Gender preference among pregnant women in Nepal

INTRODUCTION

There has been decline in fertility and explicit preference for smaller families in most parts of East and South Asia.¹ Aborting a fetus of the undesired sex or halting childbearing only after achieving the desired sex composition of children is common practice in this region.² The sex selective abortion inflates the sex ratio at birth and lowers fertility.³ Sex ratio is a significant social indicator measuring status and equity between male and female in the society indicating gender preference. Nepalese society is patriarchal and son-preferred. Sex-selective abortion is rising and is one of the major causes of imbalances in the sex ratio in Nepal.^4,5 Legalization of abortion in 2002 in Nepal has further led to the sex selective abortion and activists estimate that around 50,000 unborn babies are aborted in Nepal every year after parents find out through ultrasound scans that they are girls.⁶

Gender preference affects contraceptive use, pregnancy rates, average number of siblings, sex distribution of children, birth intervals, and duration of postpartum abstinence.^7-10 Further, the risk of postnatal depression in the mothers having baby girls is higher because such mothers often experience domestic violence.¹¹ There is limited research about women’s preferences for sex of the child across castes, ethnicities, and ecological areas of Nepal, even though this is an important issue in addressing the unbalanced sex ratio and managing the overall demographic transition in the country.¹² Hence, the study was carried out to estimate the gender preference and identify associated factors among the pregnant women in Kathmandu, Nepal.

METHODS

Study area and setting

The research was conducted in Ante-Natal care out-patient department (ANC-OPD) of Paropakar Maternity and Women’s Hospital in Kathmandu for the duration of 6 months (September 2018 to February 2019). This is a tertiary public maternity hospital with high flow of the pregnant women coming for the ANC and delivery care. The hospital receives more than 5000 ANC visits monthly, and the hospital record showed that there was a total of 68111 ANC visits in the year 2018.

Study design and sampling

This is a hospital based cross-sectional study. A sample of 821 pregnant women were included in study. Pregnant women of any gestation week who came for the antenatal visit in ANC OPD of the Paropakar Maternity and Women’s Hospital were randomly selected. About 17 samples were recruited each day over the two months duration.

Data collection procedure

Pretested semi structured questionnaire was used as the research tool and data was collected via face to face interview method. The questionnaire included three sections consisting of socio-demographic information, obstetric characteristics and gender preference related information. the questionnaire was first prepared in English. We adopted the Coombs’ scale questions to seek gender preference. The questions were translated into Nepali, pre-tested and used.

Study variables and data analysis

Dependent variable is gender preference and was assessed using the Coombs scale where women were asked questions regarding their sex preferences for children, each question offering different scenarios. Respondents were given to choose their ideal scenario if they had an opportunity to bear only three children throughout their lifetime. A minimum odd number was used to maximize the biases of respondents’ gender preferences. Using these questions, each respondent’s IS position was estimated.

First, respondent was asked, “if you were to have only three children, would you most like to have three girls, one boy and two girls, two boys and one girl, or three boys.” If the respondent answered three boys, she was in the IS 7 position (extreme boy preference), and if she answered three girls, she was in the IS 1 position (extreme girl preference).

If she answers either one boy and two girls or two boys and one girl, she was further asked what her next preference would be excluding the combination she chose (second scenario). If she answered two boys and one girl in her first question, she was further asked what her alternate preference would be if she could not have that proportion. If her second choice was three boys, she was IS 6 (relatively strong boy preference).

If instead she answered two girls and a boy, she was again questioned as to what she would want if she could not have that proportion (third scenario). If she answered three boys, she was IS 5 (weak boy preference), and if she answered three girls, she was IS 4, indicating a preference for balance. The same procedure was followed if the respondent answered two girls and a boy to the initial question to identify her as IS 2 to IS 4.^13,14

In this way, we categorize the gender preference in three categories: son preference (IS 7 to 5), balanced (IS 4) and daughter preference (3 to 1). Independent variable included were age, ethnicity, religion, working status, marital duration, type of family, type of marriage, marital duration, parity, age at first pregnancy, previous child history, and history of abortion/miscarriage of the respondent. The data was analysed using descriptive statistics and bivariate analysis (Chi-square test) in SPSS version 21.

