1. Gerlock, April A. PhD, ARNP, PMHNP-BC, PMHCNS-BC
  2. Grimesey, Jackie L. PhD
  3. Pisciotta, Alisa K. MSW
  4. Harel, Ofer PhD


Background: Men seeking care for posttraumatic stress disorder (PTSD) are believed to have high rates of relationship conflict and intimate partner violence (IPV). But little is known about screening for IPV perpetration in this population.


Objective: In phase one of a two-phase study of male veterans treated for PTSD, the primary objective was to determine how many veterans' records showed documentation that they'd been screened for IPV perpetration. The secondary objective was to count the total number of screenings and to determine whether an initial screening affected future screenings.


Methods: For this retrospective cohort study, a stratified random sample of 10% (N = 507) of all male veterans receiving treatment for PTSD at a U.S. Department of Veterans Affairs health care facility in a five-year period (November 2002 to November 2007) was selected and more than 70,000 progress notes were reviewed. The presence or absence of a documented screening for IPV perpetration in each record was noted and a Spearman rank correlation test to determine the relationship between the documentation of a first screening and future screenings was performed.


Results: Of the 507 records examined, 120 (24%) showed documentation of screening for IPV perpetration. Of those, 73 (61%) showed positive results for IPV perpetration, and 61 (51%) showed more than one screening. Documentation of screening was most likely to have occurred at the veteran's initial appointment (71%) and in an outpatient mental health setting (72%); IPV perpetration was determined most often as the result of a provider's inquiry (45%). There was a total of 415 screenings, including 356 in records in which there was more than one screening. The documentation of a single screening for IPV perpetration was significantly correlated with the documentation of subsequent screenings and with IPV perpetration determination (Spearman rank correlation = 0.611, P < 0.001). Also, veterans with documented IPV perpetration and high rates of relationship conflict accessed the health care system twice as often as those without such documentation.


Conclusions: In the sample analyzed, fewer than a quarter of male veterans with PTSD had a documented screening for IPV perpetration. Also, because those identified as IPV perpetrators accessed the health care system at a higher rate than those not so identified and repeated screenings were associated with a higher rate of IPV perpetrator determinations, health care providers should be made more aware of opportunities for screening for IPV in this population.


Keywords: documentation, intimate partner violence, partner abuse, perpetrators, posttraumatic stress disorder, screening, veterans


Article Content

The impact of intimate partner violence (IPV) on victims' health has been widely acknowledged, but less is known about its effects on the health of the perpetrators. Intimately violent men often seek general medical care, at times for injuries resulting from their perpetration of IPV.1, 2 In recent years, health care providers have been encouraged to screen male patients for IPV perpetration as well as victimization,3 and when a provider asks, male perpetrators sometimes acknowledge their violent behavior.4, 5 There's limited research that specifically looks at providers' documentation of screening for IPV victimization,6, 7 and we know of no retrospective research examining documentation practices for screening for IPV perpetration.


The American Medical Association and the Joint Commission recommend routine screening for IPV victimization in emergency and primary care settings.8 Researchers have reported conflicting findings in studies examining whether or not screening actually decreases rates of IPV victimization,9-15 yet some research has demonstrated that screening and intervention result in an increase in safety-promoting behaviors in victims.16, 17 And an Institute of Medicine report released this year, Clinical Preventive Services for Women: Closing the Gaps, recommends "screening and counseling for interpersonal and domestic violence" as one of eight preventive services for women.18 For these reasons, we believe that IPV screening should be considered part of a prevention model that includes intervention when IPV victimization is identified. And while there's currently no mandate for screening for IPV perpetration, it's now considered an important health care issue and screening is starting to be recommended.3, 6 A general screening of men and women is likely to capture both IPV perpetration and victimization-but clinicians may be unprepared to respond to a perpetrator's revelation of abuse.


