By Author
  By Title
  By Keywords

December 2018, Volume 68, Issue 12

Research Article

Ethnic background and alcohol use of the spouse emerge as major risk factors for domestic violence: An observational study from Turkey

Yusuf Adnan Guclu  ( Tepecik Education and Research Hospital Family Medicine Clinic )
Huseyin Can  ( Izmir Katip Celebi University, Faculty of Medicine, Department of Family Medicine, Izmir, )


Objective: To assess the prevalence and risk factors of domestic violence in a multi-cultural region of Turkey.
Methods: The cross-sectional study was conducted in Tepecik district of Izmir, Turkey, from February to June 2013, and comprised women aged 15 years or above. A 21-item questionnaire was used to collect data through interviews. The primary outcome variable was history of domestic violence.
Results: There were 602 subjects with a mean age of 42.8±14.3 years. The mean age of spouses was 45.5±13.4 and the mean marriage duration was 19.6±4.4 years. The prevalence of any kind of violence within the preceding month was reported by 355(59%) women. Alcohol use of spouse had an odds ratio of 2.7 compared with non-users (p=0.007). Compared with Turkish background, Kurds had a 3.2-fold risk of being exposed to violence within the preceding year (p=0.010).
Conclusion: Independent of other demographic factors, race and alcohol use of the spouse were important factors in domestic violence.
Keywords: Domestic violence, Women health, Ethnic origin, Alcohol use. (JPMA 68: 1782; 2018)


According to the World Health Organisation (WHO), violence is defined as the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community, that either results in or has a high likelihood of resulting in injury death, psychological harm, mal-development, or deprivation.1 Domestic violence has been defined as threat to members of the family by using either physical or psychological force.2 The typology of domestic violence can be further sub-grouped as forcing to have sexual intercourse or doing other activities, restriction from meeting family or friends, restraining from work or school, or insulting.3 Domestic violence is observed in all varieties of human societies independent of educational, religious, economic or cultural backgrounds.4 The proportion of everpartnered women who had ever experienced physical or sexual violence, or both, by an intimate partner in their lifetime, ranged from 15% to 71%, with most sites falling between 29% and 62%.1 Also in Turkey, violence against women is relatively common with lifetime physical or sexual abuse reported by women as 41.9%.5 Biological, psychological as well as social reasons have been blamed in the aetiology of domestic violence.6 The current study was planned to assess the prevalence and risk factors of domestic violence in a multi-cultural region of Turkey and to look for the distribution of risk factors in this specific population. Our primaryhypothesis was that there is no difference in exposure to domestic violence concerning demographic factors.

Subjects and Methods

The cross-sectional study was conducted in the Tepecikdistrict of Izmir, Turkey, from February to June 2013.Women aged 15 years or above were included in the study. An address-based simple random sampling method was utilised. The list of families living in thedistrict was obtained from the local governor (mukhtar). Two physicians, one male and onefemale, visited eachaddress to ensure privacy and ease of responding. All women in the household meeting the inclusion criteria were invited to participate. Sample size calculation was based on the main outcome variable which was history of domestic violence. We needed 298 cases to detect a difference in domestic violence history between five ethnic groups (degree of freedom= 4) with an alpha error of 5%, effect size of 0.25 (medium), and a power of 95%.7 After adjusting for cluster effect with a constant of 1.5, the sample size stood at 447. The primary outcome variable of the study was “history of domestic violence”. Other variables studied were age, age of spouse, educational status of self and spouse, marital status, age at first marriage, duration of marriage, number of children, marriage type, employment status of self and spouse, smoking status of self and spouse, alcohol use of self and spouse, gambling habits of self and spouse, family type, and ethnic origin. Questions regarding marriage and spouse were only put to the married participants. Domestic violence exposure was assessed using a 9-item tool developed by the Turkish Ministry of Health 5 Validated for Turkish, this tool evaluates domestic violence among women aged 15 years and above within the preceding one year and within the preceding one month. It asks eight questions about short (1-month) and long-term (1-year) psychological, physical, economic, and sexual violence, plus one question about fear from household members. We used relevant WHO reference1 in the development of the tool, which included English-Turkish and back translation. Using these variables we calculated two additional variables: "any violence within the last one year" and "any violence within the last one month". The questionnaire was in Turkish language. Participants with languagebarrier were excluded. Participants were approached in their houses. Data collection was performed via face-to-face interview using pre-tested questionnaire with 27 items. Each data collection process lasted around 15 minutes. Interviewers were trained for this purpose. Ethical approval was obtained from the Izmir Tepecik Research and Education Hospital Ethics Committee and verbal consent was taken from each participant before starting the interview. Descriptive analyses were expressed as frequenciesand percentages for categorical variables and mean ± standard deviation (SD) for numerical variables. Bivariate comparisons were performed using the Student\\\'s t-test and Mann Whitney U test, when necessary. Chi square analysis was also done. A binary logistic regression model was used to test the combined effects of the studied variables on the presence of any violence exposure. All the independent variables were entered into the model using a stepwise conditional method. Significance levelwas set at p<0.05.


