|Year : 2019 | Volume
| Issue : 4 | Page : 298-304
Occurrence and predictors of abortion among women of the reproductive age group in a block of Darjeeling District, West Bengal, India
Pallabi Dasgupta1, Romy Biswas2, Dilip Kumar Das3, Jayanta Kumar Roy4
1 Assistant Professor, Department of Community Medicine, RG Kar Medical College and Hospital, Kolkata, West Bengal, India
2 Professor and Head, Department of Community Medicine, Coochbehar Government Medical College and Hospital, Coochbehar, West Bengal, India
3 Professor and Head, Department of Community Medicine, Burdwan Medical College and Hospital, Burdwan, West Bengal, India
4 Ex-Assistant Professor, Department of Community Medicine, North Bengal Medical College and Hospital, Siliguri, West Bengal, India
|Date of Web Publication||18-Dec-2019|
Dr. Romy Biswas
Department of Community Medicine, Coochbehar Government Medical College and Hospital, Coochbehar, West Bengal
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: Despite being a sensitive and less explored issue, abortion is a major preventable cause of maternal morbidity and mortality affecting millions of women in developing countries. Objectives: The study aimed to determine the occurrence, nature, and predictors of abortion among women in the reproductive age group in Naxalbari block of Darjeeling district. Methods: A community-based cross-sectional study was conducted in Naxalbari block of Darjeeling district, West Bengal, India, from May 2015 to April 2016, among 420 women aged 15–49 years selected from 30 villages by cluster sampling technique. A predesigned, pretested interview schedule validated in the local vernacular was used. Binary logistic regression was used for finding out predictors of abortion among ever-pregnant women. Results: Lifetime occurrence of abortion was 33.6%. Among total 178 events of abortion, 51.7% were spontaneous and 48.3% induced. Majority of spontaneous abortions events were attended by a doctor (73.9%). About 59.3% of induced abortions were illegal, and unwanted pregnancy was major reason (62.4%) for induced abortion. Private facilities and over the counter drugs were preferred. Lower education, nuclear family, number of children <2, not having male child, domestic violence during pregnancy were significant predictors of abortion. Conclusions: Illegal abortions were highly prevalent in the area. Unwanted pregnancies hint toward unmet needs of family planning. Stigma and poor awareness were the root cause of not visiting a health facility in case of abortions. Building up of better infrastructure, better orientation of frontline workers, non-judgemental and confidential services will attract women to in government facilities.
Keywords: Abortion, epidemiology, illegal, India
|How to cite this article:|
Dasgupta P, Biswas R, Das DK, Roy JK. Occurrence and predictors of abortion among women of the reproductive age group in a block of Darjeeling District, West Bengal, India. Indian J Public Health 2019;63:298-304
|How to cite this URL:|
Dasgupta P, Biswas R, Das DK, Roy JK. Occurrence and predictors of abortion among women of the reproductive age group in a block of Darjeeling District, West Bengal, India. Indian J Public Health [serial online] 2019 [cited 2020 Sep 29];63:298-304. Available from: http://www.ijph.in/text.asp?2019/63/4/298/273356
| Introduction|| |
Abortion is a preventable cause of maternal morbidity and mortality. Deaths following unsafe abortions are mostly due to hemorrhage, infection, and sepsis. Ending this silent pandemic is an urgent public health and human rights imperative. Between 2010 and 2014, about 56 million-induced abortions occurred worldwide each year. Around 22 million were estimated to be unsafe abortions. Nearly 97% of unsafe abortions take place in developing countries.
Unsafe abortions also place a financial burden on the health system. In 2006, estimated 680 million US dollar (USD) was globally spent treating serious consequences of unsafe abortion. An additional 370 million USD needed to fully meet the unmet need for treating complications.
In India, 8% maternal deaths are attributed to unsafe abortions. Every 2 h, one woman dies of complications of unsafe abortion. District level household survey-4 shows abortion percentage to be 4.4% in West Bengal and 8.1% in Darjeeling District.
Since the legalization of abortion in India with medical termination of pregnancy (MTP) Act, 1971, there has been a decline in the maternal mortality indicators over the years. Strategies under National Population Policy and Reproductive, Maternal, Newborn, Child, and Adolescent health  initiative gave considerable focus toward comprehensive abortion care.
