Abstract
Objectives. We examined the prevalence of intimate partner violence (IPV) and its association with social deprivation in England.
Methods. We used multivariable logistic regression to investigate IPV correlates among 21 226 men and women aged 16 to 59 years in the 2008 nationally representative cross-sectional British Crime Survey.
Results. Lifetime IPV was reported by 23.8% of women and 11.5% of men. Physical IPV was reported by 16.8% and 7.0%, respectively; emotional-only IPV was reported by 5.8% and 4.2%, respectively. After adjustment for demographic confounders, lifetime physical IPV experienced by women was associated with social housing tenure (odds ratio [OR] = 2.3; 95% confidence interval [CI] = 2.0, 2.7), low household income (OR = 2.2; 95% CI = 1.8, 2.7), poor educational attainment (OR = 1.2; 95% CI = 1.0, 1.5), low social class (OR = 1.5; 95% CI = 0.3, 1.7), and living in a multiply deprived area (OR = 1.4; 95% CI = 1.1, 1.7). Physical IPV experienced by men and emotional IPV experienced by either gender were generally not associated with deprivation factors.
Conclusions. Physical and emotional IPV are very common among adults in England. Emotional IPV prevention policies may be appropriate across the social spectrum; those for physical IPV should be particularly accessible to disadvantaged women.
The World Health Organization highlights intimate partner violence (IPV) as a pressing public health and human rights issue.1 The World Health Organization defines IPV as physical or sexual violence, emotional abuse, or controlling behavior by a current or former intimate partner.2 In nearly 50 populationwide surveys globally some 10% to 69% of women report having ever experienced physical abuse by an intimate partner.2 In the United Kingdom, a 2009 review found that the prevalence of lifetime IPV against women was 13% to 31% in community studies, and 13% to 41% in clinical populations.3 The health impact of IPV extends beyond mortality and direct injury4 to poor overall self-rated health, mental health problems, and gynecological and sexual health problems.5,6 The annual UK domestic violence cost was estimated at £ 16 billion in 2008.7
Often, IPV is seen in terms of physical or sexual violence perpetrated by men against women.1 As a consequence, most UK and international IPV prevention policies are targeted at women.1,8 However, some recent national surveys found a near-equal prevalence of physical IPV reported by men and women, mainly in North America and New Zealand, and there are calls for more services for male victims.9,10 Many have argued that this finding reflects measurement artifact, ignoring important differences in the nature and context of abuse.11 Further detailed examination of IPV as reported by both genders in national studies is needed to inform this debate.
Policies for prevention of IPV also tend to emphasize that all women are at risk, regardless of their socioeconomic background.1,12 This is perhaps in an effort to decrease stigma associated with IPV. However, many studies show that both male perpetrators and female victims of physical IPV are more likely to come from disadvantaged backgrounds.13–15 The association with social deprivation depends on the broader social context, with more empowered women being at higher risk in some settings.16 There is little evidence on the socioeconomic profile of male victims. Finally, although central to the World Health Organization’s definition of IPV, emotional abuse is not well-described in the current literature among either gender. This is an important knowledge gap, as emotional abuse has a significant public health impact that can be as great as that of physical abuse.17,18 Understanding how social deprivation is associated with different types of abuse in both sexes will inform the need for targeted versus universal interventions.
The British Crime Survey (BCS), a large national victimization survey in England and Wales, provides a detailed assessment of IPV. Home Office BCS reports examined associations between social deprivation and IPV victimization during the past year,19–21 but did not examine associations with lifetime IPV or IPV subtypes. To our knowledge there are no IPV studies using BCS data in the peer-reviewed scientific or public health literature, and only 1 UK national study using data other than the BCS, focusing on physical IPV.3,22 We used BCS data to describe the prevalence of both recent and lifetime IPV among men and women in England, and to explore whether different types of lifetime IPV were associated with social deprivation among either gender.
To facilitate interpretation of our empirical findings we generated hypotheses in advance, following our review of theoretical and empirical literature. We expected to find that (1) women would report a higher prevalence of all types of IPV than men, particularly severe, prolonged, and controlling types of abuse11,23; (2) social deprivation would be associated with being a victim of lifetime IPV in both men and women13,15; and (3) social deprivation would be more strongly associated with being a victim of physical than of emotional lifetime IPV.17,24
METHODS
We performed a secondary analysis of 2008–2009 BCS data. The BCS is a nationally representative cross-sectional survey of victimization in people aged 16 years or older living in private residential households in England and Wales. It was first carried out in 1982 and has been conducted annually since 2001. It aims to measure the experiences of being a victim of personal or property crime in the preceding 12 months. The survey has a complex design, with clustering, stratification, and unequal sampling probability, as detailed in Home Office technical reports.25 The sample size is calculated to include at least 1000 participants per police force area. Lay interviewers collected the data. One adult per household was randomly chosen to participate in the survey. Participants aged 16 to 59 years were invited to complete modules on “Domestic Violence, Sexual Victimisation and Stalking” using computer-assisted self-interviewing. Participants completed the questionnaires by themselves using interviewers’ laptops, after which their answers were hidden, to ensure privacy and confidentiality. If participants asked for interviewer help with these self-completion modules, the Domestic Violence module was omitted.
Eligibility criteria for this secondary analysis were (1) English resident (because sampling procedures and area deprivation measures differed in England and Wales) and (2) age 16 to 59 years (the IPV questionnaire was limited to this age group). Exclusion criteria were (1) data unavailable on the primary outcome or (2) data unavailable on the Primary Sampling Unit (because this was needed to carry out design-based analysis).
Measurement
We defined IPV as any emotional, physical, or sexual partner abuse. We defined “partner” as any current or former boyfriend or girlfriend, or husband or wife. We assessed IPV for 2 time periods: “lifetime” (since the age of 16 years) and “recent” (in the preceding 12 months). For the main analysis, we assessed the association of social deprivation with “any lifetime IPV” (defined as any emotional, physical, or sexual abuse since age 16 years). For the subgroup analyses, we assessed the association of social deprivation with 2 nonoverlapping IPV subtypes: “lifetime physical IPV” and “lifetime emotional-only IPV.” We defined “physical IPV” as being the victim of physical partner abuse, with or without coexisting emotional or sexual abuse. We defined “emotional-only” IPV as being the victim of emotional-only abuse, without coexisting physical or sexual abuse. Only a small number of IPV victims experienced sexual abuse without coexisting physical abuse (n = 171; 0.8% of study sample; 3.8% of those reporting lifetime IPV) and so we ran no subgroup analysis for this outcome.
We measured individual or household deprivation in 4 domains: (1) housing tenure, (2) equivalized household income, (3) respondent’s educational attainment, and (4) social class. We measured local area deprivation in 2 domains: (1) multiple deprivation and (2) crime deprivation. We operationalized each measure as an ordinal categorical variable:
Housing tenure, categorized as “owners,” “private renters” (i.e., renting from a private landlord), and “social renters” (i.e., renting a Council Housing property—a government-owned property with subsidized rents). In England, 65% of Council House tenants are in receipt of Housing Benefit, a top-up benefit for people on low income.
Household income: The BCS data provide total household income from all sources (including employment, pensions, and benefits). We calculated equivalized household income quintiles using the square root scale, whereby income is divided by the square root of the number of household residents.26
Educational attainment: respondent’s highest educational attainment, categorized as “degree or diploma,” “apprenticeship or advanced levels” (usually obtained at end of secondary school, at age 19), “General Certificate of Secondary Education (GCSE) or equivalent” (usually obtained during secondary school, at age 16), and “none.”
