Intimate Partner Violence Against Older Women in Germany

Abstract

Violence against women is a recognized human rights and public health issue, with significant impacts on women’s life and health. Until now, several studies, most of them relying on small scale samples, have explored the prevalence and health effects of intimate partner violence against older women, whereas few have examined what actually puts older women at risk of intimate partner violence. This study is based on a secondary analysis of the first national survey on violence against women in Germany, looking at the prevalence and associated factors for physical and for sexual violence by the current partners of women aged 50 to 65 and women aged 66 to 86 years. The prevalence of violence in women’s current relationships was 12% and 5%, respectively. In both age groups, women who had experienced violence during childhood and nonpartner physical or sexual violence after the age of 16 had higher odds of experiencing current partner violence. Current partner violence was associated only with women and their partner’s level of education and women’s vocational training among women aged 66 to 86 years. Relationships where one or both partners drank heavily in recent months were associated with higher odds of violence among women aged 50 to 65. Future studies on intimate partner violence need to recognize that women above reproductive risk are also at risk of current partner violence.

Keywords

battered women elder abuse predicting domestic violence

Introduction

Violence against women is a recognized human rights and public health issue, with significant impacts on women’s life and health. In recent years, an increasing number of studies have started to explore women’s experience of violence at different ages—including the extent to which older women experience partner abuse.

Although violence against older women is often only considered in the context of care and dependency (Phillips, 2000), several reviews have shown that violence against elder women is mainly perpetrated by intimate partners (see, for example, Penhale, 2005). Although several studies suggest that the prevalence of intimate partner violence is lower among older women (Burazeri et al., 2005; Helweg-Larsen, Sørensen, Brønnum-Hansen, & Kruse, 2011; Rennison & Rand, 2003) when compared with younger women, comparisons of this kind can only be made cautiously, as these studies are mainly of small scale with differing designs, measurements of violence and age categories. In addition, apart from the 1992 National Violence against Women Survey (Greenfeld et al., 1998) and the 1993 and 1999 National Crime Victimization Survey in the United States (Rennison, 2001), few studies consist of nationally representative data.

Intimate partner violence against older women is often viewed in the context of women’s perceived greater vulnerability and dependency on their partner. However, this is refuted by several small scale studies that show fairly similar factors associated with intimate partner violence among older and younger woman (Phillips, 2000).

In Germany, as in most (Organisation for Economic Co-operation and Development) OECD countries, life expectancy is rising rapidly although birth rates decline. It is estimated that by 2050, 40% of the working-age population will be 50 to 64 years old, and the age structure of the German population in 2050 will be substantially older and smaller than nowadays (Christensen, Doblhammer, Rau, & Vaupel, 2009). Although mortality is higher for men than for women at all ages, women experience more physical ill health than men (Christensen et al., 2009). Numerous studies have shown that intimate partner violence has significant adverse physical and psychological health outcomes, especially among older women (Fisher, Zink, & Regan, 2011; Fisher & Regan, 2006; Mouton, 2003), regardless of whether such violence is a continuation of violence that has occurred throughout the history of the relationship, or whether it has commenced de novo in older age. Addressing intimate partner violence among older women is therefore not only important in itself but also has clear implications for their health, well-being, and physical functioning. To develop appropriate and effective intimate partner violence interventions targeting older women, more needs to be known about its prevalence and the factors associated with increased risk.

In Europe, several national representative studies have investigated the prevalence of intimate partner violence and the factors associated with it (Hagemann-White, 2001). Unfortunately, only a few of them surveyed women above the age of 50 (Burazeri et al., 2005; Helweg-Larsen et al., 2011; Papadakaki, Tzamalouka, Chatzifotiou, & Chliaoutakis, 2009; Piispa, 2004; Zorrilla et al., 2010). Among these studies, the age categories for women above the reproductive age vary widely or are not even provided. Apart from their overall findings that the lifetime prevalence of intimate partner violence reduced with increasing age, none of these studies specifically investigated what puts older women at risk of intimate partner violence, with an apparent assumption that associated factors would be similar across age groups. In addition, the two most recent prevalence studies of elder abuse in Europe did not fully consider intimate partner violence in later life, but rather focused on the wider context of elder abuse (Luoma et al., 2011; Soares et al., 2011). A further European study on intimate partner violence against older women did not explore either prevalence or factors associated with it (Nagele, Bohm, Toth, & Goergen, 2010).

For these reasons, a key remaining gap is representative, population-based data to enable comparisons of prevalence rates and factors associated with intimate partner violence across generations (Rennison & Rand, 2003) This article helps addressing this gap by using the German national representative survey on violence against women to explore the prevalence and factors associated with intimate partner violence among women aged 50 to 65 and women aged 66 to 86 years.

