Gender Bias in Survival and Nutritional Status of Children: Review of Available Evidence
Item
- Title
- Gender Bias in Survival and Nutritional Status of Children: Review of Available Evidence
- Creator
- Leela Visaria
- Date
- 1988
- extracted text
-
com:..--
•
cat
47/1,(first Iloor)3t. Marks Road
BANGALORE-5C° 001
Background Paper-VI.
XIV ANNUAL MEET OF MFC.
GENDER BIAS IN SURVIVAL AND NUTRITIONAL STATUS OF
CHILDREN : REV JEWOF AVAILABLE EVIDENCE.
By
L e e 1 a
V isaria.
INTRODUCT ION;
Women are regarded as both biologically stronger and
physiologically superior to men. Biologically, the presence
pair of chormosomes protects women against chromosome-linked
recassive disorders and makes them less susceptible to infec
tious diseases (Waldron, 1983). Physiologically, women are
reported to be more efficient than men; for a given quantum
of work, they require somewhat less protein and energy than
men. (Rivers, 1982). These innate differences between the
sexes have been regarded as responsible for the lower female
mortality observed in most parts of the world today.
In the developed countries, the sex differences in
mortality have been widening. By 1983, a difference of 7
to 8 years in the life expectancy at birth between males and
females was not at all uncommon and was reported by U.S.A.,
France, Finland and Australia (United Nations, Demographic
Year Book, 198b-). The once-common maternal deaths have been
eliminated. At the same time, the biological disadvantage
of men is aggravated by a stressful life style and accidents,
which account-for a significant proportion of deaths. The
observed high male-female differences in mortality at ages
3?-7b- in countries ■ such as Finland, France, U.S.A, and the
U.S.S.R. are attributable to the higher incidence of
cardiovascular and respiratory diseases (including lung
cancer) and accidents among men (Lopez, 1983).
11;
\
1
In sharp contrast to this general pattern is the
situation reported by the populations of the Indian subcont
inent where males enjoy lower mortality than females almost
from birth until about the end of the reproductive period
of the latter. This has been an important factor contribu
ting to the anomalous excess of males in the population
reported by the decennial censuses for nearly a century now.
The age specific death rates based on the recent large data
sets such as the sample registration system (SRS) in India
i
Contd.. .2...
2 -
have confirmed the excess female mortality suggested by the
estimates of life expectancy at birth and at other ages,
based on the census age data.
The reasons and the mechanism underlying the emergence
of this pattern in the Indian subcontinent are an important
theme of social research. To understand, document and analyse
fully the magnitude and modalities of differentials throughout
the life span is a difficult task. This paper, therefore,
limits itself to exploring the available literature on some
of the possible manifestations of the differentials in
treatment between young boys and girls which lead to higher
mortality among girls. The treatment received by boys and
girls in the early years of life determines their expectat
ions from the family as well as the rest of society in later
life and has a critical influence on the way they raise the
next generation.
The SRS data have pointed to a clear well established
sex differentials in infant and child mortality. (Visaria,
1985). It has also been shown that during the neonatal period,
when the biological factors are important, the sex different
ials in mortality are non-existent or male mortality is
higher than female mortality.
After the first month, when
factor? which are environmental or extraneous and subject
to manipulation come into play; female mortality has been
found significantly higher compared to male mortality. This
paper is an attempt to document the available evidence on
what causes such differentials. Also, the issue has to be
examined in a wider perspectiv e. Is ife that the female
children suffer from relative disadvantage in much the same
way as female adults do ?.
i
!
pI
PREFERENCE FOR SONS > D THE NEGLECT OF DAUGHTERS.
The literature is replete with statements that in
the Indian subcontinent there is a strong preference for
sons for economic and religious reasons. Sons are valued
because they can provide economic support to the parents in
the old age. Also, income earning help can be expected
from sons and their spouses but not from the daughters.
Sons are considered essential for religious reasons because
Contd...3.•.
i
1
- 3 they alone can perforin shraddha rites believed to take the
ancestors towards heaven. On the other hand, daughters have
to be given away in marriage as Kanyadan and become somebody
else's property. Their marriage also involves expenditure
which has n®-. returns. In a male dominated society male
-progeny is important for the continuation of family.
A society in which this basic philosophy is prevalent,
birth of a son is welcomed and a daughter.is unwanted. An
unwanted daughter is discriminated against or.neglected from
the beginning of her life to such an extent that her
inherent biological advantages not only disappears, but is
negated (Among others, see; Visaria, 1971). In other words,
the environmental factors and culturally influenced
behaviour counteract the inherent advantages of females and
result in their excess mortality.
