Domestic violence: prevalence and association with gynaecological symptoms


Dr S. W. Lindow, Women and Children's Hospital, Hull Royal Infirmary, Anlaby Road, Hull, HU3 2JZ, UK.


Objective  To determine the prevalence of domestic violence in a population of women attending a gynaecology outpatient clinic in the United Kingdom and also to investigate whether women who reported domestic violence were more likely to complain of certain gynaecological symptoms.

Design  Questionnaire survey.

Setting  A gynaecology outpatient clinic in a North of England Hospital.

Sample  Nine hundred and twenty consecutive clinic attenders.

Methods  Anonymous confidential questionnaire given to women.

Main outcome measures  Disclosure of a past history of domestic violence and gynaecological complaints.

Results  Nine hundred and twenty consecutive women were included and 825 questionnaires were returned (90% response rate). The prevalence of physical abuse was 21% (171/825). Thirty-four (4%) had experienced violence in the past year. Domestic violence is three times less common in women over 50 years old. Ex-husbands (32%) and ex-boyfriends (29%) were the main perpetrators. Forty-eight percent women who had experienced physical violence also had forced sexual activity. Of the 15 presenting symptoms reported by the women, lower abdominal pain, dysmenorrhoea, dyspareunia, smear abnormalities, cancer worries and bowel symptoms were significantly more common complaints in the group who reported domestic violence. The women with domestic violence also had significantly more consultations; however, the duration of their symptoms was not significantly different.

Conclusion  The prevalence of domestic violence in a cohort of women who attended the gynaecology outpatient clinic in a North of England Hospital was 21%. Women who are subjected to domestic violence tend to have more consultations and are more likely to complain of certain symptoms.


Domestic violence includes emotional, sexual and economic abuse as well as physical violence inflicted on a spouse or partner by the other.1,2 Worldwide, domestic violence has been reported to affect one in three women in their lifetime and one in nine women annually.3 In 2002, a WHO report on violence and health reported that nearly 70% of female murder victims were killed by their partners or former partners. In the British Crime Survey, 23% of women aged 16–59 have been physically assaulted by a current or former partner, and two women are killed every week.4

A study in general practice conducted in the Hackney district of London showed a prevalence of 41% (425/1035) with 17% experiencing violence within the past year.5 Apart from the physical injuries, intimate partner violence can result in chronic health problems.6–18 Persistent or recurrent gynaecological symptoms could be an important presentation of domestic violence.9,16,18 The victims of intimate partner violence are three times more likely to have gynaecological problems and the odds are higher with the severity of physical assault.9 A Canadian population-based study reported that on average, battered women saw a medical professional about three times more often than non-battered women.6

Domestic violence appears to be under-reported as women are reluctant to disclose experiences of violence at home. It has been shown that the use of a structured screen can improve the detection of domestic violence,19 and this has been welcomed by majority of women.20

With respect to referral gynaecological practice in the United Kingdom, it is not precisely known how many gynaecological problems have their origin in domestic violence. The aim of this study was to calculate the prevalence of domestic violence in gynaecology outpatient attenders and also to investigate whether certain gynaecological conditions are related to domestic violence by completion of a simple questionnaire.


The local ethics committee approved this study. All the women attending the gynaecology outpatient clinic over a period of six weeks (19 May 2003 to 27 June 2003) at Hull Royal Infirmary were offered to participate in the study. Nine hundred and twenty consecutive women were seen in the clinic during this period and an anonymous and confidential questionnaire was given to these women.

In the gynaecology clinic, all patients were taken to an enclosed room for weight estimation; partners or accompanying persons were not allowed to enter. A questionnaire and a closed reply box were already placed in the room. Women were under no pressure to complete the questionnaire and no checks were made to ensure it was completed. The women were requested to place a blank questionnaire in the box if they did not wish to participate and were asked to state any reasons for this decision. The questionnaire was confidential in that no accompanying persons were told that the questionnaire was offered or completed. The reply box was emptied at the end of each day.

The questionnaire used was a modified version of Abuse Assessment Screen19,21 with guidance from the Department of Health directive on domestic violence.2 Questions were also asked about the nature and duration of their symptoms and consultations.

Statistical analysis was performed using SPSS version 9.5.


Of the 920 consecutive women who attended the clinic during the study period, 95 did not fill the questionnaires (23 women did not have their reading glasses, 15 women had language difficulties and 57 did not state any reason), giving a response rate of 90%.

One hundred and seventy-one women (21%) stated that they had been physically assaulted in the past; 34 of these women (4%) had experienced violence in the past year. One hundred and eight women reported that they had suffered emotional violence from somebody close to them (24%). Domestic violence is three times less common in women in the 50–59 and the above 60 years old age group (χ2 for whole group = 28.1, P < 0.0001; Table 1).

Table 1.  Details of 820 women who filled the domestic violence questionnaire. Values are given as n (%).
 Physical violence (n= 171)No violence (n= 649)
  • *

    χ2= 15.5, P < 0.001.