Access to safe, legal abortion is a fundamental right of women. However, many physical, economic, social, and policy factors limit access to competent care. These are poor availability of services, high cost in private sectors, stigma, conscientious objection of health-care providers, provision of misleading information, etc. The availability of modern contraception can reduce but never eliminate the need for abortion. Various reasons for seeking abortion reported ranges from proximate causes such as the desire to limit family size or space pregnancies, preference for a son, medical reasons to distal determinants such as poverty, violence, and belief system.
Facility-based studies and hospital records cover only reported abortions in India. Indirect estimates of abortion mainly depend on the ratio of induced abortions to live-births, ill-timed pregnancies, and age-specific fertility rates. National surveys cover only ever-married women. In South India, Bihar, Jharkhand, Orissa, and Maharashtra,,, there are few published community-based studies. In these studies, findings might not have made distinction between induced and spontaneous abortion.
In this context, the present study was conducted with the objectives of determining the occurrence, nature, and predictors of abortion among women in the reproductive age group in the Naxalbari block of Darjeeling district.
| Materials and Methods|| |
Study design, population, and area
A community-based cross-sectional study was conducted from May 2015 to April 2016 in the Naxalbari block of Darjeeling district, in West Bengal, India. Darjeeling, the gateway of North-East India, is surrounded by Sikkim, Kishanganj district of Bihar, and the country of Nepal. It has structured government healthcare delivery system, with North Bengal Medical College as the tertiary referral institution.
Study participants were women of 15–49 years who were residents of the study area for at least 1 year.
Considering the anticipated proportion of abortion event as 45.7% (based on an earlier study ), 95% confidence level (CI), 15% relative precision, and design effect 2 the minimum required sample size was calculated to be 420.
Cluster sampling technique was applied to select the subjects. 30 clusters (villages) were identified by probability proportional to size. Cluster size was 14. In each identified cluster, the sampling frame of women of 15–49 years' age was prepared beforehand with the help of frontline workers and 14 women were selected by simple random sampling (without replacement). Unwilling women, seriously ill, those withdrawing from study or women unavailable even after three visits were considered nonresponders. Nonresponders per cluster were replaced from the sampling frame of the same cluster.
Tools and techniques
Data were collected by interviewing study women and reviewing relevant medical records. A predesigned interview schedule consisting of background characteristics, risk factors of abortion, pregnancy, abortion history, i.e., lifetime occurrence of abortion, nature (spontaneous or induced) of abortion, reasons for induced abortion, place, person, and method of abortion was prepared. Before starting the study, the schedule was validated in local vernacular by initial translation, back-translation, and re-translation, followed by pretesting among a convenient sample of 30 women of reproductive age group residing in a village of different block.
Variables: Operational definition/descriptions and data collection
Background characteristics included age (in completed years), religion, education (years of schooling), type of family (nuclear or joint), marital status, occupation (working or homemaker), number, and gender of children. Risk factors for abortion were contraception usage (of preceding 1 month), addiction, frequent travel (traveling more than 1 hour by walking/cycling/by other vehicles such as motorcycles, auto-rickshaw, motor cars on a regular basis), any sort of moderate or heavy work, accidental injury/domestic violence (physical) during pregnancy.
Abortion was defined as expulsion of the products of conception before the fetus is viable before the 28th week of gestation. Spontaneous abortion is the natural loss of products of conception without any intervention. In this study, inevitable, incomplete, septic, and missed abortion were also included under spontaneous abortion. Abortion brought on intentionally by medication or instrumentation due to any cause is induced abortion. Legal aspects of abortion were perceived depending on variables such as gestational age, place, person conducting-induced abortion. According to the provisions of the MTP act  as per available guidelines at the start of data collection, the responses elicited by the women were classified as much as possible. Any abortion done beyond the provisions of MTP act was considered illegal.
Ethical clearance was obtained from the Institutional Ethics Committee of North Bengal Medical College. After necessary approval from respective authorities, data was collected at households of selected participants with prior informed consent maintaining privacy, confidentiality, and anonymity.