Veterans returning from serving in Iraq and Afghanistan have high rates of posttraumatic stress disorder (PTSD), major depressive disorder and depressive symptoms, and traumatic brain injury.19 Suicide is a risk among veterans20 and has been found to exist whether or not veterans were affiliated with the U.S. Department of Veterans Affairs (VA), making it a concern for both VA and non-VA facilities.21 A risk of suicide may be associated with a risk of homicide when IPV is also present.22-25


Historically, veterans with PTSD have been found to have a higher incidence of IPV perpetration than veterans without PTSD.26-28 Of the veterans and active duty military personnel attending a batterers' intervention program, those with PTSD had a greater frequency and intensity of IPV perpetration than those without PTSD.29 Close to half (53%) of the veterans returning from Iraq and Afghanistan who were receiving care at a VA clinic "endorsed" at least one act of physical aggression in the prior four months.30 Of veterans referred for a mental health evaluation, 75% reported family readjustment problems.31 Among those with "current or recent past partners," 60% reported mild-to-moderate IPV within the previous six months.


The setting for our study was a VA health care system in the Pacific Northwest, but many reservists and National Guard members returning from Iraq and Afghanistan receive care from civilian providers. We sought to examine documentation practices in order to provide information on the settings in which IPV perpetration is identified and by whom. Also, with more and more health care systems using computerized records, we sought to determine whether documentation can affect future assessments and identification of IPV perpetration.


We undertook this study to answer the following five questions about veterans seeking treatment for PTSD:


* Was an IPV perpetration screening or assessment documented in the medical record? (We defined IPV screening as documentation that IPV perpetration is or isn't present and IPV assessment as documentation that an assessment was conducted to determine the presence of other factors, such as harm risk and interventions.)


* If yes, when was it conducted, in which clinic, and by whom?


* If yes, how was the presence or absence of IPV determined (through provider inquiry, self-disclosure, third-party report)?


* Is there a difference in health care access for veterans in whom IPV perpetration has been documented?


* Does documentation of IPV perpetration affect the detection and documentation of IPV in later health care visits?




In this two-phase study, we sought to examine variables associated with the accurate detection of IPV perpetration by providers at our facility. Phase one of the study, reported here, involves a cohort-design, retrospective record review of veterans receiving treatment for PTSD. (Phase two involves interviews and surveys of more than 400 male veterans and their partners, as well as surveys of mental health providers about their awareness of abuse in the veterans' relationships.) The study was approved by the University of Washington's institutional review board.


Sample. For phase one, we conducted a five-year retrospective review of the computerized records of a stratified random sample of veterans in our VA system. We stratified by sex and diagnosis; only men receiving treatment for PTSD from November 1, 2002, to November 1, 2007, were eligible for inclusion. From a list of all male veterans receiving PTSD treatment at several clinics-roughly 5,600 men-we chose every 10th name to arrive at our full sample of 507 patient records.


Initial data gathering. Two of us (JLG and AKP) began to track the number of progress notes (each note corresponded to one health care encounter). For the last 403 records reviewed, we identified 59,524 unique health care visits. We extrapolated that number to the full sample of 507 records and estimated that there were 74,885 notes in the sample.


Data algorithm. Because providers used various terms, such as domestic violence, partner abuse, or intimate partner violence, in the notes reviewed, and no screening or assessment tool existed to document the presence of either IPV victimization or perpetration during the study period, we developed a decision tree (see Figure 1) to ensure that we captured all documented IPV screenings and assessments.

Figure 1 - Click to enlarge in new windowFigure 1. The IPV Perpetration Coding Decision Tree

IPV is defined as physical or sexual violence against an intimate partner or behaviors that establish a threat of such violence (such as violence against others or self, stalking, display of weapons) AND a pattern of psychologically abusive and coercive behaviors toward an intimate partner. Psychological abuse can include name calling, threats, intimidation, financial control, abuse of children or pets, and blaming the victim.32


Categorizing data. We developed the first three research questions to identify patterns in documentation practices according to the provider's setting and discipline, as well as how and when screening or assessment for IPV perpetration was conducted (that is, at an initial or follow-up visit). During the pilot phase of record review, we added two questions: one to track health care access among veterans in whom IPV perpetration was documented and one to determine whether a single documentation of IPV perpetration affected future screening, assessment, and documentation.


