Intimate Partner Violence and Abuse Among Active ...

VIOLENCE ARTICLE Campbell AGAINST et al.WOMEN / ABUSE/AMONG September ACTIVE 2003 DUTY 10.1177/1077801203255291 MILITARY WOMEN

Intimate Partner Violence and Abuse Among Active Duty Military Women JACQUELYN C. CAMPBELL MARY A. GARZA ANDREA CARLSON GIELEN PATRICIA O’CAMPO JOAN KUB Johns Hopkins University

JACQUELINE DIENEMANN University of North Carolina, Charlotte

ALISON SNOW JONES Wake Forest University

EIMAN JAFAR Johns Hopkins University

In a sample of 616 active duty military women, 30% reported adult lifetime intimate partner violence (IPV), defined as physical and/or sexual assault. The prevalence of IPV during the time of military service was 21.6%. Lifetime prevalence of any abuse, including emotional abuse and/or stalking, was 44.3%. Risk factors for IPV (lifetime and while in the military) from multivariate logistic regression were separated or divorced marital status (odds ratio = 5.23, 6.17, respectively), being widowed (odds ratio = 3.57, 4.57), having one child (odds ratio = 2.12, 2.49) or three or more children (odds ratio = 2.72, 3.34), and being enlisted personnel rather than officers (odds ratio = 2.45, 2.77). These prevalence rates and risk factors were similar to a demographically comparable civilian sample from the same geographic area. Existing military policies and programs should be examined and enhanced to maintain military readiness as well as reduce military women’s risk of harm. Keywords: active duty military women; intimate partner violence

The prevalence of intimate partner violence (IPV) has seldom been measured in the military, although the issue is recognized as serious (Department of Defense, 2001; Mercier & Mercier, 2000). There have been compilations of the incidence of spouse abuse VIOLENCE AGAINST WOMEN, Vol. 9 No. 9, September 2003 DOI: 10.1177/1077801203255291 © 2003 Sage Publications

1072

1072-1092

Campbell et al. / ABUSE AMONG ACTIVE DUTY MILITARY WOMEN

1073

(physical and psychological aggression) reported to the Family Advocacy Program and thus represented in the central registries of the various services (e.g., McCarroll et al., 1999). However, the only population-based active duty military (ADM) self-report survey published to date in a peer-reviewed journal is that of Heyman and Neidig (1999), conducted with a large sample of Army-enlisted married males (28,345) and females (2,772). In spite of its comprehensiveness and excellent response rate, the study did not report demographic risk factors for female victims separately from males in the analysis. The survey was also limited to perpetration, to the U.S. Army, and to married soldiers. To determine the prevalence and demographic risk factors for victimization among all ADM women across services, a survey was conducted of those residing in the greater Washington, D.C., metropolitan area. BACKGROUND IPV is defined by the Centers for Disease Control as physical and/or sexual assault or threats thereof between sexually intimate partners (Saltzman, Fanslow, McMahon, & Shelley, 1999). The military addresses domestic violence as violence or threats between marital partners, and some of the services include emotional abuse (Defense Task Force on Domestic Violence, 2001). Spouse abuse has been recognized as a serious problem in the military, affecting the readiness, safety, health, and quality of life of military personnel (Department of Defense, 1992, 2001; Mercier & Mercier, 2000). The military has adopted a zero tolerance policy, and the Department of Defense (Defense Task Force on Domestic Violence, 2001) and each service branch has issued appropriate policies regarding IPV (e.g., Defense Task Force on Domestic Violence, 2001; Department of Defense, 1992; Mollerstrom & Patchner, & Milner, 1992). Each service has also instigated several initiatives to address and prevent spouse abuse among its personnel (Caliber Associates, 1996). AUTHORS’ NOTE: This research was supported by Grant No. R01 DAMD17-96-1-6310, U.S. Army Medical Research. Patricia Modrow, U.S. Department of the Army, was the military principle investigator for this study.

1074

VIOLENCE AGAINST WOMEN / September 2003

To accurately assess the extent of the problem, and thereby appropriately respond, it is necessary to go beyond official reports, which will always be underestimates of the true extent of problem behavior. For example, the official reports of spousal abuse in the U.S. Army indicate a past-year prevalence in 1996 between 8.0% and 10.5% of married couples (McCarroll et al., 1999), whereas the Heyman and Neidig (1999) survey of married Army couples found 22.8% of the active duty males and 31.1% of the active duty females reported perpetrating physical violence against a wife or husband in the prior year. The participants in the Heyman and Neidig study were randomly sampled from 38 Army installations within the United States from 1990 to 1994, and the survey was completely anonymous. That study was also strong in its careful comparison with the civilian 1985 National Family Violence Survey (Straus & Gelles, 1990), weighted to be comparable and eliminating unmarried couples, both using the Conflict Tactics Scale (Straus, 1979; Straus & Gelles, 1990). The comparison revealed a similar prevalence of moderate husbandto-wife physical assault but a significantly higher prevalence of severe husband-to-wife physical assault, as well as moderate and severe wife-to-husband physical assault, in military families. Another important and somewhat comparable study to this investigation was conducted using a sample (N = 411) of hospitalized and outpatient female veterans from Minneapolis (Murdoch & Nichol, 1995). These veterans reported a 28% lifetime prevalence of IPV in an anonymous survey. In another nationally representative telephone survey of female veterans, prevalence of physical and/or sexual assault by any perpetrator during military service (48%) was reported, but violence perpetrated by an intimate partner was not reported separately (Sadler, Booth, Nielson, & Doebbeling, 2000). Demographic risk factors that have been frequently identified in population-based civilian research for female IPV victimization include being divorced or separated and having low income and education (e.g., Dearwater et al., 1998; Greenfield et al., 1998). Minority ethnicity has sometimes been identified as a risk factor for IPV (e.g., Greenfield et al., 1998; Straus & Gelles, 1990); however, its power generally decreases substantively or disappears when income and education are controlled (e.g., Dearwater et al., 1998; Jones et al., 1999; Rosen, Parmley, Knudson, & Fancher,


