Prevalence and factors associated with psychological violence against married women in Thiruvananthapuram district, Kerala

INTRODUCTION

Violence against women (VAW) can occur in many forms over the world. Throughout the life span, female gender is prone to face one or the other forms of violence. Gender inequality manifests as female foeticide, female infanticide, abuse of the girl child, domestic violence in the family, mental torture, physical violence, social harassment etc. Domestic violence including partner and non-partner violence, comprises ‘physical, psychological, sexual and economic abuse’. Domestic violence is a global issue affecting 30% of the population.¹ The burden in our country is higher than the global prevalence data. Various national studies show that, up to 70 per cent of women have experienced physical and/or sexual violence from an intimate partner in their lifetime.² The burden in India was 37.2% and 31.1% from study conducted by National Family and Health Survey in 2005-06 and 2015-16 respectively.³

Domestic violence is traumatic and incapacitating. Female partners often feel they have nowhere and nobody to help, especially in societies where it is not easy for women to leave their husbands and live alone. In most cultures, women used to bear violence against them in calm for fear of adverse repercussions. The position of women in India is not equal to that of men in terms of access, contribution and reward. Neglected and abused women live under constant fear, threat and humiliation even in her shelter which actually should be a source of cherish and emotional bonding. Family is the basic unit of society. When family becomes a source of inequality, exploitation, and violence, it can create personality disorders and psycho-somatic problems among the victims.

Eliminating all forms of violence against women and girls (VAWG) was adopted as a target 5 on gender equality and empowerment of women in 2015 sustainable development goals (SDG). Studies are essential for building evidence-based prevention and monitoring progress to SDG. Further, data will help to plan about the future allocation of resources and the implementation of preventive strategies.

Among the sub domains of domestic violence, psychological violence constitutes a big share. Multisite household survey report in 2000 showed 65.1% whereas 2001 study showed 64.9% had experienced psychological violence in Thiruvananthapuram district.^4,5 There is no recent published data of psychological violence in the study setting.

Assessing the prevalence will give a picture of this public health problem prevailing in the community and also help to compare the present burden with the previous studies done in the district. Assessing the factors associated with psychological violence also is needed for taking effective control and preventive measures. The measures taken will finally improve the health outcomes not only of the females but also of the whole community and future generations. Hence this study measured the prevalence of psychological violence and identified associated factors, among married women aged 18-55 years in Thiruvananthapuram district.

METHODS

Study setting, subjects and sampling

A community based cross-sectional study was conducted among ever married women aged 18-55 years, in Thiruvananthapuram district between January and May 2017. Sample size calculated was 270. A study done in 2000 by INCLEN has showed a prevalence of psychological violence in Thiruvananthapuram as 43.5%. Thiruvananthapuram district was the study setting. It has 6 taluks namely Thiruvananthapuram, Neyyattinkara, Chirayinkeezhu, Nedumangadu, Varkala and Kattakkada with 120 villages.⁶ Samples were taken from 1 urban village and 5 rural villages to cover all the taluks. Urban village was taken from the Thiruvananthapuram taluk. Multistage cluster sampling was done. Each village from the six taluks of the district were the sampling units. 45 samples from each cluster was studied. When more than one eligible subject was present in a house one participant was selected randomly. Subjects who were not willing to give consent, for whom privacy could not be ensured and those who have had any serious medical or psychiatric illness were excluded.

Data collection

Tool used was a semi structured questionnaire adapted from the WHO multi country study on domestic violence and modified for local contexts to collect socio demographic features, household details, marital factors, habits of the partner, family environment of the respondent. Of the 11 items of the 20-item domestic violence questionnaire (DVQ) which was validated in Malayalam (local language) was used for measuring psychological violence.⁷ Each house selected was visited by the investigator and enquired the details given in the questionnaire by conducting face to face interview. To get community support ASHA (accredited social health activists) workers were also included when required in the data collection process. All participants were interviewed in the Malayalam language. Respondents were informed the purpose and objectives of the study and informed consent obtained from them. The participants were informed that the inclusion in the study will be voluntary and confidentiality was assured.

