Contents - Previous - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

Women and nutrition: Reflections from India and Pakistan

Meera Chatterjee and Julian Lambert

In 1987 India and Pakistan had a total population of 913.1 million, which is considerably greater than that of all 49 countries of Africa (587.9 million). Of these, 440 million were women, 200 million being of childbearing age. Uttar Pradesh, India's most populous state, alone had a population of 128.5 million, which was significantly greater than that of Africa's largest country, Nigeria, the total population of which was 101.9 million.

This paper discusses the nutrition situations of women in India and Pakistan; occasional inferences are also drawn from Bangladesh. The nutritional status of women and their nutrition-related roles are clearly interrelated. Through these diverse roles, women influence the nutritional status of individual household members (e.g., through child care) and of the household as a unit (e.g., by earning). As they are members of the households in which they acquire, cook, serve, consume, and store food, their own nutrition status is the effect of the exercise of these roles and of the ensuing household nutritional status.

While women's nutrition status is an integral part of their household's nutrition profile, it is also a cause of the household's nutrition status, since performance of nutrition-related roles depends, for example, on women's energy level. Socio-economic and socio-cultural factors (e.g., income, literacy, traditional beliefs) simultaneously influence both women's nutrition status and their nutrition-related roles. On the Indian subcontinent, the apparent contradiction between women's primary responsibility for household nutrition (e.g., food preparation, health care) and their own serious malnutrition renders a simultaneous examination of these two aspects particularly interesting.

Women's nutritional status

A discussion of women's nutrition status can deal with it both in terms of absolute levels and in relation to that of men. The latter approach would include issues of discrimination between males and females in nutrition-related matters such as feeding and health care and consequent gender differences in nutritional status. These aspects are, of course, intimately related.

Nutritional levels

In India, the National Nutrition Monitoring Bureau (NNMB) collected data on household and individual food consumption, and individual nutrition status (judged by anthropometric and clinical indicators) during the late 1970s and early 1980s in ten major states of the country on a sample basis. In a representative year, 1979 (there were no discernible secular trends in these data), 41% of households were calorie inadequate in the national aggregate, and 79% were short of both calories and protein [1]. Calorie inadequacy in the different states varied from 23% of households in Andhra Pradesh in the south to more than 65% in Uttar Pradesh in the north. The percentages of individuals who were calorie inadequate were consistently higher at both the national level (46%) and the state level (28%-70%), suggesting that in a proportion of households (variable across states), whereas total food availability was adequate for all members combined, distribution was inappropriate so that some individuals bore a disproportionate burden of deficit. This is likely to be true also of those households where food (calorie) availability is inadequate in the aggregate. From other evidence we know that the burden falls on children and possibly on women.

Table 1 shows average daily calorie intakes among adults in Pakistan. Approximately 60% of the adult population consume less than the recommended daily allowance (RDA), and around 40% consume less than 80% of the RDA [2].

TABLE 1. Distribution of adults in Pakistan (percentages) by level of calorie intake (percentage of recommended daily allowance)

Source: Ref. 2.

TABLE 2. Percentages of recommended levels of iron consumed by 13-18-year-olds in selected states of India

Source: Ref. 1.

Gender differentials

The Indian NNMB data reveal a mixed picture of male/female differences in mean calorie and protein intakes. In 1982 girls 13-16 years old consumed much less than boys and only two-thirds of their recommended calorie intakes (in all states surveyed except Karnataka) [3]. While girls 16-18 years old fared slightly better than boys, they were still below requirement. Earlier data (1979) showed no significant differences in the calorie adequacy of males and (non pregnant, non-lactating) females over 18 years old [1]. While no data were given in this year for pregnant women, lactating women were more calorie inadequate than their non-pregnant, non-lactating counterparts.

An examination of percentile values of calorie intakes expressed as percentages of recommended levels among those 13-16, 16-18, and over 18 years old (in pooled 1975-1978 data) reveals no major differences between males and females [1]. Only among pregnant and lactating women were these values lower than for males, as well as lower than for non pregnant, non-lactating women. On the other hand, the data from the Pakistani Planning Commission do not reveal any consistent differences in the distribution of adult males and females in different calorie consumption, nor among pregnant, lactating, and other adult females.

Micro-nutrient deficiencies

Iron

Pooled 1974-1978 data on food consumption from the Indian NNMB [1] show that girls between 13 and 18 years of age obtain lower percentages of the recommended levels of iron than do boys of the same ages (table 2). With the onset of menarche, young girls are highly susceptible to anaemia in the absence of adequate dietary iron. The prevalence of anaemia among women in India is extremely high, as shown by a study conducted by the Indian Council of Medical Research in four areas of the country [4].

Anaemia is also common in pregnancy, frequently caused by a combination of low iron intake and poor absorption, exacerbated by malaria and hookworm infections. In some poor communities in India, 80%90% of pregnant women may be anaemic, while nationally the figure may be greater than 50%. A national survey in Pakistan in 1976 found more than 54% of pregnant women to be anaemic [2]. Between 40% and 50% of maternal deaths may be associated with anaemia or the resultant increased dangers of haemorrhage. In Varanasi 119 pregnant women were treated with 60 mg of iron and 500 mg of folic acid daily for 100 days and improved their haemoglobin levels by 1.6 g % (compared with a fall of 0.3 g % in a non-treated group) [5]. Birth weights increased significantly in the supplemented group (mean 2.91 kg) over the placebo-treated control group (mean 2.59 kg).

Iodine

An estimated 200 million people in India and Pakistan are at risk of iodine-deficiency disorders. Of all micronutrient deficiencies, a shortage of iodine in pregnancy has the most far-reaching consequences. In some seriously affected areas in the Himalayas, 80% of the population suffer from goitre, and up to 10% of newborns may be cretinous.

Iodine-deficiency disorders also have a significant effect on rates of spontaneous abortions, still-births, and infant and neonatal deaths. A single injection of iodized oil given at the start of pregnancy in Zaire had dramatic effects on perinatal and infant mortality, and on the birth weights and development quotients of the children (table 3) [6]. If administered early in pregnancy or preferably before conception, iodized oil has demonstrable results. Given that cretinism is irreversible, and that an iodized oil injection is inexpensive (35 cents) and is effective for five years, eradication of iodine-deficiency disorders can be high on a list of priorities.

TABLE 3. Effects of injections of iodized oil given to women during pregnancy (Zaire)

Synergistic effects of gender and poverty on nutrition status

The effects of gender and poverty on nutrition status may be synergistic. An economic analysis of malnutrition among young children in Punjab revealed that, while gender was the most statistically significant determinant, differences between men and women were especially great among the lower socio-economic group or caste [7]. Nutritional status among the higher-caste landowners was better on the whole, and the gender differential was also smaller.