1075

2002b). In 1997 to 1998, a civilian sample of women enrollees in a health maintenance organization (HMO) in the metropolitan Washington, D.C., area was interviewed, with 35.5% of those women reporting adult lifetime IPV and 44.3% reporting adult lifetime intimate partner abuse when emotional abuse was also included (Jones et al., 1999). The present study used similar methods to determine the prevalence of IPV in a demographically similar sample of ADM women. METHOD SURVEY PROCEDURES

Data for the study were collected from January 1998 to October 2000. Using the Defense Enrollment Eligibility Reporting System (DEERS) database of ADM women, a random sample of 16,540 tri-service (Air Force, Army, and Navy, includes Marines) ADM women in the greater metropolitan Washington, D.C., area was contacted with an introductory letter. The letter described the study as one that would address “the health effects of stress and women’s relationships.” The letter was intentionally vague about the purpose of the study, being mindful that intimate partners, especially controlling intimate partners, might read such a letter even if addressed to the woman. Any direct references to domestic violence, therefore, might have compromised abused women’s safety. To minimize the risk of abusive men reading the letter, work addresses were used whenever possible. Unfortunately, the DEERS database does not routinely provide work addresses because they change frequently in the military due to deployment and other changes in work assignments. Women were asked for a work address to which a consent form could be sent, again because of safety concerns. A sample of 1,830 of the women receiving the letter indicated initial consent by requesting full consent forms. Reminder postcards were sent to the other 14,710 women, yielding 349 additional consent form requests. A total of 2,179 participants indicated initial consent for the study. The 13.2% response rate was very similar to the 11.5% overall response rate of the civilian sample from an HMO, used as a comparison group to this study, that used a similar introductory letter and request for response to indicate initial consent (Jones et al., 1999).

1076


Of the ADM women receiving the long (four pages) consent form, 36% (n = 779) signed and returned the form. Each service and each installation in the study has its own military institutional review board (IRB) to provide oversight and human participant protection, and all IRBs reviewed our proposed protocol. Any requirements from any of the IRBs had to be included before final approval. The final IRB requirements necessitated a fourpage written consent form with explicit reference to domestic violence as the topic of the study, a witness signature, and a statement that the individual survey research results could be reviewed by a woman’s commanding officer. In the civilian sample part of the study, the civilian IRB approved a verbal telephone consent rather than a written consent, and only 3% of those reached by phone refused participation. The statement limiting confidentiality of the study results was one of the most significant deterrents to consent in the military sample. Several ADM women called or wrote to tell the investigators that the reason they did not participate in the study was because of the statement saying that their responses could not be kept confidential from their command. Of the ADM women who completed the consent form, 79% (N = 616) were reached and completed full interviews, a completion rate similar to that in the civilian sample (78% of those consenting). The consenting ADM women who could not be reached for interviews had been deployed overseas, had been transferred to an unknown location, or had left the military. All consent forms included civilian and military local and national referral phone numbers for domestic violence services. The telephone interview protocol also included help screens for domestic violence queries and any signs of potential danger, as well as a reminder of domestic violence referral sources. Interviewers were trained and experienced domestic violence interviewers. INSTRUMENTS

Women were screened for abuse using a modified version of the Abuse Assessment Screen. The Abuse Assessment Screen has been widely used and has substantial support for reliability and validity in several studies (Soeken, Parker, McFarlane, & Lominak, 1998). The modification was that women were asked


1077

TABLE 1 Partner Abuse Screening Questions Question

Reply

1. Have you ever as an adult been emotionally abused by a husband, boyfriend, or female partner?

Yes

No

Don’t Know

Refused

2. Have you experienced any emotional abuse in the past 12 months, that is, since September of last year?

Yes

No

Don’t Know

Refused

3. Have you ever as an adult been physically abused by a husband, boyfriend, or female partner?

Yes

No

Don’t Know

Refused

4. Okay, this question is worded a little different. Have you ever been hit, slapped, kicked, pushed, or shoved or otherwise physically hurt by a current or previous husband, boyfriend, or female partner?

