Draft
ANALYSIS OF TRENDS
IN SEX RATIO AT BIRTH OF HOSPITALISED DELIVERIES IN THE STATE OF DELHI
Joe Varghese
Vijay Aruldas
Christian Medical
Association of India
In Collaboration with
Office of the
Registrar General of India
Summary
Increasing
masculinity child sex ratio in different parts of India has been an important
aspect noted in the last few census reports of India. The misuse of medical
technology for the identification of the sex of the child before birth and
selective abortion of female fetuses is considered as the major reason for that.
The State of Delhi is one of the affected areas in the country with severe
demographic imbalance in child sex ratio.
The study is an attempt to identify the emerging pattern of sex ratio at
birth [SRB] of hospitalised deliveries in the state of Delhi and various
demographic and socio-economic factors affecting it based on certain available
hospital data.
In the first stage of the study, SRB of eight large hospitals for ten-year period from 1993 to 2002 is calculated to examine any emerging trend. In the next stage, certain socio-economic and demographic variables are correlated with the SRB estimates from about 11,268 birth information of the year 2000 and 2001 available with one of the hospital. Births are taken as the unit of analysis and SRB is used as the indicator for sex selective abortions.
Ten-year study of SRB of hospitalised deliveries shows sharp increasing trends in masculinity from the beginning of the study period 1993/94 till 1997/98 period and thereafter showing stabilisation. Existence of intensive son preference is evident from the estimation of SRB according to the order of birth and sex composition of the previous children. Most of the sex selective abortions are occurring for the second or higher order of birth when the previous children are females. Evidences show that among parents who have more than high school education, sex selective abortions are more widely practiced by those who have post- graduation or professional qualifications. The probable explanation for this unexpected finding is the decreasing levels of fertility among the highly educated mothers forcing families to resort to sex selective abortions to have their desired number of sons in a small family. Result also shows that women who are ‘not working outside home’ tend to undergo more sex selective abortions compared to those who are employed suggesting the influence of women’s economic empowerment and autonomy in shaping the family organization strategies.
Only interventions focusing
far-reaching social changes in improving value of girl child, increasing female
autonomy and women’s opportunity in employment etc are likely to change the
current scenario.
Introduction
Sex Ratio At Birth [SRB]
refers to the ratio of male to female children born in a specific period or all
the children ever born to a cohorts of women. In all human populations, there
is a fairly stable Sex Ratio At Birth observed in countries with good vital
registration which is approximately 104 to 106 boys per hundred girls [Census
of India, 2001]. This advantage is because of the conception more boys than the
girls even though there is increased male foetal loss than female fetuses
during gestation period. Evidences suggest that unless there is conscious
effort at intervention by humans, the sex ratio at birth will not change even
over a century [Visaria L, 2002]. However in some regions of the world
especially in the south and East Asia the strong preference for sons have
distorted the SRB. The abnormal SRB have been found to be due to widely prevalent
sex selective practices.
The strong patriarchal
values in certain societies are reflected in their practices. Son preference
has been one of the most evident manifestations of patriarchal society which
depict the powerlessness of women within such societies. 1980 onwards, in
countries like China where there is strong population control programmes, the
fertility decline intensified the manifestation of son preference [Zhao 2000].
In India too with the declining fertility combining with the persistence of
strong preference for sons, parents are taking steps to ensure the birth and
survival of the sons compared with the females.
Over the years, the son
preference in India had worked against the female sex particularly in their
infancy and early childhood. She is discriminated against in many ways –
ranging from abandonment of girl children, fewer months of breast feeding, less
of nurturing and play, lesser medical treatment if falls ill etc- all working
against the very existence of girl children. The level of discrimination
comparatively reduces the chances of survival girl children is clearly evident
by the differential child mortality among boys and girls [Agnihotri, 2001,
Miller 1989, Das Gupta,1987]. Dreze and
Sen [1995] have pointed out that the persistence of gender inequality and
female deprivation are among India’s serious social failures. Today, with the
technological advancement in medical diagnosis this discrimination begins even
before her birth. Various medical technologies have been put into practice to
identify the sex of the child before the birth and selective abortion, if found
female. Of the various medical technologies, ultra sound machines are the most
misused one in the sex selective practices [Ganathra, 2001]. Evidences shows
growing incidence of pre-birth elimination of the girl children in India [Sabu
George,1998].