In this study, gender had a highly significant effect on calorie intake among the study population as a whole, and on diarrhoeal disease rates among the economically better-off, with females having lower intakes and exhibiting higher rates of infection. Although both high and low socio-economic groups discriminated against females in breast-feeding practices, girls in the better-off households consumed as many calories as and more protein, iron, and supplementary food than boys because food resources were not scarce among these families. Among the poor, however, discrimination against female children, coupled with inadequate purchasing power, meant that young girls had lower calorie intakes and consumed less supplementary food and less solid food than boys. This resulted in considerably higher mortality levels among low-caste female infants (196 per 1,000 live births) than among males (125 per 1,000). documented in a contiguous study are [8]. The greater vulnerability of girls may be due to differences in the care and upbringing of sons and daughters: "These differences reflect an economic, as well as a cultural premium placed on living sons. . . [while] daughters are considered unproductive and an expensive economic drain, particularly the cost of dowry when they marry" [7]. This socio-cultural attitude considerably influences household nutritional care of girls in their natal homes and is carried on to varying degrees when they marry (daughter-in-law, wife. and mother).

Deterioration in nutrition status as females grow older as a combined result of socio-cultural, economic, and biological processes has been documented (cross-sectionally) [9]. Gender differences in adult nutrition status also appear to be exacerbated by poverty, as has been demonstrated in Tamil Nadu [10].

Regional variations

In addition to the national surveys cited above, smaller studies of household dietary intake in different parts of the subcontinent provide information on differences between males and females by age, socio-economic status, region, and season. A mapping of these studies shows that nutritional equity is lower in northern India and improves toward the south. While in Rajasthan, a northern state, all children under 12 years old and adult women were deprived of their fair nutritional shares relative to adult males, as well as to the applicable RDAs, the gender differential disappeared among adults in the western states of Gujarat (except for lactating women) and Maharashtra, and the southern states of Andhra Pradesh and Tamil Nadu [11]. Even in the latter areas, however, preschool girls tended to be worse off than boys, and some gender differences were seasonal. In most instances, adolescent girls appeared to be as well off as or sometimes even better off than their male counterparts.

In the east (including Bangladesh), the situation was similar to that in Rajasthan, but the deprivation of women appears to be economically rather than culturally mediated, as work plays a significant role in female undernutrition (see below). Regional variations also occur in Pakistan, with higher rates of malnutrition among women in Baluchistan and Sind than in Punjab and the North-west Frontier Province. In essence, the social and economic value of women underlies regional variations in their nutrition status and in sex-based differences in nutrition.

Socio-economic differences

While not permitting an examination of gender differences within each socio-economic category, the NNMB's dietary intake surveys document lower food consumption in households without land than in those with land, among those who possessed land but did not grow crops in the reference year than among those who did grow crops, among labourers and "others" (village artisans and petty businessmen) than among cultivators, and among scheduled castes and tribes than among non-scheduled groups [1]. Thus, women in these groups are likely to be at the greatest risk of malnutrition.

Dietary intake versus energy expenditure

Some evidence exists that while women get a disproportionately small share of household food, they may expend a larger proportion of household energy. A few studies [12-14] suggest that women work longer hours and expend more energy than men. One estimated that women had a shortfall of 100 calories a day on average if their physical activity in paid and unpaid domestic work was considered all together, while men had a surplus of 800 calories (in 560 households in six Karnataka villages) [13].

Economic crises

In times of extreme scarcity, females" access to food is more circumscribed than that of males. A survey of some flood-torn West Bengal villages in 1978 showed that females of all ages up to 72 years had higher rates of malnutrition than males [15]. The female/male ratio of malnourished children 0-5 years old was 1.07. If only moderately and severely malnourished children were considered, the ratio rose to 1.40, and it was 1.50 among severely malnourished children alone. These data suggest that females were both more at risk of malnutrition and more severely affected by it. The authors also pointed to the synergistic effect of impoverishment and gender bias, which has been documented in Bangladesh as well [16].

Demographic influences on gender discrimination

Discrimination appears especially acute in the case of girls born into families who already have a living child, particularly among low socio-economic groups [17]. This finding is supported by the observation of greater mortality risk among girls of high birth order [18]. The shorter period of breast-feeding of girl children in turn increases the likelihood of ovulation and subsequent conception by their mothers. The increased risk of malnutrition and mortality of these female children is accompanied by further nutritional depletion of their mothers. Shorter intervals after the birth of females than of males have been documented [19].

Effects of development and the demographic transition

A comparison of data from studies in Punjab in the 1970s and 1980s reveals that the social discrimination against young girls in matters of nutrition has persisted despite agricultural growth and economic development in the area. One study found that 24% of female children in privileged families were malnourished, compared to 74% in underprivileged families [17]. Among male children, the figures were 14% in privileged and 67% in underprivileged families. The finding that the gender differential in food consumption among children from birth to four years of age was higher among landed than among land-less families [18] is instructive. Furthermore, the selective discrimination practiced against daughters who are of second or higher birth order. particularly among the better-off, who aspire to smaller families, suggests that the demographic transition that has occurred in the region may have worsened the status of female children, as mothers continue to be under severe pressure to bear and nurture sons.

Even development at the micro-level may have negative repercussions. An in-depth study of two villages in West Bengal [20] demonstrated that the inequitable distribution of development benefits exacerbated prior differences between males and females. After land reforms in one of the villages, the proportion of people owning land increased, and undernourishment among children from birth to five years of age decreased. Despite better overall nutrition levels, however, this village showed sharper gender differentials in anthropometric status, as only boys' nutrition improved, while that of girls remained on par with girls in the underdeveloped village. In sum, the economic benefits accrued selectively to boys. The finding of this study that a supplementary feeding programme helped partially to overcome the sex bias in nutrition is significant for policy and programmes.

Consequences of female nutritional deprivation

Physical underdevelopment

A major consequence of girls' nutritional deprivation in early childhood and adolescence is their failure to achieve full growth potential. A majority of girls from low-income families reach adolescence about 12-15 cm shorter than their well-to-do peers in the same society [21]. The NNMB data on Indian women show that between 12% and 23% of those 20-24 years old in the states surveyed were less than 145 cm tall and between 15% and 29% weighed less than 38 kg [1]. The percentages were even higher among younger girls: 49% were under the reference height and 67% under the reference weight among 15-year-olds who had not yet completed their adolescent growth spurt, but who might yet marry and bear children at this early age. Girls who bear a child before the close of the adolescent growth spurt may remain physically underdeveloped and hence are at greater risk of obstetric complications, obstructed labour, or maternal death as well as of bearing low-birth-weight infants. During pregnancy, women's access to food is often restricted through the taboos and rituals observed in traditional Indian and Pakistani households. Besides low maternal prepregnancy weight and inadequate weight gain during pregnancy, low birth weight is also related to low maternal haemoglobin levels, so that the high prevalence of anaemia adds to the negative outcomes of childbearing.