Social class: We used the 2001 National Statistics Socio-Economic Classification—an occupational classification of the household reference person—i.e., the person in whose name the property is owned or rented. We categorized this into “high social class”—comprising managerial, professional, and higher technical occupations, “intermediate social class”—comprising intermediate, small employers, and lower supervisory or technical occupations, and “low social class”—comprising semiroutine or routine occupations and the never employed.
Multiple area deprivation: We used the English Multiple Deprivation Index, categorized into quintiles. This index is based on deprivation in 7 domains: education or training, employment, income, health or disability, housing or services, living environment, and crime. It is commissioned by central government, and measured at the Lower Super Output Area level—a socially homogeneous area with a mean population of 1500.
Crime area deprivation: We used the English Crime Deprivation Index, categorized into quintiles. This index is based on levels of police-recorded crime and antisocial behavior, and is measured at Lower Super Output Area level.
Potential individual confounders were age, ethnicity, marital status, and living with children (as suggested by the literature). Ethnicity was self-defined as White, mixed, Asian or Asian British, Black or Black British, Chinese, or other. For this study we categorized ethnicity as White and “ethnic minority,” with the latter comprising the 5 non-White categories. Potential area confounders were region, urban or rural residence, and inner city or non–inner city residence.
Statistical Analysis
We carried out a design-based analysis, using the svy suite commands in Stata, version 11.0 (StataCorp, College Station, TX). We defined data as a 2-stage stratified sample (with individuals clustered within administratively defined medium-sized areas), weighted for unequal sampling probability and nonresponse. All descriptive analyses reported weighted point estimates and robust standard errors. Hypothesis tests were based on adjusted Pearson’s tests (for bivariate analyses), or adjusted Wald tests (for multivariate logistic regression analyses).
We stratified all analyses by gender. We first assessed the association between sociodemographic factors and IPV in unadjusted analyses. We used separate multivariable regression models to examine the association of each of the 6 deprivation measures with “any lifetime IPV,” and with “lifetime physical IPV” or “emotional-only” IPV. We adjusted the association between each individual deprivation measure and IPV for individual-level confounders but not for the other individual deprivation measures, because individual deprivation factors have complex temporal and causal interrelationships. We adjusted the association between each area deprivation measure and IPV for area-level confounders (region, urban or rural, inner city or non–inner city) and individual sociodemographic confounders, but not the other area deprivation measure. We used a final model retaining all sociodemographic factors independently associated with IPV to summarize demographic associations. For exposures with missing values in more than 5% of the sample, we estimated crude associations with the missing category to assess potential for bias. Because IPV was measured over the lifetime, with differing exposure times for different age groups, we carried out a sensitivity analysis by age group for important positive findings.
RESULTS
The BCS survey response rate was 77.1%. A total of 27 644 BCS participants were eligible for this study, of whom 21 226 (76.8%) were included, after exclusions because of refusals (n = 1468; 5.3%), IPV module omission (n = 3150; 11.4%), or missing data (n = 1800; 6.5%; Figure A, available as a supplement to the online version of this article at http://www.ajph.org). Study sample characteristics and inclusion rates are shown in Table 1. Inclusion rates were significantly lower in older people, ethnic minority groups, and more socially deprived individuals or households.
TABLE 1—
Study Sample Characteristics and Inclusion Rates: 2008 British Crime Survey
| Characteristic | Study Sample (n = 21 226), Weighted % (Unweighted No.) | Eligible Participants (n = 27 644) Included in the Study Samplea (n = 21 226), % |
| Gender | ||
| Female | 54.2 (11 503) | 76.6 |
| Male | 45.8 (9723) | 76.8 |
| Age,b y | ||
| 45–59 | 35.1 (7443) | 73.1 |
| 35–44 | 29.2 (6205) | 77.4 |
| 25–34 | 21.9 (4643) | 78.6 |
| 16–24 | 13.8 (2935) | 82.7 |
| Race/ethnicityb | ||
| White | 91.5 (19 431) | 78.4 |
| Racial/ethnic minority | 8.4 (1791) | 62.7 |
| Missing data | 0.02 (4) | … |
| Educationb | ||
| Degree or diploma | 40.2 (8541) | 82.5 |
| A level or apprenticeship | 19.1 (4047) | 79.9 |
| GCSE or equivalent | 26.8 (5684) | 80.1 |
| None | 11.4 (2419) | 57.4 |
| Missing | 2.5 (535) | … |
| Social class (HRP occupational grade)b | ||
| High | 43.9 (9313) | 82.9 |
| Intermediate | 30.5 (6470) | 76.4 |
| Low | 22.5 (4770) | 67.8 |
| Missing | 3.2 (673) | … |
| Household income quintilesb | ||
| Top quintile | 22.1 (4693) | 85.9 |
| 2 | 20.7 (4389) | 83.5 |
| 3 | 18.1 (3846) | 81.8 |
| 4 | 12.1 (2580) | 75.2 |
| Bottom quintile | 11.4 (2411) | 67.7 |
| Missing | 15.6 (3307) | 63.3 |
| Housing tenureb | ||
| Owner | 67.8 (14 387) | 79.4 |
| Private renter | 13.6 (2886) | 66.8 |
| Social renterc | 18.4 (3905) | 76.5 |
| Missing | 0.23 (48) | … |
| Multiple Deprivation Index quintileb | ||
| Least deprived | 21.5 (4568) | 80.6 |
| 2 | 22.3 (4731) | 80.2 |
| 3 | 21.0 (4468) | 78.6 |
| 4 | 18.5 (3918) | 75.0 |
| Most deprived | 16.7 (3541) | 68.5 |
Note. GCSE = general certificate of secondary education; HRP = household reference person.
Inclusion rate was calculated for missing values if these comprised 5% or more of eligible sample.
P for difference in inclusion rate < .001.
Renting a government-owned property with subsidized rents, usually reserved for low-income households on state benefits.
Prevalence, Nature, and Severity of Intimate Partner Violence
Lifetime IPV was reported by 23.8% of women and 11.5% of men (age- and ethnicity-adjusted odds ratio [AOR] = 2.4; 95% confidence interval [CI] = 2.2, 2.6), and IPV in the preceding 12 months was reported by 4.3% of women and 2.6% of men (AOR = 1.7; 95% CI = 1.4, 2.0). For subtypes of lifetime IPV, 18% of women and 8% of men reported emotional IPV; 17% of women and 7% of men reported physical IPV; and 4% of women and 0.5% of men reported sexual IPV. Women had greater relative odds for lifetime than for recent IPV (Table 2). The nature of abuse experienced differed by gender, with women having the greatest relative odds for more severe and controlling specific abusive experiences (Table 2).