Trends in juvenile sex
ratio in India
The data on child sex
ratio provides a broad indicator of the ground realities as they exist in the
fabric of our society in its attitude and outlook towards the girl
child.[Census of India, 2001] The
Juvenile sex ratio in India as published by the last few census reports show
enormous masculinity. According to the 2001 census report, while the overall
sex ratio has increased from 927 females per 1000 males to 933 females per 1000
males, the Child Sex Ratio of 0 – 6 has reduced from 945 to 927girl children
per 1000 male children. The child sex ratio at birth of 927 for the country as
a whole is less than the universal sex ratio at birth. Of the total of 577
districts in the country, 48 districts showed inordinately low levels of child
sex ratio of below 850. In the 1991 census report not a single district showed
child sex ratio below 800. It is shocking to note that in next 10 years, 16
districts fell under this category. There were as many as forty eight districts
where child sex ratio is under 850 during 2001, while there was only one
districts in 1991. It is difficult to comprehend that as many as 456 districts
constituting 79 per cent of all districts in the country have registered a
decline of child sex ratio between 1991-2001. Of these, in seventy districts
the decline is in the order of over fifty points.
The State of Delhi is one
of the severely affected areas with severe demographic imbalance in child sex
ratio. The over all child sex ratio in Delhi is 865 with 6 out 9 of its
districts showing a drastic drop [more than 50 points] over the past one
decade.
Factors determining the
son preference
Son preference is deeply
entrenched and has its social and economic basis. Various earlier studies and
the census figures are suggestive to a certain extent the association of son
preference to various socio-cultural, economic and certain demographic factors.
In south Asian countries such as India, South Korea and China not only do sons
have important roles in rituals, they may be the only source of support for the
parents for the old age [Zhao,2000; Chen et al, 1981; Dasgupta, 1987].
One anticipated
correlation was that between the rising women’s status in terms of entry into
education and employment and reduced son preference or dependence. However a
number of demographic researches show that daughter discrimination continues to
occur in populations where women enjoy education and employment [Croll, 2002].
A study on the impact of son preference among north Vietnamese community
suggests ‘more empowered female adults are more likely to resort to modern
strategies in order to have a son, such as sex selective abortions’ [Belenger,
2002]. The national census data shows that most regions characterized by the
adverse child sex ratio are the advanced regions of India in terms of per
capita income as well as literacy level [Census of India, 2001]. The data also
shows more adverse child sex ratio in urban areas, though the urban areas are
characterized by higher literacy level especially among the females, more
employment opportunities for women etc.
A study of the sex ratio
at birth [estimated indirectly from SRS data] of select six states in India
indicates its association with the socio-economic conditions, total fertility
rate and mother’s mean age at fertility [Dutta P, 2001]. The examination of the
patriarchal societies of Asia noted that the combination of fertility decline
and son preference triggers the incentive for sex selective abortion
[Zhao,2000, Bairagi, 2001]. The effect of the sex composition of previous
children on subsequent fertility reveals the intensity of son preference in the
Chinese society [Wen 1992]. Various studies looking at the India context are
also showing that when the fertility declines and the preference for male
children remains strong, parents still take the steps to ensure the birth and
survival of male children [Sudha and Irudaya Rajan: 1998, Clark and Shelly,
2000]. A study of female foeticide in rural Haryana by Sabu and Dahiya [1998]
had pointed at the linkages between sex of the living children with the family
organising strategies. The assessment of sexual preferences of women in NFHS 2
also reveals the same. It shows among women with two living children, the
proportion wanting more children is far greater among those with two daughters
[53%] than those with two sons [17%] [Remez L, 2001]. Another assessment of NFHS 2 data reveals the
association of high sex ratio at birth with geographic region, child’s birth
order and mothers number of living sons and two socio economic characters –
mother’s education and mother’s media exposure [Retherford, 2003]. A community
based study of the reasons of induced abortions shows that husbands of the
women undergoing sex selective abortions were less educated than the husbands
of other abortion seekers [Ganathra et all, 2001].
The study is an attempt to
identify the emerging pattern of sex ratio at birth of hospitalised deliveries
in the state of Delhi and various demographic and socio-economic factors
affecting it based on certain available hospital based data. The study is to
evaluate the impact of son preference under conditions of social and economic
development, changes in fertility etc on sex ratio at birth.
Objectives of the study
1.
To understand
the trends in sex ratio at birth of last 10 years among hospital deliveries in
Delhi
2.
To identify the
effect of different demographic and socio-economic factors on the sex ratio at
birth.