High maternal mortality

In developing countries overall, childbirth accounts for some 25% of deaths in women of childbearing age, compared with 1% in the United States. Worldwide, WHO has estimated that 250 women die every four hours because of problems associated with childbirth. In India, pregnancy and childbirth accounted for around 12.5% of deaths among rural females between 15 and 45 years of age in 1986 [22]. While the aggregate national maternal mortality rates are estimated to be around 500 and 600 per 100,000 live births in India and Pakistan respectively, rates of over 1,000 have been recorded in certain parts of these countries. These rates contrast with the range of 1 to 15 maternal deaths per 100,000 live births in developed countries. In comparison with this up to 1,000-fold difference, the difference between the lowest and highest recorded national infant mortality rates in the world is around 25-fold.

High maternal mortality in India and Pakistan is a reflection of women's undernutrition, poor health status, and high fertility. Poverty, low rates of female literacy, and poor access to or use of health services are some of the underlying factors. Several common causes of maternal deaths are related to malnutrition, particularly to anaemia, while other serious causes, such as toxaemia and septicaemia, reflect the inadequate health care available to women in the ante-natal, intra-natal, and post-natal periods (see below). Some research in India has found that for each maternal death there were 16.5 cases of illness related to pregnancy and childbirth, most of which were not treated.

Low birth weight and high infant mortality

In both India and Pakistan an estimated 25%-30% of babies weigh under 2,500 g at birth, and low birth weight is a significant factor underlying their high mortality [22]. A retrospective study of more than 10,000 perinatal deaths revealed that 75% were associated with weights of less than 2,500 g [23].

Poverty exacerbates the problem of low birth weight, as poor women have both a nutritional handicap and inadequate access to food during pregnancy. For example, a 35.5% frequency of low birth weight was found in babies born to poor, short women, compared with 24% among those born to poor women over 145 cm tall and 15% among babies of better-off, taller women [9]. A detailed study of 5,914 live-born infants in Pelotas, Brazil, disclosed that although low birth weight was more common in those of low income mothers, it was a much more important determinant of infant mortality than income per se [24].

In India and Pakistan, low food intake during pregnancy is a major problem. Numerous studies have found that women consume little or no extra food during pregnancy, and may even consciously limit their intake in the fear of developing large foetuses, which would make labour more difficult, given their small pelvic sizes. Food taboos often deprive women of protein and iron sources, as well as reducing calorie availability. Foetal growth in India is similar to that among Caucasians until the last five to six weeks of pregnancy, when foetal weight gain slows dramatically [25]. In addition, the average gestational age for Indian infants is 38.5 weeks.

Skewed sex ratio

The summary outcome of the higher mortality of females is the sex ratio in both India and Pakistan: 933 and 904 women per 1,000 men respectively in 1981. In India women have higher mortality rates than men up to 35 years of age, the most significant difference occurring from birth to age five. The large number of deaths in this age group account for most of the skewedness in the sex ratio, and malnutrition is a significant underlying factor in many of these deaths.

Women's nutritional roles

Care of female children

Gender differences in nutritional status in childhood initiate women's nutritional handicap and are also evidence of the problems experienced by women (mothers) in exercising their nutrition-related roles, specifically child care and feeding. Although children's nutrition status is clearly the outcome of a host of factors, starting with the nutritional status of pregnant women, gender differences are established during the breast-feeding and supplementation stages. Micro-studies have observed anthropometric differences in infancy [26] that persist through childhood [27]

Girl infants are breast-fed less frequently, for shorter durations, and over shorter periods than are boys [17; 28; 29], a situation that may be exacerbated among the poor for social and economic reasons. A study in Tamil Nadu found that, while male children were breast-fed for five months longer than females on the average, male children in landed families were breast-fed almost ten months longer than female children in agricultural labour households [10]. Weaned earlier, young girls may not receive the required quantities of supplementary food [7].

Discrimination against female children in the quality of food given was shown in cultures as diverse as those of Tamil Nadu [30] and Punjab [18]. Male children receive larger quantities of cereals, fats, milk, and sugar than females. Higher calorie and protein intakes by males of all ages also were documented in Bangladesh [31]. The difference in feeding girls is accompanied by lower levels of health care (discussed below), so that they are simultaneously exposed to higher rates of malnutrition and longer periods of more severe morbidity, ultimately resulting in their significantly higher mortality.

Women's roles as providers

Women affect the household's nutrition by acquiring food through work, as well as by preparing it. Thus, their employment, income, and decision-making power vis-a-vis the disposal of their income on the one hand and their ability to cook and serve appropriate quantities of food to individual household members (based on nutritional knowledge and autonomy in 'kitchen" decision making) on the other are important determinants of their nutrition-related roles. Therefore, the performance of these roles is related to women's social and economic status.

Women's social status

Women's inadequate dietary intake and poor nutritional status are largely a result of the fact, established by anthropological observation, that women and girls eat last and least, a reflection of the inferior social status they are accorded in Indian society throughout their lives. We can explore their social status through two important manifestations: marriage and childbearing patterns, and education levels.

Early marriage and high fertility

Marriage and childbearing affect women's nutritional status directly, as well as indirectly through associated socio-cultural norms and practices. They also affect women's education and employment, which exert considerable influence on household nutrition. Indian women have one of the lowest mean ages of marriage in the world - 18.3 years [32], with lower averages in rural areas and in some states (particularly in the north). These low ages are reflected in the proportions of girls married among younger age groups: almost 8% of those 10-14 years old and 44% of those 15-19 years old. Marriage is almost universal by the age of 24 years among women in most states of the country. In Pakistan, the mean age at marriage in 1981 was 20.4 years.

Among the correlates of age at marriage, female literacy is paramount, while other factors such as general literacy, per capita income, level of urbanization, non-agricultural employment, and mass media are also important [33]. At the state level, women's participation in agriculture has a negative correlation with their age at marriage because higher participation rates are indicative of more traditional communities. Thus, where women are married early, they are not only deprived of schooling and the benefits this may bring to nutritional awareness, but they are exposed to the double energy demands of grueling agricultural work and of early, frequent, and prolonged childbearing.