TABLE 2—
Prevalence of and Odds Ratios for Lifetime and Past-Year Intimate Partner Violence, by Gender: 2008 British Crime Survey
| Lifetime IPV |
IPV in the Past Year |
|||||
| Abuse Type and Nature | Women (n = 11 503), % (No.) | Men (n = 9723), % (No.) | OR (95% CI) Adjusted for Age and Ethnicity | Women (n = 11 503), % (No.) | Men (n = 9723), % (No.) | OR (95% CI) Adjusted for Age and Ethnicity |
| Any IPV (emotional, physical, or sexual) | 23.8 (3159) | 11.5 (1254) | 2.4 (2.2, 2.6) | 4.3 (601) | 2.6 (273) | 1.7 (1.4, 2.0) |
| Emotional IPV | 18.2 (2484) | 7.5 (811) | 2.7 (2.5, 3.0) | 3.2 (444) | 1.7 (175) | 1.9 (1.5, 2.3) |
| Frightened you by threatening to hurt you or someone close to you | 10.3 (1431) | 1.1 (129) | 10.2 (8.2, 12.7) | 1.1 (169) | 0.2 (21) | 7.0 (4.0, 12.1) |
| Repeatedly belittled you to the extent that you felt worthless | 11.6 (1607) | 2.8 (326) | 4.6 (3.9, 5.3) | 1.7 (234) | 0.4 (59) | 4.0 (2.8, 5.8) |
| Prevented you from having your fair share of the household money | 6.1 (861) | 1.9 (220) | 3.3 (2.7, 4.0) | 0.8 (114) | 0.3 (31) | 2.5 (1.5, 4.2) |
| Stopped you from seeing friends and relatives | 8.5 (1209) | 5.0 (529) | 1.7 (1.5, 2.0) | 1.2 (175) | 1.3 (116) | 1.0 (0.7, 1.3) |
| Physical IPV | 16.8 (2250) | 7.0 (795) | 2.7 (2.4, 3.0) | 2.0 (301) | 1.3 (136) | 1.6 (1.3, 2.1) |
| Choked or tried to strangle you | 5.5 (571) | 0.5 (55) | 11.1 (7.8, 15.9) | 0.4 (64) | 0.1 (5) | 7.2 (2.5, 20.5) |
| Threatened to kill you | 3.6 (517) | 0.7 (79) | 5.2 (3.9, 7.0) | 0.3 (39) | 0.1 (12) | 2.0 (0.9, 4.4) |
| Pushed you, held you down, or slapped you | 13.7 (1834) | 3.1 (363) | 4.9 (4.3, 5.7) | 1.4 (216) | 0.5 (56) | 2.7 (1.8, 4.0) |
| Used some other kind of force against you | 5.0 (702) | 1.1 (133) | 4.7 (3.8, 6.0) | 0.4 (64) | 0.1 (17) | 4.2 (2.3, 7.7) |
| Threatened you with a weapon—for example, a stick or a knife | 2.6 (376) | 1.2 (144) | 2.2 (1.7, 2.7) | 0.1 (25) | 0.1 (13) | 1.1 (0.5, 2.5) |
| Used a weapon against you—for example, a stick or a knife | 1.5 (213) | 0.7 (93) | 2.0 (1.5, 2.6) | 0.1 (13) | 0.1 (7) | 1.0 (0.3, 3.5) |
| Kicked, bit, or hit you with a fist or something else, or threw something at you | 9.9 (1378) | 5.1 (580) | 2.0 (1.8, 2.3) | 0.9 (145) | 0.9 (98) | 1.0 (0.7, 1.4) |
| Sexual IPV | 4.3 (571) | 0.5 (48) | 8.7 (5.9, 12.9) | 0.4 (54) | 0.1 (15) | 2.9 (1.3, 6.2) |
| Penetrative (attempted or actual penetrative sexual assault) | 2.4 (321) | 0.2 (16) | 13.9 (7.1, 27.1) | 0.2 (22) | 0.1 (5) | 3.4 (0.7, 15.9) |
| Nonpenetrative (indecent exposure, unwanted sexual touching, sexual threats) | 3.7 (493) | 0.4 (37) | 9.5 (6.1, 14.8) | 0.2 (39) | 0.1 (12) | 2.0 (0.8, 4.6) |
Note. CI = confidence interval; IPV = intimate partner violence; OR = odds ratio. Abusive experiences presented in descending order of OR for lifetime IPV.
For those with a lifetime history of IPV, any physical IPV was reported by 71% of female and 60% of male victims, emotional-only IPV by 25% of female and 37% of male victims, and sexual abuse by 18% of female and 6% of male victims (Figure B, available as a supplement to the online version of this article at http://www.ajph.org). When we compared male and female IPV victims, women were significantly more likely to report all 3 abuse subtypes (11.8% vs 1.3%; P < .001), to report having suffered abuse for more than 5 years (16.4% vs 4.9%; P < .001), and to report mental health problems as a result of the abuse (36.6% vs 16.4%; P < .001; Table A, available as a supplement to the online version of this article at http://www.ajph.org).
Intimate Partner Violence and Social Deprivation
In women, and after adjustment for confounders, any lifetime IPV and physical IPV were associated with educational attainment, housing tenure, household income, social class, and multiple area deprivation, but not crime area deprivation (Tables 3 and 4, respectively). When we compared the most- and least-deprived categories, the strongest associations were found between physical IPV and (1) housing tenure (AOR = 2.3; 95% CI = 2.0, 2.7; for social renters compared with owners) and (2) household income (AOR = 2.2; 95% CI = 1.8, 2.7; for the bottom vs top income quintiles). Lifetime emotional-only IPV was not associated with any deprivation factors (Table 5). In men, and after adjustment for confounders, any lifetime IPV and emotional-only IPV were associated with social housing tenure only (Tables 3 and 5, respectively). Physical IPV was not associated with any deprivation factors (Table 4).
TABLE 3—
Any Lifetime Intimate Partner Violence, by Demographic and Social Deprivation Factors in Women and Men: 2008 British Crime Survey
| Women |
Men |
|||||||||
| Variable | Prevalence, % (No./Total No.) | Crude OR (95% CI) | P | AOR (95% CI)a | P | Prevalence, % (No./Total No.) | Crude OR (95% CI) | P | AOR (CI)a | P |
| Individual social deprivation factors | ||||||||||
| Housing tenancy | <.001 | <.001 | <.001 | <.001 | ||||||
| Owner (Ref) | 20.4 (1737/7672) | 1.00 | 1.00 | 10.1 (742/6715) | 1.00 | 1.00 | ||||
| Private renter | 26.0 (639/1966) | 1.37 (1.20, 1.57) | 1.55 (1.33, 1.80) | 14.0 (313/1939) | 1.45 (1.21, 1.75) | 1.45 (1.21, 1.75) | ||||
