younger children (<2 yr) than older and supports our findings in case of slum children. The NFHS data on nutritional status (in terms of weight for age and height for age) show that prevalence for wasting was higher (57.5%) than that for stunting (50.8%) in rural children. In contrast, we find that in slum children prevalence of stunting (55%) was higher, though statistically not significant, as compared to wasting (51%). Thus, there seem to be difference with respect to breastfeeding practices, morbidity and patterns of under-nutrition in slum and rural children. Our study additionally provided data on changes over time in nutritional status of slum children. It revealed that 40-50% children below 2 yr age further deteriorated their nutritional status during the study period, while most chidren above 2 yr age remained in their same lower grade of nutritional status. The undernutrition thus sets in at an early age, which increases risk of infections leading to further deterioration of nutritional stutus. High morbidity profile significantly affected their growth velocity for weight which remained low compared to the standard, till five years age. In case of height, lower velocity upto 3 years age resulted in deficit in heights by at least 10 cm by the age of five years. Since the velocities remained similar beyond 3 years, there appeared to be no scope for catch-up growth. This highlights the need for concentrating the efforts on preventive measures to reduce morbidity especially among younger children in slums. Nutrition interventions alone are unlikely to improve nutritional status of slum children. In conclusion, analysis attempted in this study brings out important implications, both for prevention and intervention programs to be planned for combating the problem of under-nutrition in urban slums. Poor environmental conditions giving rise to infections play major role in deteriorating the nutritional status of pre-school children in slum community. Inter-vention programs planned for rural children may not be applicable in totality to slum children, as morbidity profile, the clinical profile and types of malnutrition were typically different. It is, therefore, essential that attempts are made further to examine the etiology of malnutrition in slum communities for planning appropriate programs.
This work was carried out under the project ‘Study of trends in undernutrition among urban slum children and their implications for nutrition policy’ funded by Science and Society program of Department of Science and Technology, New Delhi. The help of DST project staff for field work is acknowledged. Authors are also grateful to Dr. A.D. Agate, Director, Agharkar Research Institute, for providing facilities. Thanks are also due to Dr. (Mrs.) Kanade for help in organizing the field work initially. Contributors: SR was the Principal Investigator for the sponsored project and was also responsible for its design, overall implementation and interpretation of results. Based on these, she drafted the present paper and will act as a guarantor. SBJ was responsible for executing and superivising of the Project in the field, data collection and its analysis required for this paper. RSK actually participated in data collection and was also responsible for data entry. Funding:
Department of Science and Technology, New Delhi.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