When no documentation of an IPV screening or assessment was present, the record was coded as "No-Doc" (no documentation). When documentation of an IPV screening or assessment was present (or the note indicated that the provider had discussed the use of physical force in the veteran's intimate relationship or the consequences of it, such as an order for protection or jail time), the record was coded as "Yes-Doc" (yes documentation). Within the Yes-Doc group we created two further categories: "Yes-Doc: no IPV perpetration present" and "Yes-Doc: yes IPV perpetration present." A third category, created for instances when we couldn't code the record as Yes-Doc or No-Doc, included descriptions of relationship conflict (for example: "For the first years of our relationship, I was brutal to my wife and kids"). Notes in this category were vague in details such as the use of physical force, the target of the physical force, or the chain of events. This third category was called the "high relationship conflict" group.


In the rare case that a male veteran was identified as an IPV victim, it wasn't counted as documentation for the purposes of our study unless IPV perpetration was also noted.


We had ongoing discussions about how providers determined the presence of IPV perpetration, in order to ensure consistency in both the review and our categorization. After the sample was randomly selected and the records reviewed, one of us (AAG) completed a blinded review of a random selection of 10% of the records to establish an interrater agreement of 90%.


Statistical analyses were conducted using SPSS software, version 15.0.



Question 1: Was an IPV perpetration screening or assessment documented in the medical record? Of the 507 records reviewed, 363 (71%) showed no documentation of screening or assessment of IPV perpetration, and 120 (24%) showed such documentation-59 (49%) showing a single documentation and 61 (51%) showing more than one. Twenty-four (5%) records were unclear and were categorized as "high relationship conflict." Within the 120 records that showed documentation, 73 (61%) indicated that IPV perpetration had occurred. There were 415 documented screenings or assessments, including 356 in records in which there was more than one screening.


Question 2: If an IPV perpetration screening or assessment was documented, when was it documented, in which clinic, and by whom? Within the 120 records containing documentation of a screening or assessment of IPV perpetration, documentation occurred most often at the initial appointment (71%, n = 85), with the remainder at a follow-up visit (29%, n = 35) (see Table 1). The documentation took place at a range of clinics, including mental health, primary care, and dental. Most of it occurred in the outpatient (72%, n = 87) and inpatient (18%, n = 22) mental health settings. The providers most frequently documenting were mental health counselors (38%, n = 46) and physicians (18%, n = 22) (see Table 2).

Table 1 - Click to enlarge in new windowTable 1. When and Where an IPV Screening or Assessment Was Documented and How IPV Perpetration Was Determined (n = 120)
Table 2 - Click to enlarge in new windowTable 2. Who Documented the IPV Perpetration Screening or Assessment (n = 120)

Question 3: If an IPV perpetration screening or assessment was documented, how was the presence or absence of IPV perpetration determined (through provider inquiry, self-disclosure, third-party report)? In the Yes-Doc group, it was difficult to discern whether a determination of IPV perpetration was the result of the provider's inquiry or the veteran's self-disclosure. For this reason, 41 of the records were coded as "unclear." Most determinations of IPV perpetration were made as the result of provider inquiry (45%, n = 54), followed by veteran's self-disclosure (13%, n = 16) (see Table 1). The remainder were based on reports by a third party (8%, n = 9), such as the veteran's wife or a community agency. All third-party reports were the result of legal involvement.


Question 4: Is there a difference in health care access for veterans in whom IPV perpetration has been documented? Because each note indicated a health care contact, it was possible to track health care access. An analysis of variance (ANOVA) was performed among groups; results showed that veterans in the Yes-Doc: Yes-IPV and high-relationship-conflict groups were accessing the health care system at twice the rate of those in the Yes-Doc: No-IPV or No-Doc groups (F = 8.529, df = 405, P < 0.001).