Ethical considerations

Written informed consent was taken from all respondents for the interview and the purpose of the study and procedures were clearly explained to them. Ethical approval was taken from the Institutional Review Committee (IRC Ref. No. 187/075/076-IRC) of B. P. Koirala Institute of Health Sciences, Dharan.

RESULTS

Sample characteristics

The demographic and obstetric characteristics of the respondents were shown in (Table 1). The mean age of respondent was 24.84±4.48 years. Majority were from hilly region 75.4%, were Hindu 79.5, were literate 92.7% and lived in joint family 66.5%. More than one-third of the respondents belonged to disadvantaged Janajatis 38.48% and were upper caste 35.07%. The mean age of marriage of the respondent was 20.40 (SD=3.67) years.

More than half 55.81% of the respondents chose their partners themselves (love marriage) and had marital duration of less than five years 62% at the time of interview. Majority of the respondents were in their third trimester 61.88% and were primiparous 60.3%.

Among multiparous women, majority 80.4% had babies within 2 years of marriage and more than half 58.9% had at least one son. About 15% of the respondents had history of abortion. Among those who had history of abortion, 26% had induced abortion.

Gender preference and associated factors

In the first scenario when the respondents were asked to choose one gender combination if they were to have only three children, more than half 56.2% replied two sons and one daughter. Table 2 shows the choices of gender combinations in subsequent scenario.

The distribution of respondents according to IS scale was shown in (Figure 1) and about 40% of the respondents had son preference (IS 7-5), 33% had balanced (IS 4) and 27% had daughter preference (IS 3-1).

In bivariate analysis (Table 3), the respondent characteristics significantly associated with gender preference were ecological region (p=0.007), education (p=0.009), marital duration (p=0.044), and parity (0.032). Among the son preference, higher percentage is in hills (69.5%) followed by Terai (15.4%).

*Chi-square test of association.

There was high son preference in the Terai region (50% vs 18%) compared to Hill region (36.5% VS 30.9%). Similarly, more illiterate respondent has high son preference (58.3% vs 18.3%) as compared to literate respondents (38.1% vs 28.2%). Women having more than one parity preferred son (54.7 % vs 22.6%) compared to primiparous women (36.1% vs 30.3%).

DISCUSSION

Based on Coombs’ scale and methods, this study found that there was slightly son bias as 40% of the respondents fall into IS 7-5 position. Further, ecological region, education status, marital duration, and parity were found to be significantly associated with the gender preference.

Girl’s preference in this study 27% is slightly higher than the study undertaken in Indian city of Lucknow 22%, whereas balanced gender preference 33% was higher in this study compared to Lucknow study.¹⁵ Another study in Nepal carried out among the primiparous women found that majority has not any gender preferences while only 13% wanted son, which is lower than this study.¹⁶ This might be due the primiparous status of respondents who may not care about the sex of the first child.

The commonest reasons for son or daughter preference were psychological (mainly all siblings are of the other sex) and social (help in household tasks, lower risk of divorce and old age care/support). The economic causes were limited to son preference (help in or inherit family business/land and contribution to family income). The commonest reasons for son or daughter preference were psychological (mainly all siblings are of the other sex) and social (help in household tasks, lower risk of divorce and old age care/support).

Daughters were more preferred in hilly region while son were preferred more in terai region. This might indicate cultural and social circumstances of sons and daughters in hills and terai as well as the women status. In terai area of Nepal, there is still more persistent poverty, gender bias and dowry system.⁵ These findings were consistent with the analysis of data from national level demographic and health survey of 2016 which concluded that sons were more preferred in terai region as compared to hilly regions.¹² Another study found that Muslims have more son preference as compared to Hindu but this was not the case in this study.¹⁷

In this study, illiterate respondent had high son preference compared to literate respondents. Previous studies reported similar association.^5,18 Education empowers the women on gender disparity and educated women are more likely to have neutral preferences. Respondents with marital duration of more than 10 years preferred more sons. Higher son preference with increase in marital duration is apparently associated with parity. Son preference has increased with parity in this study and this might be related to have desire of more sons or at least one son with no previous history of son. Such parity effect has been documented, with prominent son preference for the 3^rd or 4^th child.¹⁹

This study was carried out among special population, the pregnant women coming to ANC visit in a tertiary public hospital. Though, a variety of pregnant women of ethnicity, occupation and literacy visits this popular maternity hospital, it still may not represent whole section of community and not generalizable. There may be some chances of conscious falsification of information by the respondents or social biasness. So, true magnitude of gender preference is hard to determine, however, this study assessed women’s wish to have sons and girl’s combination reflecting their gender preferences.