Study variables

The exposure variables collected were i) Socio demographic features of both respondents and their partners which included age in completed years, spousal age difference, education, spousal educational difference, employment, income per month, religion and place as rural/urban; ii) Habit of the partner were collected like alcoholic consumption(categorised as: never drink, sometimes drink, very often drink and always drink, reported based on the perception of respondents), cigarette smoking (categorised as: ever smoked or not), any history of anti-psychotic drug intake by husband if the respondent knows, husband’s extra marital affair as perceived by the respondent, husband reaching late at home (reported based on the perception of respondents as daily late, frequently late, occasionally late and rarely late), husband’s habit of joining for meals with the partner and whether husband gave expected care during pregnancy, if applicable; iii) Household details collected were house condition (categorised as: pucca, semipucca and katcha) based on the quality of materials used to make roof, floor and exterior wall), house typology as individual/flat/slums, household structure as nuclear/ extended, total number of members living in the home, total number of bedrooms in the home and overcrowding index which was computed by dividing the number of members by number of bedrooms and a value more than 2 was considered as high and up to 2 as low; iv) Marital factors studied were age at marriage, number of living children, marital status (categorised as: married, divorced or separated and widowed), type of marriage (categorised as: marriage without the consent of parents (love marriage) and marriage with the consent of parents (arranged marriage), number of marriages respondents had, consanguineous marriage or not, marital duration (categorised as: 0-5 years, 5-10 years and >10 years), respondents’ willingness to get married, husband’s satisfaction with marriage (based on respondent’s perception), in laws’ satisfaction with marriage (based on respondent’s perception), and whether respondent has received any pre-marital counselling or not; v) Family environment details collected were the frequency of contact with own parents or family of origin (categorised as: <once a week and ≥once a week), support from own family in post marital life, possession of any property by respondent, witnessed noticeable quarrel between parents of the respondents and partners and about the decision maker of family affairs (categorised as: husband, in laws, husband and wife and wife).

The main outcome variable was life time psychological violence. Psychological domain has multiple items to measure the violence. In this study scoring of 1 to 8 was given for each item. 1 to 6 scores measure the presence of current violence for the past 12 months. As score increases from 1 to 6, the severity of violence also increases. Score of 7 measures past history of violence, whereas a score of 8 measures absence of violence which was coded as 0. Score from 1 to 7 indicates presence of life time psychological violence and was assigned a code of 1. Finally, prevalence of outcome variable was measured using code ‘0’ and ‘1’.

Psychological violence was measured using eleven questions. The questions used were: i) Did your husband limit you to see or interact with your friends? ii) Did your husband restrict your freedom to see or cooperate with your relatives? iii) Did your husband feel angry or suspicion while you talk to some other men? iv) Did your husband purposely ignore you without any sex? v) Did your husband threaten to hurt or harm you? vi) Did your husband humiliate you in front of others? vii) Did your husband accuse you of being unfaithful? viii) Did your husband treat you like a servant? ix) Did your husband exclude you from decision making? x) Did your husband keep away from home without informing you? xi) Did your husband keep silent with his relatives while they insult you?

Data analysis

All data were entered into the Microsoft excel sheet and analysed using Statistical Package for Social Sciences (SPSS) trial version. The categorical variables have been summarised as frequencies and proportions. The quantitative variables have been summarized as mean and standard deviation. Bivariable analysis using Pearson chi-square test was performed for all the categorical variables. If any of the cells in contingency table had expected values less than 5, Fishers exact test was used, instead of Chi square test. The odds ratio and its 95% CI were used as a measure of strength of association. 95% CI was also computed for the outcome proportions using Winpepi software and taking into account the cluster effect. Binary logistic regression modelling was performed to find the predictors of outcome. P value <0.05 was used to indicate statistical significance. Binary logistic regression was used to predict lifetime psychological violence which was the outcome. A significance level less than 0.05 was used for assigning co-variates to the regression model. Backward Conditional method of regression was employed for discarding non significant exposure variables.