Question 5: Does documentation of IPV perpetration affect the detection and documentation of IPV in later health care visits? A Spearman rank correlation test was performed to compare records that had one documented screening or assessment (n = 59) with records that had more than one (n = 61). The documentation of a single screening was significantly correlated with documentation of future screenings and with IPV perpetration determination (Spearman rank correlation = 0.611, P < 0.001). Records containing a documentation that had been "copied and pasted" from a previous note weren't counted.


Other findings. When we looked at records showing documentation of more than one screening or assessment for IPV perpetration, we found in some cases that the results had changed. For example, within the 73 records that showed positive results for IPV perpetration, seven had at one point been coded "no IPV," six had been coded "vague," and four had been coded both "no IPV" and "vague." (There's a chance that these changing responses reflect the unreliability of perpetrators' self-disclosure of their violent behavior, as noted in the discussion of limitations, below.)


We also found that veterans who had more than one documented screening or assessment for IPV perpetration tended to self-disclose more often than those who had only one documented screening or assessment and to have their perpetration of IPV identified on a follow-up visit, identified in inpatient and outpatient mental health settings, and documented by a physician. In the records showing multiple screenings or assessments, the number of additional screenings or assessments ranged from two to 18 per record.



While most records didn't show documentation of a screening or an assessment for IPV perpetration, many provided rich descriptions of relationships, indicating that patients gave staff opportunities to ask about IPV. Despite these opportunities, staff may have failed to ask further questions to determine the presence of IPV and factors that affect dangerousness (such as suicidal or homicidal ideation). Although such questions may have been asked, without documentation the responses can't be shared with other providers. Continuity of care is a significant issue in health care systems that provide a range of services, as well as in those that have electronic medical records, which make information available to all providers in any setting.


Screening for IPV victimization can be a challenge for providers,33 who may worry about identifying IPV incorrectly. But when either suicidal or homicidal ideation is present, not screening for IPV can be dangerous.22-25 Our findings suggest that documenting the results of a screening or assessment can have an impact on other providers at later health care visits-suggesting in some cases that one provider asking about IPV perpetration can set off a cascade of screenings at future appointments.


Our findings also suggest that health care providers may improve the detection of IPV perpetration in veterans with PTSD by screening or assessing for it more than once, in particular at follow-up visits. It has been our experience in working with this population that patients reveal sensitive information about themselves, such as substance abuse or IPV, over time as part of an ongoing therapeutic relationship. We believe this merits further study.


The estimated annual costs of care given to victims of IPV range from nearly $4 billion to more than $7.6 billion.34 The cost of care given to perpetrators is unknown. The high rate of health care use by men documented as perpetrators in this study indicates a need for further study. While these veterans were randomly selected from PTSD programs, the study was able to track all of their health care visits by reviewing every progress note. Access of all types of clinics was significantly higher for the group with documented IPV perpetration and high relationship conflict. Thus, the health care system may also be seriously impacted by the perpetrators' increased need for all services. IPV determination offers an opportunity for providers to discuss with patients the impact of IPV on their health and the health of others, as well as assess for risk factors that could affect the safety of the veteran and family members (or others), while providing critical information.


We cannot say for certain why the men with documentation of IPV and high relationship conflict were accessing care at twice the rate of the men with no documented IPV or documented No-IPV. One theory is that IPV perpetration has significant health consequences for the perpetrator in the form of physical (lacerations, cardiovascular disorders) and psychological (depression, suicidality) illnesses.2


Even without standardized protocols, documentation of an IPV perpetration screening or assessment existed in 24% of the records reviewed, with 61% of those records indicating the presence of IPV perpetration. In a similar study of physicians documenting IPV victimization in women known to be experiencing IPV,33 only about 15% of women had documented screenings-despite physician awareness of the ongoing study-demonstrating that VA health care providers' assessment for IPV perpetration may already exceed that of their civilian counterparts.