Yes

No

Don’t Know

Refused

5. Have you ever, as an adult, been forced into sexual activities by a husband, boyfriend, or female partner?

Yes

No

Don’t Know

Refused

two separate questions relative to emotional and physical abuse rather than a single question that combines the two forms of abuse. Prevalence of IPV while in the military was calculated using the years women reported the abuse to have occurred and the years they reported being in the military. Women also were asked to report their educational level, ethnicity, age, marital status, number of children, and whether they were enlisted personnel or an officer. RESULTS PREVALENCE ESTIMATES

Adult lifetime physical and/or sexual assault among ADM women was 30% (see Table 2). In addition, 117 women reported emotional abuse and/or stalking. The combination prevalence of all forms of adult lifetime abuse was 44.3%. Figure 1 details the overlap of the types of violence encountered by women in this sample over their lifetimes. The majority of the women

1078

VIOLENCE AGAINST WOMEN / September 2003 TABLE 2 Prevalence of Intimate Partner Violence (IPV) in Active Duty Military Women by Sociodemographic Characteristics

Characteristic

n

Total sample 616 Age group 21 to 29 95 30 to 39 244 40 to 49 252 50 to 56 25 Race White European 466 African American 97 Other minority 50 Marital status Married 410 Separated or divorced 77 Never married 119 Widowed 10 Education High school graduatea 36 Some college 161 4 years of college 150 Postgraduate 267 Household income < $30,000 85 $30,000 to $50,000 134 $51,000 to $80,000 180 > $80,000 192 Percentage household income contributed by respondentb 25% to 50% 252 51% to 75% 121 > 75% 238 Number of children in household None 298 1 140 2 123 ⱖ3 53 Military type Enlisted 271 Officer 345

% of Total Sample

% Lifetime IPV

% IPV During Military Service

100.0

29.9

21.6

15.4 39.6 40.9 4.1

23.2 29.5 32.9 28.0

17.9 24.2 20.6 20.0

76.0 15.8 8.2

26.0† 44.3 36.0

27.8 20.0 22.0

66.6 12.5 19.3 1.6

25.9† 59.7 21.0 70.0

18.5† 45.5 12.6 70.0

5.9 26.2 24.4 43.5

33.3*** 40.4 28.7 24.0

25.0† 34.2 19.3 15.0

14.4 22.7 30.5 32.5

37.6* 32.1 31.7 23.4

28.2 23.1 22.8 16.7

41.2 19.8 39.0

26.2* 27.3 35.3

18.7 19.8 26.1

48.5 22.8 20.0 8.6

23.2† 40.0 26.8 49.1

15.1† 32.1 18.7 37.7

44.0 56.0

39.5† 22.3

30.6† 14.5

a. Includes 1 trade school graduate and 1 woman who did not complete high school. b. Three reported contributing less than 25%; these women were included in the 25% to 50% category. *p < .1. ***p < .01. †p ≤ .001, by chi-square testing.

1079


sexual 14 (5.9%)

emotional

8 (3.3%)

52(21.8%)

55 (23%)

83 (34.4%)

2(.8%)

physical 25 (10.5%)

Figure 1:

Overlap of Lifetime Physical, Sexual, and Emotional Abuse From an Intimate Partner (n = 239) NOTE: Style of Venn diagram presentation from Ellsberg (2000).

encountered more than one type of abuse. The largest single group (34.4%) reported physical and emotional abuse, while 21.8% reported physical, emotional, and sexual abuse from an intimate partner. Of the physically abused women, 33.0% also reported sexual assault by an intimate partner. Few women reported only sexual assault (3.3%). Of the 239 ADM women who reported lifetime abuse, 23.0% said that the only form of abuse they had experienced was emotional abuse, while 10.5% said they had been the victims of physical assault only. Adult lifetime and military service prevalence of sexual and/or physical abuse are presented in Table 2. Overall, the women reported an adult lifetime prevalence of 30% and a military service prevalence of 22%. Of those who were abused during military service and responded to the question (88%), 18.5% were abused by a civilian, 43.2% were abused by an ADM service member, and 38.4% were abused by retired military personnel. Enlisted personnel represented 62.9% of the military abusers (both active and retired) and officers 37.1%.