Early marriage is tantamount to early childbearing because young married women are under considerable societal and familial pressure to prove their fertility. This is demonstrated by prevailing age-specific marital fertility rates. In 1978, 17.5% of rural and 19.7% of urban females 15-19 years old bore a child [34]. Out of a 30-year reproductive span, an Indian woman spends on average 16 years in pregnancy and lactation.

Fertility and mortality

Early, frequent, and prolonged childbearing are associated with high risks and malnourishment and mortality for both mothers and infants, as discussed above. First births and those beyond the fourth one are particularly problematic, the former also being related to low maternal age and low age at marriage. Many first births occur to mothers under age 20; girls who are married young are likely to have many infants. Infants born to women married under age 18 have almost twice the risk of death of those of women married after age 21 [35]. According to the 1979 nationwide survey of infant and child mortality [35], the high rural infant mortality reflected the high proportion of births (20%-25%) of birth order of five or more. Maternal depletion and anaemia are among the factors explaining these relationships.

Another known correlate of high infant mortality is rapid childbearing, or closely spaced births. In India the mortality rate among children born within 1.5 years of a previous birth was almost twice that of those born after an interval of 1.5 to 2.5 years [36].

With longer intervals, mortality decreased further. A survey by WHO of 6,000 women showed that the mortality of infants born within a year of a previous birth was twice as high as that of infants born two years after a previous birth; the rates were 200 and 100 per 1,000 births respectively [37]. A spacing of between one and two years resulted in an infant mortality rate of 145, while a spacing of three to four years reduced the sate even further to 80. Two-year spacing between births could reduce India's aggregate infant mortality by 10% and child mortality by 16%. It would also reduce maternal deaths by lowering maternal nutritional depletion and susceptibility to disease. Similar observations were made in Pakistan, where infant mortality was 179 for a birth interval of less than two years, compared with 92 for an interval of more than two years [37].

The studies cited also pointed to another factor important to the survival of infants: mortality was considerably higher among infants whose mothers had lost a previous child. Child loss shortened the interval between births, and thus most likely aggravated the nutrition status of the mothers. It is also related to maternal competence, an intermediate variable through which socio-cultural and socio-economic factors such as education and employment exert additional influence on women's nutrition status, morbidity, and mortality (see below). The compound effects of women's low education, low employment, and early marriage are a large number of births, closely spaced births, more nutritionally depleted women, and a large number of maternal and child deaths.

Effects of women's status on female child survival

There are some discernible relationships between the survival of female children and mothers' characteristics. The sex ratio of children ostensibly born and surviving varies by the age of their mothers, being particularly low among mothers under 15 years of age [32]. Female children of young mothers are at particular risk of death because young women are under the greatest pressure to produce sons. The extent of this pressure is also related to women's economic status. Son preference in India correlates inversely with amount of female earnings across the states [38].

Where women's economic status is low, their social status is enhanced by mothering sons. As mentioned above, daughters of second or higher birth order are particularly at risk of death where women's status is low. The sex difference in mortality was more influenced by birth order than by economic or education levels per se [18].

Women's education, fertility, and mortality

Numerous factors, including rural or urban residence, religion, caste, occupation, education level, house hold expenditure per capita, and age at marriage influence fertility, with the effects of the last three being particularly profound. Illiterate women have considerably higher fertility than do literates - more than twice as high among those 15-19 years old and 30%50% higher among other age groups in rural areas, with as much variation among urban groups [34]. While their higher fertility is in part due to a lower mean age at marriage, other factors, such as a desire for a large family and high child mortality, are also important. Literacy results in fewer higher-order births, with considerable differences between women having less than primary-level education and those with five or more years of schooling [34].

Female education is also related strongly and inversely to infant mortality. Using data from the 1979 survey of infant and child mortality, it was established that female education and household economic status were important household-level factors explaining mortality variations [39]. In India as a whole, illiterate mothers experienced 145 infant deaths per 1,000 live births, while those with some education had an infant mortality rate of 101, and for those with primary education the figure was 71 per 1,000. In rural areas, infants of illiterate women had a mortality more than double that of those whose mothers had a primary-school education (132 and 64 per 1,000 live births respectively), while in urban areas the difference was reduced to two-thirds higher among illiterate women (81) than among those who had attended primary school (49). This suggests that other urban factors, such as the availability of health services (see below), can partially offset the detrimental effects of female illiteracy on infant mortality. Poverty may be an overriding negative factor, as analysis of the infant mortality rates of different states shows that, while the figures decrease with increasing female education, the relationship does not hold in some poor states such as Assam and Andhra Pradesh. The relationship between female literacy and infant mortality has been found to hold good in Pakistan as well.

The mechanisms whereby women's education results in lower child mortality have been the subject of some speculation. Child health and survival are enhanced by better hygiene, improved nutrition and feeding practices (for both the child and the mother), and timely medical intervention; and education may improve women's practice of any of these. Schooling may enable women to take independent decisions and act on them. Educated women may have greater roles in household decision making, and be permitted by other household members to pursue appropriate strategies. In Bangladesh, household decision making does indeed change with the education of women, with greater shares of household resources becoming available to them and to children [40]. Clearly, the effects of women's education on their own nutrition status and on that of their children is exerted through their roles as providers of household health and nutrition care.

In fact, when it comes to child nutrition, mothers' knowledge may be more important than income. A study of the relationship between child nutrition and factors such as family income, maternal education, and birth order revealed that income was not the only constraint on nutrition status, even in the poorest group [41]. Maternal education had a significant influence, as literate mothers used scarce resources better for their children's welfare than did illiterate mothers with higher income. A study of two villages in West Bengal showed that children with literate mothers fared better than those with illiterate mothers in terms of nourishment [20]. These authors suggested, however, that literacy and prosperity went hand in hand, as the beneficial effects of mothers' education were greater in the village where there was a higher degree of "urban integration."

In the evaluation of Punjab villages, women's education was associated with reduced child mortality, but it increased the discrimination against girls of higher birth order, perhaps because more educated women desire smaller families with only one daughter [18]. In Punjab, education may also increase rather than decrease the dowry required to marry off daughters because it does not clearly enhance the economic value of women.

Women's economic status

That women's nutritional status and health are related to their economic status is demonstrated both by macro- and regional-level analyses and by microlevel (household) data [42]. The north-south dichotomy in nutritional levels and differences discussed above suggests that, where females have high economic value, they receive larger shares of food and health resources; where their economic value is lower, they remain at considerable disadvantage. Regional analyses of health indices other than nutritional status, such as mortality rates or sex ratios, further substantiate this relationship.