Limitations. Our study was limited to male veterans enrolled in PTSD treatment at a VA facility; therefore, findings may not be generalizable to other populations or to settings that don't have computerized medical records. Also, this study took place within a health care system with a history of educating providers on IPV and providing interventions to batterers for several decades, ending in 2000. In addition, all subjects received a full mental health assessment, and they were most likely to be screened or assessed for IPV perpetration by mental health care providers. If the sample had been drawn from a primary care or other specialty clinic, the rate of documentation of IPV screening or assessment may have been significantly lower.


Another limitation may have been the unreliability of IPV perpetrators in reporting their own abusive behavior. Perpetrators minimize and deny the abuse they commit, lie about it, and justify it.32, 35 Among a group of active-duty military personnel and military veterans, this pattern of minimizing abuse was noted in a matched-pair analysis of reports of IPV from men in batterer's intervention and their partners.36 (For this reason, corroborating information from the victim is often sought37; however, we did not seek it for this study.)


Also, retrospective record reviews don't necessarily reflect providers' awareness of health care issues or interventions.



Our findings show a low rate of screening and assessment for IPV perpetration in male, treatment-seeking veterans with PTSD-a population believed to have high rates of relationship conflict. The early detection of IPV should be the goal of screening protocols, followed by the use of assessments and targeted interventions (see "Ask a Few More Questions" in this issue). Documented IPV perpetration can alert other providers, offering an opportunity for further assessment of its impact on the patient and his family members.


Despite its limitations, this study highlights the importance of providers being prepared to conduct screenings for IPV perpetration in this population, as well as to perform further assessment and to know how to respond when perpetration is detected.




1. Coben JH, Friedman DI. Health care use by perpetrators of domestic violence. J Emerg Med 2002;22(3):313-7. [Context Link]


2. Gerlock AA. Health impact of domestic violence. Issues Ment Health Nurs 1999;20(4):373-85. [Context Link]


3. Kimberg LS. Addressing intimate partner violence with male patients: a review and introduction of pilot guidelines. J Gen Intern Med 2008;23(12):2071-8. [Context Link]


4. Jaeger JR, et al. Screening male primary care patients for intimate partner violence perpetration. J Gen Intern Med 2008;23(8):1152-6. [Context Link]


5. Jaeger JR. Results of an IPV screening program for males in the health care setting. Boston: Family Violence Prevention Fund; 2004. [Context Link]


6. Cronholm PF. Intimate partner violence and men's health. Prim Care 2006;33(1):199-209. [Context Link]


7. Sims C, et al. Detecting intimate partner violence: more than trauma team education is needed. J Am Coll Surg 2011;212(5):867-72. [Context Link]


8. American Medical Association. American Medical Association Diagnostic and Treatment Guidelines on Domestic Violence. Arch Fam Med 1992;1(1):39-47. [Context Link]


9. Chapin MG, Mackie CF. Research evidence to update practice guidelines for domestic violence screening in military settings. Mil Med 2007;172(7):ii-iv. [Context Link]


10. Family Violence Prevention Fund. Health screening study misleading, experts say. Speaking Up 2009 Aug 17. [Context Link]


11. Family Violence Prevention Fund. The Family Violence Prevention Fund's review of the U.S. Preventative Services Task Force draft recommendation and rationale statement on screening for family violence. San Francisco: Futures Without Violence; 2004. [Context Link]


12. Lachs MS. Screening for family violence: what's an evidence-based doctor to do? Ann Intern Med 2004;140(5):399-400. [Context Link]


13. MacMillan HL, et al. Screening for intimate partner violence in health care settings: a randomized trial. JAMA 2009;302(5):493-501. [Context Link]


14. Moracco KE, Cole TB. Preventing intimate partner violence: screening is not enough. JAMA 2009;302(5):568-70. [Context Link]


15. U. S. Preventive Services Task Force. Screening for family and intimate partner violence: recommendation statement. Ann Intern Med 2004;140(5):382-6. [Context Link]


16. McFarlane J, et al. An intervention to increase safety behaviors of abused women: results of a randomized clinical trial. Nurs Res 2002;51(6):347-54. [Context Link]