Binary logistic regression was done to find predictors of abortion among ever-pregnant women where, ever-abortion status (Yes [=1] or No [=0]) were the dichotomous dependent variables and background variables such as age, religion, type of family, marital status, education, occupation, gender, and number of children were independent variables. Similarly, binary logistic regression was done to find associations between risk factors such as contraceptive usage, addiction, heavy work, frequent travel, accidental injury, and domestic violence during pregnancy and ever-abortion status. Analysis of the data was performed using IBM SPSS version 20, Armonk, New York, USA, IBM Corp. A P < 0.05 was taken as significant.
| Results|| |
A total of 420 women aged 15–49 years were interviewed from 30 selected villages in Naxalbari block. The mean age of the participants was 26.4 ± 7.1 years (range, 15–49). Majority were currently married (346, 82.4%), Hindu (367, 87.4%), belonged to joint family (221, 52.6%), educated with ≥5 years schooling (312, 74.3%). Majority were homemakers (252, 60.0%) by occupation, others were engaged in unskilled work such as daily wage laborers, agricultural activities, and tea garden work.
Of 420,339 (80.7%) women became ever pregnant. Among the ever-pregnant women, 141 (41.6%) had any abortions in their lifetime. Hence, the lifetime occurrence of abortion among all women was 33.6% (141/420). The total number of abortion events among ever-aborted women was 178; 92 (51.7%) were spontaneous abortion, 86 (48.3%) were induced abortion. Mean abortion events per woman were 1.3 [Figure 1]. Only 122 (29.0%) women were aware of legal status of abortion.
Majority of spontaneous abortion events were attended a private health facility (45.7%), assisted by a doctor (73.9%), and products of conception were evacuated using surgical methods (56.5%). Majority (58.2%) of induced abortion were done outside a health facility at quacks' place, home, etc; 14% by untrained providers. Induced abortions were mostly done by medical methods (68.6%) with over the counter drugs accounting for 44.9% cases. Thus, majority (51 [59.3%]) of induced abortions were illegal [Table 1]. In the previous year 2014, there were 28 abortion events; thus, the abortion percentage was 6.7% (28/420) for that year. Of 28 events, 17 were spontaneous abortions and 11 were induced abortions with only 3 events legal.
|Table 1: Nature of abortion care, timing, and legality of events (n=178)|
Click here to view
Fear of loss of confidentiality was major reason for not visiting any health facility in case of induced abortion. Lack of awareness or feeling unnecessary to visit was major reason cited in the case of spontaneous abortion events [Figure 2]. Out of 86 induced abortion events, unwanted pregnancies (54, 62.4%), medical causes (6, 7.1%), fetal abnormality (6, 7.1%), socioeconomic (5, 5.9%), and other reasons (15, 17.6%) like pregnancy before marriage, alcoholic husband, being unaware of pregnancy, having grown-up kids, etc., were the cited reasons for induced abortions.
|Figure 2: Radar chart depicting reasons for not visiting health facility in case of induced (n = 50) and spontaneous abortions (n = 16) events. In case of spontaneous abortion events, 16 cases, out of 92, no health facility was visited. Other reasons for not going to a health facility in spontaneous abortion cases were-local customs, other kids at home. In the case of induced abortion events, 50 cases, out of 86, no health was visited. Other reasons for not going to a health facility in induced abortion cases were-early stage of pregnancy, uneventful previous experience of self-induced abortion, easy availability of pills.|
Click here to view
Women more than 30 years age, Hindu, currently married staying with husband, lesser years of schooling, working women, not having a female child had higher odds of abortion. These associations were not significant in the adjusted analysis. Belonging to nuclear family (adjusted odds ratio [AOR], 1 [referent] vs. AOR, 0.48; 95% CI, 0.29–0.80), not having male child (AOR, 1.79; 95% CI, 1.01–3.17) and number of children <2 had significantly higher odds (AOR, 1 [referent] vs. AOR, 0.39; 95% CI, 0.20–0.78) of having abortion. The model here explained between 7.7% (Cox and Snell R2) and 10.4% (Nagelkerke R2) of the variance in having ever-abortion, correctly classifying 63.1% of cases. The model fitted the data (Hosmer–Lemeshow test Chi-square, 6.778; P > 0.05) and Omnibus test Chi-square, 27.259; P < 0.001) [Table 2].