However, Kynch and Sen raise two important methodolo
gical problems with the economic causation model of neglect
of daughters. First, even if the ecohomic factors are
important, they may operate not through direct calculation
□f personal utility or gains but through creating a
'perception bias' in favour of male children over female
children. That is, they operate with certain firm social ob
ligations and rules and can involve many compromises. The
second issue is that the observed bias against the female
children in the group data does not necessarily imply
discrimination within a given family, meaning that in the
same family the female child may not get less to eat than
the male child. In the group data, the male children
may, all the same, emerge as better fed, if the mother of a
baby son receives more support and more to eat or feed
her child than the mother of a baby daughter. (Kynch and
Sen, 1983). A better understanding of the inter-family
patterns of discrimination is very necessary to fully
understand the mechanisms of neglect.
EARLY SEX DECTECTION OF THE FETUS AHL) ABORTING OF FEMALE
FETUS ;
-------------------------The discrimination can begin from the time a women
conceives. -According to folklore in the subcontinent, some
individuals,, typically older women, can predict the sex
Contd.. A...
- 14- -
of the child of a pregnant woman on the basis of the way she
walks or carries herself, the shape of her portruding stomach,
the degree of lethargy orfatigue felt by the w*men, etc. The
importance attached to such predictions and faith in their
veracity can affect or alter the behaviour of the pregnant
woman or of other members of the family towards the fetus in
a way which is detrimental to the unborn female child. If
a pregnant woman, believed to carry a female child is
neglected in the form of being underfed or not given medical
care when ill, would probably give birth to an infant with
a low birth weight or a premature baby.
Despite the folklore about predicting the sex of the
fetus, whose accuracy is at best questionable, I believe
that the "knowledge" does not really alter the behaviour
of the concerned individuals, in terms of leading to abort
ing the fetus or neglecting the health of the pregnant women
because of the supposed sex of the expected child. The fear
of having to regret later must loom large to warrant any
possible risk.-l/
However, one has no hard evidence at
all on this subject.
The modern developments in chromosomal mapping apd
detection of chromosomal abnormality in the early development
of the fetus through the technique of amniocentesis have
made it possible to know the sex of the fetus accurately.
By 1986, this procedure has become available in most of
the major cities of India, although it is quite expensive
costing about Rs. 500/- much more than what a majority of
the Indians can afford. However, the procedure seems to be
used primarily for identifying the sex of the unborn
child rather than for detecting the chromosomal abnormali
ties and in the event of a female fetus, abortion is
sought under the MTP act of 1971 (Medical termination cf
pregnancy) without revealing the result of amniocentesis
or the purpose for which fetus is aborted. Feminists,
activist and women’s organisations have, in recent years.
_1/
It is possible that the sex-composition of the earlier
children consciously or unconsciously influences the
care given to a pregnant woman. A thorough anthropolo
gical study of care of pregnant women during each
pregnancy would be quite interesting and useful to
design interventions which are likely to work in
real situations.
Cond....5«• •
1
- 5 taken up this issue and are engaged in a debate on the ethicality of amniocentesis and its contribution to female
foeticide. (Chhachhi & Sathyamala, 1983? Balasubrahmanian,
19825 Kumar, 1983; Bubey, 1983; Patel, 1986).
There are practically no good statistics available
on the number, characteristics and the objectives of the
users of this method. /According to one study undertaken in
India during 1976-77, out of 700 women who had the sex of
their expected child identified, U-50 were carrying female
children. Of them VjO (or more than 95 percent) underwent
abortion. Of the 250 women with amale fetus no one opted
for abortion. (Ramanama and Bambawala, 1982) The represen
tativeness of this study cannot be assessed but given the
reported widespread preference for two boys and one girl,
an extensive recourse to amniocentesis can have alarming
implications for the sex ratio at birth and of population.
Whether the resulting shortage of eligible brides would
then raise their value is debtable because atleast the past
experience points to a long standing coexistence of a
marked deficit of females and discrimination against them
throughout atleast North India.
AVAILABILITY ANb ALLOCATION OF FOOL:
<
One of the most glaring forms of discrimination can be
practised in allocation of food or even in breastfeeding.