| Social renterb | 37.3 (777/1833) | 2.32 (2.03, 2.66) | 2.09 (1.80, 2.42) | 15.9 (198/1053) | 1.69 (1.37, 2.08) | 1.69 (1.37, 2.08) | ||||
| Household income quintiles | <.001 | <.001 | .4 | |||||||
| Top quintile (Ref) | 19.4 (465/2184) | 1.00 | 1.00 | 12.6 (340/2509) | 1.00 | … | ||||
| 2 | 24.8 (572/2205) | 1.37 (1.15, 1.63) | 1.38 (1.16, 1.65) | 12.9 (302/2184) | 1.03 (0.84, 1.28) | … | ||||
| 3 | 23.8 (545/2061) | 1.29 (1.09, 1.54) | 1.26 (1.06, 1.51) | 11.0 (211/1785) | 0.86 (0.69, 1.08) | … | ||||
| 4 | 27.9 (497/1545) | 1.61 (1.34, 1.93) | 1.44 (1.19, 1.74) | 11.2 (138/1035) | 0.88 (0.68, 1.14) | … | ||||
| Bottom quintile | 38.0 (731/1648) | 2.55 (2.14, 3.03) | 2.05 (1.71, 2.47) | 13.4 (135/763) | 1.08 (0.82, 1.42) | … | ||||
| Missing | 16.4 (349/1860) | 0.82 (0.67, 1.00) | … | 8.4 (128/1447) | 0.64 (0.49, 0.83) | … | ||||
| Social class | <.001 | <.001 | .43 | |||||||
| High (Ref) | 21.0 (1177/4959) | 1.00 | 1.00 | 11.2 (530/4354) | 1.00 | … | ||||
| Intermediate | 25.5 (950/3372) | 1.29 (1.14, 1.46) | 1.29 (1.14, 1.47) | 12.4 (423/3098) | 1.12 (0.94, 1.33) | … | ||||
| Low | 27.6 (921/2782) | 1.43 (1.26, 1.63) | 1.33 (1.16, 1.51) | 11.9 (277/1988) | 1.07 (0.88, 1.30) | … | ||||
| Educational attainment | <.01 | .02 | .1 | |||||||
| Degree or diploma (Ref) | 21.9 (1134/4597) | 1.00 | 1.00 | 11.3 (497/3944) | 1.00 | … | ||||
| A level | 23.9 (526/1814) | 1.12 (0.96, 1.30) | 1.22 (1.04, 1.42) | 11.0 (286/2233) | 0.97 (0.79, 1.18) | … | ||||
| GCSE or equivalent | 25.5 (1024/3463) | 1.22 (1.07, 1.38) | 1.19 (1.05, 1.36) | 13.2 (319/2221) | 1.19 (0.99, 1.44) | … | ||||
| None | 25.5 (396/1360) | 1.22 (1.04, 1.43) | 1.15 (0.97, 1.36) | 10.1 (118/1059) | 0.89 (0.67, 1.16) | … | ||||
| Area social deprivation factors | ||||||||||
| Multiple area deprivation quintilesc | <.001 | .03 | .48 | |||||||
| Least deprived (Ref) | 19.7 (551/2472) | 1.00 | 1.00 | 6.3 (155/2096) | 1.00 | … | ||||
| 2 | 21.3 (630/2517) | 1.10 (0.93, 1.31) | 1.09 (0.90, 1.31) | 6.8 (196/2214) | 1.1 (0.8, 1.4) | … | ||||
| 3 | 23.6 (652/2380) | 1.26 (1.06, 1.49) | 1.18 (0.98, 1.43) | 6.8 (166/2088) | 1.1 (0.8, 1.4) | … | ||||
| 4 | 27.0 (663/2103) | 1.51 (1.28, 1.78) | 1.35 (1.12, 1.64) | 6.8 (139/1815) | 1.1 (0.8, 1.5) | … | ||||
| Most deprived | 28.3 (663/2031) | 1.61 (1.36, 1.91) | 1.24 (1.01, 1.52) | 8.3 (139/1510) | 1.3 (1.0, 1.8) | … | ||||
| Crime area deprivation quintilesc | <.001 | .64 | .62 | |||||||
| Least deprived (Ref) | 21.7 (620/2597) | 1.00 | 1.00 | 7.2 (186/3018) | 1.00 | … | ||||
| 2 | 23.3 (628/2398) | 1.10 (0.94, 1.29) | 1.08 (0.91, 1.29) | 6.5 (176/2771) | 0.9 (0.7, 1.2) | … | ||||
| 3 | 22.2 (638/2381) | 1.03 (0.88, 1.20) | 1.01 (0.85, 1.21) | 6.8 (160/2602) | 0.9 (0.7, 1.2) | … | ||||
| 4 | 23.7 (619/2134) | 1.12 (0.95, 1.33) | 1.08 (0.89, 1.31) | 6.7 (130/2388) | 0.9 (0.7, 1.3) | … | ||||
| Most deprived | 28.5 (654/1993) | 1.44 (1.22, 1.69) | 1.14 (0.94, 1.39) | 7.8 (143/2372) | 1.1 (0.8, 1.5) | … | ||||
| Demographic potential confounders | ||||||||||
| Age, y | <.001 | <.001 | ||||||||
| 45–59 (Ref) | 23.6 (1026/3962) | 1.00 | … | 9.4 (381/3481) | 1.00 | … | ||||
| 35–44 | 28.0 (1043/3368) | 1.26 (1.11, 1.42) | … | 13.6 (443/2837) | 1.52 (1.27, 1.81) | … | ||||
| 25–34 | 24.9 (769/2621) | 1.08 (0.94, 1.24) | … | 13.5 (283/2022) | 1.50 (1.22, 1.85) | … | ||||
| 16–24 | 16.5 (321/1552) | 0.64 (0.53, 0.77) | … | 10.2 (147/1383) | 1.10 (0.86, 1.41) | … | ||||
| Race/ethnicity | <.001 | <.001 | ||||||||
| White British (Ref) | 24.7 (2974/10 556) | 1.00 | … | 12.3 (1195/8875) | 1.00 | … | ||||
| Not White British | 15.6 (184/946) | 0.56 (0.46, 0.69) | … | 5.8 (58/845) | 0.44 (0.31, 0.62) | … | ||||
| Marital status | <.001 | <.001 | ||||||||
| Married or cohabiting (Ref) | 20.1 (1372/6661) | 1.00 | … | 10.0 (620/5904) | 1.00 | … | ||||
| Single | 24.7 (961/3048) | 1.30 (1.15, 1.47) | … | 11.9 (383/2923) | 1.22 (1.02, 1.45) | … | ||||
| Separated, widowed, or divorced | 44.7 (826/1793) | 3.22 (2.82, 3.69) | … | 28.4 (251/894) | 3.58 (2.91, 4.40) | … | ||||
| Living with children | <.001 | .02 | ||||||||
| No (Ref) | 22.0 (1544/6247) | 1.00 | … | 12.2 (846/6218) | 1.00 | … | ||||
| Yes | 26.1 (1615/5256) | 1.25 (1.13, 1.38) | … | 10.4 (408/3505) | 0.84 (0.73, 0.97) | … | ||||
| Overall | 23.8 (3159/11 503) | 11.5 (1254/9723) | ||||||||
Note. AOR = adjusted odds ratio; CI = confidence interval; GCSE = general certificate of secondary education; IPV = intimate partner violence; OR = odds ratio.
OR for individual deprivation measures crudely associated with any lifetime IPV were adjusted for age, ethnicity, marital status, and living with children.
Renting a government-owned property with subsidized rents, usually reserved for low-income households on state benefits.
OR for area deprivation measures crudely associated with any lifetime IPV were adjusted for the factors in footnote “a,” and in addition for housing tenure, household income, social class, and educational attainment in women, and for housing tenure in men. There was no confounding by region, inner city, or rural residence.
