Mother and Child Health Care
Child Caring and Feeding Practices
Food Distribution within the Household
Education
Socio-economic Status
Gender Issues in Child Care
Maternal and child care, along with household food security, adequate health services and a healthy environment is a third necessary (but in itself insufficient) precondition for adequate nutrition. In fact, ‘care’ may be considered as a pivotal link between these two other conditions, representing the behavioural component of intra-household decision-making and resource use. It refers to the provision in the household and the community, of time, attention and support to meet the physical, mental and social needs of the growing child and other family members. In the child nutrition context, most importantly it involves the optimal use of household resources for child feeding, protection from infection, and care for the sick child. While the issue of ‘caring capacity’ refers to all the household members -- male and female -- who are potential caretakers of children, in practice in India, the main responsibility for child care lies with the mother (who often also has a major role as an income-earner). Her capacity to manage the many competing demands on her time will govern the degree to which she can maintain a clean household environment, feed her children, care for them when sick as well as providing and preparing food for other household members.
In turn, this capacity will be governed by the quantity, control and use of resources such as food, income, time and knowledge. The control of such resources at a societal level may be influenced by factors such as her socio-economic and educational status. We thus start with an examination of the feeding and caring practices in the household before examining the possible basic causes relating to the role of women in Indian society and their educational status and literacy levels relative to men.
This integral precondition for adequate nutrition may not be as well recognised by policy-makers, compared to food and health. Partly this may be as a result of its many linkages with the areas of food and health, and partly as little data is routinely collected to reveal the gender dimensions of nutrition problems. These are discussed in the concluding section.
The type of care received at child birth is often critical for the health and survival of both infant and mother. A significant proportion of neo-natal deaths is attributed to poor birth practices. During 1987, only about 32% of births in rural areas and 74% in urban areas were in institutions or attended to by trained personnel (Registrar General of India 1979-86). Traditional birth attendants are unable to attend to complications and health professionals are contacted too late. Both these factors point to the need to identify mothers at risk during the prenatal period.
Recent reports show that tetanus toxoid immunisation coverage is 77% of pregnant women in India (EPI 1990). Started in 1960, and boosted in the second half of the 1980s by the immunisation mission, this intervention is picking up as part of ante-natal care (see Figure 18). The national average of tetanus toxoid coverage however masks variations between states ranging from 16% in Assam to 99% in Kerala.
Even though abortion has been legalised in India since 1972, mortality and morbidity due to illegal abortions and birth attention by incompetent persons in unhygienic conditions remain a major problem, mainly because of ignorance of the law and inaccessability of professional services in rural areas. Only around half a million pregnancy terminations were performed through the health services in the fiscal year 1987-88 which is around 9% of the induced abortions likely to have been performed during the same period. Since the inception of this formal facility in 1972, 5.8 million abortions have been performed under it; less than the total number of induced abortions likely to happen in one year (UNICEF 1990 p15). Induced abortions in fact reflect an unmet need of women for family planning, and highlight gaps between demand for family planning on one hand and availability, accessibility and actual use of services on the other.
India is the first country to launch an official family planning programme to control population. However, the programme has not had the desired impact. The trend in the percentage of couples protected by various methods of family planning are shown in Figure 19. The target fixed by the National Health Policy is to cover 60% couples by 2000 AD, while the present protection is 41% (all methods). Efforts must be made through different channels of mass media to propagate the acceptance of a small family norm.
Figure 19. Family Planning - Couples using contraceptives
Source: GOI Ministry of Health and Family Welfare
Child feeding practices such as breastfeeding, weaning and feeding sick, anorexic children have a bearing on the nutritional status of the child. A comprehensive study on infant and child feeding practices carried out by the ICMR in six different regions of the country - Coimbatore, Gandhigram, Hyderabad, New Delhi, Pune and West Bengal (ICMR 1990) - indicated that a great majority of women started to breastfeed their new-borns on the third day after delivery, while liquid/semi solid supplements were rarely given to infants before six months (except in West Bengal).
