Prevalence and Determinants of Physical Violence and Its Impact on Birth Outcomes During Pregnancy in India: Evidence From a Nationally Representative Survey

Abstract

Physical violence during pregnancy can have negative impact on health status of mother and fetus. Hence, the current study was done to determine the prevalence and determinants of physical violence and its impact on birth outcomes during pregnancy in India. We have analyzed the most recent National Family Health Survey 4 data (NFHS-4) gathered from Demographic Health Survey (DHS) program. Stratification (urban/rural) and clustering (villages/census enumeration blocks [CEBs]) in the sample design was accounted using svyset command. In total, 62,165 ever pregnant women aged 15 to 49 years were included. Prevalence of physical violence during pregnancy in India was 3.3%. Husband/partner (2.7%) was the person most commonly responsible. Women who were widowed/separated/divorced (aPR = 1.88), belonging to the poorest quantile (aPR = 2.32), women who were employed (aPR = 1.42), women in the Southern states (aPR = 3.24), and women whose husband/partner has lesser educational qualification (adjusted prevalence ratio [aPR] = 2.02) had significantly higher prevalence of physical violence during pregnancy (p < .001). Women who faced physical violence had significantly higher proportion of miscarriage (4.3%), abortion (3.3%), and stillbirth (1.1%) when compared with women who did not face any violence (4.1% had miscarriage, 1.8% had abortion, and 0.5% had stillbirth; p < .001). These findings show the importance of providing general supportive measures and strengthen the existing punitive legislations to prevent the violence during pregnancy.

Keywords

physical violence India pregnancy reproductive health maternal health

Introduction

Gender-based violence against women is acknowledged as a notable human rights violation and is deeply rooted with gender inequality worldwide (Kaur & Garg, 2008). Domestic violence is one of the most common forms of gender-based violence against women. Both developed and developing countries have been increasingly acknowledging and reporting the domestic violence against women (Bailey, 2010; Ress, 2005). World Health Organization (WHO) reported that one in three women throughout the world face physical or sexual violence by partner or non-partner (WHO, 2017). Prevalence is highest among the South East Asian region (SEAR) as 37.7% of women face domestic violence. India, one of the SEAR counties, contributes significantly to this burden (WHO, 2017). It is more serious if it is inflicted during the period of pregnancy. Domestic violence during pregnancy is defined as any violence inflicted by spouses or any other household members during pregnancy. A systematic review on domestic violence during pregnancy showed a prevalence of 28.4% for emotional abuse, 13.8% for physical abuse, and 8% for sexual abuse, respectively (James et al., 2013). A review conducted in Africa reported that domestic violence during pregnancy ranges from 23% to 40% (Shamu et al., 2011). Studies conducted in other parts of the world such as Pakistan, Turkey, Tanzania, Ethiopia, Nicaragua, and USA also reported higher prevalence of domestic violence during pregnancy ranging from 10% to 30%, whereas study conducted in Belgium (3.4%) and Vietnam (3.1%) reported lesser prevalence (Bailey, 2010; Belay et al., 2019; Deveci et al., 2007; Fikree et al., 2006; Mahenge et al., 2013; Roelens et al., 2008; Salazar et al., 2009; Tho Nhi et al., 2019).

In India, several small-scale studies (Chhabra, 2007, 2008; Purwar et al., 1999; Raj et al., 2011; Singh et al., 2008; Varma et al., 2007) and one large-scale study (Mahapatro et al., 2011) have been conducted to determine the prevalence of domestic violence during pregnancy. These studies showed that the prevalence of women experiencing physical, psychological, or sexual violence during pregnancy ranged from 13% to 60%. This variation might be due to the relatively smaller sample size and methodological limitations with these studies. However, the large-scale study addressing the violence during pregnancy was done a decade before. It is important to know the current situation to implement appropriate supportive measures. Measures taken till now in India were the categorization of domestic violence as a criminal offense under IPC 498-A (Golder, 2016). The civil protections were offered to the victims of domestic violence under the Protection of Women from Domestic Violence Act 2005 (PWDVA; Golder, 2016). It was primarily meant to offer protection to wives and female partners from facing domestic violence at the hands of husbands or male live in partners or relatives. However, despite these acts and measures, domestic violence against women continues to challenge and threaten the women empowerment in India.

Domestic violence among nonpregnant women is more likely to affect the sexual, reproductive, and mental health of women (WHO, 2017), whereas the domestic violence during pregnancy is associated with miscarriages, preterm labor, stillbirths, birth injuries, and increased risk of maternal, infant, and under-five mortality (Asling-Monemi et al., 2003; Bacchus et al., 2001; Reardon et al., 2002). In addition, women who suffer from domestic violence may not seek proper antenatal care (Gashaw et al., 2019). Studies have shown that domestic violence can make pregnant women 16% more likely to deliver low birthweight babies (WHO, 2017). Apart from low birth weight, evidences also show that women facing domestic violence has 2 times higher risk of having preterm delivery and 37% higher risk of having small for gestational age babies (Berhanie et al., 2019; Donovan et al., 2016).