TABLE 4—
Lifetime Physical Intimate Partner Violence by Demographic Social Deprivation Factors in Women and Men: 2008 British Crime Survey
| Women |
Men |
|||||||
| Variable | Prevalence, % (No./Total No.) | Crude OR (95% CI) | P | AOR (95% CI)a | P | Prevalence, % (n/N) | Crude OR (95% CI)b | P |
| Individual social deprivation factors | ||||||||
| Housing tenancy | <.001 | <.001 | .08 | |||||
| Owner (Ref) | 13.9 (1185/7672) | 1.0 | 1.0 | 6.4 (496/6715) | 1.0 | |||
| Private renter | 18.3 (463/1966) | 1.4 (1.2, 1.6) | 1.7 (1.4, 2.0) | 8.0 (185/1939) | 1.3 (1.0, 1.6) | |||
| Social renterc | 28.5 (596/1833) | 2.5 (2.1, 2.8) | 2.3 (2.0, 2.7) | 8.3 (113/1053) | 1.3 (1.0, 1.7) | |||
| Household income quintiles | <.001 | <.001 | .19 | |||||
| Top quintile (Ref) | 13.1 (312/2184) | 1.0 | 1.0 | 8.0 (235/2509) | 1.0 | |||
| 2 | 17.4 (405/2205) | 1.4 (1.1, 1.7) | 1.4 (1.2, 1.7) | 8.2 (193/2184) | 1.0 (0.8, 1.3) | |||
| 3 | 16.6 (383/2061) | 1.3 (1.1, 1.6) | 1.3 (1.1, 1.6) | 7.1 (138/1785) | 0.9 (0.7, 1.1) | |||
| 4 | 20.6 (365/1545) | 1.7 (1.4, 2.1) | 1.6 (1.3, 1.9) | 5.7 (73/1035) | 0.7 (0.5, 1.0) | |||
| Bottom quintile | 28.4 (540/1648) | 2.6 (2.2, 3.2) | 2.2 (1.8, 2.7) | 7.7 (83/763) | 1.0 (0.7, 1.3) | |||
| Missing | 11.1 (245/1860) | 0.8 (0.7, 1.0) | … | 4.5 (73/1447) | 0.5 (0.4, 0.8) | |||
| Social class | <.001 | <.001 | .20 | |||||
| High (Ref) | 14.3 (823/4959) | 1.0 | 1.0 | 7.2 (359/4354) | 1.0 | |||
| Intermediate | 17.7 (658/3372) | 1.3 (1.1, 1.5) | 1.3 (1.1, 1.5) | 7.6 (261/3098) | 1.1 (0.9, 1.3) | |||
| Low | 20.4 (685/2782) | 1.5 (1.3, 1.8) | 1.5 (1.3, 1.7) | 6.2 (162/1988) | 0.8 (0.7, 1.1) | |||
| Educational attainment | <.001 | <.001 | .11 | |||||
| Degree or diploma (Ref) | 15.1 (777/4597) | 1.0 | 1.0 | 7.3 (332/3944) | 1.0 | |||
| Apprenticeship or A level | 15.9 (361/1814) | 1.1 (0.9, 1.3) | 1.2 (1.0, 1.4) | 6.3 (181/2233) | 0.9 (0.7, 1.1) | |||
| GCSE or equivalent | 19.0 (772/3463) | 1.3 (1.2, 1.5) | 1.3 (1.1, 1.5) | 7.8 (197/2221) | 1.1 (0.9, 1.4) | |||
| None | 18.6 (291/1360) | 1.3 (1.1, 1.6) | 1.2 (1.0, 1.5) | 5.5 (68/1059) | 0.7 (0.5, 1.0) | |||
| Multiple area deprivation quintilesd | <.001 | .05 | .48 | |||||
| Least deprived (Ref) | 12.9 (363/2472) | 1.0 | 1.0 | 6.3 (155/2096) | 1.0 | |||
| 2 | 15.2 (441/2517) | 1.2 (1.0, 1.5) | 1.2 (1.0, 1.5) | 6.8 (196/2214) | 1.1 (0.8, 1.4) | |||
| 3 | 16.6 (454/2380) | 1.3 (1.1, 1.6) | 1.2 (1.0, 1.6) | 6.8 (166/2088) | 1.1 (0.8, 1.4) | |||
| 4 | 18.9 (480/2103) | 1.6 (1.3, 1.9) | 1.4 (1.1, 1.8) | 6.8 (139/1815) | 1.1 (0.8, 1.5) | |||
| Most deprived | 21.1 (512/2031) | 1.8 (1.5, 2.2) | 1.4 (1.1, 1.7) | 8.3 (139/1510) | 1.3 (1.0, 1.8) | |||
| Crime area deprivation quintilesd | <.001 | .11 | .62 | |||||
| Least deprived (Ref) | 14.1 (412/2597) | 1.0 | 1.0 | 7.2 (186/3018) | 1.0 | |||
| 2 | 16.3 (437/2398) | 1.2 (1.0, 1.4) | 1.2 (1.0, 1.5) | 6.5 (176/2771) | 0.9 (0.7, 1.2) | |||
| 3 | 15.5 (446/2381) | 1.1 (0.9, 1.3) | 1.1 (0.9, 1.3) | 6.8 (160/2602) | 0.9 (0.7, 1.2) | |||
| 4 | 17.4 (459/2134) | 1.3 (1.1, 1.5) | 1.2 (1.0, 1.5) | 6.7 (130/2388) | 0.9 (0.7, 1.3) | |||
| Most deprived | 21.3 (496/1993) | 1.6 (1.4, 2.0) | 1.3 (1.0, 1.6) | 7.8 (143/2372) | 1.1 (0.8, 1.5) | |||
| Demographic potential confounders | ||||||||
| Age | <.001 | <.001 | ||||||
| 45–59 (Ref) | 16.6 (709/3962) | 1.0 | … | 6.0 (251/3481) | 1.0 | |||
| 35–44 | 20.7 (783/3368) | 1.3 (1.1, 1.5) | … | 9.2 (304/2837) | 1.6 (1.3, 1.9) | |||
| 25–34 | 17.9 (553/2621) | 1.1 (0.9, 1.3) | … | 7.6 (172/2022) | 1.3 (1.0, 1.6) | |||
| 16–24 | 9.9 (205/1552) | 0.6 (0.4, 0.7) | … | 4.9 (68/1383) | 0.8 (0.6, 1.1) | |||
| Race/ethnicity | <.001 | <.001 | ||||||
| White British (Ref) | 17.5 (2117/10 556) | 1.0 | … | 7.4 (757/8875) | 1.0 | |||
| Not White British | 10.7 (132/946) | 0.6 (0.4, 0.7) | … | 3.9 (37/845) | 0.5 (0.3, 0.8) | |||
| Marital status | <.001 | <.001 | ||||||
| Married or cohabiting (Ref) | 14.4 (978/6661) | 1.0 | … | 6.1 (392/5904) | 1.0 | |||
| Single | 16.9 (691/3048) | 1.2 (1.1, 1.4) | … | 6.4 (227/2923) | 1.1 (0.8, 1.3) | |||
| Separated, widowed, or divorced | 31.9 (581/1793) | 2.8 (2.4, 3.2) | … | 20.1 (176/894) | 3.9 (3.0, 4.9) | |||
| Living with children | <.001 | .13 | ||||||
| No (Ref) | 15.2 (1068/6247) | 1.0 | … | 7.3 (528/6218) | 1.0 | |||
| Yes | 18.9 (1182/5256) | 1.3 (1.2, 1.5) | … | 6.4 (267/3505) | 0.9 (0.7, 1.0) | |||
| Overall | 16.8 (2250/11 503) | 7.0 (764/9723) | ||||||
Note. AOR = adjusted odds ratio; CI = confidence interval; GCSE = general certificate of secondary education; IPV = intimate partner violence; OR = odds ratio.
In women, OR for individual deprivation measures crudely associated with physical IPV were adjusted for age, ethnicity, marital status, and living with children.
In men, none of the deprivation factors were associated with physical IPV at the 5% significance level, so no adjusted ORs are presented. There was no evidence for any negative confounding using the adjustments in footnotes “a” and “d” (results not shown).
Renting a government-owned property with subsidized rents, usually reserved for low-income households on state benefits.
In women, OR for area deprivation measures crudely associated with physical IPV were adjusted for the factors in footnote “a,” and in addition for housing tenure, household income, social class, and educational attainment. There was no confounding by region, inner city, or rural residence.