Method

The survey “Health, Well-Being and Personal Safety of Women in Germany” was conducted by the Centre for Women’s Studies at the University of Bielefeld in cooperation with the Institute for Applied Social Sciences under commission to the Federal Ministry for Family Affairs, Senior Citizens, Women and Youth in 2003. It consists of 10,264 female respondents between the age of 16 and 86 (Müller & Schröttle, 2003).

Registration lists from 250 randomly chosen communities throughout Germany were used to randomly select women to participate. The selected women received a personalized letter inviting them to participate in the study. The letter detailed the randomized selection process, data privacy laws, and the voluntary nature of participation in the survey. Female interviewers who had prior experience and training in conducting surveys on sensitive matters and violence then contacted them by phone and conducted the interviews in the women’s home. After the interview, women were provided with information on local and national violence services and a free telephone calling card to thank them for their participation. More details on the study methods, sampling and response rate can be found elsewhere (Müller & Schröttle, 2003).

The survey had two stages. First, women participated in a face-to-face interview, which focused on experiences of violence by any perpetrator. After completion, women were asked to complete a written questionnaire on their own. The written questionnaire specifically asked about psychological, physical, and sexual partner violence and childhood abuse (Müller & Schröttle, 2003). The face-to-face interviews lasted on average 64 min and the written questionnaire an additional 18 min. Interview duration increased with women’s age. Older women found it more difficult to respond to the written questionnaire, due to health-related issues, such as tiredness, visual impairment or other disabilities or illnesses, with 11% of women above the age of 75 refusing to fill out the written questionnaire and more than 60% asking for assistance to do so (Müller & Schröttle, 2003).

The national survey among all age groups had a response rate of 52%. Women cited lack of interest, general refusal to participate in survey research, and lack of time as the main reasons for nonparticipation at the initial phone contact, and only to a very small extent concerns about the survey topic, their old age and prevention in participation by a third person, such as partners, children, or grandchildren (Müller & Schröttle, 2003). Six percentage of women above the age of 65 requested the interview to be conducted outside the home to ensure privacy; the overall rate was higher at 12% (Müller & Schröttle, 2003). In 14% of the interviews someone else was present in the home and in 4% the interview was interrupted, even if it was only for a short time. Interruption of the interview by a partner was more frequent among women above the age of 55. Although interviewers were advised to stop the interview in the presence of other people, the authors of the original study stated that this was not always possible (Müller & Schröttle, 2003).

A comparison with the Micro Census 2001, a survey with mandatory participation, showed that participation bias lead to an underrepresentation of younger and older women and women with low educational degrees, an overrepresentation of women from urban areas, former East Germany, and a regional variety in participation. In this analysis, the weightings from the Micro Census were used to correct for these imbalances.

Measurement of Intimate Partner Violence

A modified version of the revised Conflict Tactics Scale (CTS) was used to measure physical and sexual intimate partner violence (Straus, Hamby, Boney-McCoy, & Sugarman, 1996). The CTS inquired about specific acts ranging from being pushed away angrily and slapped in the face to being strangled or injured with a weapon (for details see Table 1). In the modification, phrases like “so that it hurt or I became frightened” or “that it could have injured me” were added to the specific acts to make certain that only acts that constitute physical assaults were included (Müller & Schröttle, 2003). The five questions on sexual violence included acts from forced or attempted forced sexual intercourse to forced petting and forced watching and reenacting of pornographic material. For all acts of physical and sexual intimate partner violence, women had the answer options no, once, and more than once, followed by the question if the act occurred in the last 12 months. After administering the CTS, all women who experienced at least one act of intimate partner violence at least once were asked again if any of these acts occurred in the last 12 months or in the last 5 years. For this analysis, all women who reported experiencing at least one act of physical and/or sexual violence at least once at any point in their current relationship in either the face-to-face interview or in the written questionnaire were considered to have experienced partner violence by a current partner.

Table 1.

Questions Used to Measure Physical and Sexual Intimate Partner Violence (Modification of the CTS)

Questions of physical violence acts
1. Someone has pushed me away angrily.
2. Someone has given me a light slap in the face.
3. Someone has bitten or scratched me so that it hurts or I became frightened.
4. Someone has twisted my arm until it hurts.
5. Someone has kicked me painfully, pushed or grabbed me hard.
6. Shoved me so hard that I stumbled or fell.
7. Someone has given me a hard slap in the face or hit me with an open hand.
8. Someone has thrown something at me that could have injured me.
9. Someone has hit me with an object that could have injured me.
10. Someone has seriously threatened to assault or injure me.
11. Someone has made serious threats to kill me.
12. Someone has hit me with a fist so that it hurts or I became frightened.
13. Someone has thrashed me or beaten me up.
14. Someone has strangled me or tried to smother me.
15. Someone has scalded or burned me on purpose with something hot.
16. Someone has threatened me with a weapon, for example, a knife or pistol.
17. Someone has injured me with a weapon, for example, a knife or pistol.
18. Someone has assaulted me physically in another way that hurts me or made me afraid.
Questions of sexual violence acts
1. Someone forced me to have sexual intercourse and penetrated me with his penis or another object against my will.
2. Someone attempted to penetrate me with his penis or with another object against my will, but this attempt was not completed.
3. Someone forced me into intimate physical contact, stroking, petting, or the like.
4. I was forced to take part in other sexual acts or practices that I did not want.
5. Someone forced me to watch pornographic films or pictures, and to act them out, although he or she knew I did not wish to do so.