CONCLUSION

The wish to have son is prevalent among the respondents followed by a mixed gender of children. Educational status, ecological region, and parity explain this differential wish. Women’s education and autonomy is probably the significant factor to tackle with gender bias.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the institutional ethics committee

REFERENCES

1. 1. Dey I, Chaudhuri RN. Gender preference and its implications on reproductive behaviour of mothers in a rural area of West Bengal. Indian J Community Med. 2009;34(1):65-7.
2. 2. Paul A, Dasgupta S, Bhattacharya S. Gender Preference among the Mothers attending Antenatal Clinic of Burdwan Medical College, West Bengal. Indian J Matern Child Heal. 2014;16(1).
3. 3. Bongaarts J. Fertility and reproductive preferences in post-transitional societies. Popul Dev Rev. 2001;27:260-81.
4. 4. Frost MD, Puri M, Hinde PRA. Falling sex ratios and emerging evidence of sex-selective abortion in Nepal: Evidence from nationally representative survey data. BMJ Open. 2015;3(5):1-8.
5. 5. Lamichhane P, Harken T, Puri M, Darney PD, Blum M, Harper CC, et al. Sex-selective abortion in Nepal: a qualitative study of health workers’ perspectives. Womens Health Issues. 2011;21(3):37-41.
6. 6. Slaughter of our unborn daughter. Nepali times. Available at http://archive.nepalitimes.com/news.php?id=19627#.XmnW_KgzaM9. Accessed on 14 September 2012.
7. 7. Khan MA, Khanum PA. Influence of son preference on contraceptive use in Bangladesh. Asia-Pacific Popul J. 2000;15(3):43-56.
8. 8. Rai P, Paudel IS, Ghimire A, Pokharel PK, Rijal R, Niraula SR. Effect of gender preference on fertility: cross-sectional study among women of Tharu community from rural area of eastern region of Nepal. Reprod Health. 2014;11(1):15.
9. 9. Rahman M, Vanzo DJ. Gender Preference and Birth Spacing in Matlab, Bangladesh. Demography. 1993;30(3):315-32.
10. 10. Calhoun LM, Nanda P, Speizer IS, Jain M. The effect of family sex composition on fertility desires and family planning behaviours in urban Uttar Pradesh, India. Reprod Health. 2013;10(1):48.
11. 11. Cankorur VS, Duman B, Taylor C, Stewart R. Gender preference and perinatal depression in Turkey: a cohort study. PLoS One. 2017;12(3):1-13.
12. 12. Sapkota JB, Paudel DB, Neupane P, Thapa RB. Preference for Sex of Children Among Women in Nepal. Glob Soc Welf. 2019;6(2):69-78.
13. 13. Coombs LC. Preferences for Sex of Children Among U.S. Couples. Fam Plann Perspect. 1977;9(6):259-65.
14. 14. Coombs CH, Coombs LC, Mcclelland GH, Coombs LC. Preference Scales for Number and Sex of Children. Popul Stud (NY). 1975;29(2):273-98.
15. 15. Srivastava A, Singh J, Singh O, Singh V, Singh N. Gender Preference, Attitude and Awareness of Young eligible couples towards Prenatal Sex Determination in Lucknow District. Natl J Community Med. 2014;5(1):148-52.
16. 16. Thapa M, Bajracharya J. Gender Preference in Current Pregnancy among Primigravidae. Nepal J Obstet Gynaecol. 2017;12(1):36-9.
17. 17. Kanyadi S, Kulkarni R. Determinants of gender preference among women of Belagavi: a cross sectional study. Int J Community Med Public Heal. 2017;4(12):4733.
18. 18. MB P, Dhanpal S, Lokanath H. A study of gender preference, knowledge and attitude regarding prenatal diagnostic techniques act among pregnant women in an urban slum of Bengaluru. Int J Community Med Public Heal. 2015;2(3):282-7.
19. 19. Hank K, Kohler HP. Gender Preferences for Children in Europe: Empirical Results from 17 FFS Countries. Demogr Res. 2000;2(1).