Ethical considerations

Protocol was prepared and sent to institutional research and ethical committees of Government Medical College, Thiruvananthapuram. Research and ethical committees had cleared the protocol. Informed consent from participants was sought for study. Confidentiality was maintained throughout study. Minimum requirements for asking about partner violence like protocol, questionnaire, privacy, and confidentiality were ensured.⁸

RESULTS

Socio demographic factors of respondents and partners

Mean (SD) age of respondents was 40.75 years (9.0) and partners was 45.07 years (9.7). Most of the respondents (45.8%) were 5-10 years younger than their partners. Majority of the respondents and their partners (26.7% each) had degree as their educational qualification. 137 out of 270 respondents (50.7%) showed same level of education as that of their partners. 51.9% of the respondents were unemployed; out of the employed 30.7% were in the government sector. Among the partners 33% were employed in private sector and 3% were unemployed. 50.7% of the females had no income whereas 15 respondents (5.6%) had monthly income of rupees above 50000. 35.6% male partners were earning monthly income of rupees 5000-20000. More than half of the respondents [139 (51.5%)] belonged to Hindu. 83.3% belonged to rural area and 16.7% belonged to urban area.

Habit of partner

Most of the respondents [165 (61.1%)] reported their partners had never drank alcohol. More than two third partners were nonsmokers. 76 out of 270 (28.1%) partners were smokers and 194 (71.9%) were nonsmokers. 7 out of 270 (2.6%) respondents said their husbands had history of intake of anti-psychotic drug intake. 7 out 270 (2.6%) respondents said their husbands were having extramarital relations. 5 (1.9%) participants said their husbands were always late at home, 43 (15.9%) said it was frequent, 69 (25.5%) said it was occasional and 153 (56.6%) said it was rare. 184 (68.1%) responded that their partners would join meals and 86 (31.9%) responded that their partners were not used to join for meals. 190 (70.4%) said they received the expected care during pregnancy from partners; 69 (25.6%) did not get the expected care.

Household details

Mean (SD) number of members living in a house was 4.32 (1.6). It varies from a minimum of 1 to a maximum of 11. The distribution of respondents’ house condition, typology, structure and overcrowding index is as shown in Table 1.

Marital details

The distribution of variables such as marital status of respondent, type of marriage, number of marriages of respondent, consanguineous marriage, marital duration, willingness to marriage, husbands’ and in laws satisfaction and pre-marital counselling is as shown in Table 2.

Family environment

235 (87%) study subjects used to communicate at least once a week with their own parents or family of origin whereas 35 (13%) for less than once a week. 215 (79.6%) opined they would have the support of their own family members at needy times; 52 (19.3%) would not have the support whereas 3 (1.1%) were not sure about it. 165 (61.1%) respondents were possessing any kind of property whereas 105 (38.9%) did have no property. 195 (72.2%) of the participants told they had not witnessed noticeable quarrel between own parents whereas 69 (25.6%) had experienced. 172 (63.7%) of the participants told they had not witnessed noticeable quarrel between partners’ parents whereas 16 (5.9%) had experienced. More than half [142 (52.6%)] of the family affairs were controlled by husbands alone whereas 89 (32.9%) by both husband and wife, 25 (9.2%) by in laws and 14 (5.1%) by wives alone.

Psychological violence

Prevalence of psychological violence in the life time obtained was (68/270) 25.2% (95% CI 20.2-30.6). The item wise prevalence of different categories of psychological violence is as shown in Figure 1.

Bi variable analysis of factors associated with psychological violence

The main outcome was to assess the life time prevalence of psychological violence. Secondary case control analysis was done to find out the factors associated with prevalence of psychological violence. The variables tested were socio demographic variables, habit of the partner, household details, marital history, and family environment. The protective factors which were found significant include habit of the partner like joining for meals (OR=0.44), care given during pregnancy (OR=0.18) and house hold factors such as good housing condition (OR=0.47), marital factors like respondent’s willingness to marriage (OR=0.39), husband’s satisfaction with marriage (OR=0.04) and family environment such as support from own family (OR=0.32). The risk factors which were found significant include poor education of respondent and partner with an odds ratio of 1.8 and 2 respectively, habit of the partner like alcoholism (OR=2.7), smoking (OR=2), extramarital affairs (OR=7.8), coming late at home (OR=3.2), house hold factor like nuclear family with an odds ratio of 2.2. Others included marital factors like separated/widowed (OR=4.9), in laws’ satisfaction with marriage (OR=1.8) and family environment such as less communication with own family (OR 4.5), own parents and partner’s parents quarrel with an odds ratio of 2 and 5.6 respectively. The results of bivariable analysis is shown in Tables 3-7.