Note: CTS = Conflict Tactics Scale.

Conceptual Framework

The investigation of factors associated with physical and/or sexual intimate partner violence against older women in this study is based on an ecological framework, which has been widely used before in violence against women research (Heise, 1998; Krug, Dahlberg, Mercy, Zwi, & Lozano, 2002). The associated factors examined in this study are presented in Figure 1. They include the respondent’s childhood experiences, with a focus on experiences of violence, women’s current vulnerabilities, the couple’s socioeconomic and household characteristics, and women’s social support and neighborhood. The factor “vocational education and training” captures whether an individual has finished 2 to 3 years of vocational training for a job or has a university degree—important qualifications for getting stable employment in Germany.

Figure 1

Conceptual framework on factors for intimate partner violence in current relationships, outlining the measurement of factors analyzed in this study

Analysis

This analysis looks separately at women aged 50 to 65 and 66 to 86 years at the time of the survey. The rationale for categorizing women into these two age groups was to capture the potential generational differences occurring among these two groups of women. The younger age group is one in which women are mostly still actively involved in the labor market and who were teenagers in the late 60s and 70s, whereas the older group consists of those women who are likely to have had a more traditional upbringing and are now in retirement.

For the analysis of associated factors we only included women with current partners to ensure that the data collected about women’s partners related to their current abuse status. This restriction unfortunately excludes women who were widowed, had left their abusive partners or had experienced partner violence during or after their separation. In total, 2,616 women aged 50 to 65 and 1,644 women aged 66 to 86 participated in the face-to-face survey. Of these women 2,030 and 779, respectively, were currently in a relationship and were thus included in the analysis. Those who had experienced partner violence by a prior but not their current partner were excluded from the analysis of associated factors to avoid diluting associations.

To analyze the data, we calculated descriptive statistics and explored possible associations between factors by conducting cross tabulations and chi-square statistics. For variables with low cell counts, Fisher Exact Tests were used to establish significance. We then estimated crude odds ratios using bivariate logistic regression. Statistical significance is considered at the 5% level.

Despite the large sample size, the low number of women aged 66 to 86 reporting physical and sexual violence did not allow us to conduct a multivariate analysis. Similarly, low cell count for some factors among women in this age group might have resulted in nonsignificant findings. Furthermore, the cross-sectional nature of the data cautions the interpretation of the associations as suggesting causality, as they might also be consequences of the violence. Data were missing for less than 5% of respondents for most variables, and women with missing data were excluded from analyses with that variable. All data were analyzed using STATA 11.

Results

Of the 2,030 women aged 50 to 65, 11% had a high school or university education, and 81% had received occupational training and education. These percentages were 6 and 62, respectively, for the 779 women aged 66 to 86. In both age groups less than 4% were not of German nationality, more than 77% considered themselves to be religious, and more than 83% were in a relationship lasting longer than 20 years. Some 23% of women aged 50 to 65 lived with their children in the same household, compared with 6% of the women aged 66 to 86.

Prevalence

The lifetime prevalence of physical and sexual partner violence was 23% among women aged 50 to 65 and 10% among women aged 66 to 86. Restricting the sample to women who were currently in a relationship, 14% of women aged 50 to 65, and 5% of women aged 66 to 86 years reported physical or sexual violence by their current partner (For details, see Table 2).

Table 2.

Prevalence of Intimate Partner Violence According to Age

	Total		Women Aged 50 to 65		Women Aged 66 to 86
	N	%	N	%	N	%
Lifetime physical or sexual violence, or both by any partner
No	3,491	82%	2,005	77%	1,486	90%
Yes	769	18%	611	23%	158	10%
Physical or sexual partner violence, or both, in the last 12 months
No	4,202	99%	2,565	98%	1,637	100%
Yes	58	1%	51	2%	7	0%
Physical or sexual partner violence, or both, in the last 5 years
No	4,158	98%	2,526	97%	1,632	99%
Yes	102	2%	90	3%	12	1%
Experienced physical or sexual, or both, violence in a current relationship^a
No	2,318	89%	1,601	86%	717	95%
Yes	291	11%	251	14%	40	5%

Restricted to women in current relationships.