However, there are three different forms of domestic violence (physical, emotional, and sexual) and each has their own impact and determinants. It has not been well explored in India and only limited knowledge is available in this regard (Koski et al., 2011). Violence can act through direct or indirect mechanisms. Physical violence acts in direct way and determinants will differ compared with the sexual or emotional violence which affects in an indirect way. Previous evidences have shown education, marital status, behavioral habits, socioeconomic status, and abuse before pregnancy as the significant determinants of physical violence during pregnancy (James et al., 2013; Naved & Persson, 2008; Pool et al., 2014). However, these determinants can be setting-specific, and finding out whether these determinants hold true for the culturally and socially diverse population like India needs to be seen. This is particularly important as the pregnant women who are likely to be struck or hit in the abdomen have higher risk of having preterm labor, fetal injury, or even fetal death (Asling-Monemi et al., 2003; Bacchus et al., 2001; Connolly et al., 1997; Petersen et al., 1997; Valladares et al., 2005). This is one of the reasons why the physical violence during pregnancy is an important public health issue though the prevalence is lesser compared with nonpregnant women (NFHS-4, International Institute for Population Sciences [IIPS] & ORC Macro, 2017).

Although longitudinal research is an ideal one to check this impact, it is important to provide at least a background knowledge and develop hypothesis for further research with the available nationwide survey data in India. Hence, this study was done to determine the prevalence and determinants of physical violence during pregnancy in India. We also explored the impact of physical violence on birth outcomes such as miscarriage, abortion, or stillbirth.

Method

Study Setting, Design, and Population

India is the second most populous country in the world with more than 1,210 million people officially residing in it (Chandramouli, 2011). This South Asian country is divided into 30 states and six union territories (UT). Each state and UT is further divided into districts. Districts are subdivided into census enumeration block (CEB) and wards in the urban area and villages/taluk in the rural area.

We have analyzed the most recent NFHS-4 (2015–2016) data gathered from Demographic Health Survey (DHS) program. Initially, a proposal was submitted to DHS to conduct a study on physical violence during pregnancy after which authorization to use data was obtained. Subsample of ever pregnant reproductive age group (15–49 years) women who participated in NFHS-4 survey was taken as study population for the current analysis.

Sample Size and Sampling Technique

The NFHS procedure included two-stage sampling approach for the selection of villages and CEBs, respectively, in rural and urban areas. The survey utilizes the census data of 2011 as the sampling frame. Probability proportional to size (PPS) sampling was used to select villages within each rural stratum. In each of the rural stratum, six equal substrata were made by crossing three substrata, each made based on number of households in each village, with two substrata, each made based on the percentage of population belonging to scheduled castes or scheduled tribes (SCs/STs). Primary sampling units (PSUs) were sorted within each explicit sampling stratum as per the literacy rate of women aged 6 or more years. Final sample PSUs were selected with PPS sampling. Similarly, in urban areas, CEBs were arranged as per the percentage of SC or ST population, and sample CEBs were selected by PPS sampling.

In all the selected urban and rural PSU, complete household listing and mapping operation were conducted before the survey. Selected PSUs with an approximate number of 300 households were divided into segments of 100 to 150 households. Two of these segments were selected randomly for survey by systematic sampling with probability proportional to segment size. Therefore, clusters in NFHS-4 can be either PSU or a segment of PSU. Twenty-two households were selected randomly by systematic sampling in all the selected urban and rural clusters during the second stage of sampling.

Development and validation of household questionnaire, data collection procedure, and data validation have been published as a separate report elsewhere (NFHS-4, IIPS & ORC Macro, 2017). To monitor the burden of physical violence, NFHS-3 and 4 included a domestic violence module consisting of questions related to physical, emotional, and sexual violence faced by women of reproductive age group. In NFHS-4, additional questions on physical violence faced during pregnancy were included to measure its burden (NFHS-4, IIPS & ORC Macro, 2017). Questions related to emotional and sexual violence during pregnancy was not gathered in the survey. Women of reproductive age group between 15 and 49 years were eligible to participate in the survey in all the selected households. Among females, 723,875 women of reproductive age group were identified to be eligible for the survey, out of which 699,686 women completed the questionnaire with response rate of 97%.

Data Variables and Data Sources

The first research question for the current study is to find the prevalence and determinants of physical violence during pregnancy in India. To find the determinants, independent variables included were sociodemographic characteristics such as age, education, occupation, wealth index, marital status, type of residence, religion, and husband education and occupation status and state. Dependent variable was the history of physical (physical) violence during pregnancy among ever pregnant reproductive age group women. The second research question is to explore the impact of physical violence on birth outcomes. Here, the independent variable is physical violence during pregnancy and the dependent variable is the adverse birth outcomes (miscarriage, abortion, or stillbirth)

Data collection was done in accordance with the WHO guidelines on the ethical collection of information related to physical violence WHO (2001). Only one woman per household was selected randomly for the module, and it was not implemented if privacy could not be maintained (i.e., some adult was trying to listen, or came into the room, or interfered in any other way). Questionnaires were translated to 18 languages to account for the linguistic diversity of the country. Practice interviews were conducted in local language for the data collectors during training. This is done as there might be circumstances where modification of the wordings in questions is required to fit the culture and local dialects without changing the meaning of question during this process. Field teams were arranged in such a manner that the data collector is working in the area in which their language is commonly spoken. Information on physical violence during any pregnancy was obtained using the following two questions:

“Has anyone ever hit, slapped, kicked or done anything else to hurt you physically while you were pregnant?”—Yes or No response

“If yes, who has done any of these things to physically hurt you while you were pregnant?”—Open ended question

Ethical Considerations

NFHS data set was available for download through data distribution system of DHS. All the data sets in DHS are accessible for free and downloaded for further use after registration process. Informed consent for all the respondents was obtained during the survey. The result obtained in the current study is based on the secondary analysis of existing NFHS survey data and does not contain patient name or any other identifiers.

Statistical Analysis

All the analyses were performed using STATA 14.2 (StataCorp, College Station, TX, USA). Questions related to physical violence were not asked to all the participants. In total, 83,397 women were questioned for the physical violence module and 79,729 completed the module. Out of these, 62,165 women were ever pregnant in their lifetime and they were included in the final analysis. Special weights were provided to adjust for the selection of one woman per household and also to ensure that the subsample selected for physical violence module was nationally representative. These weights were included in the analysis to account for the differential probabilities of participation and selection. Clustering at the level of PSUs (villages in rural and CEBs in urban) and stratification based on urban/rural area of residence in the sampling design were also accounted, after which svyset command was used to declare the NFHS data sets as survey type from two-stage cluster sampling: the selection of villages and census enumeration areas based on a probability proportionate to area size and random selection of households from the complete list of households within the selected villages and enumeration areas. Point estimates were reported with 95% confidence interval (CI). Poisson regression model was performed to obtain the prevalence ratio (PR) for the independent variables such as age, education, occupation, wealth index, marital status, type of residence, religion, state, and husband education and occupation (Barros & Hirakata, 2003). Here, Poisson regression was done instead of commonly used logistic regression as the logistic regression provides only odds ratio while PR is estimated directly by the Poisson regression (more applicable to be reported for cross-sectional surveys). In addition, it is difficult to interpret an odds ratio for a cross-sectional study as there is confusion between risk or odds leading to incorrect quantitative interpretation. Here, robust error variance correction was not done as the outcome is rare; hence, the variance obtained in the Poisson model is similar to the binomial model (ideally used to obtain PR, but convergence is a problem). The above-mentioned covariates were selected after going through the previous literature on similar studies and also after obtaining the public health expert’s opinion in this regard. Unadjusted and adjusted PR with 95% CI was reported. Variables with a p value of less than 0.05 were considered statistically significant and considered into the multivariable regression model. Impact of physical violence during pregnancy on birth outcomes was assessed using chi-square test and p < .05 is considered statistically significant.

Results

Sociodemographic characteristics of the study participants are described in Table 1. Majority (19.5%) of the study participants belonged to the age group between 25 and 29 years followed by 30 to 34 years (18.8%). Almost 95% were currently married and about 75.6% were Hindu by religion. Majority (42.5%) had education up to secondary level. Majority of the husband/partner of the participants (52.6%) also had education up to secondary level. Participants were almost equally distributed across all the five quantiles of wealth index. About one fourth of the participants were employed and more than two thirds belonged to rural area of residence. Almost 95% of the husband/partner of the participants was employed. Majority of the participants belonged to the Central part of India (23.4%) followed by North India (21.6%).

Table 1.

Sociodemographic Characteristics of Ever Pregnant Women Aged 15 to 49 Years Covered in NFHS-4 in India (N = 62,165).