1080


The demographic characteristics of the women are presented in Table 2. Women between the ages of 21 and 29 reported the lowest lifetime and military service prevalence (23% and 18%, respectively). Women between the ages of 40 and 49 reported the highest adult lifetime prevalence (33%), yet not during military service (21%). African American women reported the highest adult lifetime prevalence (44%, p ≤ .000), nearly twice that of White European women (26%). The same pattern was not observed for abuse during military service. In fact, the contrast was in the opposite direction, but the difference was not significant (28% among White women vs. 20% among African American women). Widowed women reported the highest adult lifetime prevalence (70%), including lifetime and during military service. However, the sample (n = 10) was quite small. Separated or divorced women reported the next highest lifetime (60%) and military service (46%) prevalence. Single women reported the lowest prevalence, across lifetime (21%) and military service (13%) categories. Women with some college education reported the highest prevalence (p < .001), regardless of timing. Lifetime abuse was 40%, and abuse during military service was 34%. Women with postgraduate degrees reported the lowest prevalence, regardless of timing (24% lifetime and 15% during military service). An inverse linear trend was observed for lifetime and during military service prevalence related to household income. The highest prevalence was for individuals with an income less than $30,000, while the lowest prevalence represented individuals reporting an income greater than $80,000. A positive linear trend was observed relative to the percentage household income contributed by the respondent in both groups. Women who contributed more than 75% reported the highest adult lifetime and during military service prevalence, while women contributing less than 50% reported the lowest. Women with no children reported the lowest lifetime and military service prevalence, while women with one child or three or more children reported the highest prevalence. Enlisted women reported almost twice the rate of lifetime and military service physical and/or sexual abuse as did officers (p < .001). Among enlisted personnel, 39.5% reported lifetime abuse in contrast to 22.3% of officers. The rates for abuse during military service were 30.6% of enlisted personnel and 14.5% of officers.


1081

LOGISTIC REGRESSION ANALYSIS

Multivariate logistic regression of risk factors of IPV is presented in Table 3 for prevalence for adult lifetime and during military service. The effects of race, education, and income are not significant when other characteristics are controlled. Marital status, number of children, and military rank were strongly associated with both lifetime and military service prevalence. Women between the ages of 40 and 49 experienced an elevated adult lifetime risk of 48%. In contrast, the (nonsignificant) elevated risk was 66% for women between the ages of 50 and 59 during military service. Separated and divorced women had a strongly elevated prevalence, regardless of timing (odds ratio = 5.23 lifetime and 6.17 during military service, p < .001). The pattern observed relative to the number of children in the household persisted; women with one child or three or more children were more than twice as likely to experience physical and/or sexual abuse compared to women with no children (p < .001). Enlisted women were about three times more likely to be physically and/or sexually abused than officers, controlling for all other demographic differences. DISCUSSION A substantial proportion of this sample of ADM women experienced IPV during military service. Of the ADM women, 22% reported physical and/or sexual assault, and 36% reported some type of abuse, including emotional abuse, from an intimate partner while serving in the military. Although these ADM women clearly are not protected from abuse from an intimate partner while in the military, the differences between lifetime and military service prevalence rates indicate that some (9.1%) had managed to end the violence before their military service. Among this sample of ADM women, reported past-year prevalence was only 0.5%. This rate is extremely low and of questionable validity. The lack of confidentiality of the survey results because of the consent process imposed on the study is probably the most important concern (see also Rosen, Knudson, Brannen, Fancher, & Kilgore, 2002). Currently abused women may have opted out of the survey when notified that their commanding officer could review the survey research records. It is also noteworthy

1082

VIOLENCE AGAINST WOMEN / September 2003 TABLE 3 Multivariate Logistic Regression of Risk Factors for Adult Lifetime Prevalence of Intimate Partner Violence in a Sample of Active Duty Military Women Lifetime Prevalence

Characteristic

Odds Ratio

Confidence Interval

Age group 21 to 29 1.07 0.56, 2.05 40 to 49 1.48* 0.93, 2.35 50 to 59 1.33 0.43, 4.11 30 to 39a Race African American 1.24 0.72, 2.15 Whitea Marital status Separated or divorced 5.23† 2.36, 11.59 Never married 0.88 0.41, 1.91 Widowed 3.57* 0.85, 15.01 Marrieda Education High school graduate 1.26 0.52, 3.04 College plus 0.95 0.51, 1.77 Some collegea Household income < $30,000 0.68 0.31, 1.50 $30,000 to $50,000 0.87 0.48, 1.60 > $80,000 1.10 0.60, 1.99 $51,000 to $80,000a Percentage household income contributed by respondent 25% to 50% 0.74 0.35, 1.54 51% to 75% 0.98 0.46, 2.08 > 75%a Number of children in household 1 2.12*** 1.25, 3.59 2 1.15 0.64, 2.06 ≥3 2.72*** 1.29, 5.71 Nonea Military type Enlisted 2.45*** 1.28, 4.68 Officera

Military Period Prevalence

Odds Ratio

Confidence Interval

1.13 1.22 1.66

0.54, 2.36 0.71, 2.08 0.48, 5.72

0.72

0.37, 1.39

6.17† 0.78 4.57**

2.57, 14.86 0.32, 1.92 1.07, 19.48

0.98 0.63

0.37, 2.62 0.31, 1.26

0.54 0.80 1.36

0.22, 1.33 0.41, 1.56 0.68, 2.72

0.57 0.85

0.25, 1.30 0.36, 2.02

2.49*** 1.20 3.34***

1.37, 4.54 0.60, 2.41 1.46, 7.67

2.77***

1.31, 5.85

a. Reference category. *p < .1. **p < .05. ***p < .01. †p ≤ .001.

that this sample of ADM women is more highly educated, has a higher proportion of officers, and has a lower proportion of minority ethnic group membership than ADM women in general.