Of those women who experienced violence by their current partner, 60% of the 50-to 65-year-old women and 82% of the 66- to 86-year-old women reported only low levels of violence. These included acts like being pushed, shoved, bitten, scratched or hurt in an undefined physical way. Medium levels of violence, capturing acts like being pulled by the arm or hair, kicked, flung around, slapped hard, hit with an object or having something thrown at them, were reported by 21% and 9%, respectively. Of the women aged 50 to 65 years old, 15% and 9% of the 66 to 86 years old, reported severe levels of violence. Severe levels of violence were acts like receiving serious threats of being attacked or killed, being hit with a fist, beaten up, choked, burned and threatened, or actually hurt with a weapon. Four percentage of the 50 to 65 years old and none of the women in the older group reported only sexual violence.

Associated Factors

Factors, which are significantly associated with current partner violence among women aged 50 to 65 and 66 to 86 are displayed in Table 3, whereas details on factors, which did not show a significant association are only provided in the text. As is evident from Table 3, witnessing of parental violence initiated by the father, physical punishment during childhood, and sexual abuse in childhood were significantly associated with violence by a current partner in both age groups. Whether women grew up with both parents or not was not significantly associated with current partner violence (50- to 65-year-olds: 85% vs. 79%, p < .08; 66- to 86-year-olds: 88% vs. 83%, Fisher Exact: p < .5).

Table 3.

Factors Associated With Physical and Sexual Violence by a Current Partner According to Age (Weighted Data^a)

	Women Aged 50 to 65						Women Aged 66 to 86
	No IPV		IPV				No IPV		IPV
	n	%	n	%	Unadj. OR	CI	n	%	n	%	Unadj. OR	CI
Total	1,601	86	251	14			717	95	40	5
Woman’s witnessing of parental violence
No	1,447	88	205	80			678	94	33	75
Father initiated	95	7	28	12	2.03**	[1.22, 3.38]	27	4	6	22	7.56***	[2.28, 25.13]
Mother initiated	21	2	7	3	2.49	[0.93, 6.64]	5	1	1	3	3.65	[0.40, 33.13]
Both parents initiated	38	3	11	5	2.19*	[1.00, 4.80]	7	1	0	0	1.00	[1.00, 1.00]
Woman’s physical punishment in childhood
No	623	37	50	20			348	48	8	16
Yes	978	63	201	80	2.41***	[1.66, 3.49]	369	52	32	84	4.85**	[1.82, 12.92]
Woman’s child sexual abuse
No	1,516	94	206	81			692	97	35	87
Yes	85	6	45	19	4.01***	[2.59, 6.21]	25	3	5	13	4.09*	[1.39, 12.05]
Woman’s physical/sexual nonpartner violence (>15 years old)
No	1,352	84	161	65			642	90	30	75
Yes	249	16	90	35	2.78***	[1.91, 3.84]	75	10	10	25	2.89*	[1.21, 6.90]
Woman’s level of education
Primary	680	63	92	57			488	81	18	53
None/still pupil	22	2	5	2	1.56	[0.52, 4.69]	27	3	1	1	0.31	[0.04, 2.53]
Secondary	611	25	101	29	1.32	[0.95, 1.82]	134	11	8	26	3.82*	[1.32, 11.08]
High school	88	3	15	3	1.16	[0.61, 2.20]	34	3	2	4	1.94	[0.41, 9.28]
University	197	7	38	10	1.51	[0.97, 2.36]	34	2	11	16	10.03***	[4.02, 25.03]
Woman has occupational training
No	254	20	34	17			239	40	2	4
Yes	1,347	80	217	83	1.20	[0.78, 1.84]	478	60	38	96	17.9***	[4.01, 79.8]
Partner’s level of education
Primary	643	55	94	49			432	71	15	50
None/still pupil	12	1	3	1	1.74	[0.41, 7.35]	7	1	1	3	3.30	[0.37, 29.60]
Secondary	470	21	64	21	1.13	[0.77, 1.67]	106	11	5	9	1.17	[0.36, 3.85]
High school	74	4	16	5	1.35	[0.71, 2.58]	26	3	2	3	1.26	[0.26, 6.08]
University	367	19	72	24	1.46*	[1.00, 2.13]	127	13	16	35	4.00**	[1.60, 9.99]
Marital status
Married	1,526	96	232	91			700	98	40	100
Single	15	1	6	3	3.21*	[1.14, 8.99]	5	1	0	0	1.00	[1.00, 1.00]
Divorced or separated	59	3	13	6	1.87	[0.89, 3.93]	11	1	0	0	1.00	[1.00, 1.00]
Heavy drinking
Neither	549	36	54	21			315	45	13	27
Her only	266	17	33	13	1.35	[0.79, 2.30]	142	20	3	4	0.37	[0.09, 1.52]
Him only	583	34	121	48	2.34***	[1.58, 3.46]	195	27	15	41	2.57*	[1.03, 6.44]
Both	203	13	43	18	2.24**	[1.36, 3.70]	65	8	9	28	5.34***	[1.96, 14.5]
Having visitors at home
Yes	1,412	90	212	85			612	83	31	80
No	182	10	38	15	1.58*	[1.02, 2.46]	105	17	9	20	1.26	[0.52, 3.03]
Feeling socially included
Yes	1,485	93	214	84			660	92	34	85
No	102	7	36	16	1.35***	[1.15, 1.59]	50	8	6	15	1.30	[0.93, 1.83]
Area of residence
Urban	1,057	67	188	75			469	66	26	67
Rural	544	33	63	25	0.66*	[0.47, 0.93]	248	34	14	33	0.93	[0.42, 2.08]

Note: IPV = Intimate partner violence.

n presented are not weighted.

p < .05. **p < .01. ***p < .001.