TABLE 5—
Lifetime Emotional-Only Intimate Partner Violence: Prevalence, Crude Odds Ratio, and Adjusted Odds Ratio by Social Deprivation Factors in Women and Men in 2008 British Crime Survey
| Women |
Men |
|||||||||
| Variable | Prevalence, % (No./Total No.) | Crude OR (95% CI) | P | AOR (95% CI)a | P | Prevalence, % (No./Total No.) | Crude OR (95% CI) | P | AOR (95% CI)a | P |
| Individual social deprivation factors | ||||||||||
| Housing tenancy | .01 | .28 | <.001 | <.001 | ||||||
| Owner (Ref) | 5.5 (468/7672) | 1.0 | 1.0 | 3.3 (231/6715) | 1.0 | 1.0 | ||||
| Private renter | 5.7 (136/1966) | 1.1 (0.8, 1.4) | 1.0 (0.8, 1.3) | 5.7 (123/1939) | 1.8 (1.3, 2.3) | 1.6 (1.2, 2.2) | ||||
| Social renterb | 7.8 (160/1833) | 1.5 (1.1, 1.9) | 1.2 (0.9, 1.6) | 7.0 (79/1053) | 2.2 (1.6, 3.0) | 2.2 (1.6, 3.1) | ||||
| Household income quintiles | .04 | .5 | .38 | |||||||
| Top quintile (Ref) | 5.3 (127/2184) | 1.0 | 1.00 | 4.2 (99/2509) | 1.0 | … | ||||
| 2 | 6.3 (144/2205) | 1.2 (0.9, 1.6) | 1.2 (0.8, 1.6) | 4.4 (104/2184) | 1.1 (0.7, 1.5) | … | ||||
| 3 | 6.3 (143/2061) | 1.2 (0.9, 1.6) | 1.2 (0.9, 1.6) | 3.6 (68/1785) | 0.9 (0.6, 1.3) | … | ||||
| 4 | 6.0 (108/1545) | 1.1 (0.8, 1.5) | 1.0 (0.7, 1.4) | 5.0 (61/1035) | 1.2 (0.8, 1.8) | … | ||||
| Bottom quintile | 8.2 (161/1648) | 1.6 (1.2, 2.1) | 1.3 (0.9, 1.7) | 5.6 (50/763) | 1.3 (0.9, 2.1) | … | ||||
| Missing | 4.0 (81/1860) | 0.7 (0.5, 1.0) | 3.5 (51/1447) | 0.8 (0.5, 1.3) | … | |||||
| Social class | .68 | .04 | .07 | |||||||
| High (Ref) | 5.7 (298/4959) | 1.0 | … | 3.7 (160/4354) | 1.0 | 1.0 | ||||
| Intermediate | 6.3 (244/3372) | 1.1 (0.9, 1.4) | … | 4.5 (154/3098) | 1.2 (0.9, 1.6) | 1.2 (0.9, 1.6) | ||||
| Low | 6.1 (203/2782) | 1.1 (0.9, 1.3) | … | 5.4 (109/1988) | 1.5 (1.1, 2.0) | 1.4 (1.1, 2.0) | ||||
| Educational attainment | .91 | .29 | ||||||||
| Degree or diploma (Ref) | 5.8 (295/4597) | 1.0 | … | 3.7 (154/3944) | 1.0 (0.0, 0.0) | … | ||||
| Apprenticeship or A level | 6.1 (138/1814) | 1.0 (0.8, 1.4) | … | 4.2 (97/2233) | 1.1 (0.8, 1.6) | … | ||||
| GCSE or equivalent | 5.5 (218/3463) | 0.9 (0.8, 1.2) | … | 4.9 (116/2221) | 1.3 (1.0, 1.8) | … | ||||
| None | 5.8 (90/1360) | 1.0 (0.7, 1.3) | … | 4.6 (50/1059) | 1.3 (0.8, 1.9) | … | ||||
| Area social deprivation factors | ||||||||||
| Multiple area deprivation quintiles | .36 | .19 | ||||||||
| Least deprived (Ref) | 5.8 (160/2472) | 1.0 | … | 3.0 (68/2096) | 1.0 | … | ||||
| 2 | 5.3 (164/2517) | 0.9 (0.7, 1.2) | … | 4.2 (97/2214) | 1.4 (0.9, 2.1) | … | ||||
| 3 | 5.7 (163/2380) | 1.0 (0.7, 1.3) | … | 4.7 (107/2088) | 1.6 (1.1, 2.4) | … | ||||
| 4 | 6.8 (155/2103) | 1.2 (0.9, 1.5) | … | 4.4 (89/1815) | 1.5 (1.0, 2.2) | … | ||||
| Most deprived | 5.6 (122/2031) | 1.0 (0.7, 1.3) | … | 4.8 (72/1510) | 1.6 (1.0, 2.5) | … | ||||
| Crime area deprivation quintiles | .48 | .91 | ||||||||
| Least deprived (Ref) | 6.5 (173/2597) | 1.0 | … | 3.9 (86/3018) | 1.0 | … | ||||
| 2 | 6.0 (170/2398) | 0.9 (0.7, 1.2) | … | 4.3 (97/2771) | 1.1 (0.8, 1.6) | … | ||||
| 3 | 5.8 (164/2381) | 0.9 (0.7, 1.2) | … | 4.6 (90/2602) | 1.2 (0.8, 1.7) | … | ||||
| 4 | 5.1 (130/2134) | 0.8 (0.6, 1.0) | … | 4.3 (85/2388) | 1.1 (0.8, 1.6) | … | ||||
| Most deprived | 5.6 (127/1993) | 0.9 (0.6, 1.2) | … | 4.0 (75/2372) | 1.0 (0.7, 1.6) | … | ||||
| Demographic potential confounders | ||||||||||
| Age, y | .53 | .02 | ||||||||
| 45–59 (Ref) | 5.8 (266/3962) | 1.0 | … | 3.3 (124/3481) | 1.0 | … | ||||
| 35–44 | 6.2 (221/3368) | 1.1 (0.9, 1.4) | … | 4.2 (132/2837) | 1.3 (0.9, 1.8) | … | ||||
| 25–34 | 6.1 (184/2621) | 1.0 (0.8, 1.3) | … | 5.5 (108/2022) | 1.7 (1.2, 2.4) | … | ||||
| 16–24 | 5.0 (93/1552) | 0.9 (0.6, 1.2) | … | 4.5 (69/1383) | 1.4 (1.0, 2.0) | … | ||||
| Race/ethnicity | .03 | <.001 | ||||||||
| White British (Ref) | 6.1 (725/10 556) | 1.0 | … | 4.6 (415/8875) | 1.0 | … | ||||
| Not White British | 3.7 (39/946) | 0.6 (0.4, 0.9) | … | 1.5 (18/845) | 0.3 (0.2, 0.6) | … | ||||
| Marital status | <.001 | <.001 | ||||||||
| Married or cohabiting (Ref) | 4.7 (328/6661) | 1.0 | … | 3.7 (221/5904) | 1.0 | … | ||||
| Single | 6.4 (225/3048) | 1.4 (1.1, 1.7) | … | 4.6 (139/2923) | 1.3 (1.0, 1.7) | … | ||||
| Separated, widowed, or divorced | 11.2 (211/1793) | 2.5 (2.0, 3.2) | … | 8.1 (73/894) | 2.3 (1.6, 3.3) | … | ||||
| Living with children | .57 | .11 | ||||||||
| No (Ref) | 5.7 (400/6247) | 1.0 | … | 4.5 (300/6218) | 1.0 | … | ||||
| Yes | 6.0 (364/5256) | 1.1 (0.9, 1.3) | … | 3.7 (133/3505) | 0.8 (0.6, 1.0) | … | ||||
| Overall | 5.8 (764 /11 503) | 4.2 (433/9723) | ||||||||
Note. AOR = adjusted odds ratio; CI = confidence interval; GCSE = general certificate of secondary education; IPV = intimate partner violence; OR = odds ratio.
OR for individual deprivation measures crudely associated with emotional IPV were adjusted for age, ethnicity, and marital status.
Renting a a government-owned property with subsidized rents, usually reserved for low-income households on state benefits.