Two particular aspects of women's economic value have been related to health status: labour-force participation and inheritance of property, including payment of dowry. It was hypothesized that the greater demand for female labour in the rice-growing southern region of the country supports higher female survival rates, compared with wheat cultivation in the north [43]. On the basis of an analysis of sex ratios in different regions, a clear relationship was noted between high labour-force participation by women 1534 years old and young girls' survival [44]. In some areas where female work participation was low, however, such as West Bengal, Assam, Orissa, and Kerala, the sex ratios were not unfavourable to females, as culture protected them despite economic backwardness.

Further evidence of a relationship between women's work and survival was provided by a two-stage regression analysis of an all-India sample of rural households [45]. The authors first demonstrated a correlation between rainfall and female employment, then a negative correlation between female employment and the male/female survival ratio. Higher female employment in wetland cultivation decreased the difference in survival of males and females. Female employment was more significant than present wealth or parents' education status in explaining variations in sex-specific survival. Significantly, a rise in male employment increased the difference between boys' and girls' survival in favour of boys.

Regional differences in female survival are also related to the payment of dowry, a practice that is more prevalent in the north than in the south. High dowry and marriage expenses are associated with adverse sex ratios [44]. There is also an inverse correspondence between female work participation and dowry - the lower the former, the higher latter. Thus, in areas where female work participation is low, a daughter's value is considered to be below the cost to parents of her upbringing and marriage, including dowry payments. (The value of domestic labour is not considered, as it is perceived to accrue to marital and not to natal families.) The minimal value attached to young girls underlies decreased investment in their education in areas where female work participation and survival are low, marriage costs high, and early marriage and early childbearing the norms.

Effects of women's employment through income

There is some evidence that women's employment has the potential to benefit household nutrition through increasing household income. Daily nutritional adequacy in agricultural labourer households in Kerala was related more to women's employment than to men's [46]. It was estimated that on days when both the male head of household and his wife were employed, the shortfalls in terms of calories were 11% and 20% respectively, while on days on which the woman was unemployed, they increased to 26% and 50%. A strong association was described between child nutrition and mother's income in low-income households, but no significant association with father's income [47]. Female children were particularly dependent on their mother's wages.

Women's decision-making power

Women's employment may also exert influences on household nutrition through increased status, power, autonomy, or decision-making ability. It appears that women spend their earnings preferentially on goods and services that improve the health of children, implying an increase in their decision-making power. For example, a study of women participants in Maharashtrats Employment Guarantee Scheme reported that the nutrition status of children was better when women received the cash or grain payments directly [47]. In addition, where women exercise control over their wages, the money was spent on food and other basic needs [48].

Women's participation in wage work alone may not guarantee them greater decision-making power. Men were seen to make market decisions relating to food in 60% of households in three Tamilian villages, and they made joint decisions with their wives in another 15%, leaving women primarily responsible only in 25% of households [11]. In the majority of households women had some say in the qualitative issues (e.g. choice of ingredients), but not much regarding quantities, regardless of whether or not they participated in wage work.

Other characteristics of female employment

Seasonality

Some characteristics of women's employment, such as seasonality, have important implications for household nutrition. For example, pregnant or lactating women often lose weight during peak work (low food) seasons, and infants may be summarily weaned at such times [49]. As women's agricultural work tends to be seasonal because of its task-specificity, households that are dependent on their wages for nutritional adequacy would be especially vulnerable.

Seasonal variations in availability often exaggerate differences in food intake between men and women. When more food is available, it appears to be preferentially allocated to males, thus increasing the gap. In households with low average food availability, women and children are especially at risk during lean periods and may fall below the survival line, as the shortfalls in caloric intake would be exceedingly drastic. Even among slightly better-off households, discrimination in the allocation of food renders females more susceptible to malnutrition. The availability of off-season employment and food-for-work programmes may mitigate these detrimental effects of seasonality to some extent.

Child care

Other employment-related factors, such as the location of work sites relative to homes, the time spent in work and in travel, the energy cost and ergonomic nature of the work, and the provision of child-care facilities at work places, may also greatly affect women's nutrition status and roles. These factors may, in fact, determine the extent and nature of the trade-offs between women's productive and domestic roles.

A detailed study of an Andhra village revealed that employment in the household in general, and of female members specifically, was a major factor affecting the energy intake of children under six years of age and the equitable allocation of food resources [50]. Female employment was a more significant determinant than income or amount of land owned. Female labour participation contributed significantly to the dietary intake of children. The authors hypothesized that working women had more say in food distribution within the family, "as male members saw them as more competent."

On the other hand, they also suggested that while malnutrition in poor households was due to the lack of food, among medium and big farm families, women who hired and supervised labour had little time for child care, which precipitated some of the severe malnutrition among children in such households. In this context, the contention that women in landless and small farm families must work because men cannot meet household expenses entirely is significant because of the consequent lack of choice afforded the poor family [48]. Among the poor in particular, the additions to household nutrition made possible by women's employment may not totally offset the detrimental effects on child feeding and care due to women's absence from the home.

Women's employment in the organized sector, in the context of inadequate protective legislation or lack of enforcement of existing legislation, has been considered inimical to breast-feeding because of mothers' time away from home [37]. Only among certain occupational groups, and for very limited numbers, are there viable arrangements to permit women to take care of the nutritional needs of their infants and young children. In India, mobile crèches are available in the construction industry in a few metropolitan areas, and legislation calls for crèches or day care centres to be provided in the plantation sector, factories, and mines, but these remain inadequate. The large majority of women workers in agriculture are not served by such facilities, although in areas where the Integrated Child Development Services Scheme (ICDS) has been established, a part-time alternative exists. The attendance of infants and toddlers at ICDS anganwadis is low, however, due to structural and social constraints. The situation of self-employed urban women (vendors, petty traders, domestic servants, etc.) is difficult, as they lack access to institutionalized child care, as well as the traditional joint family system. Unfortunately there is little or no information on the effects of different female occupations on the nutritional status of children, of the women themselves, or of other household members, or on the effects of the support services provided for the few occupational groups mentioned or of different child-care situations on overall home nutrition management.

In sum, while indications imply that women's work can bring about improvements in household nutrition, particularly if women have control over their wages, numerous unanswered questions remain. Does women's employment enhance nutrition and health, and, if so, under what conditions? Who benefits and how, and what is the process by which improvements are brought about? Are women's wages preferentially allocated to nutrition and health care, so that one could argue that their wages improve family wellbeing more than men's? Are working women more knowledgeable about nutritional needs, food values, and so on? Do they make more or better decisions in favour of nutrition? What are the disadvantages?