Although women’s physical disability (50-to 65-year-olds: 7% vs. 6%, p < .7; 66- to 86-year-olds: 10% vs. 13%, Fisher Exact: p < .8) and involvement of illegal drug use in the last 5 years (50- to 65-year-olds: 1% vs. 1%, p < .9; 66- to 86-year-olds: 1% vs. 2%, Fisher Exact: p < .4) was not significantly associated with current partner violence among women in both age groups. Educational level and occupational training and education of women and educational level of their partners were only significantly associated with current partner violence for women aged 66 to 86 years. For both age groups, the partner’s occupational training and education was not associated with current partner violence (50- to 65-year-olds: 96% vs. 97%, p < .64; 66- to 86-year-olds: 92% vs. 100%, Fisher Exact: p < 0.16). Further analyses that looked at the relative impact of education and occupational training and education found that whether one of the partners had a higher level of education than the other was not significantly associated with current partner violence, although women aged 66 to 86 had lower odds of violence by their current partner if only their partner had occupational training and education but they did not (35% vs. 7%; OR = 0.2, CI [0.04, 0.53]; p < .005).

Heavy drinking, capturing frequent drinking in the last year or drinking more than five units of alcohol at one time in the last month, was strongly associated with current partner violence among both age groups. Among women aged 50 to 65, the odds of experiencing current partner violence were similar whether only the partner or both, the woman and her partner, were heavy drinkers. The odds of experiencing current partner violence were much higher among women aged 66 to 86 if both were heavy drinking than if only the partner was a heavy drinker. For both age groups, no significant association emerged with current partner violence and lack of employment, neither for women (50- to 65-year-olds: 59% vs. 54%, p < .2; 66- to 86-year-olds: 98% vs. 98%, Fisher Exact: p < .7) or their partner (50- to 65-year-olds: 59% vs. 54%, p < .2; 66- to 86-year-olds: 98% vs. 96%, Fisher Exact: p < .7) and women (50- to 65-year-olds: 3% vs. 4%, p < .6; 66- to 86-year-olds: 2% vs. 7%, Fisher Exact: p < .2) or their partner (50- to 65-year-olds: 4% vs. 5%, p < .5; 66- to 86-year-olds: 1% vs. 0%, Fisher Exact: p = 1) not being of German nationality.

Household characteristics and factors at the social support and community level, such as the number of children (50- to 65-year-olds: p < .6; 66- to 86-year-olds: Fisher Exact: p = 1), relationship duration (50- to 65-year-olds: p < .2; 66- to 86-year-olds: Fisher Exact: p < 0.5), women’s lack of activities outside the home (50- to 65-year-olds: 10% vs. 9%, p < .7; 66- to 86-year-olds: 25% vs. 15%, Fisher Exact: p < .3), women’s perception of not having people close to them (50- to 65-year-olds: 39% vs. 42%, p < .5; 66- to 86-year-olds: 70% vs. 66%, Fisher Exact: p < .7), and living in an anonymous neighborhood (50- to 65-year-olds: 35% vs. 40%, p < .3; 66- to 86-year-olds: 37% vs. 52%, p < .2) and living in East versus West Germany (50- to 65-year-olds: 19% vs. 15%, p < .08; 66- to 86-year-olds: 21% vs. 10%, Fisher Exact: p < .06) were not significantly associated with current partner violence. Exceptions were the three factors of being single versus married, barely having visitors at home and feeling socially excluded, which increased the odds of experiencing current partner violence among women aged 50 to 65. Living in a rural area also reduced the odds of experiencing current partner violence, so geographical location may act as a protective factor.

Conclusion

To the best knowledge of the authors, this is the only study investigating the factors associated with the occurrence of intimate partner violence among women above reproductive age, using nationally representative data and separating older women into two different age groups to capture their different life experiences.