Justification
of the study
The widespread misuse of
medical technologies for the selective elimination of the female foetuses had
evoked the civil society response for long in India. Unfortunately, a law
enacted in 1996 to regulate pre-natal diagnostic techniques and prevent sex
selection was very laxly implemented resulting in the rampant malpractice. Very
recently, for the last three years, the country had witnessed an
intensification of campaign against sex selection by the civil society with the
judicial intervention following a public interest litigation by Cehat, Masum
and Sabu George and also with the alarming revelation of demographic profile
[0-6 sex ratio] brought out by the national census 2001. The Law implementation
machinery was also revamped as evident by the increase in the registration of
the ultra-sound machines though out the country. In the state of Delhi, both
the civil society action and the government law enforcement mechanisms were on
an upsurge over this period. Whether this has changed the ground realities in
terms of reduction in the elimination of female fetuses need to be further
probed. Studying the trend in the sex ratio at birth from different hospitals
over the last 10 years is expected to provide this information.
Much of the evidence on the
spread of sex selective abortion in India is anecdotal. There is no reliable
statistics on the practice at either state or national level [Retherford, Roy
2003]. The main determinants in the child sex ratio are sex selective
mortality, sex selective migration and sex selective omission on enumeration
and sex ratio at birth. The first three reasons can confuse in drawing
conclusions on the actual severity of sex selective abortions. In a state like
Delhi selective migration of families coming only with the male children are
often sited as one reason for the skewed child sex ratio [Times of India,
2002]. Visaria [2002] opinions that the contribution of sex selective abortion
on the sex ratio at birth requires a careful analysis of data from various
sources, such as SRS, census and birth records from hospitals or institutions.
It was further stressed that more data is needed on the extent of female
foeticide, the demographic and socioeconomic status of women who undergoes sex
selective abortions. Information on sex ratio at birth is likely to reflect
more light in to the actual scenario. According to Sudha and Raja [1998], in
India available data help us to understand the juvenile sex ratio rather than
sex ratio at birth. They too emphasis the need to have future research
examining the demographic behaviors in India from a gendered perspective that
scrutinises the nexus between cultural and economic factors and household
organisation and strategies. Vina Mazundar in comparison of studies on sex
selection practices using foeticide and infanticide have pointed out the lack
of information on cast, culture and life styles of those who practice foeticide
practices.
Data and Methods
The study is based on the data available with the birth
records of select large hospitals in Delhi. Births are the unit of analysis and
sex ratio at birth [SRB] based on the hospital birth is the indicator for sex
selective abortions. The study is designed in two stages. In the first stage of
the study each year’s SRB of eight hospitals for the previous 10 years will be
calculated and plotted on a graph to examine any emerging trend. The birth data
was collected from three public sector hospitals and five private hospitals
belonging to various geographical areas of the state. The state of Delhi, being
well connected with roads and public transport system, it is assumed that
people do not have geographical preference in accessing any of these large
hospitals.
In the next stage of the study, socio-economic and
demographic variables were correlated with the SRB estimates from about 12,000
birth information of the year 2000 and 2001 available with one of the hospital.
The hospital is purposely selected for the study because of the availability
and accessibility of the data.
Based on the existing
evidences and data availability the study conceptualises the relationship
between sex ratio at birth with the following predictor variables.
In the analysis of the data, births are the unit of analysis.
SRB for various categories were calculated as number of boys born divided by
number of girls born based on the indexed birth of hospital records during the
2000 –01 period. SRB is used as an indirect indicator for sex selective
abortions. Each of the variables is correlated with the SRB to identify the
relation between them. Finally logistic regression is applied to analyse the
effects of selected demographic and socio-economic variables on sex ratio at
birth. For the logistic regression, sex ratio at birth is taken as binary coded
as 1 if, birth is male and 0 if it is a female.
Our estimates of SRB based on hospital data are in many cases
substantially affected by sampling variability and therefore highly
approximate. Difference between SRBs that are not statistically significant
must be interpreted with caution.
Results
Part 1
Trends in the
Sex Ratio at Birth of Hospitalised deliveries of Delhi
The trend is calculated from the data of hospitalised
deliveries of eight large hospitals of Delhi for ten year period between the
year1993 and 2002. These hospitals belong to various geographical locations in
the state of Delhi. Analysis is pertaining to a total of 375499 births in ten
years of which 249964 are from the government hospitals and 125535 are from the
private hospitals. The SRB of total births of the study hospitals in ten year
is 115, that of the government hospitals is 113 and private hospitals is 120
boys per 100 girls. The total number of births in each year is as given in the
table no. 1. On an average, the data is pertaining to about 37550 births per
year. As per the SRS data [2002], the birth rate for the state of Delhi is 20.3
for the year 1996-1998. Applying the same crude birth rate to the population of
Delhi according to the 2001 census [13,782,976], it can be estimated that
279794 births are taking place in Delhi every year. Since only 59% of births
are taking place in health facilities [NFHS– 2] the sample is roughly about 23%
of hospitalised deliveries of Delhi.