1083

These demographic differences are also true of the general population of ADM women in the D.C. area in comparison to ADM women in general. A low, but clearly higher, rate of recent IPV (4.5%) was reported in the comparison HMO sample, from the same geographic area with similar rates of employment and education (Jones et al., 1999). The written introduction and initial recruitment, measures, and telephone interview methods for both samples were the same. Even so, the comparisons between the two groups must be interpreted with caution. The reporting period for the HMO sample was 2 years rather than 1, and the women in the HMO sample were assured confidentiality (Jones et al., 1999). In addition, there was a higher proportion of African American women in the civilian sample (48%) than in the military sample (16%), although the rest of the demographics were not significantly different between the two samples. Adult lifetime prevalence of intimate partner physical and/or sexual assault of the ADM women was slightly, but only marginally, lower than the comparison group of civilian women (30% vs. 36%) (Jones et al., 1999). The lifetime prevalence rates are within the range of 25% to 39% reported in most population-based and health care–setting surveys (Coker, Hall-Smith, McKeown, & King, 2000; Dearwater et al., 1998; Jones et al., 1999; Plichta, 1996; Quillian, 1996; Tjaden & Thoennes, 2000). ADM women are on average better educated than the population as a whole and are employed. This ADM sample was particularly well educated and, obviously, more likely to be employed than other samples of U.S. women. Similarly, lifetime abuse (including emotional abuse, but not stalking) in the civilian sample was slightly higher but very similar (44.3% vs. 38.8%) to the military sample. These rates of abuse are also similar to those found in civilian population-based and health care–setting surveys of domestic violence. For instance, Tjaden and Thoennes (2000) reported prevalence among women of stalking by an intimate partner as 4.8%, while Coker et al. (2000) reported that 12.5% of their sample reported only emotional abuse or psychological threats from an intimate partner in their lifetime. Our findings included an additional 8.9% of ADM women reporting only emotional abuse and an additional 6% reporting only stalking. The

1084


stalking finding is the first in a military population and should be investigated further in future research. OVERLAP IN TYPES OF ABUSE

The overlap of emotional, physical, and sexual abuse found in the study is similar to findings in prior IPV research. Between 30% and 45% of civilian women who were physically abused also reported forced sex (Campbell & Soeken, 1999; Tjaden & Thoennes, 2000). Similarly, one third of female veterans reporting physical assault by an intimate partner also reported being sexually assaulted (Murdoch & Nichol, 1995). Nearly all of the physically abused women also reported emotional or psychological abuse in a sample of civilian women (Coker et al., 2000), and Rosen, Parmley, Knudson, & Fancher (2002a) found psychological abuse significantly related to psychological distress in another sample of ADM women. The overlap in types of abuse supports the argument for a broad definition of domestic violence within the military. A high rate of IPV among all types of unmarried ADM women was reported during military service, including those never married (12.6%), separated or divorced (45.5%), and widowed (70.0%). The data demonstrate the need for military policies that include unmarried military service members. The Defense Task Force on Domestic Violence (2001) has recommended the adoption of a broad definition of domestic violence to include the following: the use, attempted use, or threatened use of physical force, violence, a deadly weapon, sexual assault, stalking, or the intentional destruction of property; behavior that has the intent or impact of placing a victim in fear of physical injury; or a pattern of behavior resulting in emotional/psychological abuse, economic control, and/or interference with personal liberty(s) that is directed toward a current or former spouse, a person with whom the abuser shares a child in common, or a current or former intimate partner. RISK MARKERS FOR IPV

The risk factors for IPV found in the study were nearly the same as those found in the demographically and geographically similar


1085

civilian sample (Jones et al., 1999). The middle-aged group (40 to 49) in the ADM sample was associated with a slightly increased risk (adjusted odds ratio = 1.48) for lifetime prevalence of IPV. The increased risk may be correlated to older women having lived longer and, consequently, having more chances to be abused. This age risk marker was not present for IPV during military service or in the civilian sample. In both samples, divorce or separation significantly increased risk. The most protective marital status was being married in the civilian sample (Jones et al., 1999). For women serving in the military, remaining single was more protective. Divorce or separation may precede or follow the IPV, and longitudinal studies suggest that both scenarios occur (e.g., Campbell, Miller, Cardwell, & Belknap, 1994). Risk of homicide and attempted homicide also increases for abused women after separation in civilian samples (Browne, Williams, & Dutton, 1998; Campbell, Webster, Koziol-McLain, & McFarlane et al., in press). The other difference between the military and the civilian samples was related to ethnicity. Significantly more African American ADM women reported lifetime prevalence of IPV than did White ADM women (44% vs. 26%), but only 20% of the African American women said they were victimized during military service, in comparison to almost 28% of the White women. In the civilian sample, African American ethnicity slightly (odds ratio = 1.34, p < .05) increased the risk for IPV, even controlling for education and income. Because there were no significant effects of ethnicity in the military multivariate analysis, and because there were relatively few African American ADM women in the sample, we cannot be sure of this effect. In fact, the proportion of African American ADM women in the sample (16%) is far less than the proportion of African American ADM women in the Washington, D.C., area (34%). However, if the same result was observed in a larger, more representative sample, it would suggest that military service appears to be a way for African American women to gain protection against IPV. That the apparent increased risk for lifetime IPV among African American women compared to White women became nonsignificant in the multivariate analysis supports other findings (e.g., Dearwater et al., 1998; Jones et al., 1999) that increased risk of abuse associated with minority ethnicity tends to decrease or totally disappear when income and education are controlled. More research is needed that examines within-