Sociodemographic Characteristics	Frequency, N (Unweighted Proportion, %)	Weighted Proportion [95% CI]
Age category (in years)
15–19	930 (1.5)	1.8 [1.6, 2.0]
20–24	7,625 (12.3)	12.8 [12.4, 13.2]
25–29	13,195 (21.2)	19.5 [19.1, 19.9]
30–34	13,211 (21.2)	18.8 [18.4, 19.2]
35–39	11,106 (17.9)	17.6 [17.2, 18.0]
40–44	8,441 (13.6)	15.2 [14.8, 15.6]
45–49	7,657 (12.3)	14.3 [13.8, 14.7]
Marital status
Currently married	59,199 (95.2)	94.7 [94.4, 95.0]
Widowed/Separated/Divorced	2,966 (4.8)	5.3 [5.0, 5.5]
Religion
Hindu	46,585 (74.9)	75.6 [74.7, 76.5]
Muslim	8,164 (13.1)	14.0 [13.3, 14.7]
Christian	4,382 (7.1)	6.4 [5.9, 7.0]
Others^a	3,034 (4.9)	4.0 [3.6, 4.3]
Education status
No formal education	21,160 (34.0)	33.7 [33.0, 34.4]
Primary	9,238 (14.9)	14.5 [14.1, 14.9]
Secondary	26,317 (42.3)	42.5 [41.8, 43.1]
Higher	5,450 (8.8)	9.4 [8.9, 9.8]
Ethnicity
Caste	49,860 (80.2)	81.8 [81.0, 82.6]
Tribe	9,046 (14.6)	12.5 [11.8, 13.2]
No caste/tribe	2,788 (4.5)	5.0 [4.5, 5.5]
Don’t know	219 (0.3)	0.4 [0.3, 0.6]
Not recorded	252 (0.4)	0.3 [0.2, 0.4]
Wealth index
Poorest (I quintile)	12,104 (19.5)	17.5 [16.8, 18.1]
Poorer (II quintile)	13,250 (21.3)	19.7 [19.2, 20.3]
Middle (III quintile)	12,974 (20.9)	20.4 [19.9, 20.9]
Richer (IV quintile)	12,376 (19.9)	20.7 [20.1, 21.3]
Richest (V quintile)	11,461 (18.4)	21.7 [20.8, 22.6]
Current occupation
Unemployed	46,289 (74.5)	75.0 [74.4, 75.6]
Employed	15,876 (25.5)	25.0 [24.3, 25.6]
Residence
Urban	18,284 (29.4)	33.0 [31.7, 34.3]
Rural	43,881 (70.6)	67.0 [65.7, 68.2]
State
North	13,340 (21.5)	21.6 [20.5, 22.6]
Central	14,068 (22.6)	23.4 [22.4, 24.4]
East	10,909 (17.5)	17.2 [16.3, 18.2]
Northeast	8,307 (13.4)	12.8 [12.0, 13.6]
West	6,249 (10.0)	10.1 [9.4, 11.0]
South	9,292 (15.0)	14.9 [13.9, 15.9]
Husband’s education status
No formal education	12,160 (19.6)	18.7 [18.2, 19.2]
Primary	9,410 (15.1)	14.9 [14.4, 15.3]
Secondary	32,576 (52.4)	52.6 [52.0, 53.2]
Higher	7,838 (12.6)	13.5 [13.0, 14.0]
Don’t know	181 (0.3)	0.3 [0.2, 0.4]
Husband’s occupation status
Unemployed	2,444 (3.9)	4.2 [3.9, 4.4]
Employed	59,066 (95.0)	94.7 [94.3, 94.9]
Don’t know	655 (1.1)	1.1 [1.0, 1.3]

Note. NFHS-4 = National Family Health Survey-4; CI = confidence interval.

Includes Sikh, Buddhist, Jain, Jewish, and others.

Table 2 shows the details related to physical violence during pregnancy among the study participants. Prevalence of physical violence during pregnancy among ever pregnant women aged 15 to 49 years in India was 3.3% (95% CI: 3.1%–3.5%). Husband/Partner (2.7%) was the person most commonly responsible for the physical violence during pregnancy.

Table 2.

Physical Violence During Pregnancy Among Ever Pregnant Women Aged 15 to 49 Years in India Covered in NFHS-4 (N = 62,165).

	Frequency (Unweighted Proportion, %)	Weighted Proportion [95% CI]
Physical Violence During Pregnancy
Someone hurt respondent during pregnancy	2,098 (3.4)	3.3 [3.1, 3.5]
No one hurt respondent during pregnancy	60,067 (96.6)	96.7 [96.4, 96.9]
Person responsible for physical violence during pregnancy
Husband/Partner	1,653 (2.7)	2.5 [2.4, 2.7]
Former Husband/Partner	195 (0.3)	0.4 [0.2, 0.4]
In-law	101 (0.2)	0.2 [0.1, 0.2]
Other relatives	84 (0.1)	0.1 [0.09, 0.2]
Father/Step-father	45 (0.07)	0.1 [0.06, 0.2]
Brother/Step-brother	34 (0.05)	0.06 [0.04, 0.09]
Family friend	33 (0.05)	0.05 [0.03, 0.07]
Current/Former boy friend	8 (0.01)	0.02 [0.008, 0.08]
Teacher	8 (0.01)	0.03 [0.009, 0.02]
Others^a	30 (0.05)	0.05 [0.04, 0.08]
Outcome of pregnancy among women who faced during physical violence (n = 2,098)
Miscarriage	108 (5.1)	4.3 [3.3, 5.6]
Abortion	73 (3.5)	3.3 [2.5, 4.4]
Stillbirth	25 (1.2)	1.1 [0.7, 1.8]
No response	1,892 (90.2)	91.2 [89.5, 92.6]

Note. NFHS-4 = National Family Health Survey-4; CI = confidence interval.

Includes Own friend, stranger, employer, police, and religious leader.

Determinants of physical violence during pregnancy among ever pregnant women aged 30 to 49 years in India are described in Table 3. There was no significant difference with respect to age group as the prevalence of physical violence was almost equal across the age groups. Women who were widowed/separated/divorced had higher chance of facing physical violence during pregnancy (aPR = 1.88, p < .001) when compared with those who were currently married. Women having no formal education had higher prevalence of physical violence during pregnancy when compared with women with higher educational qualification. However, it was not statistically significant in the adjusted analysis. Participants in the poorest quantile had 2.32 times more prevalence of physical violence during pregnancy (aPR = 2.32, p < .001) when compared with those in richest quantile. Women who were employed had higher proportion of participants facing physical violence during pregnancy (aPR = 1.42, p < .001) when compared with those who were unemployed. Women in the rural area faced more physical violence during pregnancy when compared with women living in urban areas. However, it was not statistically significant in the adjusted analysis (p = .07). Women in the Southern states have significantly higher prevalence of physical violence during pregnancy (aPR = 3.24, p < .001) when compared with women in North-Eastern states. Women whose husband/partner has lesser educational qualification had significantly higher prevalence of physical violence during pregnancy when compared with women with husband/partner having higher educational qualification and this was statistically significant (p < .001).