Of the 687 subjects approached, 85(12.4%) were excluded for various reasons and the final sample stood at 602(87.5%) (Figure).

The mean age of the participants was 42.8±14.3 years. Mean age of spouses was 45.5±13.4. Mean marriage duration was 19.6±4.4 years. Mean age of marriage was 21.2±13.4 years, and the mean number of children was 2.3±1.3 (Table-1).

One-year domestic violence prevalence ranged from 72(12%) sexual to 304(50.5%) psychological. The prevalence of any violence in the preceding one month was 355(59%) (Table-2).

There were significant differences between some categorical variables and exposure to violence (p<0.05) (Table-3).

According to the logistic regression model, alcohol use of the husband andethnic origin were significant independent variablespre dicting any domestic violence with a Nagelkerke R square of 10.0% and an overall predictive percentage of 62.6%. Alcohol use of the spouse had an OR of 2.7 compared with non-users (p=0.007; 95% confidence interval [CI]: 1.3 - 5.5). Compared with Turkish background, Kurds had a 3.2 fold risk of being exposed to any violence within the preceding year (p=0.010; 95% CI: 1.3 - 7.8).


Domestic violence in the studied population was relatively high, ranging from 47.4% to 75.3% among different ethnic groups. Although there are many factors significantly associated with domestic violence, alcohol use of the spouse and ethnic origin were the two independent factors affecting exposure in our sample. Demographic findings of the studied sample are similar to those of the general adult women in Turkey.8 Concerning the thnic origins of the participants, the proportion of Kurds and other ethnic backgrounds are similar to the general Turkish population. However, Roman population in the Tepecik district is denser than most other regions.9 It is reported that violence has an increasing trend in the Turkish community.10 Even violence to healthcare workers has become prevalent.11 Researchers have developed scales to measure physician-related healthcare perceptions among Turks.12 We observed relatively high domestic violence figures in our sample. However, the world\\\'s scorecard in this field is not brilliant either. One WHO study has revealed prevalence ranging from 15% to 71%.1 Also, previous studies from Turkey have consistently shown high domestic violence figures in a range of 32.6% to 52%.13-15 Observing the rise in violence claims, the Turkish Ministry of Family and Social Policies has adopted an action plan to stop domestic violence against women16 with its aim defined as "to take necessary actions by enabling collaboration of all partners involved to stop any violence against women". Still a study among women in an eastern province of Turkey in 2015 showed prevalence as high as 89%.17 However, the method of this measurement isthrough a different scale, which reminds one of the wellknown importance of validated measurement tools.18 The effects of most of thestudied independent variables on domestic violence have been inv estigated before. Among United States women, significant racial and ethnic differences were observed in association with intimate partner violence.19 Many studies have demonstrated that domestic violence is more common among married women.14,15,20,21 Divorced women have higher prevalence of domestic violence,5 which suggests that partner abuse is among the reasons for divorce. The reason why married women are more exposed to violence compared to singles is indicative of the fact that spouses are the principal source of violence. It is a traditional knowledge that "domestic violence knows no boundaries". However, some populations are more prone to this malady than others. Raised educational levels make women less fearless, and so does economic
independence.5,13 Also, substance abuse, including tobacco and alcohol use, is related to higher risks of practising violence. 1-3 The number of traditional extended families is decreasing with time.22 Still one of every five families in our sample was of extended type. Although we would expect that the wisdom and experience of extended family should be guarding against violence, this appears not to be the case. Also, literature has proposed a similar relationship.2 Probably there are confounding factors such as education affecting this finding. Findings of the regression model, demonstrating that race and alcohol are the two main independent risk factors for domestic violence, supports this judgement.Failing to take precautions on time to prevent domestic violence may result in damage to the whole society.23 To achieve international goals against domestic violence, a multinational and inter-sectoral collaboration is vital.Informing and educating services for avoiding domestic viole nce must be planned and implemented. The gap between societies with regard to educational status, economic independence and women rights must narrow down. The method of data collection was a limitation of the current study. Responses might have changed if we had conducted data collection confidentially without the presence of other household members. However, we should only expect higherrates of domestic violence in response to direct questioning.24 Another issue to mention is the time gap between data collection and manuscript preparation, which was due to health problems of the primary author. As hypothesised, demographicfactors were found to play asignificant ro le in domestic violence. However, most important to no te are the effects of race and alcohol use, which independent from other variables interfere with the outcome. Although alcohol use is a well-known risk factor for domestic violence, race should be taken into consideration as well in the efforts to combat this never vanishing ailment. Domestic violence does not know any boundaries and it seems not to be declining despite all the efforts given by
official and civil initiatives.