Our analyses showed that there was a 23% prevalence of women aged 50 to 65 who had ever experienced physical and/or sexual intimate partner violence. This is substantially higher than the 10% ever prevalence reported by women aged 66 to 86. A similar pattern emerged for current partner violence, with a prevalence of 14% and 5% for these age groups. Although these findings support existing evidence that rates of intimate partner violence are lower among older women (Rennison & Rand, 2003), comparisons with other studies are difficult due to the different age categories, definitions of violence and perpetrators considered. Although the lower rates of current partner violence in the last 5 years may represent a general decrease of criminal behavior, including violence with increased age (Laub & Sampson, 2003) or the change of physical and sexual violence to financial or emotional abuse suggested by others (Rennison & Rand, 2003), there is little evidence to believe that women above the age of 65 have experienced less violence in their lives than younger women. Another probable explanation is that the lower prevalence of current partner violence among older women results from recall biases reflecting social desirability. Despite qualitative evidence that women remember abusive situations that occurred 40 to 50 years ago well, especially when linking them to the age of their children (Grunfeld, Larsson, Mackay, & Hotch, 1996), they are probably less likely to recall an event that occurred 30 compared with 2 years ago (Walby, 2005) unless they are experiencing major physical and/or cognitive impairment, which may prevent accurate short-term recall as well. In addition, older women’s socialization and their perception of their current economic and social status might also influence their willingness and ability to disclose violence by their current partners (Rennison & Rand, 2003). There may also be some reluctance by women in the older cohort to admit or disclose abuse, either through a survey or more qualitative research (Nagele et al., 2010), and it is also possible that perceptions of abuse and violence may vary between generations.

The findings on the influences of childhood experiences of violence, including witnessing of violence initiated by the father, physical punishment, and sexual abuse correspond to findings of several prospective studies from New Zealand and the United States (Ehrensaft, 2007; Fergusson, Boden, & Horwood, 2008), as well as of studies looking specifically at factors associated with violence among older women (Penhale, 1999; Teaster, Roberto, & Dugar, 2006). Childhood experiences of violence therefore seem to be important factors associated with intimate partner violence not only across countries but also across different age groups. Another established factor associated with intimate partner violence that also emerged in this study is alcohol misuse. Existing studies suggest that heavy alcohol use may result in intimate partner violence by increasing marital conflict and disinhibition, raising individual’s levels of aggression, and interfering with cognitive abilities (Klostermann & Fals-Stewart, 2006). Qualitative studies among older women further suggest that abusive partners often perceived their use of alcohol as a permission to let down their self-control and abuse their partner while women said they felt less able to defend themselves when they have been drinking (Brandl, 2000; Penhale & Porritt, 2010; Teaster et al., 2006).

An interesting difference between the two main age groups in this study emerged regarding education and occupational training, with women above the age of 65 years having an increased risk of current partner violence if they or their partner had a high level of education or if they had occupational training and education. No such associations emerged for women below the age of 66. As the percentages showed, having an education higher than the primary level and having received occupational training and education was less common among women in the older age group, which implies that older women with such training or higher education are more likely to have found good employment. It is also likely to have resulted in higher levels of empowerment and independence from their partner for these women, apart from having more liberal ideas. Higher levels of empowerment can lead women to challenge certain aspects of traditional sex roles, which carry an increased risk of violence when the level of equality is not yet high enough in the general population for protective effects to predominate (Jewkes, 2002).

Marital status only increased the risk of current partner violence among the women aged 50 to 65 and not among the older group of women. One potential reason for this difference is that nearly all partnered women aged 66 to 86 were married, as cohabitation or divorce was seldom a possibility among the older age group in earlier phases of their relationships and that abuse did not usually previously constitute a reason for leaving a relationship (Penhale & Porritt, 2010; Teaster et al., 2006). Our findings on marital status and the lack of association with relationship duration do not support claims made in another study (Phillips, 2000) that wife abuse in the older years may arise in new marriages that occur among aging couples because these new marriages can be complicated by conflicts among grown children or concerning the use and distribution of assets and resources from previous marriages, or because these new marriages can often escalate quickly into stressful care giving situations.

Our study has several limitations. First, the overall participation rate was 52% and refusal may have biased the sample, especially as it was linked to age. This also relates to participation in the written questionnaire, which older women were more likely to refuse or demand assistance with due to health-related issues. Furthermore, the interviews with older women were more often interrupted by third persons. Second, low reporting of physical and sexual violence among women aged 66 to 86 did not permit the exploration of multivariate associations and may have led to nonsignificant associations due to small numbers. It is assumed that the low reporting of intimate partner violence among this group is linked to reporting biases, which will be discussed in more detail below.

Third, all the information in this survey is based on women’s accounts alone and we were therefore unable to explore men’s experiences of violence during childhood, societal acceptance of violence or laws and services available to abused women and used by them in different regions. Finally, the cross-sectional design of the study limits the extent to which we can establish the order of effect and distinguish risk factors from consequences of abuse. This is especially the case with factors at the societal level, as social isolation is known to be a risk factor as well as a consequence of current partner violence and this appears to affect women across the life-course, not just those of younger ages.