Table No. 1 Total Number
of births in the study hospitals each year
Year
|
Number
of births
|
1993
|
34707
|
1994
|
36064
|
1995
|
37262
|
1996
|
36512
|
1997
|
37701
|
1998
|
35317
|
1999
|
39852
|
2000
|
40039
|
2001
|
39229
|
2002
|
38816
|
The trend in SRB over the ten year period is verified
by calculating the floating average of three years and plotted on the graph as
given below [Figure 1].
Figure
1 Three-year
floating average SRB of hospitalised deliveries of Delhi [ between 1993- 2002]
The above figure shows increasing SRB trend from 1993-95 period [111 boys per 100 girls] onwards up to the period of 1996-98 where it reached 117 boys per 1000 girls. There after SRB shows slight reversal of the trend. The data shows that for the next few years it remains between 116 to 118 boys per 100 girls.
Part 2
Correlates of
Sex Ratio at Birth of hospitalised deliveries
The
results are pertaining to the detailed birth information for the year 2000 and
2001 that was collected from one of the study hospital. There were 11267 births
occurred in the hospital in the two-year study period, SRB of which is 1.24. The SRB for the year 2001 is 1.18 and that of
the year 2000 is 1.30. All the following
analysis are done for both the years together. The birth information are taken
from the labour room records which are entered in the just before and after the
delivery as reported by the mother.
SRB by birth order
and sex of previous children
Sex Ratio at Birth by number of existing children
indicates the intensity of sex selection in the higher order of births. It can
be seen from the figure 3 that the SRB for the first order of birth is 107,
which falls slightly above the normal range showing that only minimal sex
selection happening for the first child. However for the 3rd and
above order of birth, SRB is significantly masculine at 247 boys per 100 girls.
This is other words explains that there are only about 30 percent chance for a
girl to be born in 3+ birth orders.
Figure 3 SRB by Birth Order
Table
1 SRB by Birth Order
Birth Order
|
N
|
SRB
|
1
|
5929
|
107
|
2
|
4142
|
138
|
3+
|
1196
|
147
|
SRB by birth order may conceal the intensity of the
sex selective abortions, if the sex selective abortions are happening to
eliminate both boys and girls to achieve ideal sex composition of
children. Therefore to capture the realistic
picture of sex selective abortions that are occurring, it is necessary to
examine SRB by both birth order and sex composition of the existing children.
Table 1 shows that, as expected the sex composition of the existing children is
an important determinant for the sex of the next child. The difference is very
conspicuous for the second child depending on whether the first child is a boy
or a girl. The SRB for the second order birth for those who have already one
male child is 1.04 [959 girls per 1000 boys], which is within the range of
normally occurring SRB. However when the first child is a girl, the SRB for the
second order of birth is as high as 1.85 [542 girls per 1000 boys]. Similarly,
SRB is as high as 4.56 [219 girls per 1000 boys] for the 3rd order
birth when both the earlier children are girls.
Table 1 SRB by
sex of the previous children
|
Birth Order |
Sex composition
of existing Children |
N |
Number of female
birth per 1000 male birth |
SRB Hospitalised
Deliveries |
SRBNFHS –2[Delhi] |
|
2nd
order |
One male child |
2091 |
959 |
1.04 |
1.08 |
|
One female child |
2075 |
542 |
1.85 |
1.21 |
|
|
3rd
Order |
One male child
and one female child |
391 |
558 |
1.79 |
1.03 |
|
Two male
children |
161 |
894 |
1.12 |
0.87 |
|
|
Two female
children |
474 |
219 |
4.56 |
1.56 |
Age of parents
In the sample the average
age of mother is 25. 73 years and that of the father is 29.19. The table 4
shows direct relation of high sex ratio at birth with the increasing age of
parents. However this is inconclusive as the increase in SRB could be due to high
SRB in higher birth orders as suggested in the figure 2.