Table 3.

Determinants of Physical Violence During Pregnancy Among Ever Pregnant Women Aged 15 to 49 Years in India (N = 62,165).

Characteristics	Total	Physical Violence During Pregnancy (%)*	Unadjusted Prevalence Ratio [95% CI]	p Value	Adjusted Prevalence Ratio [95% CI]	p Value
Age category (in years)
15–19	930	4.0	Ref.	—	{Omitted from the model}
20–24	7,625	3.1	0.72 [0.36, 1.41]	.34
25–29	13,195	3.2	0.67 [0.34, 1.32]	.25
30–34	13,211	3.5	0.75 [0.39, 1.46]	.40
35–39	11,106	3.3	0.79 [0.40, 1.54]	.48
40–44	8,441	3.5	0.77 [0.39, 1.53]	.46
45–49	7,657	3.8	0.81 [0.41, 1.61]	.55
Marital status
Currently married	59,199	3.2	Ref.	—	Ref.	—
Widowed/Separated/Divorced	2,966	7.6	2.33 [1.95, 2.78]	<.001	1.88 [1.56, 2.25]	<.001
Religion
Hindu	46,585	3.6	1.47 [1.04, 1.94]	.02	{Omitted from the model}
Muslim	8,164	3.1	1.19 [0.84, 1.68]	0.32
Christian	4,382	2.6	1.06 [0.71, 1.57]	0.76
Others^a	3,034	2.4	Ref.	—
Education status
No formal education	21,260	4.5	2.81 [2.07, 3.82]	<.001	1.28 [0.88, 1.86]	.19
Primary	9,238	4.0	2.42 [1.73, 3.39]	<.001	1.29 [0.88, 1.88]	.19
Secondary	26,317	2.6	1.59 [1.16, 2.17]	.004	1.07 [0.76, 1.53]	.68
Higher	5,450	1.5	Ref.	—	Ref.	—
Wealth index
Poorest (I quintile)	12,104	5.2	3.07 [2.37, 3.97]	<.001	2.32 [1.67, 3.24]	<.001
Poorer (II quintile)	13,250	3.8	2.12 [1.62, 2.78]	<.001	1.75 [1.26, 2.43]	.001
Middle (III quintile)	12,974	3.6	2.05 [1.56, 2.69]	<.001	1.65 [1.20, 2.26]	.002
Richer (IV quintile)	12,376	2.5	1.42 [1.07, 1.88]	.01	1.18 [0.86, 1.60]	.29
Richest (V quintile)	11,461	1.7	Ref.	—	Ref.	—
Current occupation
Unemployed	46,289	2.82	Ref.	—	Ref.	—
Employed	15,876	5.0	1.76 [1.56, 1.97]	<.001	1.42 [1.26, 1.60]	<.001
Residence
Urban	18,284	3.0	Ref.	—	Ref.	—
Rural	43,881	3.5	1.20 [1.01, 1.44]	.03	0.83 [0.68, 1.01]	.07
State
North	13,340	2.0	1.07 [0.83, 1.39]	.58	1.38 [1.06, 1.79]	.01
Central	14,068	4.1	2.17 [1.73, 2.71]	<.001	2.06 [1.65, 2.60]	<.001
East	10,909	4.2	2.11 [1.65, 2.70]	<.001	1.90 [1.47, 2.45]	<.001
Northeast	8,307	1.9	Ref.	—	Ref.	—
West	6,249	2.7	1.29 [0.96, 1.72]	.08	1.49 [1.10, 1.99]	.008
South	9,292	5.1	2.81 [2.16, 3.66]	<.001	3.24 [2.50, 4.20]	<.001
Husband/Partner’s education level
No formal education	12,160	5.0	3.53 [2.57, 4.84]	<.001	1.74 [1.21, 2.50]	.003
Primary	9,410	4.5	1.21 [1.15, 1.27]	<.001	2.02 [1.40, 2.91]	<.001
Secondary	32,576	3.0	2.09 [1.53, 2.86]	<.001	1.58 [1.13, 2.21]	.007
Higher	7,838	1.2	Ref.	—	Ref.	—
Don’t know	181	4.4	4.62 [1.83, 11.64]	.001	2.57 [0.98, 6.76]	.06
Husband/Partner’s occupation status
Unemployed	2,444	4.1	1.26 [0.95, 1.68]	.11	{Omitted from the model}
Employed	59,066	3.3	Ref.	—
Don’t know	655	3.2	1.31 [1.22, 1.40]	.79

Note. CI = confidence interval; Ref. = reference category.

Includes Sikh, Buddhist, Jain, Jewish and others.