|Table 2: Background predictors of abortion among ever-pregnant women (n=339)|
Click here to view
Majority of ever-pregnant women did not have any sort of addiction (260, 76.7%); others were addicted to betel-leaf, supari, tobacco, and alcohol. About 109 (32.2%) of women were engaged in some sort of heavy work like carrying weight for agricultural works, tea gardening, construction or drawing water from wells or carrying water from distance. Some (57, 16.8%) had to travel frequently [Table 3]. About 111 (32.7%) did not use any type of contraception. About 25 (7.4%) women experienced domestic violence during pregnancy and 56 (16.5%) were accidentally injured during pregnancy. Domestic violence was significantly (AOR, 4.73; 95% CI, 1.79–12.51) associated with abortion. The model here explained between 4.3% (Cox and Snell R2) and 5.8% (Nagelkerke R2) of the variance in having ever-abortion, correctly classifying 62.2% of cases. The model fitted the data (Hosmer–Lemeshow test Chi-square, 2.989; P > 0.05 and Omnibus test Chi-square, 14.857; P < 0.05) [Table 3].
| Discussion|| |
During 2010–2011, about 6, 20, 472 MTPs were performed by 12,510 approved institutions in India. About 60% MTPs were performed in Assam, Maharashtra, West Bengal (10%), Tamil Nadu, Uttar Pradesh, and Haryana. Health on March report of West Bengal, (2015–2016) shows a total of 54,422 abortions done in the state with 681 reported from Darjeeling district. Block-level data for 2014–2015 of Naxalbari reported 15 abortions and 12 still births. However, the present study shows the total number of abortions among 420 women in 2014 as 28. The abortion percentage (6.7%) in the present study is higher than the state percentage but lower than district statistics. There might be under-reporting of block data due to nonreporting of MTPs done at private institutions or self-induced abortions. Reporting of late trimester abortions as stillbirths can be another possible cause.
The lifetime occurrence of abortion ranged from about 33.5% to 45.7% in various studies conducted globally.,, A Pakistan study  found mean abortions per women as 1.4; 76.9% spontaneous and 23% induced abortions. In Purba Medinipur district  of West Bengal, 46.2% of reported abortions were spontaneous and 53.8% were induced.
Most induced abortions are usually performed first trimester in India , suggestingawareness regarding the complications of late abortions. In India, 21% of reported births in 5 years from 1999 to 2004 were unplanned. About 63.2% of MTP seekers reasoned that completed family size was behind their decision to terminate pregnancy, in a study conducted in Jamnagar. Similar findings were observed elsewhere.,
The findings in the present study regarding place of abortion suggest towards the stigma associated with induced abortions. Health on March 2015–2016, West Bengal  shows only 38.4% of all reported MTP taking place in public facilities. In the Purba Medinipur study, 48.2% of MTP took place in nursing homes and 46.2% at residence. In Madhya Pradesh, a multi-center study showed 23% of women visited private facilities, 9% government doctor and rest 68% went to chemists, trained birth attendant, nurse at first for induced abortion. Poor and erroneous knowledge regarding legal aspects of MTP as reported in various studies add to the stigmatization of abortion in India.,
The present study finding depicts situation of illegal abortions in the area. A facility-based study in Darjeeling  depicts nearly 71.9% of septic abortion cases to be conducted by untrained persons. The surgical method was adopted in 33.3% cases and indigenous method in 26.3% cases. Uncertified persons such as traditional birth attendants, nurses, midwives, pharmacists or quacks are still the major abortion service providers in India. Pharmacists of Delhi, revealed almost 20% clients come without prescription for MTP. Over the counter drugs has advantages of convenience, anonymity, hasty transaction, easy accessibility, and cost-saving. Shrivastava andYadav. reported 89.5% patients taking medicine at home prescribed by either family members (37%) or chemists (42.38%).
Background predictors of abortion
The proportion of having abortion progressively decreased with increasing age in various studies. In Purba Medinipur, majority of abortions were in 20–29 years (61.2%). MTP-seekers at Jamnagar were mostly aged 20–30 years. The present study showed increased likelihood of abortion with increased age in the adjusted analysis. Such findings might be due to the consideration of lifetime occurrence of abortion. Whereas, most of the other studies were conducted among a cross-section of women seeking abortion care.
Likelihood of abortion among Hindu women was slightly more in this study. Muslim religious beliefs often prelude abortion. However, in Pakistan  findings showed strong association of induced abortion with nonMuslim families. Since the overall proportion of women belonging to Hindu religion was more in the present study, the findings cannot be generalized.
Marriage has a strong effect on fertility and the married constitutes significant majorities of all abortion clients. They report to facilities more approvingly. However, unmarried/separated women have problems of confidentiality, nonconsensual sexual relations, premarital sex, partner or family insistence, lack of partner support. However, these issues could not be elicited in present study due to overall less proportion of unmarried women.