Table
4 Age of
parents
|
Age of mother [Number of
birth] |
SRB |
Number of female
birth per 1000 male birth |
Age of father [Number of
birth] |
SRB |
Number of female
birth per 1000 male birth |
|
Less than 24
years [3371] |
1.11 |
902 |
Less than 27
years |
1.16 |
860 |
|
Between 24 and
28 [ 5631] |
1.26 |
792 |
Between 27 and
32 |
1.23 |
814 |
|
More than 28 [2325] |
1.58 |
634 |
More than 32 |
1.6 |
627 |
Table 3 describes SRB of
first born by mother’s age. SRB of younger mothers [less than the sample
average age of mother for the first order birth] for the first child falls
within the normal range of universal SRB. However there is increase in SRB for
more aged mothers for the first order of birth showing that some sex selective
abortions of girls are happening even for the first order of birth, when the
mothers age at first birth is increasing.
SRB is increased from 1.05 to 1.08 when the first order birth for the higher
age group mother, which is categorised above and below the sample average.
Table
3 SRB for the first order of birth by mother’s
age
|
|
N |
SRB |
Number of female
birth per 1000 male birth |
|
Age of the
mother less than 23.94 years |
2769 |
1.05 |
949 |
|
Age of the
mother more than 23.94 years* |
3205 |
1.08 |
926 |
* Average age of the mother 23.94 years
Sex Ratio at Birth
and Religion
The table 4 illustrates
the SRB of current birth according to mother’s religion. All the major
religions except Muslim religion show higher than universal sex ratio at birth.
The severity is stronger among the ‘other religions’ category compared to
Hindus and Muslims community. However it should be noted that 66.2% of the
information of the current birth and 67.8% of the existing children in the
‘other religions’ group is that of the Sikh community. The similar trends are
followed when we consider the over all sex ratio of all existing children,
though with much less severity. The lesser severity sex ratio for all the
children reveals family organisation practices to achieve the desirable sex
composition of children.
|
Religion |
N |
Number of female birth per
1000 male birth |
SRB |
Number of existing
children including the current birth |
Sex Ratio of existing
children |
|
Hindu |
10308 |
781 |
1.28 |
16492 |
1.06 |
|
Muslim |
442 |
982 |
1.02 |
826 |
0.86 |
|
Other Religions |
500 |
713 |
1.40 |
772 |
1.23 |
Table 4 Sex Ratio at Birth according to religion
Education of Parents
Education
of parents is taken as a continuous variable and is calculated as the minimum
number of years of study required in reaching the reported educational
qualification. The sample shows the average years of education of the mother is
12.59 years and that of father is 13.26 years.
The
level of education appears to have an impact on the SRB. SRB of current births
of those parents who have more than high school education is 1.24 [806 female
birth per 1000 male birth]. At the same time SRB of parents with less than 10
years of education is 1.35 [739 female birth per 1000 male births]. A separate
detailed examination SRB by education of father, mother and both parents
together shows similar trends as the level of education goes up. It is
interesting to note that the only category that has normal SRB is when both the
parents have minimal education. It also shows that SRB is highest among those
with medium levels of education [high school education].
Analysis
of SRB of hospital deliveries shows that among parents who are educated [more
than 10 years of education] SRB is increasing with increase in education. While
SRB of the graduate parents is 1.22 [822 girls per 1000 boys] for parents with
postgraduate qualification it is 1.30 [769 girls per 1000 boys].
Table 5 SRB
by
education of parents
|
|
Mother |
Father |
Both Parents |
|||||||||||||||
|
N |
SRB |
No. of female birth per 1000 male
births |
N |
SRB |
No. of female birth per 1000 male
births |
N |
SRB |
No. of female birth per 1000 male
births |
||||||||||
0 to 7years of schooling
[up to middle school complete] |
897 |
1.27 |
787 |
415 |
1.15 |
869 |
271 |
1.02 |
978 |
|||||||||
|
8 to
10 years of schooling [<middle
school to high school complete] |
2383 |
1.33 |
755 |
1059 |
1.43 |
702 |
1059 |
1.42 |
703 |
|||||||||
|
11 to
15 years of schooling [<high school to graduate] |
6177 |
1.25 |
803 |
7047 |
1.23 |
814 |
4795 |
1.22 |
822 |
|||||||||
|
More
than 15 years of education [higher
than graduate education] |
1897 |
1.28 |
780 |
1820 |
1.30 |
769 |
948 |
1.30 |
769 |
|||||||||
Occupation of Parents
The table 6 shows different pattern of SRB for father and mother according to their different occupational status. Mother’s better employment status has a positive impact on SRB [as reflected by the declining masculinity of SRB] as it moves from mothers who are not working outside home to those are employed in high end job professional job.