A collaborative study on contemporary patterns of breast feeding conducted by WHO (1981) indicated that the prevalence of breastfeeding in the rural regions was related to socio-economic background. In general, it was more common in rural than urban areas, and within the urban population it was more prevalent among the poor than the economically advantaged. As many as 95% of infants belonging to poor rural and poor urban mothers were breastfed, even at 15 months of age, while corresponding percentages for high and middle income groups were 18% and 51%. In another study (GOI 1975-88), comparison of child feeding practices in ICDS (Integrated Child Development Services) and control areas had shown that nearly 85% mothers started breast feeding their newborn children within 6 hrs after delivery. The report indicated that though there was no significant difference in the breast feeding practices between ICDS and control areas, delayed weaning was more common in non-ICDS children (28% vs 59%). The ICDS is reviewed in Part II.
The four-state ICDS study undertaken by NIN (Sarma et al. 1990) has indicated that nearly 50% of the mothers did not alter the diets of their children when they were sick. Quality of intranatal care is another important indicator of health service availability and utilisation. The ICDS report shows that 76% of the deliveries in ICDS areas were conducted by trained personnel compared to 49% in non-ICDS areas. The immunisation coverage of children between 12-24 months also showed that the ICDS children in most of the states were better protected. In the states of Tamil Nadu and Maharashtra, however, the coverage was more or less similar in ICDS and non-ICDS areas. Similar results were observed in the four-state ICDS study undertaken by NIN where immunisation coverage, antenatal care, massive dose vitamin A administration, folifer distribution etc., was two to three fold better in ICDS areas.
Analysis of dietary data to assess intra-family distribution of food has shown that in 50% of the households surveyed, levels of energy adequacy did not differ between preschool children, adult men and women. Either all of them were consuming adequate amounts (31% of households) or inadequate amounts (19% of households). When intakes were corrected for requirements, it was observed in NNMB surveys that the average calorie intake levels of women were close to 94% of their RDI as against 85% in men. This is contrary to the general belief that women get least. However, in 59.4% of the households, the diets of preschool children were deficient in energy when judged on the most conservative cut-off of -2SD RDI.
The average food/nutrient intakes of preschool children were assessed by oral questionnaire method in the NNMB repeat surveys. The intake of cereals and to some extent fruits and sugar, showed an increase between 1975 and 1990, and no remarkable changes were seen in the consumption of other foods. The average daily per capita intake of calories among children 1-3 years old during 1988-90 was 908 kcal as against 834 kcal in the seventies, although the RDI is 1240 kcals (NNMB 1991); corresponding figures for 4-6 year olds were 1260 kcal and 1118 kcal, while the RDI was 1690 kcals. The increase over time in both cases has been mainly due to increased consumption of cereals rather than other foods. It shows that, although overall household food consumption has not changed much in the last 15 years, there has been some preferential allocation to children -- probably via an increased awareness of the nutritional needs of growing children by parents, along with the beneficial effects of direct nutrition interventions initiated during this time period. There was no notable difference in food consumption between boys and girls -- a finding in line with the lack of gender differentials in anthropometry and IMRs.
Resource allocation
Infrastructure development
Female literacy outcomes
The total expenditure on education increased from 1.2% of the GNP in 1950-51 to about 4% in 1986-87, against an optimal target level of 6%. The real per capita expenditure also increased steadily, particularly in the late 1980s, from $ 0,68 in 1975-76 to $ 1.64 in 1989-90 (see Figure 16).
The share of education as a proportion of plan outlay in the public sector has shown a declining trend through the plan periods. Within the education sector, the share of elementary education has been falling from 56% in the first five-year plan to 29% in the seventh plan; higher education has benefited at the cost of primary education. There is a need to step up the resources for formal and non-formal education. Assuming that 70% of the 6-14 age group will be provided education through formal and 30% through non-formal channels, the annual per capita cost for universal elementary education has been estimated at Rs. 103 (formal) and Rs. 54 (informal) by the year 2000.