  • **

    χ2= 28.1, P < 0.0001.

Marital status*
Single28 (16)93 (14)
Married63 (37)340 (52)
Divorced23 (13)48 (7)
Stable relationship53 (31)155 (24)
Not recorded4 (2)13 (2)
<206 (4)29 (4)
21–3044 (26)142 (22)
31–4067 (39)194 (30)
41–5039 (23)113 (17)
51–608 (5)109 (17)
>605 (3)61 (9)
Not recorded2 (1)1 (0.2)

Slaps, push, kick, punch and threats are the kinds of assaults commonly reported (Table 2). Forty-eight women (28%) had been to the hospital for their injuries and 16 women (9%) had suffered permanent injury. Thirty-four of these women (20%) had been physically attacked in the past year and one woman was assaulted 50 times and another one 20 times. Another woman stated that the number of attacks were ‘too many to remember’.

Table 2.  Details of domestic (physical) violence in 171 women. Values are given as n (%) or mean [range].
  • *

    Identified perpetrators: husband, ex-husband, boyfriend, ex-boyfriend, father, stepfather, brother, grandfather, step grandfather, uncle, family friends and strangers.

Ex-husband54 (32)
Ex-boyfriend50 (29)
Boyfriend26 (15)
Father20 (12)
Husband17 (10)
Mother6 (4)
Stranger2 (1)
Son1 (0.5)
Daughter2 (1)
Others10 (6)
Pattern of violence
Threat106 (62)
Slap/push133 (78)
Punch/kick108 (63)
Objects (knife, etc.)54 (32)
Fracture/burns29 (17)
Hospital admission48 (28)
Permanent Injury16 (9)
Forced sexual activity*82 (48)
No. of times in the past year1.1 [0–50]
Afraid of anyone listed30 (18)

Forty-eight percent women who had experienced physical violence also had forced sexual activity (Table 2).

Of the 15 presenting symptoms, the majority was not more common in the domestic violence group; however, there were significantly more women complaining of lower abdominal pain, dysmenorrhoea, dyspareunia, smear abnormalities, cancer worries and bowel symptoms in the group who reported domestic violence (Table 3).

Table 3.  Presenting symptoms in women with and without a history of domestic violence. Values are given as n (%). Women complained of more than one symptom in many cases.
 Domestic violence (n= 171)No domestic violence (n= 649)χ2Significance (P)
  • *

    χ2= 10.1, P= NS.

Termination of pregnancy24 (14)62 (10)2.9NS
Lower abdominal pain36 (21)86 (13)6.5<0.01
Abdominal discomfort24 (14)77 (12)0.6NS
Vaginal pain22 (13)57 (9)2.6NS
Dysmenorrhoea28 (16)69 (11)4.3<0.05
Dyspareunia29 (17)47 (7)15.2<0.001
Abnormal uterine bleeding51 (30)206 (32)0.2NS
Fertility problems17 (10)43 (7)2.2NS
Ovarian cyst11 (6)56 (9)0.9NS
Smear abnormality49 (29)122 (19)8.0<0.01
Contraceptive advice6 (4)12 (2)1.7NS
Worry about cancer14 (8)28 (4)4.2<0.05
Prolapse10 (6)41 (6)0.1NS
Stress incontinence10 (6)23 (4)1.9NS
Bowel symptoms13 (8)13 (2)13.8<0.001
Expectation of cure*
Definitely yes49 (29)128 (20)  
Probably yes30 (17)141 (22)  
Possibly yes23 (13)121 (19)  
Probably not27 (16)92 (14)  
Definitely not8 (5)17 (3)  
Not sure32 (19)124 (19)  
Not recorded2 (1)26 (4)  

The women with domestic violence also had significantly more consultations in the past year with their problems [median (range): 3.0 (0–20) vs 2.0 (0–20) number of consultations, Mann–Whitney U test 36,673, P < 0.0001]. The duration of the problem was not significantly different [median (range): 18 (0–240) vs 18 (0–360) months, Mann–Whitney U test 24,814, P= NS].


This study involved 920 consecutive women who attended the gynaecology outpatient clinic. To get a true reflection of the prevalence of domestic violence and to avoid potential retaliation by the partners, anonymity and confidentiality were maintained. Women were provided with a separate room to complete the questionnaire before their consultation; thus, the opportunity for waiting room discussions among women was minimised. This response rate (90%) indicates that women are comfortable in the conditions provided by the study about answering questions relating to domestic violence.

Although the detection rate would possibly increase with repeated interviews at multiple visits,20 the Abuse Assessment Screen has been shown to be effective in the past.19

Definitions of domestic violence can vary considerably and may be based on different personal relationships and different degrees or types of violence (physical, sexual, emotional, etc.), which affect the results of prevalence studies. The results obtained from prevalence studies are also dependent on data collection methods.

The prevalence of domestic violence in unselected gynaecology outpatient attenders has never been assessed in the United Kingdom before, and referral patterns may make this setting different from other medical practices. The prevalence of 21% is consistent with many other population-based or general practice studies, although this study was conducted in a hospital setting. Straus and Gelles22 have reported that 28% of respondents had experienced some kind of physical violence and over 12% had experienced severe violence at some point in their marriage.