For those with missing household income values, the prevalence of any IPV and its subtypes was lower than that found in the highest income households in both sexes (Tables 3–5). For all positive associations between social deprivation and IPV, the strength of the association did not vary by age group, except for a stronger association between low household income and IPV in younger compared with older women.
Intimate Partner Violence and Demographics
In the final models retaining sociodemographic factors independently associated with IPV, being in the group aged 35 to 44 years, separated or divorced, and White British were associated with any lifetime, physical, and emotional-only IPV in both men and women (except for a lack of association between age and emotional-only IPV in women; Table B, available as a supplement to the online version of this article at http://www.ajph.org).
DISCUSSION
In line with our expectations, we found that women in England reported a higher prevalence of emotional, physical, and sexual IPV than did men. Compared with male victims, female victims reported more severe, multiple, and prolonged abuse. Also in line with our hypotheses, social deprivation was generally not associated with being a victim of emotional-only IPV among either gender. Individual and area social deprivation were associated with being a victim of any IPV and of physical IPV among women but, contrary to our hypotheses, not generally among men. The exception to this overall pattern was an association in men between social housing tenure and any lifetime IPV and emotional-only IPV.
Strengths of this study included analysis of a large national sample with detailed validated measures of IPV and sociodemographics, with careful adjustment for confounders. Selection bias might have affected our findings as there were significantly lower inclusion rates for those from deprived backgrounds. The BCS might have elicited inaccurate responses from participants because of safety or social stigma concerns, and misreporting may have differed by sociodemographic characteristics. Confounding of area deprivation by residual or unmeasured individual deprivation is possible. In this cross-sectional study, the direction of causality is uncertain, with plausible mechanisms for bidirectional associations between most social deprivation measures and IPV. We measured IPV across the lifespan, whereas we measured the social deprivation measures at the time on the interview. The latter can vary considerably over time, further limiting causal inference. The findings may not generalize to people aged older than 60 years, or to other sociocultural settings. Risk factors were measured for victims only, and there were little data available on partners’ characteristics.
Our findings are consistent with government BCS reports, which revealed an association between past-year IPV and social housing tenure in both genders (consistent with our findings), and a more consistent association between IPV and area deprivation in women than in men.19–21
Our first hypothesis—that women would report more IPV than would men—was supported by the data. The magnitude of the association between IPV and gender differed by abuse subtype, specific experiences, and timeframe. The smallest gender differences were found for recent, less severe physical and emotional abuse (Table 2). In line with our findings, a recent Canadian study revealed that although men and women were equally likely to experience minor physical violence without controlling behavior, women were much more likely to experience prolonged, severe controlling IPV associated with fear.23 Some of the inconsistencies in the literature on gender and IPV may be associated with differences in the nature or timeframe of IPV under study.11 Screening questions that only inquire about less severe forms of recent physical abuse, or are not specific about abuse type, may not find a large difference in report levels between men and women.
To our surprise, we observed an association between social deprivation and lifetime IPV among women but not among men. Three US national studies examined social gradients in both male and female victims of physical IPV. The National Violence Against Women Survey revealed an income gradient in women only,14 whereas 2 other national surveys found an income gradient in both genders.13,15 Physical IPV appears to be consistently associated with individual social deprivation in women, but findings for men differ between studies. One plausible explanation for the lack of association in some settings between deprivation and IPV victimization in men might be that the causal mechanisms that link social deprivation with IPV are more relevant to female victims. There is epidemiological and qualitative evidence that women are more likely to suffer from prolonged, severe abuse in the context of controlling behavior by men, especially where socioeconomic resources are limited,27 whereas men are more likely to suffer from less severe abuse in the context of low-level aggression in everyday relationship conflicts.11 This suggests that social deprivation is more likely to both precede and follow on from IPV among women. Low socioeconomic status of female victims’ households may reflect risk factors in their male partners. In a recent study, the association between women’s household socioeconomic status and IPV was potentially mediated by the male partner’s low education level, substance misuse, and experiences of childhood violence.28
The presence of a clear social gradient for physical IPV in women suggests that efforts to improve identification and prevention measures may need to be designed so that they are particularly accessible for women from disadvantaged homes. Economic empowerment of women may be a useful intervention, although a recent systematic review found that increased financial autonomy for women could increase or decrease the risk of IPV depending on the social context. In particular, greater financial autonomy for women could increase a woman’s risk of partner abuse in settings where male partners felt threatened by this.29 More generally, there is evidence that levels of interpersonal violence are higher in societies with greater social inequality, and that this is turn may be explained by the predominance of aggressive or competitive social strategies found in these hierarchical societies, compared with affiliative social strategies used in more equal settings.30 Tackling domestic violence may therefore require political and socioeconomic measures that tackle inequality in society as a whole.
In the United States, the National Alcohol Survey found that the effects of area deprivation on IPV differed by ethnicity (being stronger for Black than White couples) but not by gender.31 By contrast, we found that area social deprivation was associated with IPV among women only. Living in a poor area (independent of the socioeconomic status of an individual’s home) could increase IPV because of poor social cohesion limiting social control of IPV,32 or because of higher background levels of community violence.33 Although we had no data on social cohesion, our data did not support the latter hypothesis because we found little evidence of an association between local area crime levels and IPV. Both mechanisms may be more relevant to abuse by men against women, as men are more likely to be involved in nondomestic violence, and women may be more dependent on the protective effect of social cohesion. Although IPV is seen as violence occurring in the private domain, the associations with area deprivation suggest it may be appropriate to target interventions at the community level.
Our third hypothesis, that physical IPV would be more strongly associated with social deprivation than emotional-only IPV, was largely supported. Our findings were similar to those of 2 recent studies of women attending primary care clinics in Spain and North America in which low income or unemployment was associated with physical abuse but not with emotional-only abuse.17,24 The reasons for this have not been explored, but we suggest that among higher SES groups the stigma barriers for physical abuse do not apply to the more subtle and hidden forms of emotional abuse. In addition, socially disadvantaged groups have a higher prevalence of a number of risk factors more specific for physical abuse, such as substance misuse and exposure to physical childhood abuse.34 Social gradients may be steeper in severe abuse, and physical abuse may simply be a marker of abuse severity. However, the public health impact of emotional-only abuse can be significant, with a Japanese study reporting that the health status of those suffering emotional abuse only did not differ on most indicators from those suffering emotional with physical or sexual abuse.35 Although the current evidence base does not support universal screening for IPV,3 the lack of a social gradient in emotional abuse suggests that it may be appropriate to use universal rather than targeted preventative interventions for this abuse subtype.
In England, IPV was more prevalent and severe in women than in men. In general, there was a social gradient for physical IPV in women only, and not for emotional-only IPV in either gender. This suggests that interventions across the social spectrum are appropriate for emotional IPV prevention, but that policies for physical IPV prevention should particularly address the needs of women from deprived backgrounds.
Acknowledgments
H. Khalifeh is supported by a Medical Research Council Population Health Scientist Fellowship (G0802432/1). L. M. Howard is supported by the National Institute for Health Research (NIHR) Programme Grant for Applied Research scheme (RP-PG-0108-10084) and is also supported by the NIHR Biomedical Research Centre based at South London and Maudsley NHS Foundation Trust/King’s College London.
Material from Crown copyright records made available through the Home Office and the UK Data Archive has been used by permission of the Controller of Her Majesty’s Stationery Office and the Queen’s Printer for Scotland.
Note. Those who carried out the original analysis and collection of the data bear no responsibility for the further analysis or interpretation of it.
Human Participant Protection
This secondary analysis was granted ethical approval by the London School of Hygiene and Tropical Medicine Ethics Committee.