As of 1987-88, there were 543,677 primary, 141,014 middle and 71,305 secondary and higher secondary schools in India. The growth of educational facilities during the last one and half decades has been steady (GOI 1987).
Inter-state variations in school enrolment indicate that gender differentials are pronounced, though the percentage of scheduled castes/tribes out of the children enrolled at primary level is at par with, or higher than, the percentage of scheduled castes/tribes in the total population (UNICEF 1990). The rural-urban divide also shows up sharply in school enrolment and as a gender differential.
Available information on retention rate in primary classes indicate that it is quite low, particularly for girls. For example, in 1985-86, the retention rate for class V was 56.9% for boys and 51.0% for girls (CSO 1989). The drop-out rate worsens as girls move from lower to higher classes. Existing measures therefore need to be strengthened in order to have continuous education for girl children.
The new educational policy adopted by the Government in 1986 accords a very high priority to universalisation of education to ensure essential minimum education to all children up to the age of 14 years.
Changes in literacy rates during the 1980s are impressive for both males and females (Registrar General of India 1991). According to the 1991 census, about 52% of India’s population is literate. At the time of independence in 1947, the female literacy rate was a mere 6%. Over the years however, there has been a steady improvement in the rate, although the absolute number of female illiterates has increased from 215 million in 1971 to 242 million in 1981. This backlog is estimated to have further swollen to 253 million in 1988, notwithstanding the rise in the female literacy level reported by the 1987-88 National Sample Survey. Of the 340.5 million illiterates above 5 years in India in 1981, as many as 200.3 million were women. Of them 170.7 million live in rural areas. In other words, more than half of the total illiterates in India in 1981 were rural females, and this proportion remained throughout the 1980s. The female rural literacy rate in the 10-14 year age group was 36.4%, but it declined progressively with increasing age and is only 8.6% in the above-35 age group (UNICEF 1990 p88).
Overall, female literacy rates improved threefold from 1961 to 1991 (from 13.3 to 39.4%) while for males the level has gone up from 34.4% to 63.8% during the same period. Clearly however the female rate is still much lower than that of males (Registrar General of India 1991) - in fact the female rate in the early 1990s can be seen to approximate the male literacy rate of a quarter a century ago.
The regional variation with regard to female literacy is striking. In the 1981 census, the highest female literacy rate (5 years and above) of 78.9% is registered by Kerala (UNICEF 1990 p102). By contrast, among the 14 most populous states, four states viz., Rajasthan (28.4%), Madhya Pradesh (32.3%)- Uttar Pradesh (31.4%) and Bihar (30.2%) ranked the lowest. These states accounted for half the illiterate rural women in India (as well, in fact, as the majority of India’s poor).
There have been efforts to correlate female literacy with age at marriage, fertility rates and child mortality. In rural areas, a higher proportion of married women are illiterates as compared to urban areas. Further, among illiterates, around two-thirds of women got married before reaching the age of 18 years, suggesting a positive correlation between age at marriage and level of education. Available data also suggest an inverse correlation between a woman’s level of education and her fertility.
Child mortality rates are about five times higher among illiterate mothers compared to graduates (Registrar General of India 1989). Better child survival among the educated group may be due to several factors such as better hygiene, improved nutrition and feeding practices, and timely medical intervention. A study conducted by NIN (Brahman et al. 1988) showed that, controlling for income, the energy content of the diet of children whose mothers were literate tended to be better than those whose mothers were illiterate. There are other case studies showing that maternal education has a significant influence on the nutritional status of the children (Walker and Ryan 1990).
Women’s literacy and their use of health facilities go hand in hand. Krishnan (1985) examined overall death rates in terms of literacy, doctor, hospital and bed population ratio, per capita income and % per capita expenditure on medical and health services. He observed that literacy was the most important factor while health services also had some explanatory power.