The age specific limitations found in this study are shared with other studies investigating intimate partner violence against older women, with the main issue being the low reporting of intimate partner violence by women above the age of 65. In addition to the above mentioned reporting biases, such as recall difficulties, social desirability, and increased dependencies of older women, other potential reasons that affect the underreporting of intimate partner violence among older women are that they are more likely to underreport less severe acts of intimate partner violence and their consequences, especially if they have lived in violent partnerships for a long time (Piispa, 2004). Due to their upbringing, exposure, and the shame and isolation associated with intimate partner violence, older women are less likely than young women to talk openly with outsiders about their intimate relationships and might find it harder to report violence to a researcher (Hagemann-White, 2001). Furthermore, because intimate partner violence against older women often becomes mainly psychological and less physical or sexual, older women might believe that it is now acceptable and therefore not mention past experiences of physical and sexual intimate partner violence (Fisher & Regan, 2006). Some older women may also equate violence and abuse solely with physical and sexual acts and, therefore, do not consider other situations to constitute violence and abuse and so do not report them (Nagele et al., 2010).

In this survey, as in others, it was more difficult for researchers to conduct interviews with older women in complete privacy, as their partners, their children, or their caretakers were often present for at least some part of the interview. This is concerning, as the perpetrators of violence against older women are often those close to them and it is consequently likely to be unsafe for them to reveal intimate partner violence, let alone that women might want to protect their privacy if they have not revealed past experiences of intimate partner violence to their families, caretakers, or anyone else (Fisher & Regan, 2006; Walby, 2005).

Survey methodologies to allow women to reveal experiences of intimate partner violence in more confidential ways have developed rapidly in recent years, such as the inclusion of a self-completion survey in addition to the face-to-face interview in this survey. Unfortunately, older women made less use of this method, which would have given them the ability to reveal intimate partner violence in a more confidential way. Due to concentration problems, tiredness after the 1-hour interview, and poor eyesight they often had to ask the interviewers for assistance in completing them (Müller & Schröttle, 2003).

Future studies including women above the age of 65 will need to consider the age specific challenges older women face when participating in surveys on sensitive issues such as intimate partner violence to ensure the accurate comparability of prevalence rates across different age groups.

Despite these limitations, this study showed that, regardless of lower incidence rates, for many women above reproductive age, violence by their current partner remains a problem in their lives. With growing evidence that violence that occurred years ago still impacts women’s health and the fact that most women are still living with their abusive partners and may thus continue to experience violence, even if this is of different types, more research is needed on how to best intervene to provide these women with appropriate support and services to meet their needs for support and assistance.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Heidi Stöckl was funded via an interdisciplinary postdoctoral fellowship from the Medical and the Economic and Social Research Councils (grant no. G0802699). Charlotte Watts and Bridget Penhale received no financial support for the research, authorship, and/or publication of this article.

Bios

Heidi Stöckl, PhD, is a research fellow at the London School of Hygiene and Tropical Medicine. She has a background in political science, sociology, and evidence-based social intervention. As an MRC/ESRC interdisciplinary research Fellow she is currently investigating different intervention models to prevent and address partner violence in antenatal care in Tanzania as well as carrying out a risk and protective factor analysis for partner violence in current relationships in Germany.

Charlotte Watts, PhD, is Sigrid Rausing professor in gender, violence, and health and head of the Social and Mathematical Epidemiology Group in the Department of Global Health and Development. Originally trained as a mathematician, with further training in epidemiology, economics and social science methods, she has more than 15 years experience in international HIV and violence research and brings a strong multidisciplinary perspective to the complex challenge of addressing HIV and violence against women.

Bridget Penhale, MSc, is a reader in Mental Health of Older People at the University of East Anglia. She has substantive clinical social work and managerial experience across a variety of hospital and community services. Her academic work is predominantly involved in research concerning older people, encompassing but not limited to elder abuse, adult protection, social care, attachment and intergenerational relationships in later life, mental health of older people, and domestic and interpersonal violence.

References

Brandl

(2000). Power and control: Understanding domestic abuse in later life. Generations, 24, 39-45.

Burazeri

Roshi

Jewkes

Jordan

Bjegovic

Laaser

(2005). Factors associated with spousal physical violence in Albania: Cross sectional study. British Medical Journal, 331, 197-201.

Christensen

Doblhammer

Rau

Vaupel

J. W.

(2009). Ageing populations: The challenges ahead. Lancet, 374, 1196-1208.

Ehrensaft

M. K.

(2007). Intergenerational transfer of intimate partner violence. In Jackson

N. A.

(Ed.), Encyclopedia of domestic violence (pp. 390-397). New York, NY: Routledge.

Fergusson

Boden

Horwood

(2008). Developmental antecedents of interpartner violence in a New Zealand birth cohort. Journal of Family Violence, 23, 737-753.

Fisher

B. S.

Regan

S. L.

(2006). The extent and frequency of abuse in the lives of older women and their relationship with health outcomes. The Gerontologist, 46, 200-209.