The answers to these questions have relevance to actions such as the promotion of appropriate employment options for women, legislation of women's wages for different types of work, provision of support services for productive and domestic work including crèche and child-care facilities, development of drudgery-reducing technologies and services such as fuel and fodder, water supply, and so on. An understanding of household decision-making processes may strengthen policy making and action in vital social arenas such as education.

Access to health care

Access to health care affects women's nutrition status and roles. As the first-level health-care providers within the household, women's knowledge of good health and nutrition practices is crucial. Although such knowledge may be gleaned elsewhere, such as in school or from older family members or other informal networks, as formal health systems become increasingly concerned with health promotion, they are important conduits of health knowledge. Health services offer the next level of support to the household when treatment is required and in the arena of preventive health. Household health and nutrition profiles subsume the health of women, which in turn determines the extent to which they can effectively carry out their roles as producers, mothers, child-mincers, and so on, all of which affect household health.

Pre-natal, intra-natal, and post-natal care affect the viability of infants and the survival of mothers; nutritional care protects growth and development and ensures better health for work and during pregnancy and lactation; family planning services address issues of birth spacing and limiting family size, which we have seen affect women's health status and roles. Experience in both India and Pakistan has shown that health care and health education are best delivered to women by women, preferably from a similar socioeconomic background.

Assessment of women's access to health care is complicated because it is not only a determinant of women's and households' nutrition and health status but may be simultaneously affected by (macro and micro) socio-economic and socio-cultural factors. Access implies both the physical availability of services and social and economic conditions that permit and enable women to use health services when in need [51; 52]. Unfortunately, few investigations of women's health needs in India have included assessments of available health services and information on health knowledge, attitudes, and practices at the household level. Available studies focus on different access of males and females to health services, particularly among children, sometimes relating this to nutrition levels or morbidity.

For example, one study found that, while females outnumbered males four to three among children suffering from kwashiorkor, more than half of hospital admissions were boys [53]. Similarly, a survey of 37,000 people in rural Maharashtra revealed that, although higher percentages of girls were ill than boys, lower percentages received medical treatment in the group under 15 years of age [54]. Girls may be taken to less qualified doctors than boys [17], and expenditure on medicines may be higher for boys, particularly among better-off families [18].

In fact, better and more timely medical care for boys may be the most important factor explaining higher survival among males than females, as early studies showed [8; 55]. These and other studies imply that households discriminate against female children in health care in a way similar to the discrimination in nutrition.

Adults' use of health services is different as well. A greater proportion of ailing women than men received no treatment, and women who were treated received mostly home remedies or traditional medical care, while men received institution-based care [54]. Hospital and clinic attendance records in both India and Pakistan invariably show a preponderance of males receiving treatment. A comparison of studies of hospital admissions in different parts of India shows considerably higher ratios of male to female admissions in northern hospitals (e.g. 2.1:1) than in southern hospitals (1.3 1), although males were clearly favoured in all areas [44]. The proportion of treatment provided to women in general is lower whether one considers out-patient attendance or hospital admissions. For example, in one study of admissions to a major hospital in Delhi, only 35% were women [28]. Similarly, larger numbers of males are treated at primary health centres in Uttar Pradesh, Gujurat, and Rajasthan - as many as five times more men than women [29; 56]. The observation that women seek medical help only at advanced stages of illness is corroborated by hospital-based data showing higher fatalities among female patients [15].

These findings are even more significant in light of reports that illness is higher among women than men, even though female morbidity is likely to be underestimated because women are "shy" about revealing illness or purposely downplay it to avoid seeking medical care. The few studies that compare the health of women and men in the same household generally report a higher prevalence of illness among women [14; 57]. One study followed 110 families over a two year period and found a significant difference in the number of illnesses suffered by adult women and men: 10.8 episodes per year compared with 6.0 [58]. Failing to find a gender difference in the incidence of disease, however, some researchers in Bangladesh suggested that higher female morbidity is largely the result of the lack of treatment of female illness [31].

The relationships between the availability of health facilities, women's use of them, and female literacy have been well established, as have their combined effects on mortality [59]. Certain areas of India such as Kerala and Goa bear further witness to the strength of the relationships. Infant mortality in Kerala was found to be lowest where access to health facilities was easiest [60; 61]. Mothers' education is a highly significant explanatory factor [62]. In sum, the state's well-developed health services and high levels of female education together explain Kerala's impressive mortality and fertility declines, the crux being that public health activities have been made effective by prevailing high literacy levels that stimulate demand, peoples' participation, and government responsiveness [63].

Prospects for intervention

As has been shown, the nutrition status of women in both India and Pakistan is all too often unsatisfactory and is a process that beings at birth and often ends in early death. Numerous causative factors and correlates have been identified. The potential interventions to address the problem of female malnutrition are similarly numerous. The range of options available to planners in various sectors is laid out in table 4. (Although the options apply primarily to India, the matrix could easily be modified to suit other countries.) Here, discussion is limited to a few key issues.

Focus on nutrition for women as women

The concept of improving women's nutrition for their own sakes, rather than just as mothers, must be fostered. There is little doubt that a woman whose basic nutritional and health needs are met will be in a better position to meet the needs of her family. Specific nutritional deficiencies, such as those of iron and iodine, must be tackled (and they can be, at low cost) with all women forming the target group. Better targeting of supplementary feeding at those most at risk of malnutrition will help to address the more intractable problem of protein-energy malnutrition, as will job-creation and literacy programmes.

Nutrition in adolescence

The nutrition status of women can be considerably influenced by attention during adolescence, with spin-off benefits also to the children they bear later. Even children who are stunted and malnourished throughout childhood can experience catch-up growth if fed adequately during their adolescent growth spurt, and can achieve an adult size almost as great as children who were better nourished in their early years. For example, one African study demonstrated complete catch-up during adolescence of a cohort of girls who at 10 years of age were 20 cm shorter than a normally nourished cohort [21]. Thus, midday-meal programmes for adolescent girls could have long-lasting benefits.

Improving female literacy, education, and health-related knowledge

The critical role of female literacy in improving women's overall health and nutritional status should be well recognized. The coincidence of girls' adolescence and dropping out of school signals the need for education systems to focus on keeping girls in schools. This may be done through the provision of special incentives, public education, and alternative forms of education. It is important to provide basic vocational skills, enhancing girls' employability and delaying marriage until they are physically prepared for childbearing.

While these are longer-term goals, in the short term, efforts specifically to improve women's knowledge of health, nutrition, and hygiene must be increased. The communication of basic nutrition information, based on proper understanding of existing knowledge, attitudes, and practices, and involving health workers, primary-school teachers, women extension officers, and other front-line workers, reinforced by appropriate use of the mass media, can help to empower women to address malnutrition.