The capacity of a mother to care for her children depends on her social status and economic activities. In many poor societies patriarchy is likely to be the main obstacle in securing a fairer distribution of work and decision-making power between adult household members.
Increases in the ratio of female to total income is expected to improve the economic status of women within the household and their control over resources. Their ability to realise their own preferences within the family (of which health and well-being of children is likely to be a priority) may consequently be strengthened. However, working outside the house may leave her little time for child care. It is a complex situation and women’s problems are difficult to capture in national surveys. There have been a few case-studies to assess the impact of women’s work and income on child nutrition. The analysis of women’s work and child survival undertaken by Rosenzweig and Schultz (1982), whose two-stage regression analysis of an all-India sample of rural households demonstrated that female employment had a significant influence on the survival of the girl child. A case study undertaken in Kerala (Gulati 1978) indicated that the nutritional adequacy of households dependent on agricultural labour was more related to women’s employment than men’s. A study undertaken by the Maharashtra Employment Guarantee Scheme (Walker and Ryan 1990) indicated that in households where women exercise control over their wages, more money was spent on food and other basic needs while men tend to spend more on liquor, cigarettes etc. These studies suggest that women’s gainful employment and decision-making power in the family influence the child health and survival (see ‘Gender Issues in Child Care’ section below).
Labour-force statistics under-report female contributions at a national level. In the Indian census of 1981 only 14 per cent of the total female population were classified as "workers" although an estimated 54 per cent of rural women and 26 per cent of urban women are engaged in work activities. The basic difficulty in reporting lies in the extremely hazy demarcations between economically productive employment and domestic work within the subsistence sector, especially as it relates to agricultural activities (though the latter should also be considered as economically productive in that it maintains the labour force).
Data on gender differences in child care and access to food and health care are contradictory. While recent NNMB data fail to show evidence of gender discrimination in food consumption (and by implication, in food access and allocation), anthropological and other evidence to the contrary has been reported from numerous studies. Miller (1981) has collated evidence from thirty one ethnographic studies from several regions in India on allocation of food, medical care, and love and affection. Of the 13 studies that report on food allocation among young children, only two studies indicate no gender differentiation in feeding practices, while 11 studies indicate caretaker bias against female children.
This bias is most often exhibited during early infancy in breastfeeding, weaning and supplementation practices, and in later years in apportionment of quality foods such as milk, butter, snacks and sweets. Most studies conclude that gender biases are small (which may be one of the reasons why these are unlikely to be picked up by generic surveys), but as one study states "in an emergency...[a daughter] is more readily expendable than a son". Twelve other studies that document gender-related food practices during adolescence present evidence for allocation of special foods to girls for periods ranging from a few days to a few months at/around the time of menstruation. Of the ten studies that report on allocation of medical care, all but one infer better medical care for boys.
Only two of the studies reviewed address the issue of sex differences in nutritional status, and both present evidence for consistently higher prevalences of underweight girls than boys. Many of the studies also suggest more pronounced gender differences in northern India, and among the less privileged classes. The report presents further evidence that sex ratios at birth are similar in all regions of the country, while juvenile sex ratios (females/males) are significantly low in North India - the region where evidence for gender discrimination is most strong (sec Table 13). While some evidence of female infanticide may explain this differential, the implication is that overt or covert marginalization in child care in North India translates into greater female mortality in the early years and a consequent rise in the sex ratio.