Weighted proportion.

Bold Values indicate significant determinants.

Age category, religion, husband/partber’s occupation status were omitted from the model as they were not statistically significant in the unadjusted model.

Coming to the impact of physical violence on birth outcomes, women who faced physical violence had significantly higher prevalence of adverse birth outcomes (4.3% had miscarriage, 3.3% had abortion, and 1.1% had stillbirth) compared with the women who did not face any violence (4.1% had miscarriage, 1.8% had abortion, and 0.5% had stillbirth) and this association was statistically significant.

Discussion

NFHS-4 data provide sufficient opportunity to study these factors as it has separate module for domestic violence. Our study uses these data to shed some light on the prevalence, determinants, and birth outcomes of physical violence during pregnancy in India. Prevalence of physical violence during pregnancy in India was 3.3%. Previous NFHS surveys (NFHS-2 and 3) did not have questions related to physical violence during pregnancy (NFHS-2, IIPS & ORC Macro, 2000; NFHS-3, IIPS & ORC Macro, 2007) Only nationally representative study related to physical violence during pregnancy was done more than a decade before (2004–2007). The findings in the current study were far lesser than the prevalence reported in that previous study where 26% of women during pregnancy face physical violence (Mahapatro et al., 2011). This shows that there has been an improvement in the attitude of husband/partner or other close relatives toward pregnant women over the past decade.

We found plenty of sociodemographic characteristics of women and their husband/partner influencing the physical violence during pregnancy. We did find a significant association between marital status and physical violence during pregnancy. Since majority of the previous studies were conducted among married women only, it is important to focus on widowed/separated/divorced women as they have significantly higher risk of facing violence as found in our study (Mahapatro et al., 2011; Purwar et al., 1999; Raj et al., 2011; Singh et al., 2008; Varma et al., 2007). We also found that women with higher educational qualification had lesser chance of facing violence during pregnancy. Similarly, women with husband/partner with higher educational attainment faced lesser violence during pregnancy. This shows the attitude of husband/partner toward their wife based on the educational status of their own or the women. However, in contrast, women who were employed faced more physical violence during pregnancy when compared with unemployed women.

Women in the poorest quantile had the highest risk of facing violence during pregnancy. This may be due to the level of frustration faced by the husband or other family members because of their own socioeconomic status and unable to meet their daily needs. State-wise distribution of point estimates showed that violence during pregnancy was least in North-eastern states (1.9%) followed by Northern states (2.0%) while it was highest among the Southern states (5.1%) followed by Eastern states (4.2%). This shows that more stringent legislative measures should be taken against violence during pregnancy by the Southern and Eastern states. Understanding the determinants alone is not enough as the potential impact of the violence has during pregnancy should be understood to create awareness specifically by telling why it is an important public health issue.

Hence, on exploring its impact on birth outcomes showed that women who faced violence during pregnancy had higher prevalence of miscarriage, abortion, and stillbirth compared with women who did not face any physical violence. Although the difference is minimal and attributability of the finding is questionable, this should serve as a hypothesis to conduct future longitudinal research. Furthermore, this finding was consistent with the previous evidences which showed that the pregnancy negatively impacts the health status of pregnant women resulting in death of the fetus (Asling-Monemi et al., 2003; Connolly et al., 1997; Mahapatro et al., 2011; Petersen et al., 1997; Valladares et al., 2005). We could not assess the impact of violence on mental health status or its influence on negative behavioral problems during pregnancy as that information was not collected during the survey. However, intersectional understanding of the factors responsible for physical violence during pregnancy and the impact it has on the birth outcomes will help as an initial boost to design a specific intervention package targeting the vulnerable population.

The study has following limitations. The cross-sectional nature of the survey makes it difficult to infer causal relationship between the exposure and outcome. The reported estimate may not be accurate and can be considered as minimum as there is chance of under reporting in spite of ensuring privacy during the interview. The NFHS-4 survey had questions related only to physical violence during pregnancy. Hence, we could not assess the burden of emotional or sexual violence during pregnancy. Future surveys (NFHS-5) should have questions related to other forms of domestic violence to ensure comprehensive assessment of this module.

Despite these limitations, this study results have several programmatic implications. Our study reports the reduction in the burden of physical violence during pregnancy compared with previous national survey. However, these findings should be interpreted in caution owing to the design of the current study and further longitudinal research in this regard might give an idea whether the years of legal actions and awareness programs have worked in favor of pregnant women or not. However, a stronger surveillance system including national surveys assessing all the three forms of violence (physical, emotional, sexual) during pregnancy is important in estimating the exact burden of domestic violence during pregnancy in India.

We also identified determinants of violence during pregnancy which can be utilized to identify the target groups who should be screened for violence during pregnancy during their visit to health care facilities. Antenatal visits provide the window of opportunity to screen for any forms of violence during pregnancy. Health care providers should make use of this opportunity to find the high-risk groups as identified in our study and refer to appropriate authorities. Further research on developing interventional package targeting the vulnerable groups identified in our study should be done.