1086


ethnic group influences as well as across-group analyses to better analyze and understand the influences of culture and ethnicity on risk and responses to IPV. As stated above, further investigation with a larger sample into the possible protective factor of military service for women of color is needed. The military affords many career and educational options to minority ethnic group women and an excellent opportunity for researchers to examine IPV among a well-educated and fully employed group of women of color. A significant risk associated with rank (enlisted vs. officer) for both lifetime and during military service IPV also was demonstrated. The finding was also documented in cases of domestic violence reported to military authorities (Caliber Associates, 1996) and in a survey of female veterans (Coyle & Wolan, 1996) but was not found in at least one other study (Mercier & Mercier, 2000). This is the first reported difference in IPV by rank in a controlled multivariate analysis of data from ADM women. The significant difference between the enlisted and officer ranks probably subsumed the bivariate differences in education and income. Although IPV was more likely to occur among enlisted women, 14.5% of female officers also reported physical and/or sexual assault by an intimate partner while in military service. The absence of confidentiality in the military system fosters the reluctance of ADM victims to report abuse to military authorities. Any such victimization is reported to the woman’s commanding officer in the current system. This lack of confidentiality may be even more of an issue for abused officers than enlisted women. Although victimization should not adversely affect a woman’s military career, there is widespread concern that it will do so. This lack of confidentiality for victims has been recognized as a serious issue that needs to be addressed in military policies on domestic violence (Defense Task Force on Domestic Violence, 2002). Three different groups of perpetrators were abusing these military women. The largest percentage (43%) also was ADM. Another large proportion (38%) was retired military. Of the active duty and retired military perpetrators, 63% were enlisted personnel, and 37% were officers. The percentages mirror the 60% enlisted and 40% officer proportion across the military. Little research has been conducted on officers as perpetrators of spousal abuse in the military, with the Heyman and Neidig (1999) study


1087

not reporting rank and the Rosen, Knudson, et al. (2002) and Rosen, Parmley, et al. (2002a, 2002b) studies restricted to enlisted personnel. The general assumption has been that military officers would not be domestic violence perpetrators (e.g., Caliber Associates, 1996). The possible interventions for domestic violence are partially dependent on the military status of the perpetrator. ADM perpetrators can have a number of sanctions applied to them, including being ordered to batterer intervention programs with severe career-threatening sanctions for not attending regularly (Defense Task Force on Domestic Violence, 2002). The military is one of the only arenas where reassaulting one’s intimate partner can result in less desirable work assignments, blockage of workplace promotions, demotions, or even the end of a career, powerful motivators for change. Civilian perpetrators present considerable challenges, necessitating memoranda of understanding between military bases and civilian courts and close working relationships with civilian domestic violence programs (Defense Task Force on Domestic Violence, 2001). Retired military perpetrators are a civilian group that has not been considered in most recommendations. Increased batterer intervention programs in the Veterans Affairs health services system would be an appropriate policy recommendation. The bivariate risk associated with women making considerably more income than their intimate partners also became nonsignificant in the multivariate analysis, although there was still more risk than for those with equal or less income than their partners. This is a marker of status inconsistency, where the female partner is of a higher educational, occupational, or income status than her male partner. Status inconsistency is also a risk factor for IPV in civilian studies (e.g., Tjaden & Thoennes, 2000). The final significant risk factor for IPV in both the multivariate and bivariate analysis was children, with three or more children presenting the strongest risk (odds ratio = 3.34). The finding is specific to this military sample and has not been reported elsewhere. The stress of responsibilities associated with parenting combined with the demands of a military career for an ADM female have been noted, and this possible risk factor should be investigated more fully in future studies. Meanwhile, the New Parent Support Program in the military should pay particular attention to the possibility of IPV among the families it serves

1088


(Defense Task Force on Domestic Violence, 2001). Additional programs, such as Families in Need of Assistance in the Navy, should also be encouraged and strengthened (Defense Task Force on Domestic Violence, 2002). LIMITATIONS

Clearly, the low response rate is a serious limitation to the study, compromising generalizabilty and sample size and, thereby, power to detect differences. Anonymous surveys are preferable for determining prevalence of IPV among ADM women. We strongly urge that anonymous surveys be conducted of IPV victimization as well as perpetration among all branches of the military. Such surveys are the only way to enable the military to assess the true (vs. reported) prevalence of IPV and monitor increases and decreases as a means of assessing the efficacy of its responses, including prevention and intervention programs. In studies such as ours that need unique identifiers to determine consequences and costs of IPV, participants cannot be anonymous. However, complete confidentiality can be assured in such research. We therefore also urge that participants in all research on domestic violence be assured complete confidentiality and that the military examine its research processes to try to diminish barriers to participation. CONCLUSIONS In spite of our limitations, the study demonstrates that a substantial proportion of ADM women reports being victimized by IPV during military service. The findings in the two samples demonstrate that IPV is equally as important for military women as for civilian women. Given the well-established negative physical and mental health effects of IPV, including post-traumatic stress disorder for military members (Engel et al., 1993; Murdoch & Nichol, 1995; Sadler et al., 2000) as well as civilians (e.g., Campbell et al., 2002; Plichta, 1996), these results are particularly distressing because of the substantial demands on military personnel to be fit for duty at all times. This becomes an issue of military readiness. The risk factors for IPV among ADM women (enlisted rank, lowest income group, primary income provider, and/or