Residing in joint family influences women's decision toward her reproductive choices by family pressure, social norms, and customs. Going for abortion in case of a woman belonging to the nuclear family is much easier due to less chance of being known by others. A study analyzing secondary data of the National Family Health Survey-2 (NFHS-2) had shown that women's autonomy influences the use of pregnancy care services. Moreover, economic support of rearing another child in case of unintended pregnancy is more likely in joint families. In contrast, in Bihar and Jharkhand, many women perceived that they have strong family social support systems for abortion if required.
There are differences in evidence to show whether lower education level increases chances of abortion or otherwise. Occurrence of abortion was higher among less-educated women suggest findings from Jamnagar, Indore. However, educational level might not influence the desire for more children. Lower education might make the women vulnerable to repeated and unwanted conceptions due to lower age at marriage, social pressure onto bear children, women's lack of decision-making power, improper contraceptive behavior.
Proportion of women having induced abortion was statistically significant in case of working females in Pakistan. In the present study, working women were mostly engaged in moderate-to-heavy physical work. This might make them more prone to abortions.
The findings, in the present study suggest that there may be more preference for male child but usually no gender preference among our study population, rather abortion is usually seen as a means of limiting number of children or unintended pregnancy. Gender preference is thought to be an important reason for seeking abortion in India. Jha et al. reported that till 2006 an estimated 10 million female fetuses were illegally aborted in India. Nationally, census data show child sex ratio dropped from 962 (1981) to 914 (2011). Health on March  report shows about 231 ultra-sonography clinics were suspended in West Bengal under the Pre-Conception and Pre-Natal Diagnostic Techniques act indirectly showing gender-selective abortion might be still prevalent in the state. However, NFHS-4 shows favorable sex ratio of 1011 in the state. It is well-known fact that sex determinations are possible at >12 weeks of gestation. Interestingly, 91.9% of induced abortion in the present study was within 12 weeks, hinting that gender preference might not be the cause of abortions in our study population. NFHS-2 analysis suggests unintended pregnancy rather than gender preferences underlie demand for most abortions.
Risk factors for abortion
Contraception failure was reported by 38.1% women as a cause for seeking abortion in Indore  and 22.3% women in Delhi. In the present study, contraception usage in preceding 1 month was taken as proxy measure of usual contraceptive behavior. Detailed variables such as the start of using, years of use, and method of use were not explored. Hence, the results might not be conclusive.
Domestic violence is a consistent and strong risk factor for abortion across a variety of settings. NFHS-4 data shows 5% of ever-married women experienced domestic violence during pregnancy in West Bengal. About 32.8% have ever experienced spousal violence. All these figures depict the scenario of women's reproductive freedom. In Tamil Nadu nonconsensual sex, sexual violence led to abortion.
Despite careful effort, some limitations could not be avoided, which should be kept in mind while interpreting the findings. Household setting and privacy during interviewing have tried to reduce the possibility of social desirability bias. Researcher asked open-ended questions which complemented data from structured questionnaires. Retrospective recording of events of abortions and related variables might have led to recall bias. A follow-up of a single cohort of women would have given a greater insight into the subject.
| Conclusions|| |
Illegal abortions are highly prevalent in the study area with preference toward private facility and over the counter drugs. Unwanted pregnancies hint toward unmet needs of family planning. Stigma and poor awareness are the root cause of not visiting a health facility in case of abortions. Abortion is usually seen as a means of limiting number of children. Building up of better infrastructure, orientation of frontline workers, nonjudgmental and confidential services will attract women to government facilities. Awareness generation about MTP act, emergency contraception, dangers of illegal abortions are needed among the women and community in general.
Indian Council of Medical Research for providing research grant.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Vlassoff M, Shearer M, Walker D, Lucas H. Economic Impact of Unsafe Abortion Related Morbidity and Mortality: Evidence and Estimation Challenges. Brighton: Institute of Development Studies; 2008. (IDS Research Reports 59).
Maternal Health Division, Ministry of Health and Family Welfare. Comprehensive Abortion Care: Provider's Manual. New Delhi: Government of India; 2014.
Ministry of Health and Family Welfare: District Level Household and Facility Survey-4; West Bengal Factsheet. International Institute for Population Science Mumbai (India); 2012-13. Available from: http://www.rchiips.org
[Last accessed on 2018 Aug 24].