$ items are detailed in methodology.

*P value<0.05 is considered significant #chi square test

*P value<0.05 is considered significant #chi square test

*P value<0.05 is considered significant #chi square test

*P value<0.05 is considered significant #chi square test

*P value<0.05 is considered significant #chi square test

Multivariable analysis to find out the predictors of lifetime psychological violence

Maximum value of Nagelkerke R square with minimum number of variables and significance of the model in the Chi square table were the criteria used for finalizing the model. The model was found to be significant with a p value of <0.001. The variables found to be significant predictors of psychological violence were care obtained during pregnancy, household structure, in laws’ satisfaction with marriage, frequency of communication with own family, and partner’s parents quarrel. The model could explain 67.8% of the variability seen in life time psychological violence (Nagelkerke R square =0.678). The results of multivariable analysis to find out the predictors of life time psychological violence is as shown in Table 8.

DISCUSSION

The study was undertaken to find out the prevalence of life time psychological violence among ever married women of Thiruvananthapuram district. A secondary case control analysis was done to study the factors associated with it which includes variables related to socio demography, household details, marital history, habit of the partner and family environment. The prevalence of life time psychological violence obtained was 25.2% (95% CI 20.2-30.6%). The most common form of psychological violence was treating as servant (15.9%).

In United Nations multi country study the reported psychological violence ranges 4.1%-27.7%.⁹ Systematic review of Indian studies shows the median and range of lifetime estimates of psychological abuse was 22% and 2-99% respectively.¹⁰ In an Indian study insulting, threatening and withholding or delaying food were the most common forms of psychological violence.¹¹

As in most other studies, poor education of both respondents and partners were associated with violence among intimate partners.¹² Among the habit of partners alcoholism, smoking and having extramarital relations found to be a major risk factor for violence. Victims of violence in Kerala showed alcoholism of the male partner as the most common cause.¹³ The main reason reported for lack of trust between partners was suspected or actual infidelity and when women question them with suspicions of infidelity, this provokes their male partners and triggers violence.¹⁴

Nuclear family seems to be a risk factor of psychological violence. It may be because some type of violence can be prevented by living with a joint family whereas in some cases joint family itself perpetrates violence.

Separated and widowed women had higher risk of victimization whereas factors such as respondent's willingness for marriage and husband's satisfaction with marriage were came as a protective factor. Though in law's satisfaction with marriage found as a risk factor, on regression it came as one of the protective factor. There was no association between love (marriage without parental consent)/arranged marriage with violence but in one study love marriage and non-registration of marriage were significantly associated with violence.¹² It can be due to the inadequate sample size to study the associated factors.

Similar to current result, one study proved history of father ever beat mother was associated with a 92% chance of violence and witnessing such quarrels of parents is also significantly associated with violence in the future as obtained from one study.^15,16

The variations in factors contributing to psychological violence might be due to the differences in the study design, settings and importantly none of them specifically looked for psychological domain of domestic violence.

CONCLUSION

Abuse in the form of psychological violence is still prevailing in the district but lesser than the previous studies. Of the psychological violence the most common form was treating as servant. Psychological violence is mainly affected by multiple components like socio demographic elements, habit of the partner, family environment, marital and household factors. Most of the independent factors are modifiable by interventions aimed at individual, family and community levels, suggesting its preventable nature. The study provides rationale for nurturing friendly family atmosphere and improving familial bond as the core strategy in the prevention of psychological violence.