Literature (stories and movels) is replete with accounts
of mothers letting a female child cry when hungry, but
picking up and feeding the male child immediately when he
lets out the first cry. Even if the mother wants to feed
her daughter, she may seek unsponken consent of her
mother-in-law who may scorn the daughter-in-law for doting
on the female child. When a mother takes time to feed
a daughter she might invite caustic remarks about her neglect
of her household duties. Statements to the effect that a
give child is breastfed for shorter period than the male
child are also found in liternature. Such assertions not
withstanding, very little evidence or hard data are
available on differentials in breast-feeding by sex in the
large number of studies on the initiation of breast feeding,
its duration, introduction of solid food, relationship
between breastfeeding and fertility etc.
Contd.. .6.,.
•. . . .l'jtioC
- 6 Of course, such studies are difficult to conduct,
First the quantity of breast milk available to an infant is
not easy to measure. The duration of time for which a baby
is fed cannot quite indicate the quantum of milk sucked.
(Technically, one can weigh/ the baby before the feed and
weigh it immediately after the feed. The difference in the
two weights would indicate the quantity of milk intake by the
child.
However, in field conditions it is very difficult
to conduct such atudies). Secondly, except in the event
of twins-, a mother normally breast-feeds only one child so
that intra-family differences in feeding between boys and
girls can be studied only subsequentially, and would be
influenced by other factors which cannot easily be controlled.
Such weighing is difficult because, subject to the
presence of the., mother near the baby, Indian infants are
fed on demand rather than on any schedule. Thirdly the
study itself might influence the behaviour of the lactating
mother and the members of her family.
Complete breast-feeding means that a nutritionally
balanced diet is available to the baby up to a certain age.
It also means that the -.anti-infective properties of breast
milk protect the child from contaminated food in an
environment where the load of contamination is quite high.2-/
If girls are not given adequate breast milk and are
weaned earlier compared to their brothers, they would
indeed suffer from all the disadvantages. Their nutrient
intake would be inadequate. They would suffer from
higher morbidity especially due to consumption of food
which is ‘often contaminated. While in Khan’s study of
21 families, theinformatits reported that they did not
discriminate between boys and
girls either in breast
feeding or in the quantity of food, he cites a prospective
study on breast feeding and weaning practices in rural
Haryana, where at...the end of the fourth month, a signifi
cantly larger proportion of daughters were put on artifi
cial milk than sons (Khan, at. el, 1983).
2_/ To the extent that breast-feeding delays the resumption ed
of ovulation, in a situation of no or very little contra
ception, longer duration of breast feeding would delay
conception and increase birth interval. Birth-spacing
can be achieved through the use of contraception as well .
Contd...?...
!
I
- 7 -
Once the child is completely weaned, there can be
differential in the availability of food between boys and
girls. There are a few studies which show that there is a
sex-bias in food allocation not only among adults, but
among children as well. A study undertaken in Matlab Thana
in Bangladesh indicated that food intake (as well as use of
health care services) was biased in favour of male children
(Chen. at.el, 1981). The caloric intake of male children in
the age .group 0-L- was 809 as opposed to 69*+ for girls, The
differences were observed in the intake of protein also;23.0
grams for boys as against 20.2 grams for females. Another
set of .data available from the same region in Bangladesh
indicated that famines or food shortages had a significant
adverse effect on the nutritional status of female children
compared to male children. (Bairagi, 1986). An improvement
in the household resources raised the nutritional level of
all children, but it benefitted male children more than
female children.
However, apart from the Matlab study, there are hardly
any studies which measure in detail the quantity of food .
intake of children. What the literature is replate with is
that boys receive preferential treatment in certain foods
which are considered more nutritions on are more valued.
Such foods are not consumed daily in majority of the
households and therefore, theycannot leave a significant
impact on the nutitional status of the children.
A study of 300 children, undertaken in Ludhiana City
in child
to investigate the influence of sex preference
feeding habits revealed that a higher proportion of
mothers with "traditional" attitudes offered vegetarian
food to their daughters and non-vegetarian food to their
sons.3_/ The percentage of girls of "traditional" mothers
suffering from malnutrition was high compared to their
2j The author devised an index of traditional and modern
attitudes on the basis of the responses of the mothers
to ten basic questions ammed at eliciting opinion,
behaviour, and practices towards child -feeding.
Contd...8...