Possibilities of utilizing the existing community women support groups such as Mahila Arogya Samithi or self-help groups to raise awareness, consequences of violence during pregnancy, and facilities available for management should be explored. Community health workers such as Accredited Social Health Activist (ASHA), Anganwadi workers, and Auxiliary Nurse Midwife (ANM) can play a crucial role as they are closer to the community than any other health care professionals in India. They can be provided training on screening, effective management, and referral of women facing violence to appropriate facilities. Coordinated efforts from the government, health sector, and community can take India a step closer to eradicating the act of violence during pregnancy.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Compliance With Ethical Standards

This study is compliant with ethical standards.

Ethical Statement

Ethical approval was not required (secondary data analysis).

Informed Consent

Informed consent was obtained during the survey.

ORCID iD

Yuvaraj Krishnamoorthy

Author Biographies

Yuvaraj Krishnamoorthy is a senior resident in the Department of Preventive and Social Medicine, JIPMER. His research focuses on answering important public health–related questions mainly on maternal and child health and noncommunicable diseases using data from national-level surveys. He is also interested in conducting qualitative research, systematic reviews, and economic evaluation for the assessment of health technologies.

Karthika Ganesh is a junior resident in the Department of Preventive and Social Medicine, JIPMER. Her research focuses on determining the burden of sexually transmitted infections among reproductive age group women in rural areas. She is also interested in performing public health research related to maternal and child health.

References

Asling-Monemi

Rodolfo

Mary

C. E.

Lars

A. P.

(2003). Violence against women increases the risk of infant and child mortality: A case-referent study in Nicaragua. Bulletin of the World Health Organization, 81(1), 10–16.

Bacchus

Bewley

Mczey

(2001). Domestic violence in pregnancy. Fetal and Maternal Medicine Review, 12, 249–271.

Bailey

B. A.

(2010). Partner violence during pregnancy: Prevalence, effects, screening, and management. International Journal of Women’s Health, 2, 183–197.

Barros

A. J.

Hirakata

V. N.

(2003). Alternatives for logistic regression in cross-sectional studies: An empirical comparison of models that directly estimate the prevalence ratio. BMC Medical Research Methodology, 3, Article 21.

Belay

Astatkie

Emmelin

Hinderaker

S. G.

(2019). Intimate partner violence and maternal depression during pregnancy: A community-based cross-sectional study in Ethiopia. PLoS ONE, 14(7), Article e0220003. https://doi.org/10.1371/journal.pone.0220003

Berhanie

Gebregziabher

Berihu

Gerezgiher

Kidane

(2019). Intimate partner violence during pregnancy and adverse birth outcomes: A case-control study. Reproductive Health, 16(1), Article 22.

Chandramouli

, Registrar General, & Census of India. (2011). Provisional population totals. Government of India.

Chhabra

(2007). Physical violence during pregnancy. Journal of Obstetrics and Gynaecology, 27, 460–463.

Chhabra

(2008). Sexual violence among pregnant women in India. Journal of Obstetrics and Gynaecology Research, 34, 238–241.

10.

Connolly

A. M.

Katz

V. L.

Bash

K. L.

McMahon

M. J.

Hansen

W. F.

(1997). Trauma and pregnancy. American Journal of Perinatology, 14(6), 331–336. https://doi.org/10.1055/s-2007-994155

11.

Deveci

S. E.

Acik

Gulbayrak

Tokdemir

Ayar

(2007). Prevalence of domestic violence during pregnancy in a Turkish community. Southeast Asian Journal of Tropical Medicine and Public Health, 38(4), 754–760.

12.

Donovan

B. M.

Spracklen

C. N.

Schweizer

M. L.

Ryckman

K. K.

Saftlas

A. F.

(2016). Intimate partner violence during pregnancy and the risk for adverse infant outcomes: A systematic review and meta-analysis. BJOG: An International Journal of Obstetrics and Gynaecology, 123(8), 1289–1299.

13.

Fikree

F. F.

Jafarey

S. N.

Korejo

Afshan

Durocher

J. M.

(2006). Original article intimate partner violence before and during pregnancy: Experiences of postpartum women in Karachi, Pakistan. The Journal of the Pakistan Medical Association, 56, 252–257.

14.

Gashaw

B. T.

Magnus

J. H.

Schei

(2019). Intimate partner violence and late entry into antenatal care in Ethiopia. Women Birth, 32(6), e530–e537.

15.

Golder

(2016). Measurement of domestic violence in NFHS surveys and some evidence. Oxfam India.

16.

International Institute for Population Sciences & ORC Macro. (2000). National Family Health Survey (NFHS-3), 1998-99: India. International Institute for Population Sciences.

17.

International Institute for Population Sciences & ORC Macro. (2007). National Family Health Survey (NFHS-3), 2005 -06: India. International Institute for Population Sciences.

18.

International Institute for Population Sciences & ORC Macro. (2017). National Family Health Survey (NFHS-4), 2015 -16: India. International Institute for Population Sciences.

19.

James

Brody

Hamilton

(2013). Risk factors for domestic violence during pregnancy: A meta-analytic review. Violence and Victims, 28, 359–380.

20.