Table 13: Sex ratios in Indian regions (females per 1000 males)
|
Region |
Sex ratio at birth # |
Juvenile sex ratio * |
|
North Zone |
962 |
847-980 |
|
East Zone |
943 |
952-1031 |
|
South Zone |
943 |
980-103 |
|
West Zone |
935 |
952-1000 |
|
Central Zone |
926 |
1000-1031 |
|
North West Zone |
935 |
901-952 |
Sources: # Miller (1981) Appendix A, * Extrapolated from Miller (1981), Figure 4 (expressed as numbers of females per 1000 males).Five explanations are offered by Kundu and Sahu (1991) for the worsening sex ratio (from 933 females per 1000 males in 1981 to 929 in the 1991 census): i) a progressive undercount of women compared to men in different censuses, ii) an increased discrimination of females (including infanticide) in providing the minimum nutrition, access to health and other amenities, iii) increase in the proportion of male selective migrants from other countries, iv) reduction in foetal wastage resulting in a decline in female-male ratio at birth, and v) female selective termination of pregnancy leading to, once again, a decrease in the sex ratio at birth.
A recent World Bank country study (World Bark 1991) has re-analyzed district-level data from Miller’s study to show that between 1961 and 1971, there was a substantial increase in the percentage of rural districts that exhibit higher female mortality. Again, this drift is many times greater in the north than in the south. Country level data also suggest that the sex ratio in India has increased between 1961, 1981 and 1991 (UNICEF, 1990).
Relatively recent evidence of female infanticide has been provided by a study carried out in North Arcot district of northern Tamil Nadu (George et al. 1992). Between April 1st 1987 and September 3.0 1989, of 759 live births (from a study population of 13,000), 56 infants died - 23 males and 33 females. Of these 19 were confirmed as infanticides, and all were girls. Thus more than half of all female infant deaths in the 12 study villages during this two and a half year period were infanticides. An infant mortality rate for the whole study population was calculated at 69 per 1,000 live births. If infanticides are subtracted out, this leaves an IMR of 46 per 1,000 live births. Put another way, in the 6 villages (of the 12) were female infanticide was practised, this was the outcome of 9.7 per cent of all female births!
These are frightening statistics, even more so when research has shown that the problem is much worse in the north of the country. The question of why it is that all the Indian states with IMRs higher than the national average (of 94) are concentrated in the north of the country demands investigation from this perspective? While the relative incidence of poverty in the north vis-a-vis the south may explain some of this, the findings above suggest that research into the degree of female infanticide in other parts of India, particularly the north, is urgent.
What underlies such an extreme manifestation of discrimination? The debate on excess female mortality in South Asia has been summarised by Harriss (1987). Differentials in child mortality and nutritional status can be considered as being related to female economic and social status which may be influenced by both material and cultural factors (Clark 1987).
Material explanations relate to the economic undervaluation of women (Bardhan 1987). This, in turn, depends on female labour demand, participation and earnings, as well as the gender distribution of inheritance rights (that governs control of property) and the exchange value of the female at marriage (reflected in dowry costs). Household income differentials may have paradoxical effects. In poorer households, women (who are less dowered) may participate more in the wage labour markets and suffer less from the adverse effects of patrilinearity. However, resources being more scarce, the results of less discrimination may nonetheless be more fatal. The World Bank study mentioned above argues further that regional differences in female access to food and health resources are determined by the economic value of women’s labour. Data from Rosenzweig and Schultz (1982) complement these conclusions with evidence for a synergistic relationship between female survival and economic productivity of adult females.
Cultural factors on the other hand determine not only the gender division of waged tasks within which market mechanisms may operate, but also systems of property ownership (Das Gupta 1987). In a tribal south Indian population, for example, increased income from female wage work was not associated with an increased female control over household food allocation decisions (Gillespie 1989). The impact of the concurrent decrease in capacity for child care was disproportionately borne by girls under the age of one, a bias that was reflected in excess female child undernutrition. Material factors in this case appear to have been less important than cultural factors in influencing gender-differentials in child care and nutrition. The process of Hinduisation which is diffusing throughout tribal societies in South India, with the practice of dowry becoming increasingly widespread and the costs of the dowered increasing, may here be reflected in the relative neglect of the nutritional and health needs of girls vis-a-vis boys.
The role of women in nutrition-relevant actions in the states of Andhra Pradesh and Tamil Nadu in South India will be investigated in Part II of this report.