References
- 1.Preventing Intimate Partner and Sexual Violence Against Women. Geneva, Switzerland: World Health Organization; 2010 [Google Scholar]
- 2.Heise L, Garcia-Moreno C. Violence against intimate partners. : Krug E, Dahlberg L, Mercy J, World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 2002 [Google Scholar]
- 3.Feder G, Ramsay J, Dunne Det al. How far does screening women for domestic (partner) violence in different health-care settings meet criteria for a screening programme? Systematic reviews of nine UK National Screening Committee criteria. Health Technol Assess. 2009;13(16):iii–iv, xi–xiii, 1–113, 137–347 [DOI] [PubMed] [Google Scholar]
- 4.Wu V, Huff H, Bhandari M. Pattern of physical injury associated with intimate partner violence in women presenting to the emergency department: a systematic review and meta-analysis. Trauma Violence Abuse. 2010;11(2):71–82 [DOI] [PubMed] [Google Scholar]
- 5.Ellsberg M, Jansen H, Heise L, Watts CH, Garcia-Moreno C, WHO Multi-country Study on Women's Health and Domestic Violence Against Women Study Team Intimate partner violence and women’s physical and mental health in the WHO multi-country study on women’s health and domestic violence: an observational study. Lancet. 2008;371(9619):1165–1172 [DOI] [PubMed] [Google Scholar]
- 6.Hegarty K, Gunn J, Chondros P, Small R. Association between depression and abuse by partners of women attending general practice: descriptive, cross sectional survey. BMJ. 2004;328(7440):621–624 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Walby S. The Cost of Domestic Violence: Update 2009. Lancaster, England: Lancaster University; 2009 [Google Scholar]
- 8.Department of Health Improving Services for Women and Child Victims of Domestic Violence: The Department of Health Action Plan. London, England: Her Majesty’s Stationery Office; 2010 [Google Scholar]
- 9.Archer J. Cross-cultural differences in physical aggression between partners: a social-role analysis. Pers Soc Psychol Rev. 2006;10(2):133–153 [DOI] [PubMed] [Google Scholar]
- 10.Archer J. Sex differences in physically aggressive acts between heterosexual partners: a meta-analytic review. Psychol Bull. 2000;126(5):651–680 [DOI] [PubMed] [Google Scholar]
- 11.Johnson MP. Conflict and control: gender symmetry and asymmetry in domestic violence. Violence Against Women. 2006;12(11):1003–1018 [DOI] [PubMed] [Google Scholar]
- 12. Department of Health. Improving Services for Women and Child Victims of Violence: the Department of Health Action Plan–Equality Impact Assessment. London: Her Majesty’s Stationery Office; 2010.
- 13.Breiding MJ, Black MC, Ryan GW. Prevalence and risk factors of intimate partner violence in eighteen U.S. states/territories, 2005. Am J Prev Med. 2008;34(2):112–118 [DOI] [PubMed] [Google Scholar]
- 14.Coker AL, Davis KE, Arias Iet al. Physical and mental health effects of intimate partner violence for men and women. Am J Prev Med. 2002;23(4):260–268 [DOI] [PubMed] [Google Scholar]
- 15.Cunradi CB, Caetano R, Schafer J. Socioeconomic predictors of intimate partner violence among White, Black, and Hispanic couples in the United States. J Fam Violence. 2002;17(4):377–389 [Google Scholar]
- 16.Burazeri G, Roshi E, Jewkes R, Jordan S, Bjegovic V, Laaser U. Factors associated with spousal physical violence in Albania: cross sectional study. BMJ. 2005;331(7510):197–201 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Coker AL, Smith PH, McKeown RE, King MJ. Frequency and correlates of intimate partner violence by type: physical, sexual, and psychological battering. Am J Public Health. 2000;90(4):553–559 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Jewkes R. Emotional abuse: a neglected dimension of partner violence. Lancet. 2010;376(9744):851–852 [DOI] [PubMed] [Google Scholar]
- 19.Coleman K, Jansson K, Kaiza P, Reed E. Homicides, Firearm Offences and Intimate Violence 2006/07. London, England: Home Office; 2007 [Google Scholar]
- 20.Finney A. Domestic Violence, Sexual Assault and Stalking: Findings From the 2004/05 British Crime Survey. London, England: Home Office; 2006 [Google Scholar]
- 21.Povey D, Coleman K, Kaiza P, Roe S. Homicide, Firearm Offences and Intimate Violence 2007/08. London, England: Home Office; 2009 [Google Scholar]
- 22.Carrado M. Aggression in British heterosexual relationships: a descriptive analysis. Aggress Behav. 1996;22(6):401 [Google Scholar]
- 23.Ansara DL, Hindin MJ. Exploring gender differences in the patterns of intimate partner violence in Canada: a latent class approach. J Epidemiol Community Health. 2010;64(10):849–854 [DOI] [PubMed] [Google Scholar]
- 24.Ruiz-Pérez I, Plazaola-Castaño J, Álvarez-Kindelán Met al. Sociodemographic associations of physical, emotional, and sexual intimate partner violence in Spanish women. Ann Epidemiol. 2006;16(5):357–363 [DOI] [PubMed] [Google Scholar]
- 25.Bolling K, Grant C, Donovan JL. 2008–2009 British Crime Survey (England and Wales): Technical Report. Vol I London, England: Home Office; 2009 [Google Scholar]
- 26.Growing Unequal? Income Distribution and Poverty in OECD countries. Paris, France: Organisation for Economic Co-operation and Development; 2008 [Google Scholar]
- 27.Jewkes R. Intimate partner violence: causes and prevention. Lancet. 2002;359(9315):1423–1429 [DOI] [PubMed] [Google Scholar]
- 28.Djikanovic B, Jansen HA, Otasevic S. Factors associated with intimate partner violence against women in Serbia: a cross-sectional study. J Epidemiol Community Health. 2010;64(8):728–735 [DOI] [PubMed] [Google Scholar]
- 29.Vyas S, Watts C. How does economic empowerment affect women’s risk of intimate partner violence in low and middle income countries? A systematic review of published evidence. J Int Dev. 2009;21(5):577–602 [Google Scholar]
- 30.Wilkinson R. Why is violence more common where inequality is greater? Ann N Y Acad Sci. 2004;1036(1):1–12 [DOI] [PubMed] [Google Scholar]
- 31.Cunradi CB, Caetano R, Clark C, Schafer J. Neighborhood poverty as a predictor of intimate partner violence among White, Black, and Hispanic couples in the United States: a multilevel analysis. Ann Epidemiol. 2000;10(5):297–308 [DOI] [PubMed] [Google Scholar]
- 32.Li Q, Kirby RS, Sigler RT, Hwang S-S, LaGory ME, Goldenberg RL. A multilevel analysis of individual, household, and neighborhood correlates of intimate partner violence among low-income pregnant women in Jefferson County, Alabama. Am J Public Health. 2010;100(3):531–539 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Koenig MA, Stephenson R, Ahmed S, Jejeebhoy SJ, Campbell J. Individual and contextual determinants of domestic violence in north India. Am J Public Health. 2006;96(1):132–138 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Fergusson DM, Boden JM, Horwood LJ. Developmental antecedents of interpartner violence in a New Zealand birth cohort. J Fam Violence. 2008;23(8):737–753 [Google Scholar]
- 35.Yoshihama M, Horrocks J, Kamano S. The role of emotional abuse in intimate partner violence and health among women in Yokohama, Japan. Am J Public Health. 2009;99(4):647–653 [DOI] [PMC free article] [PubMed] [Google Scholar]