8
brothers as wll as compared to the daughters of mother with
"modern" attitudes (Gupta, 1986). One does not know whether
and how far the caloric or nutrient value of non-vegetarian
food was different from that of vegetarian food. However,
it does reflect a social practice (in certain groups men eat
meat but.women do not) as well as higher value put on male
children. Khan also reported that boys in his study area in
Uttar Pradesh were provided foods like milk, butter, ghee
more than the girls, although in daily normal food little
discrimination was practiced (Khan, 1983). Bas Gupta’s
re study of Khanna villages in Punjab indicated that boys
received more of milk and fats with their breads compared to
girls who were given dry bread. Thus boys received food
which is nutritionally superior and also more valued
socially (Das Gupta, 1987.)
In the Khanna Tehsil it was observed that although
mothers started boys and girls on solid food at the same age
mortality among girls was significantly higher compared to
boys due to the possibility of more supplementary food being
to boys as well as "higher quality of medical care" (Wyon
and Gordon, 1971).
In a study undertaken in rural West Bengal also it was
indicated that male children received more milk or choicest
pieces of fish compared to girls who received tea instead of
milk and those parts of fish which were not valued very high.
Interestingly, such discrimination was practised more by the
upper castes and the landed and not by the landless (Warrier 5
1987.) What is quite likely is that while in the better off
families, sons are valued more, the amount of food available
is also high and therefore, girls are not likely to be star
ved, or even receipients of fewer calories.
I
It must however, be recokoned that intra-family
availability or intake of food is extremely difficult to
research. In the rural areas of the subcontinent, food
items are consumed outside the homes also. Children pick
wiia berries in certain seasons^ similar produce are also
available more or less freely £yefrom the trees and is
consumed by children. It is often stated that girls
consume berries and similar items more than boys, although
Contd
9
i
I
■■
- 9one does not know the veracity of such statements. It is
also lihely that the energy used up in walking and picking
berries etc. may be greater than the number of. .calories ’ '
obtained from such an effort. Measurement of total caloric
intake is not however easy, unless the children are const
antly under the. surveillance of the field workers,9 who in
addition, also measure the quantity of intake. At the same
time, one can with somewhat more accuracy measure the food
given to the children by the parents at home. The results,
however, have to be interpreted with great caution. Further,
even within the age group 1-U-, age would have to be controlled,
because, the needs of one or two year olds are different from
those of a four year old. One would therefore have to observe
a fairly large number of children in order to lower sampling
variability.
Thus, it is not at all surprising that most of the
available information on discrimination in food allocation
between boys and girls is inferential rather than direct and
suggest that boys receive more valued, socially as well as
higher priced, food compared to girls. Limited information
tends to indicate that discrimination in food availability
or intake is by and large confined to special occasions,
when items like ghee or butter are bought by the households,
Men and boys are given preference over women and girls.
NUTRITIONAL STATUS OF CHILDREN .
I
r
Therefore, actual growth of children measured in
terms of weight and height (or anthropometric indices) is
sometimes used for the measurement of well being of
children. The underlying reasoning is that compared to
measuring food intake 9 it is relatively easier to measure
height, weight or age.
However, weight gain is influenced
not only by food intake but also by the episodes of illness 9
their severity and duration and actual care received for
treatment of illness.
The' effect of infection on the intake and efficiency
of use of food by children has been studied through quanti
tative estimates of catabolic losses, clinical and subclinical absortive losses, changes in food intake patterns and
losses due to high infant and childhood mortality. On the
Contd...10...
I
10 -
basis of some da.ta, it has been estimated that in Bangladesh
about 9 percent of the food available to children is not used
for growth, maintenance or activity due to a heavy load
of infection (Briscoes, 1979.)
Gopalan, in order to understand whether girls suffer
relative neglect in terms of nutritional inputs, analysed the
NM1B survey data for rural populations in seven states
(Tamilnadu, Karnataka, Andhra Pradesh, Maharashtra, Orissa,
West Bengal and Uttar Pradesh). The weights of under five
children were compared to those of well-to-do1 children of
Hyderabad obtained from earlier surveys, and different
gradesfe of undernutrition as per the Gomez scale were
delineated. The conclusion of this study was that there
' 'exists a "strong sex bias against male children' . A smaller
proportion.of girls were in moderfete to severe malnutrition
grades compared to boys in all the states , except for
Orissa, However, instead of Hyderabad standard, if one uses
the conventional NCHS standard as the yardstick for
comparison, the picture changes completely; girls appear to
be decidedly worse off. (.Gopalan, 1987) This raises questions
about the standard itself 3 which is-more appropriate for
Indian populateons and so on. '
An analysis of the monthly weight data available
from our project area in Kachchhh for nearly two yeg.rs
indicates that during the first six months of life 3 children
of both sex fare well compared to the NCHS standard, After
the ago of six months, when breast milk is not adequate,
faltering of growth begins among both boys and girls. The
sox difference in the percentage of children below the third
standard deviation is relatively, small. After the age of
three years, however, a markedly higher percentage of girls
than of boys falls below the three standard deviations below
the median curve. Given the small number of children, the
sampling error can be large and the observed sex differences
are not statistically significant (Visaria, 1987.)