TABLE 4. Women and nutrition: Draft matrix for policy planning

Matrix developed by Geeta Athreya, Health and Nutrition Section, UNICEF. New Delhi.

Increasing the number of female health workers

Providing basic services to women in rural areas of India and Pakistan is a huge undertaking, given the number of women involved. Furthermore, in addition to size, the delivery system must be appropriate in order to address nutrition and health problems. Experience has shown that, to provide effective services to women, the front-line workers must in turn be women. Under these circumstances, the shortage of female health workers is a serious constraint to improving women's nutrition and health status. While this is widely recognized, many countries are still a long way from meeting their womanpower requirements in health systems, or, for that matter, in education and other key sectors. International aid agencies can play a key role in this area. For example, in Pakistan during the early drafting stages of the Eighth Five-Year Plan, the government announced plans to recruit and train one community-level worker for each of the country's 50,000 villages; all these workers would be men educated to high-school level. After receiving considerable pressure from a number of donors and UNICEF, the government revised its plans and said that at least half of the workers would be women, for whom the education qualifications would be relaxed.

Another key female health worker is the trained birth attendant. In the subcontinent, traditional birth attendants assist at half or more of all deliveries, and in some places they may visit mothers in the postpartum period for varying periods of time. They can be useful resources not only to conduct safe deliveries but also in providing pre-natal and post-natal care.

Detection and care of mothers at risk

In the arena of maternal and child health services, simple indicators of maternal malnutrition and predictors of risk of low birth weight can be usefully employed (table 5) [64]. Among the most significant are severe anaemia, poor obstetric history, current obstetric problems, and tuberculosis, which increase the chances of a woman having a low-birth-weight baby fourfold or more. All these problems can be diagnosed and managed at the level of the primary health centre, where skilled attention and limited resources should be focused on the most needy.

Food supplements during pregnancy, particularly if given during the last trimester, have been shown to have a positive impact on birth weight. For example, supplements given to pregnant women through the Indian IODS programme resulted in a mean increase of 150 g in birth weight [65]. A review of a number of studies concluded that the mean increase in birth weight resulting from a daily supplement of 500 kcal was 300 g [66]. Research in the Gambia showed that supplemention is particularly effective when given during times of food shortage. such as that caused by seasonal fluctuations in food supply [67].

TABLE 5. Predictors of risk of low birth weight

Source: Ref. 64.

Increasing female employment and income

Increasing the number of female workers could clearly do more to improve women's nutrition and health status than many other strategies. Ensuring fair wages for work done so that women can purchase adequate food for themselves and their families, improving working conditions so that they are not nutritionally draining or hazardous to health, and providing access to services such as day care, health care, and means to lighten domestic work are all important ingredients of female employment strategies.

Both India and Pakistan share the rare distinction in the developing world of having elected women leaders. It is to be hoped that this elevation of women to the highest elected positions will be translated steadily into improvements in the position of women in these countries.