ACKNOWLEDGEMENTS

I am deeply indebted to Dr. Indu P. V., Additional Professor, Department of Psychiatry, Government Medical College Kozhikkode for her constant support for my endeavour. Also thankful to Dr. Devraj R., Assistant Professor, Department of Community Medicine, Government Medical College Idukki, for his valuable guidance.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. World Health Organization. Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence. Available from: http://apps.who.int/iris/bitstream/10665/85239/1/9789241564625_eng.pdf?ua=1. Accessed on 30 October 2017.
2. 2. Facts and figures: Ending violence against women. UN Women. Available from: http://www.unwomen.org/en/what-we-do/ending-violence-against-women/facts-and-figures. Accessed on 3 November 2017.
3. 3. India Fact Sheet. National Family Health Survey 2015-2016. Available from: http://rchiips.org/NFHS/pdf/NFHS4/India.pdf. Accessed on 18 May 2020.
4. 4. Domestic Violence in India-3. A Summary Report of a Multi-Site Household Survey. Available from: https://www.icrw.org/wp-content/uploads/2016/10/Domestic-Violence-in-India-3-A-Summary-Report-of-a-Multi-Site-Household-Survey.pdf. Accessed on 18 May 2020.
5. 5. Panda PK. Domestic Violence against women in Kerala. Discussion Paper No. 86. Available from: http://www.cds.ac.in/krpcds/publication/downloads/86.pdf. Accessed on 30 October 2017.
6. 6. About District. Website of Thiruvananthapuram District - The city of the Holy Anantha. India Available at: https://trivandrum.nic.in/about-district/ Accessed on 23 November 2020.
7. 7. Indu PV, Remadevi S, Vidhukumar K, Anilkumar TV, Subha N. Development and validation of the domestic violence questionnaire in married women aged 18-55 years. Indian J Psychiatr. 2011;53:218-23.
8. 8. Responding to intimate partner violence and sexual violence against women. WHO clinical and policy guidelines. Available from: http://apps.who.int/iris/bitstream/10665/85240/1/9789241548595_eng.pdf?ua=1. Accessed on 30 October 2017.
9. 9. Jewkes R, Fulu E, Tabassam NR, Chirwa E, Dunkle K, Haardörfer R, et al. Women’s and men’s reports of past-year prevalence of intimate partner violence and rape and women’s risk factors for intimate partner violence: a multicountry cross-sectional study in Asia and the Pacific. PLoS Med. 2017;14(9).
10. 10. Kalokhe A, del Rio C, Dunkle K, Stephenson R, Metheny N, Paranjape A, et al. Domestic violence against women in India: A systematic review of a decade of quantitative studies. Glob J Public Health. 2017;12(4):498-513.
11. 11. Sinha A, Mallik S, Sanyal D, Dasgupta S, Pal D, Mukherjee A. Domestic violence among ever married women of reproductive age group in a slum area of Kolkata. Indian J Public Health. 2012;56(1):31.
12. 12. George J, Nair D, Premkumar NR, Saravanan N, Chinnakali P, Roy G. The prevalence of domestic violence and its associated factors among married women in a rural area of Puducherry, South India. J Fam Med Prim Care. 2016;5(3):672-6.
13. 13. A-Situational-Analysis-of-Domestic-Violence-Against-Women-in-Kerala.pdf. Available at: http://www.jeywin.com/wp-content/uploads/2009/12/A-Situational-Analysis-of-Domestic-Violence-Against-Women-in-Kerala.pdf: Accessed on 23 November 2017.
14. 14. Muthengi E, Gitau T, Austrian K. Is working risky or protective for married adolescent girls in urban slums in Kenya? Understanding the association between working status, savings and intimate-partner violence. PLoS One. 2016;11(5).
15. 15. Owusu Adjah ES, Agbemafle I. Determinants of domestic violence against women in Ghana. BMC Public Health. 2016;16.
16. Stöckl H, March L, Pallitto C, Garcia-Moreno C. Intimate partner violence among adolescents and young women: prevalence and associated factors in nine countries: a cross-sectional study. BMC Public Health. 2014;14.