On the other hand, Sen and- Sen Gupta observed on
the basis of two village study of nutritional conditions
of -zander five
children that there was greater prevalence
of undernourishment of various degrees among girls than
among boys. Interestingly, the village with a better
Contd...11...
*
' >
- 11 -
averall nutritional record had much sharper sex discriminati
on, implying that the economic benefits seem to have
primarily benefitted boys (Sen and Sengupta, 1983).
Ghosh, citing indirect evidence from outpatient
department of Safdarjung Hospital, New Delhi, (where although
more male children are brought in, severe malnutrition was
greater among girls) and from a study undertaken by PARE
in the Punjab, concludes that in India there is higher
level of malnutrition among females compared to males (Ghosh
198?).
DIFFERENTIALS IN HEALTH CARE:
While-;calories or food availability are very vital in
weight gain or nutritional status of children, equally
important is the suspectibility to ill health and recovery
from it. Differential treatment can also lead to lower level
of nutritional status among girls.
In rural areas in the Indian subcontinent where the
environmental infection load is very high, children are
highly susceptible to infections of upper respiratory tract 5
measles and diarrohea from an early age. According to the
matlab data from Bangladesh, an average child goes through
several episodes of illness before reaching age five. (Chen,
at. al. 1981, p.6^-). Recurring infections lead of protein
c&loric malnutrition (loss of appetite, malabsorption of
nutrients,etc.), which in turn increases the susceptibility
to infection. While some of the illnesses, which parents
consider minor, are generally treated at home with home
with home remedies, some others require medical attention.
The proverbial preference for sons can influence -the
decision of parents to provide even home remedies to boys
and not to girls. If the illness is severe, boys may be taken
to a private medical practitioner, whereas girls with similar
symptoms may be treated with home remedies for a longer time.
It is even argued that the boys are weaker or more vulnerable
and their condition would rapidly deteriorate, if not
treated, compared to girls.
Contd...12...
I
12 -
This is an area, where a few studies have been carried
out. Indirectly, the outpatient records of primary health
centres (PHC) can be examined to discern differentials by
sex and age in utilization of health care facilities. Also,
the information on causes of death and help sought before the
death can help to understand whether there are any sex/
specific differences in medical treatment.
The analysis of the medical care in fatal illnesses
in the Khanna study, indicated that for the same illness
parents provided higher quality medical care to boys than to
their daughters (Singh, Gordon and Wyon, 1962). In Maltab
area of Bangladesh, where treatment for the diarrhoeal
diseases was free, parents brought their sons to the
treatment centre far more frequently than their daughters
(Chen, at, al. 1981).
The Narangwal data on the extent of
early medical care (within the first 25- hours), in terminal
illnesses also showed that women were discriminated against,
As against 65- percent of males receiving "early care 5-8
percent of women received similar treatment (Kleiman , at.
al. 1983, pp 199 - 200).
In Khan's study area in Uttar Pradesh 9 although none
of the 21 informatits accepted that they discriminated
between boys and girls in treatment, the records of patients'
visits to the primary health centre for one week indicated
that out of 56 children brought there, 5-3 (or 77 percent)
were byys. (Khan, 1983), Further, it was observed that
when sons did not get well by the treatment of local doctors 9
they were taken to city hospitals, while girls were not.
Pettigrew has sh-'wn that in rural Punjab diarroheal affected
male and female children alike, but there was differential
in treatment s " an active interest din curing diarrhoeal upsets
or indeed any illness at all was most noticeable with
respect to male children". (Pettigrew, 1987).
Our analysis of information on medical help sought
prior to death of children in the project area of Kachchh
in ^ujarat indicates that the parents resorted to medical
help when children of either sex fell ill, and yet the number
of female infant and child deaths was two and a half times
contd.... 13..•
Position: 911 (7 views)