1089

responsibilities of parenthood) can be used as potential markers to help with early detection of IPV. Military commanders and coworkers who are alert to IPV risk factors can be better positioned to mobilize assistance for resolution of the problem before it has a negative impact on women’s performance and personal safety. Current military efforts to address the problem of IPV should continue, be strengthened, and be rigorously evaluated (Defense Task Force on Domestic Violence, 2001, 2002). The military has initiated some promising strategies to address risk factors identified in this study, such as the New Parent Support Program and Families in Need of Assistance (Caliber Associates, 1996). More strategies are needed that are consistent with both military traditions and the reality of IPV. More attention and resources directed to the issue as well as improvement of existing policies and programs to help abused women end the abuse are urgently needed to maintain military readiness and retain valuable, skilled ADM personnel. REFERENCES Browne, A., Williams, K. R., & Dutton, D. C. (1998). Homicide between intimate partners. In M. D. Smith & M. Zahn (Eds.), Homicide: A sourcebook of social research (pp. 149-164). Thousand Oaks, CA: Sage. Caliber Associates. (1996). Final report of the study of spousal abuse in the Armed Forces. Washington, DC: Author. Campbell, J. C., Miller, P., Cardwell, M. M., & Belknap, R. A. (1994). Relationship status of battered women over time. Journal of Family Violence, 9, 99-111. Campbell, J. C., Snow-Jones, A., Dienemann, J. A., Kub, J., Schollenberger, J., O’Campo, P., et al. (2002). Intimate partner violence and physical health consequences. Archives of Internal Medicine, 162, 1157-1163. Campbell, J. C., & Soeken, K. (1999). Forced sex and intimate partner violence: Effects on women’s health. Violence Against Women, 5, 1017-1035. Campbell, J., Webster, D., Koziol-McLain, J., McFarlane, J., Block, C., Campbell, D. W., et al. (in press). Risk factors for intimate partner femicide. American Journal of Public Health. Coker, A., Hall-Smith, P., McKeown, R. E., & King, M. (2000). Frequency and correlates of intimate partner violence by type: Physical, sexual, and psychological battering. American Journal of Public Health, 90, 553-559. Coyle, B. S., & Wolan, D. L. (1996). The prevalence of physical and sexual abuse in women veterans seeking care at a Veterans’ Affairs medical center. Military Medicine, 161, 588593. Dearwater, S. R., Coben, J. H., Campbell, J. C., Nah, G., Glass, N. E., McLoughlin, E., et al. (1998). Prevalence of intimate partner abuse in women treated at community hospital emergency departments. Journal of the American Medical Association, 280, 433-438.