Kaur

Garg

(2008). Addressing domestic violence against women: An unfinished agenda. Indian Journal of Community Medicine: Official Publication of Indian Association of Preventive & Social Medicine, 33(2), 73–76. https://doi.org/10.4103/0970-0218.40871

21.

Koski

A. D.

Stephenson

Koenig

M. R.

(2011). Physical violence by partner during pregnancy and use of prenatal care in rural India. Journal of Health, Population, and Nutrition, 29(3), 245–254. https://doi.org/10.3329/jhpn.v29i3.7872

22.

Mahapatro

Gupta

R. N.

Gupta

Kundu

A. S.

(2011). Domestic Violence During Pregnancy in India. Journal of Interpersonal Violence, 26(15), 2973–2990. https://doi.org/10.1177/0886260510390948

23.

Mahenge

Likindikoki

Stöckl

Mbwambo

(2013). Intimate partner violence during pregnancy and associated mental health symptoms among pregnant women in Tanzania: A cross-sectional study. International Journal of Obstetrics & Gynaecology, 120(8), 940–946. https://doi.org/10.1111/1471-0528.12185

24.

Naved

R. T.

Persson

L. A.

(2008). Factors associated with physical spousal abuse of women during pregnancy in Bangladesh. International Family Planning Perspectives, 34(2), 71–78.

25.

Petersen

Gazmararian

J. A.

Spitz

A. M.

Rowley

D. L.

Goodwin

M. M.

Saltzman

L. E.

Marks

J. S.

(1997). Violence and adverse pregnancy outcomes: A review of the literature and directions for future research. American Journal of Preventive Medicine, 13(5), 366–373.

26.

Pool

M. S.

Otupiri

Owusu-Dabo

de Jonge

Agyemang

(2014). Physical violence during pregnancy and pregnancy outcomes in Ghana. BMC Pregnancy Childbirth, 14, Article 71.

27.

Purwar

M. B.

Jeyaseelan

Varhadpande

Motghare

Pimplakute

(1999). Survey of physical abuse during pregnancy GMCH, Nagpur, India. Journal of Obstetrics and Gynaecology Research, 25, 165–171.

28.

Raj

Sabarwal

Decker

M. R.

Nair

Jethva

Krishnan

Donta

Saggurti

Silverman

J. G.

(2011). Abuse from in-laws during pregnancy and post-partum: Qualitative and quantitative findings from low-income mothers of infants in Mumbai, India. Maternal and Child Health Journal, 15, 700–712.

29.

Reardon

D. C.

Ney

P. G.

Scheuren

Cougle

Coleman

P. K.

Strahan

T. W.

(2002). Deaths associated with pregnancy outcome: A record linkage study of low income women. Southern Medical Journal, 95, 834–841.

30.

Ress

(2005). Pregnant women in poor countries face domestic violence. British Medical Journal (Clinical Research Ed.), 331(7527), Article 1228. https://doi.org/10.1136/bmj.331.7527.1228-f

31.

Roelens

Verstraelen

Van Egmond

Temmerman

(2008). Disclosure and health-seeking behaviour following intimate partner violence before and during pregnancy in Flanders, Belgium: A survey surveillance study. European Journal of Obstetrics & Gynecology and Reproductive Biology, 137(1), 37–42.

32.

Salazar

Valladares

Ohman

Högberg

(2009). Ending intimate partner violence after pregnancy: Findings from a community-based longitudinal study in Nicaragua. BMC Public Health, 9, Article 350. https://doi.org/10.1186/1471-2458-9-350

33.

Shamu

Abrahams

Temmerman

Musekiwa

Zarowsky

(2011). A systematic review of African studies on intimate partner violence against pregnant women: Prevalence and risk factors. PLoS ONE, 6(3), Article e17591. https://doi.org/10.1371/journal.pone.0017591

34.

Singh

Rohtagi

Soren

Shukla

Lindow

S. W.

(2008). The prevalence of domestic violence in antenatal attendee’s in a Delhi hospital. Journal of Obstetrics and Gynaecology, 28, 272–275.

35.

Tho Nhi

Hanh

Hinh

N. D.

Toan

N. V.

Gammeltoft

Rasch

Meyrowitsch

D. W.

(2019). Intimate partner violence among pregnant women and postpartum depression in Vietnam: A longitudinal study. Biomed Research International, 2019, Article 4717485. https://doi.org/10.1155/2019/4717485

36.

Valladares

Peña

Persson

L. A.

Högberg

(2005). Violence against pregnant women: Prevalence and characteristics. A population-based study in Nicaragua. BJOG: An International Journal of Obstetrics and Gynaecology, 112(9), 1243–1248.

37.

Varma

Chandra

P. S.

Thomas

Carey

M. P.

(2007). Intimate partner violence and sexual coercion among pregnant women in India: Relationship with depression and post-traumatic stress disorder. Journal of Affective Disorders, 102, 227–235.

38.

World Health Organization. (2001). Putting women first: Ethical and safety recommendations for research on domestic violence against women (No. WHO/FCH/GWH/01.1).

39.

World Health Organization. (2017). Violence against women. https://www.who.int/news-room/fact-sheets/detail/violence-against-women