17. McFarlane J, et al. Increasing the safety-promoting behaviors of abused women. Am J Nurs 2004;104(3):40-50. [Context Link]


18. Committee on Preventive Services for Women, Institute of Medicine. Clinical preventive services for women: closing the gaps. Washington, DC: National Academies Press; 2011. [Context Link]


19. Tanielian T, Jaycox LH, editors. Invisible wounds of war: psychological and cognitive injuries, their consequences, and services to assist recovery. Santa Monica, CA: Rand Center for Military Health Policy Research; 2008. [Context Link]


20. McCarthy JF, et al. Suicide mortality among patients receiving care in the Veterans Health Administration health system. Am J Epidemiol 2009;169(8):1033-8. [Context Link]


21. Kaplan MS, et al. Suicide among male veterans: a prospective population-based study. J Epidemiol Community Health 2007;61(7):619-24. [Context Link]


22. Hobart M. Honoring their lives, learning from their deaths: findings and recommendations from the Washington State Domestic Violence Fatality Review. Seattle: Washington State Coalition Against Domestic Violence; 2000 Dec. [Context Link]


23. Hobart M. Tell the world what happened to me: findings and recommendations from the Washington State Domestic Violence Fatality Review. Seattle: Washington State Coalition Against Domestic Violence; 2002 Dec. [Context Link]


24. Karch DL, et al. Surveillance for violent deaths-National Violent Death Reporting System, 16 States, 2005. MMWR Surveill Summ 2008;57(3):1-45. [Context Link]


25. Starr K, Fawcett J. If I had one more day[horizontal ellipsis] findings and recommendations from the Washington State Domestic Violence Fatality Review. Seattle; 2006 Dec. Fatality review reports; [Context Link]


26. Glenn DM, et al. Violence and hostility among families of Vietnam veterans with combat-related posttraumatic stress disorder. Violence Vict 2002;17(4):473-89. [Context Link]


27. Orcutt HK, et al. Male-perpetrated violence among Vietnam veteran couples: relationships with veteran's early life characteristics, trauma history, and PTSD symptomatology. J Trauma Stress 2003;16(4):381-90. [Context Link]


28. Prigerson HG, et al. Population attributable fractions of psychiatric disorders and behavioral outcomes associated with combat exposure among US men. Am J Public Health 2002;92(1):59-63. [Context Link]


29. Gerlock AA. Domestic violence and post-traumatic stress disorder severity for participants of a domestic violence rehabilitation program. Mil Med 2004;169(6):470-4. [Context Link]


30. Jakupcak M, et al. Anger, hostility, and aggression among Iraq and Afghanistan War veterans reporting PTSD and subthreshold PTSD. J Trauma Stress 2007;20(6):945-54. [Context Link]


31. Sayers SL, et al. Family problems among recently returned military veterans referred for a mental health evaluation. J Clin Psychiatry 2009;70(2):163-70. [Context Link]


32. Warshaw C, Ganley AL. Understanding domestic violence. In: Lee D, et al., editors. Improving the health care response to domestic violence: a resource manual for health care providers. 2nd ed. San Francisco: Family Violence Prevention Fund; 1996. p. 15-42. [Context Link]


33. Coker AL, et al. Missed opportunities: intimate partner violence in family practice settings. Prev Med 2002;34(4):445-54. [Context Link]


34. National Center for Injury Prevention and Control. Costs of intimate partner violence against women in the United States. Atlanta: Centers for Disease Control and Prevention; 2003 Mar. [Context Link]


35. Pence E, Paymar M. Education groups for men who batter: the Duluth model. New York: Springer Publishing Company; 1993. [Context Link]


36. Gerlock AA. A profile of who completes and who drops out of domestic violence rehabilitation. Issues Ment Health Nurs 2001;22(4):379-400. [Context Link]


37. Edelson JL, Brygger MP. Gender differences in reporting of battering incidences. Fam Relat 1986;35(3):377-82. [Context Link]


For 25 additional continuing nursing education articles on research topics, go to