References

1. Nutrition Monitoring Bureau. Report for the year 2979. Hyderabad. India: National Institute of Nutrition, 1980.
2. Planning Commission. Micro-nutrient survey, 1976-77. lslamabad, Pakistan: Planning Commission, 1979.
3. National Nutrition Monitoring Bureau. Report for the year 1982. Hyderabad, India: National Institute of Nutrition, 1984.
4. Indian Council of Medical Research Use of common salt fortified with iron in the control and prevention of anaemia: a collaborative study. Report of the Working Group on Fortification of Salt with Iron. Am I Clin Nutr 1982;35: 1442-51.
5. Agarwal DK, Agarwal KN, Mishra KP. Strategy to reduce low birth weights. In: Agarwal KN. Bhatia BD, eds. Varanasi: update growth. Varanasi, India: Department of Paediatrics, Banaras Hindu University. 1988.
6. Thilly C. Distribution of iodised oil in Zaire. Proceedings of the IDD workshop, Lima. Geneva: World Health Organization, 1983.
7. Levinson FJ. Morinda: an economic analysis of malnutrition among young children in rural India. Cornell-MIT International Nutrition Policy Series. Cambridge, Mass, USA, 1974.
8. Kielmann AA. Taylor CE, De Sweemer C, et al. Child and maternal health services in rural India: the Narangwal experiment. Vols. I and 2. Baltimore, Md, USA: Johns Hopkins University Press, 1983.
9. Ghosh S. Bhargava SK. Moriyama IM. Longitudinal study of the survival and outcome of a birth cohort. Vol. 2. Report of phase I of a research project. New Delhi: Safdarjung Hospital, 1982.
10. McNeill G. Energy under-nutrition in adults in rural south India. Progress report of funding agencies. 1984 (mimeo),
11. Harriss B. The intra-family distribution of hunger in South Asia. London: London School of Hygiene and Tropical Medicine, 19X6 (mimeo).
12. Jain D, Chand M. Rural children at work: preliminary results of a pilot study. Ind J Social Work 1979:4():311-22.
13. Batilwala S. Rural energy scarcity and under-nutrition: a new perspective. Econ Pol Weekly 1982;27:328-34.
14. Kahn ME, Dastidar SG, Singh R. Nutrition and health practices among rural woman: a case study of Uttar Pradesh, India. Presented at the International Symposium on Problems of Development of Underprivileged Communities in the Third World, New Delhi, Oct 1982.
15. Kynch J. Sen A. Indian women: well-being and survival. Camb J Econ 1983;7:363-80.
16. Bairagi R. Food crisis, nutrition and female children in rural Bangladesh. Pop Devel Rev 1986;12:307-15.
17. Das D, Dhanoa J. Cowan B. Letting them live. Future (New Delhi:UNICEF) 3rd quarter 1982.
18. Das Gupta M. Selective discrimination against female children in rural Punjab, India. Pop Devel Rev 1987; 13:77-100.
19. Haldar K, Bhattacharya N. Fertility and sex sequence of children of Indian couples. Sankhya 1969;6: 144-49.
20. Sen A, Sengupta S. Malnutrition in rural children and the sex bias. Econ Pol Weekly 1983;28:855-864.
21. Rohde J. For tomorrow's generation today. New Delhi: USAID, 1987 (mimeo).
22. Office of the Registrar General. Survey of causes of death (rural), annual report. New Delhi: Ministry of Home Affairs, 1988.
23. Metha A. Perinatal mortality survey in India. Scientific brochure of the Third International Seminar on Maternal and Perinatal Mortality, Pregnancy Termination and Sterilisation, New Delhi, Oct 1980.
24. Victoria CG, Barros FC. Vaughn JP. Mortality, morbidity and malnutrition in relation to birth weight. Proceedings of the XIII International Nutrition Congress. London: John Libbey, 1986.
25. Gruenwald P. Growth of the human foetus. Obstet Gynecol 1966;94:112-19.
26. Gopalan G. The mother and child in India. Econ Pol Weekly 1985;20:159-66.
27. CARE. Nutrition in Punjab. New Delhi: CARE, 1974.
28. Ghosh S. Discrimination begins at birth. Prepared for the UNICEF Journalists' Workshop on the Female Child, New Delhi, Oct 1985.
29. Khan ME, Dastidar Ghosh S. Bairathi S. Women and health: a case study in sex discrimination. Proceedings of ICMR Workshop on Child Health, Nutrition and Family Planning, Bangalore, Nov 1983. New Delhi: ICMR, 1983.
30. Devadas R. Kamalanathan G. A woman's first decade. Presented at the Women's NGO Consultation for Equality, Development and Peace, New Delhi, Apr 1985.
31. Chen LC, Huq E, D'Souza S. Sex bias in the family allocation of food and health care in rural Bangladesh. Pop Devel Rev 1981 ;7:55-60.
32. Office of the Registrar General. Census of India, 1981. Series 1, part 2. Special report and tables based on 5 percent sample data. New Delhi: Ministry of Home Affairs, 1983.
33. Srivastav JN. Factors in female age at marriage in India with special reference to literacy status: interstate analysis for 1981. Demography India 1986;15:228-38.
34. Office of the Registrar General. Survey report on levels, trends and differentials in fertility. New Delhi: Ministry of Home Affairs, 1982.
35. Office of the Registrar General. Survey of infant and child mortality, 1979. New Delhi: Ministry of Home Affairs, 1981.
36. Das N. An indirect approach to study the interrelationship between infant mortality and fertility. Demography India 1975;4:449-56.
37. Ghosh S. Child survival and development in the context of integrated MCH/FW services. Keynote address, Workshop on Action Learning for Effective Child Care. Indian Institute of Health Management Research, Jaipur, July 1987.
38. Bardan PK. On the economic geography of sex disparity in child survival in India: a note. Berkeley, Calif. USA: University of California, 1987 (mimeo).
39. Jain A. Determinants of regional variations in infant mortality in India. Working paper no. 20. New York: Population Council, 1984.
40. D'Souza S, Bhuiya AL. Socio-economic mortality differentials in a rural area of Bangladesh. Pop Devel Rev 1982;8:753-59.
41. Bairagi R. Is income the only constraint on child nutrition in rural Bangladesh? Bull WHO 1980;58:767-72.
42. Chatterjee M. Indian women, health and productivity. Prepared for the World Bank's review of women in development and 1989 country economic memorandum for India. New Delhi, 1988.
43. Bardhan PK. On life and death questions. Econ Pol Weekly 1974;9:1293-1304.
44. Miller B. The endangered sex: neglect of female children in rural north India. Ithaca, NY, USA: Cornell University Press, 1981.
45. Rosenzweig M, Schulta TP. Market opportunities, genetic endowment and intra-family resource distribution: child survival in rural India. Am Econ Rev 1982;72:803-15.
46. L. Profile of a female agricultural labourer. Econ Pol Weekly 1978; 13(12): A27-A36.
47. Kumar S. Role of the household economy in child nutrition at low income. Occasional paper no. 95. Ithaca, NY, USA: Cornell University Department of Agricultural Economics, 1978.
48. Mencher J. Sardamoni K. Muddy feet and dirty hands: rice production and female agricultural labour. Econ Pol Weekly 1982; 17: A149-A187.
49. Palmer I. Seasonal dimensions of women's roles. In: Chambers R. ed. Seasonal dimensions of rural poverty. London: Frances Pinter, 1981.
50. Bidinger P. Nag B. Babu P. Factors affecting intrafamily food distribution in a south Indian village. Report to the Ford Foundation. 1986 (mimeo).
51. Chatterjee M. Women access to health care: a critical issue for child health. Proceedings of ICMR Workshop on Child Health, Nutrition and Family Planning, Bangalore, Nov 1983. New Delhi: ICMR, 1983.
52. Chatterjee M. Competence and care: policies to develop health and nutrition through women in the household. Presented at IUAES/ISA/IAWS Conference on Women and the Household, New Delhi, 1984.
53. Goplan C, Naidu A. Nutrition and fertility. Lancet 1972;2:1077-79
54. Dandekar K. Why has the proportion of women in India's population been declining? Econ Pol Weekly 1975;10:1663-67.
55. Singh S. Gordon J. Wyon J. Medical care in fatal illness of a rural Punjab population: some social biological and cultural factors and their implications. Ind J Med Res 1962;50:865-80.
56. Murthy N. Reluctant patients: the women of India. World Health Forum 1982;3:315-16.
57. Jesudason V, Chatterjee M. Health status and behaviour of two rural communities: report of a sample survey in Madhya Pradesh. New Delhi: Council for Social Development, 1979.
58. Kamath KR, Feldman PSS, Rao S. Webb JKG. Infection and disease in a group of south Indian families. II. General morbidity patterns of families and family members. Am J Epidemiol 1969;80:375-83.
59. Krishnan P. Mortality decline in India, 1951-61: development vs. public health programme hypotheses. Soc Sci Med. 1975;90:475-79.
60. Krishnan TN. Demographic transition in Kerala: facts and factors. Econ Pol Weekly, 1976;118:1203-24.
61. Nair PSG. Regional variations in infant and child mortality vis-a-vis rural health services: a case study of Kerala. Presented at IASP conference. Bombay, 1980.
62. Zachariah KC, Patel S. Trends and determinants of infant and child mortality in Kerala. Discussion paper no. 82-1. Washington, DC: World Bank, 1983.
63. Chatterjee M. Implementing health policy. New Delhi: Manohar Publishers, 1988.
64. Walsh JA. Recommendations for a maternal nutrition study. Report for USAID Office of Nutrition. Washington, DC: INCS, 1987.
65. Tandon DN, Ramachandran K, Bhatnager S. Integrated child development services in India: evaluation of delivery of nutrition and health services and the effect on the nutritional status of children. Ind J Med Res 1981 ;73:385-94.
66. Sai FT Nutrition and fertility. Proceedings of the XIII International Nutrition Congress. London: John Libbey, 1986.
67. Prentice AM. Dietary supplementation of lactating Gambian women. Hum Nutr 1983;37C:53-64.

Contents - Previous - Next