Duggal R, Ramachandran V. The abortion assessment project – India: Key findings and recommendations. Reprod Health Matters 2004;12:122-9.
Ravindran TK, Balasubramanian P. “Yes” to abortion but “no” to sexual rights: The paradoxical reality of married women in rural Tamil Nadu, India. Reprod Health Matters 2004;12:88-99.
Ganatra B, Hirve S. Induced abortions among adolescent women in rural Maharashtra, India. Reprod Health Matters 2002;10:76-85.
Nojomi M, Akbarian A, Ashory-Moghadam S. Burden of abortion: Induced and spontaneous. Arch Iran Med 2006;9:39-45.
Jansankhya Sthirata Kosh (National Population Stabilisation Fund). Abortion and MTP act. Available from: http://www.jsk.gov.in/abortion.asp
. [Last accessed on 2018 Aug 24].
Government of India. The Medical Termination of Pregnancy ACT, 1971 Act No. 34 of 1971, 10th
August 1971, New Delhi: Ministry of Health and Family Welfare; 1971.
Statistics Division. Ministry of Health and Family Welfare. Family Welfare Statistics India. New Delhi: Government of India; 2011. p. 18-9.
State Bureau of Health Intelligence, Directorate of Health Services. Health on the March 2015-2016. Kolkata: Government of West Bengal; 2016.
Humayun A, Sheikh NH, Ashraf M. Abortion prevalence and socio-demographic differentials. Biomedica 2005;21:12-17.
Lamina MA. Prevalence and determinants of unintended pregnancy among women in South-Western Nigeria. Ghana Med J 2015;49:187-94.
Biswas DK, Bhunia R, Mukherjee A. High prevalence of abortion among primigravida and teen aged girls in the district of PurbaMedinipur, West Bengal; India. Indian J Public Health Res Dev 2016;7:74-9.
Sahu PC, Inamdar IF, Salve D. Abortion among married women of reproductive age group: A community based study. Int J Pharm Sci Invent 2014;3:22-8.
Shivakumar BC, Vishvanath D, Srivastava PC. A profile of abortion cases in a tertiary care hospital. Indian Acad Forensic Med 2011;33:33-9.
Gupta S, Dave V, Sochaliya K, Yadav S. A study on socio-demographic and obstetric profile of MTP seekers at Guru Govind Singh Hospital, Jamnagar Health line 2012;3:50-4.
Banerjee SK, Clark AK. Exploring the Pathways of Unsafe Abortion: A Prospective Study of Abortion Clients in Selected Hospitals of Madhya Pradesh, India. New Delhi: Ipas India; 2009.
Banerjee SK, Andersen KL, Buchanan RM, Warvadekar J. Woman-centred research on access to safe abortion services and implications for behavioural change communication interventions: A cross-sectional study of women in Bihar and Jharkhand, India. BMC Public Health 2012;12:1-13.
International Institute for Population Sciences and Population Council. Youth in India: Situation and Needs 2006-2007, National Report. Mumbai: IIPS; 2009.
Chatterjee C, Joardar GK, Mukherjee G, Chakraborty M. Septic abortions: A descriptive study in a teaching hospital at North Bengal, Darjeeling. Indian J Public Health 2007;51:193-4.
] [Full text]
Mishra A, Yadav A, Malik S, Purwar R, Kumari S. Over the counter sale of drugs for medical abortion- Knowledge, Attitude, and Practices of pharmacists of Delhi. India IJPR 2016;6:92-6.
Shrivastava N, Yadav S. The study of KAP of medical abortion in a tertiary centre. IOSR J Dent Med Sci 2015;14:1-4.
Pallikadavath S, Stones RW. Maternal and social factors associated with abortion in India: A population-based study. Int Fam Plann Perspect 2006;32:120-25.
Mistry R, Galal O, Lu M. Women's autonomy and pregnancy care in rural India: A contextual analysis. Soc Sci Med 2009;69:926-33.
Jha P, Kumar R, Vasa P, Dhingra N, Thiruchelvam D, Moineddin R, et al.
Low female[corrected]-to-male [corrected] sex ratio of children born in India: National survey of 1.1 million households. Lancet 2006;367:211-8.
Ministry of Health and Family Welfare, Government of India. National Family Health Survey 2015-16: State Fact Sheet West Bengal. International Institute for Population Science, Mumbai (India); 2016.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3]