1090


Defense Task Force on Domestic Violence. (2001). Defense Task Force on Domestic Violence initial report. Washington, DC: Author. Defense Task Force on Domestic Violence. (2002). Defense Task Force on Domestic Violence second annual report. Washington, DC: Author. Department of Defense. (1992). DOD Directive 6400.1. Family Advocacy Program (FAP). Washington, DC: Author. Department of Defense. (2001). Statement on domestic violence in military. Washington, DC: Author. Ellsberg, M. C. (2000). Candies in Hell: Research and action on domestic violence against women in Nicaragua. Umea, Sweden: Umea University. Engel, C. C., Engle, A. L., Campbell, S. J., McFall, M. E., Russo, J., & Katon, W. (1993). Posttraumatic stress disorder symptoms and precombat sexual and physical abuse in Desert Storm veterans. Journal of Nervous and Mental Disease, 181, 683-688. Greenfield, L. A., Rand, M. R., Craven, D., Klaus, P. A., Perkins, C. A., Ringel, C., et al. (1998). Violence by intimates: Analysis of data on crimes by current or former spouses, boyfriends and, girlfriends. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. Heyman, R. E., & Neidig, P. H. (1999). A comparison of spousal aggression prevalence rates in U.S. Army and civilian representative samples. Journal of Consulting & Clinical Psychology, 67, 239-242. Jones, A. S., Campbell, J. C., Schollenberger, J., O’Campo, P., Dienemann, J. A., Gielen, A., et al. (1999). Annual and lifetime prevalence of partner abuse in a sample of female HMO enrollees. Women’s Health Issues, 9, 295-305. McCarroll, J. H., Newby, J. H., Thayer, L. E., Norwood, A. E., Fullerton, C. S., & Ursano, R. J. (1999). Reports of spouse abuse in the U.S. Army central registry (1989-1997). Military Medicine, 164, 77-84. Mercier, P. J., & Mercier, J. D. (Eds.). (2000). Battle cries on the home front: Violence in the military family. Springfield, IL: Charles C Thomas. Mollerstrom, W. W., Patchner, M. A., & Milner, J. S. (1992). Family violence in the Air Force: A look at offenders and the role of the Family Advocacy Program. Military Medicine, 157, 371-374. Murdoch, M., & Nichol, K. L. (1995). Women veterans’ experiences with domestic violence and with sexual harassment while in the military. Archives of Family Medicine, 4, 411-418. Plichta, S. B. (1996). Violence and abuse: Implications for women’s health. In M. M. Falik & K. S. Collins (Eds.), Women’s health: The Commonwealth Survey (pp. 237-272). Baltimore: Johns Hopkins University Press. Quillian, J. P. (1996). Screening for spousal or partner abuse in a community health setting. Journal of the American Academy of Nurse Practitioners, 8(4), 155-160. Rosen, L. N., Knudson, K. H., Brannen, S. J., Fancher, P., & Kilgore, T. E. (2002). Intimate partner violence among US Army soldiers in Alaska: A comparison of reported rates and survey results. Military Medicine, 167, 688-691. Rosen, L. N., Parmley, A. M., Knudson, K. H., & Fancher, P. (2002a). Gender differences in the experience of intimate partner violence among active duty military soldiers. Military Medicine, 167, 959-963. Rosen, L. N., Parmley, A. M., Knudson, K. H., & Fancher, P. (2002b). Intimate partner violence among married male US Army soldiers: Ethnicity as a factor in self-reported perpetration and victimization. Violence and Victims, 17, 607-622. Sadler, A. G., Booth, B. M., Nielson, D., & Doebbeling, B. N. (2000). Health related consequences of physical and sexual violence: Women in the military. Obstetrics and Gynecology, 96, 473-480.


1091

Saltzman, L. E., Fanslow, J. L., McMahon, P. M., & Shelley, G. A. (1999). Intimate partner violence surveillance: Uniform definitions and recommended data elements, Version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Soeken, K., Parker, B., McFarlane, J., & Lominak, M. C. (1998). The Abuse Assessment Screen: A clinical instrument to measure frequency, severity, and perpetrator of abuse against women. In J. C. Campbell (Ed.), Empowering survivors of abuse: Health care for battered women and their children (pp. 195-203). Thousand Oaks, CA: Sage. Straus, M. A. (1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT) Scales. Journal of Marriage and the Family, 41, 75-88. Straus, M. A., & Gelles, R. J. (1990). Physical violence in American families: Risk factors and adaptations to violence in 8,145 families. New Brunswick, NJ: Transaction Publishing. Tjaden, P., & Thoennes, N. (2000). Full report of the prevalence, incidence, and consequences of violence against women. Washington, DC: National Institute of Justice.

Jacquelyn C. Campbell, Ph.D., RN, is the Anna D. Wolf endowed professor and associate dean for faculty affairs at the Johns Hopkins School of Nursing, with a joint appointment in the Bloomberg School of Public Health. She has been conducting advocacy policy work and research in the area of domestic violence since 1980. Mary A. Garza, Ph.D., MPH, is a postdoctoral fellow at Johns Hopkins University, Bloomberg School of Public Health, in the Department of Epidemiology. Her research interests are in minority health, domestic violence, and program evaluation. Andrea Carlson Gielen, Sc.D., is professor of social and behavioral sciences and deputy director for injury research and policy at the Johns Hopkins Bloomberg School of Public Health. Gielen’s research focuses on applying behavior change theory and health promotion interventions to the prevention of domestic violence and childhood injury. Patricia O’Campo, Ph.D., is a social epidemiologist and professor at the Johns Hopkins Bloomberg School of Public Health. For the past 12 years, she has conducted research and published numerous articles concerning women’s health. Specific areas of interest include measurement of intimate partner violence (IPV), neighborhood characteristics as risk factors for IPV, and IPV and reproductive health. Joan Kub, Ph.D., is an assistant professor in the Johns Hopkins School of Nursing with joint appointments in the Bloomberg School of Public Health and the John Hopkins School of Medicine. Her research has focused on domestic violence and substance abuse. She is currently a co-investigator on a violence prevention project focused on dating violence. Jacqueline Dienemann, Ph.D., RN, is an adjunct professor at the University of North Carolina at Charlotte College of Health and Human Services, adjunct associate professor at the Johns Hopkins University School of Nursing, and a private consultant on outcomes for family violence programs.

1092


Alison Snow Jones holds a Ph.D. in health economics and is an assistant professor in the Department of Public Health Science at Wake Forest University School of Medicine. Her research has focused on women’s and children’s health, alcoholism and alcohol abuse, and domestic violence. Eiman Jafar, Ph.D., RN, is senior project coordinator at Johns Hopkins University School of Nursing. Jafar holds a master’s degree in nursing from the University of Iowa and a Ph.D. from the University of Wisconsin–Milwaukee.