A descriptive, cross-sectional, community based study was undertaken in a slum of Kolkata under ward 132, borough 14 of Kolkata Municipal Corporation (KMC) catering a population of around 35,000. Duration of study was 6 months i.e., from July to December, 2015. The study area was an urban slum purposively chosen for the research. All the ever married women of reproductive age group (15-49 years) not having any psychiatric or serious medical illness at present were included in the study. Subjects who did not give informed consent were excluded from the study.

The subjects were interviewed with a pre-designed, pre-tested semi-structured schedule adapted from N.F.H.S Woman’s Questionnaire [9] after taking informed consent. In case of non response from any women, information was taken from the next willing woman.

Institutional ethical clearance was obtained before the study. Assurance about the anonymity and non-disclosure of the details of interview to husband/guardian were ensured. The schedule was first prepared in English and translated into Bengali by a linguistic expert keeping semantic equivalence. To check the translation, it was retranslated back into English by two independent researchers who were unaware of the first English version. Face validity of each item of the questionnaire had been checked from previous researches [1,2] in presence of public health experts. They also decided the content validity of each domain (Cronbach’s alpha 0.85). To check the reliability, it had been pretested to see clarity, absence of any ambiguity, objectivity and simplicity. Necessary corrections were made and the corrected version of the schedule had been pilot tested among 50 participants of the neighbourhood slum that was not included in the study to get the final corrected schedule. The sample size was calculated by taking the confidence interval of 95%, absolute precision of 5% and non-responsiveness 10%. Therefore, the estimated sample size was 428. Each household of the study area was allotted a number and the samples were selected by using systematic random sampling technique. Data were entered in SPSS 20.0 version and analysed through descriptive and inferential statistics.

The current research was done after interviewing 430 ever married women in the reproductive age group (mean age – 30.58±9.97 years). Majority belonged to the age group of 15-24 years (36.7%). Among the study population, 76.7% were Hindu and 23.3% were Muslims. The mean spousal age difference and duration of marriage were 5.91±2.85 years and 13.19±10.19 years respectively. Regarding educational level, 31.9% of the subjects and 27% of the husbands were educated up to below primary level. Unemployment amongst husbands was 29.1%, whereas 34.9% of the subjects were employed. Majority belonged to joint family (79.5%) and lower middle and lower socio-economic class according to Modified B.G. Prasad scale [10] for the year 2016. Alcohol consumption in any form amongst the husbands was 74.9%. Social support was not found among 43.5% of the study population. Property in the form of land, money, ornament alone or in combination was found among 53.7% of the subjects.

In the present study, the overall prevalence of any form of domestic violence was found in 255 (59.3%) of ever married women. The perpetrators were mostly the in laws including the spouse (59.6%). Emotional violence was more common than physical violence. As far as types of violence were concerned, 84.3% had emotional violence during lifetime (throughout married life), whereas 65.9% were subjected to current (within last 12 months) emotional violence. Regarding physical violence 156 (61.2%) of the subjects experienced it during lifetime; whereas 37.2% of them suffered from current physical violence. Regarding sexual violence, 58.8% of the subjects experienced it during lifetime, whereas 33.4% of them suffered from current sexual violence [Table/Fig-1]. Slapping (73.7%), kicking (62.8%), pushing (57.1%) were the most common forms of physical violence, whereas, insulting (80.8%), humiliating (73%) and threatening (60.5%) were the most common forms of emotional violence.

The occurrence of the event was reported by only 33.1% of the victims either to legal authority or friends, relatives, neighbours. The most common reason for not reporting was fear of social stigma (in 43.7%). Only, 8.45% had reported the event to the police, the remaining majority of the population had only shared it with relatives or friends.

Bivariate analyses revealed multiple factors leading to domestic violence. The event was significantly higher in the age group more than 29 years where the spousal age difference was more than 6 years, poor educational background (up to primary level), unemployment of both the partners, with low socio economic status and among Muslims [Table/Fig-2]. Despite the above covariates some other risk factors came into light by studying the opinions of the subjects regarding the possible causes of their sufferings such as not cooking, not attending household properly, talking to others, alcohol addiction of the husband, not having male child, going out without permission, dowry and property issues, not having any social support, paranoid behaviour of the husband [Table/Fig-3].

Subsequently, multivariate logistic regression was applied to find out the association of domestic violence with the factors which were statistically significant in the bivariate analysis.

Multivariate analysis revealed violence was higher among women belonging to a family of lower SES with a male child preference, when the spousal age difference was more than six years, having low educational background of husband, unemployment amongst both the spouses, not attending household activities and going outside without permission [Table/Fig-4].

Domestic violence is multidimensional and multifactorial, ergo prevalent in all societies of the world. Risk factors vary with geographical boundaries and socio-demographic pattern. Dey F et al., showed a trend of ever increasing tendency of crime against women in Kolkata [11]. The prevalence of lifetime physical and emotional violence was found to be 35.7% and 64.9% respectively in a study conducted in Kerala [12].

A study by International Centre for Research on Women (ICRW) [13] covering rural and urban slum and non-slum areas in 7 cities in India found the prevalence of physical and emotional violence to be 40.3% and 43.5%, respectively. Ramirez JC et al., found prevalence of domestic violence among urban population in India as 57% [14], whereas, Jewkes R et al., showed its lifetime prevalence to be 24.6% in South Africa [15].

The prevalence of lifetime and current domestic violence was 56% and 27%, respectively, in a study conducted in slums of Bangalore [7], whereas, in urban slum of Nagpur [8] physical violence was found to be 66%.

A study in urban slum area of Kolkata by Sinha A et al., showed overall prevalence of domestic violence to be 54%, of which 41.9% suffered from both current and lifetime physical and psychological violence [2].

In comparison with the above studies the current study showed a relatively higher prevalence of lifetime physical (61.2%) and emotional violence (84.3%).

Dowry and low education were recognised risk factors of domestic violence in previous studies [8,12], though dowry was not found as a risk factor in the current study. Low levels of education, unemployment of both the partners, alcohol consumption by the husband, lower socio-economic status and absence of social support of the wife were significant factors contributing to domestic violence as shown by Sinha A et al., and Saradamoni K et al. [2,12].

Women in every sphere of life are facing violence not only by their spouse but also by members of paternal home, sometimes by friends, teachers and in their work place too. The current study had focussed mainly on spousal violence. Further studies would help in understanding the violence against women by others apart from the spouses. There is also a need for in depth qualitative research to find out all the hidden contributory factors of violence against women.

The present study could have been hindered by wilful falsification of data by some of the participants as well as inclusion of a single slum, that too without any proper sampling technique for which it might not represent the true scenario for the whole of Kolkata.

More than half of the study subjects experienced domestic violence of any kind throughout their married life with emotional violence being the most common one. The perpetrators were mostly the in laws and the spouse. The current study showed an inverse relation of domestic violence with education of husband, employment, socio-economic status and gender preference. The event was found to be higher among the women when the spousal age difference was more and who did not attend household activities and go outside household without prior permission from the husband.

The prevalence of domestic violence in the present study was relatively higher than other studies of this kind. The women in the slums were habituated with the violent behaviour of their husbands in day-to-day life and keeping silence on the issue. They were forced to sexual relationship as it was the only means of entertainment of their partners. They have well accepted this violence as a normal phenomenon in their regular life. Therefore, they ignore this sort of violent behaviour of their husbands, thus, leading to under-reporting of the event. This issue needs to be addressed urgently through continuous IEC, appropriate utilisation of the legislations, proper reporting of the events to a legal authority, women empowerment and change in the attitude of the perpetrators towards the subjects.