Fisher

B. S.

Zink

Regan

S. L.

(2011). Abuses against older women: Prevalence and health effects. Journal of Interpersonal Violence, 26, 254-268.

Greenfeld

Rand

Craven

Klaus

Perkins

Ringel

. . . Fox

J. A.

(1998). Violence by intimates. Washington, DC: Bureau of Justice Statistics factbook, U.S. Department of Justice.

Grunfeld

Larsson

Mackay

Hotch

(1996). Domestic violence against elderly women. Canadian Family Physician, 42, 1485-1493.

10.

Hagemann-White

(2001). European research on the prevalence of violence against women. Violence Against Women, 7, 732-759.

11.

Heise

L. L.

(1998). Violence against women: An integrated, ecological framework. Violence Against Women, 4, 262-290.

12.

Helweg-Larsen

Sørensen

Brønnum-Hansen

Kruse

(2011). Risk factors for violence exposure and attributable healthcare costs: Results from the Danish national health interview surveys. Scandinavian Journal of Public Health, 39, 10-16.

13.

Jewkes

(2002). Intimate partner violence: causes and prevention. Lancet, 359, 1423-1429.

14.

Klostermann

K. C.

Fals-Stewart

(2006). Intimate partner violence and alcohol use: Exploring the role of drinking in partner violence and its implications for intervention. Aggression and Violent Behavior, 11, 587-597.

15.

Krug

E. G.

Dahlberg

L. L.

Mercy

J. A.

Zwi

Lozano

(2002). World report on violence and health. Geneva, Switzerland: World Health Organization.

16.

Laub

Sampson

(2003). Shared beginnings, divergent lives: Delinquent boys to age 70. Cambridge, MA: Harvard University Press.

17.

Luoma

M.-L.

Koivusilta

Lang

Enzenhofer

De Donder

Verte

. . . Penhale

(2011). Prevalence study of abuse and violence against older women: Results of a multicultural survey conducted in Austria, Belgium, Finland, Lithuania and Portugal. Helsinki, Finland: THL, Department of Health and Welfare.

18.

Mouton

C. P.

(2003). Intimate partner violence and health status among older women. Violence Against Women, 9, 1465-1477.

19.

Müller

Schröttle

(2003). Lebenssituation, sicherheit und gesundheit von frauen in Deutschland: Eine repräsentative untersuchung zu gewalt gegen frauen in Deutschland [Living conditions, safety, and health of women in Germany: A representative study of Violence against Women in Germany]. Baden-Baden, Germany: Federal Ministry for Family Affairs Senior Citizens Women and Youth, Koelblin-Fortuna-Druck.

20.

Nagele

Bohm

Toth

Goergen

(2010). Intimate partner violence and older women: Summary report. Gottingen, Germany: ZOOM-prospektive.

21.

Papadakaki

Tzamalouka

G. S.

Chatzifotiou

Chliaoutakis

(2009). Seeking for risk factors of intimate partner violence (IPV) in a Greek national sample: The role of self-esteem. Journal of Interpersonal Violence, 24, 732-750.

22.

Penhale

(1999). Bruises on the soul: Older women, domestic violence, and elder abuse. Journal of Elder Abuse & Neglect, 11, 1-22.

23.

Penhale

(2005). Older women, domestic violence, and elder abuse: A review of commonalities, differences, and shared approaches. Journal of Elder Abuse & Neglect, 15, 163-183.

24.

Penhale

Porritt

(2010). Intimate partner violence and older women in Europe: UK national report. Sheffield, UK: University of Sheffield.

25.

Phillips

(2000). Domestic violence and aging women. Geriatric Nursing, 21, 188-193.

26.

Piispa

(2004). Age and meanings of violence: Women’s experiences of partner violence in Finland. Journal of Interpersonal Violence, 19, 30-48.

27.

Rennison

C. M.

(2001). Intimate partner violence and age of victim, 1993-99. Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice.

28.

Rennison

Rand

(2003). Nonlethal intimate partner violence against women. Violence Against Women, 9, 1417-1428.

29.

Soares

Barros

H. D.

Torres-Gonzalez

Ioannidi_Kapolo

Lamura

Lindert

. . . Stankunas

(2011). Abuse and health among elderly in Europe. Kaunas: Lituanian University of Health Sciences Press.

30.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

D. B.

(1996). The revised Conflict Tactics Scale (CTS-2): Development and preliminary psychometric data. Journal of Family Issues, 17, 283-316.

31.

Teaster

Roberto

Dugar

(2006). Intimate partner violence of rural aging women. Family relations, 55, 636-648.

32.

Walby

(2005). Improving the statistics on violence against women. Statistical Journal of the United Nations, 22, 193-216.

33.

Zorrilla

Pires

Lasheras

Morant

Seoane

Sanchez

L. M.

. . . Durbán

(2010). Intimate partner violence: Last year prevalence and association with socioeconomic factors among women in Madrid, Spain. European Journal of Public Health, 20, 169-175.