Another US study23 estimated the lifetime prevalence of violent experiences and their relationship to health and use of health services in women aged 18–64 years. Over 4 of 10 women are likely to have experienced one or more forms of violence, including child abuse (17.8%), physical assault (19.1%), rape (20.4%) and intimate partner violence (34.5%). In the current study, we did not ask questions about child abuse.

There have been a number of studies in general practice settings. A study5 undertaken among 425 women attending general practice in Hackney, London, reported a prevalence of 41% with 17% experiencing violence within the past year. In the current study, the prevalence in the past year was 4%. Another Irish general practice-based study by anonymous questionnaire survey showed that 39% had experienced violent behaviour by a partner.24 General practice study from Israel showed a lifetime prevalence of ranging from 21% to 34% with 8% to 14% reporting abuse in the previous year.25

In the present study, main perpetrators were ex-husbands, ex-boyfriends, boyfriends and husbands. Fathers, mothers, sons, daughters, grandfathers and uncles were the other close relatives involved. In general, domestic violence is considered synonymous to intimate partner violence. On a broader basis, this report describes violence in the household, and because many young women (and their partners) still live with their parents or extended family, we believe that categorising these abuses under domestic violence would be the only way to address these issues.

In the current study, 48% of women who were physically abused reported of forced sexual activity. Ex-boyfriends, ex-husbands, boyfriends and husbands were the main perpetrators. Fathers, stepfathers, grandfathers, step-grandfathers, uncles, brothers, own friends, family friends and strangers were also mentioned in the long list. In a cross sectional household survey in one province in Zimbabwe, 26% of women who had been married reported being forced to have sex.26 In the Campbell27 series (79 battered women), 59.5% were sexually abused or raped by their partner.

In the Israeli study, the women at highest risk were older than 40 years, living alone and were uneducated.25 In our study, women over 50 years were three times less prone for physical violence (7%vs 22%).

In the study by Straus and Gelles,22 the rate of such violence was 53% among those ever divorced or separated, and 11% among those currently married or cohabitating for the first time. Hillard28 stated that the women assaulted are more likely to be divorced or separated. In the current study, married women were less likely to be subjected to abuse than single or divorced women (15%vs 23%vs 32%). Socio-economic or educational status was not included in our study.

Several studies showed that women with a recent history of domestic violence reported a poorer health status, poor quality of life and higher use of health services.7,8,10 This may result in long term consequences, and the victims may present with non-specific psychosomatic and psychosexual symptoms. These include headache, chest pain, hypertension, backache, pelvic pain and gastrointestinal symptoms such as loss of appetite, eating disorders or irritable bowel disorders. Gynaecological problems are shown to be the most consistent. The odds of having a gynaecological problem were three times greater than average for victims of spouse abuse.6 In the current study, we looked into 15 different symptoms reported by the women who filled the questionnaires (both abused and non-abused). Certain symptoms (lower abdominal pain, dysmenorrhoea, dyspareunia, smear abnormality, cancer worry and bowel symptoms) were significantly more frequent in women reporting domestic violence (Table 3).

In the present study, women requesting a termination of pregnancy were not more likely to have suffered from domestic violence (14%vs 10%), a finding contrary to Gazmararian et al.29 who found an incidence of domestic violence of 12.1% in women with unwanted pregnancies compared with 3.2% in women with intended pregnancies. Glander et al.30 found ‘relationship issues’ to be a primary reason in 16% of requests for pregnancy termination in abused women compared with 6.8% in non-abused women. The provision of termination of pregnancy services in the UK is different from that in the United States and there may be differences in the decision-making process for individual women. The duration of the symptoms was not significantly longer in abused women.

Women experiencing domestic violence access health services more frequently. A Canadian population-based study found that they were three times more likely to access emergency health services than women who had not experienced abuse.6 In the present study also, physically abused women saw their own doctor or the gynaecologist more often than non-abused women.

Domestic violence is a serious social problem. This can be the underlying cause of many repeated, non-specific symptoms. The rates of disclosure of abuse without direct questioning in healthcare setting are poor. It appears that routine screening by structured questionnaire or direct interview is the only way to elicit this problem.

Routine enquiry for domestic violence in maternity settings is acceptable to women if conducted in a safe, confidential environment by a trained health professional who should be empathic and non-judgemental.31,32

Screening by interview, not questionnaire, appears to be more effective. Training has been shown to overcome some of the other barriers to routine enquiry. Training and education of health professionals can be successful in increasing their awareness, knowledge and willingness to screen women for abuse.


Domestic violence was noted in 21% of women attending a gynaecological clinic. Women with a history of domestic violence were more likely to consult their doctors and more likely to complain of lower abdominal pain, dyspareunia, dysmenorrhoea and bowel symptoms, and have smear abnormalities and worry about cancer.

Accepted 3 April 2004