Independent of other demographic factors, race and alcohol use of the spouse were found to be important factors in domestic violence, which is a still highly prevalent health problem in Turkey.

Disclaimer: None.
Conflict of Interest: None.
Source of Funding: None.


1. World Health Organization. WHO Multi-country Study on Women’s
Health and Domestic Initial results on prevalence. [Online]. 2005 [Cited 2018 Feb 23]. Available from: URL:
2. Page AZ, İnce M. [Aile ici siddet konusunda bir derleme (A Review of Domestic Violence)]. Turk Psikol Yazilari 2008; 11: 81–94.
3. Kalokhe AS, Potdar RR, Stephenson R, Dunkle KL, Paranjape A, Del Rio C, et al. How well does the world health organization definition of domestic violence work for India? PLoS One 2015; 10: 1–16.
4. Ibiloglu AO. Aile Ici Siddet- Domestic Violence. Psikiyatr Guncel Yaklasimlar-Current Approaches Psychiatry Psikiyatr Guncel Yaklasimlar 2012; 4: 204–22.
5. Turkish Ministry of Family and Social Policies [Domestic Violence Against Women in Turkey] [Online] 2009 [Cited 2018 Feb 23]. Available from: URL:
6. George DT, Umhau JC, Phillips MJ, Emmela D, Ragan PW, Shoaf SE, et al. Serotonin, testosterone and alcohol in the etiology of domestic violence. Psychiatry Res 2001; 104: 27–37.
7. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007; 39: 175–91.
8. Turkish Statistical Institute. Population Statistics. [Online] [Cited 2018 Feb 23]. Available from: URL:
9. Mutlu S. Population of Turkey by ethnic groups and provinces. New
Perspect Turk 1995;12:33–60.
10. Yazici M. The Relationship Between Social Change With Form of Violence: USA / Europe - Turkey Comparison. Electron J Soc Sci 2013; 12: 350–69.
11. Erkol H, Gokdoǧan MR, Erkol Z, Boz B. Aggression and violence towards health care providers - A problem in Turkey? J Forensic Leg Med 2007; 14: 423–8.
12. Heybet M, Tekin O, Kahveci R, Heybet E, Yarloglues G, Sencan I, et al. [Development of a Scale for Measuring Physician Perception: Physician Related Health Care Perception Scale]. Konuralp Med J 2016; 8.
13. Mayda AS, Dilek A. Domestic violence against 116 Turkish housewives:
a field study. Women Health 2005; 40: 95–108.
14. Ergin N, Bayram N, Alper Z, Selimoglu K, Bilgel N. Domestic violence: a tragedy behind the doors. Women Health 2006; 42: 35–41.
15. Kocacik F, Dogan O. Domestic violence against women in Sivas, Turkey: survey study. Croat Med J 2006; 47: 742–9.
16. Turkish Ministry of Family and Social Policies. 2012-2015 National Action Plan to Combat Against Violence to Women. [Online] 2012
[Cited 2018 feb 23]. Available from: URL:
17. Kivrak Y, Gey N, Kivrak HA, Kokacya MH, Copoglu US, Ari M. Partner violence against women, childhood trauma, depression and quality of life: A population based-study. Anadolu Psikiyatr Derg 2015; 16: 314–22.
18. Evans M, Gregory A, Feder G, Howarth E, Hegarty K. “Even ‘Daily’ is Not Enough”: How Well Do We Measure Domestic Violence and Abuse?-A Think-Aloud Study of a Commonly Used Self-Report Scale. Violence Victim 2015; 31: 3–26.
19. Masho SW, Rozario S, Walker D, Cha S. Racial Differences and the Role of Marital Status in the Association Between Intimate Partner Violence and Unintended Pregnancy. J Interpers Violence 2018; 33: 3162-85.
20. Jewkes R, Levin J, Penn-Kekana L. Risk factors for domestic violence: Findings from a South African cross-sectional study. Soc Sci Med 2002;55: 1603–17.
21. Ellsberg M, Pena R, Herrera A, Liljestrand J, Winkvist A. Candies in hell:
Women’s experiences of violence in Nicaragua. Soc Sci Med 2000; 51: 1595–610.
22. Turkish Family with Statistics. [Online] 2013 [Cited 2018 Feb 23].
Available from: URL:
23. Wolfe D a, Jaffe PG. Emerging strategies in the prevention of domestic
violence. Future Child 1999; 9: 133–44.
24. Gazmararian J a, Lazorick S, Spitz AM, Ballard TJ, Saltzman LE, Marks JS.
Prevalence of violence against pregnant women. JAMA 1996;275: 1915–20. 

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: