Mothers' and Health Care Providers' Perspectives on Screening for Intimate Partner Violence in a Pediatric Emergency Department

Objective  To determine the attitudes, feelings, and beliefs of mothers and pediatric emergency department health care providers toward routine intimate partner violence screening.

Methods  This qualitative project employed focus groups of mothers who brought their children to a children's hospital emergency department for care, and physicians and nurses who staffed the same department. We held 6 ethnically homogeneous mother focus groups (2 white, 2 African American, and 2 Latina) and 4 provider focus groups (2 predominately female nurse focus groups and 2 physician groups: 1 male and 1 female). Professional moderators conducted the sessions using a semistructured discussion guide. All groups were audiotaped and videotaped, and tapes were reviewed for recurring themes.

Results  A total of 59 mothers, 21 nurses, and 17 physicians participated. Mothers identified intimate partner violence as a common problem in their communities, and most remarked that routine screening for adult intimate partner violence is an appropriate activity for a pediatric emergency department. However, many expressed concern that willingness to disclose might be affected by a fear of being reported to child protective services. They stressed the importance of addressing the child's health problem first, that screening be done in an empathetic way, and that immediate assistance be available if needed. Themes identified in the provider groups included concerns about time constraints, fear of offending, and concerns that unless immediate intervention was available, the victim could be placed in jeopardy. Many said they would feel obligated to notify child protective services on disclosure of intimate partner violence.

Conclusions  Intimate partner violence screening protocols in the pediatric emergency department should take into consideration the beliefs and attitudes of both those doing the screening and those being screened. Those developing screening protocols for a pediatric emergency department should consider the following: (1) that those assigned to screen must demonstrate empathy, warmth, and a helping attitude; (2) the importance of addressing the child's medical needs first, and a screening process that is minimally disruptive to the emergency department; (3) a defined, organized approach to assessing danger to the child, and how and when it is appropriate to notify child protective services when a caregiver screens positive for intimate partner violence; and (4) that resources must be available immediately to a victim who requests them.

INTIMATE PARTNER violence (IPV), commonly called domestic violence, is a critical but underaddressed health care issue in the United States. An estimated 2 to 4 million women are physically abused each year, and IPV occurs in as many as 1 of every 4 households.¹ Intimate partner violence occurs twice as often in households with children.² There is a direct connection between IPV and child health and development. The children of abused women are 6 to 15 times more likely to be abused, with concordance rates between child and maternal abuse thought to be as high as 77%.^3-5 An estimated 3 to 10 million children are exposed to violence occurring between their parents each year, causing significant social, emotional, behavioral, and cognitive problems.^4-6

Increased recognition of the IPV problem has prompted professional organizations^7-9 and regulatory agencies¹⁰ to recommend or mandate routine IPV screening in health care settings. The American Academy of Pediatrics (Elk Grove Village, Ill) has stressed the importance of the problem, stating that "the abuse of women is a pediatric issue," and encouraging the screening of mothers in pediatric settings.⁷ In particular, the pediatric emergency department (PED) has been identified as an important place to screen for IPV.³

Despite the importance of the problem, and policy statements encouraging screening, it is estimated that few screening protocols for PEDs exist, with one study finding only 4.2% of programs with specific IPV protocols in place.³ Barriers to screening in adult emergency departments and clinics have been identified, and they include health care providers' attitudes concerning time constraints, fear of offending, lack of education, and lack of resources.^11,12 A single survey examined additional possible barriers in a PED and found that lack of training, lack of experience, and a feeling that responding to IPV was not in the purview of pediatrics.³

Successful planning, implementation, and evaluation of an IPV screening and education program should take into account the attitudes and beliefs of those doing the screening and those being screened. Focus groups are particularly useful as a first step in developing such an understanding. Additionally, they serve to provide information for generating hypotheses for later quantitative studies such as surveys or observational studies.

The goal of this study was to identify the barriers and opportunities for IPV screening in a PED. Specifically, we sought to explore the attitudes, beliefs, and feelings toward routine IPV screening of those potentially being screened, mothers who bring their children to the PED for care, as well as physicians and nurses who staff the PED.

Subjects for the health care provider (HCP) focus groups included full or part-time physician and nursing staff members of the Children's Mercy Hospital (Kansas City, Mo) emergency department, a high-volume (57 000 visits per year) department in an urban Midwestern children's hospital. Nursing students, medical students, and resident physicians were not included. Two focus groups of nurses and 2 physician groups were conducted. Emergency nurse practitioners were included in the physician group because of similarities in their duties. All participants spoke English, and participation was voluntary. Each group consisted of 7 to 11 participants. Recruitment of HCP groups was done by e-mail invitation, wall posters, and word of mouth. Additionally, the primary investigator (M.D.D.) or study coordinator (J.S.-R.) made announcements about the focus groups in staff meetings. Each participant was given $50 compensation for his or her time, and participation in the focus groups was done during nonwork hours.

Subjects for the mothers group included any woman 18 to 65 years of age who was the current primary care giver of at least 1 child and had at some time brought a child to the Children's Mercy Hospital emergency department for care. For the purpose of this study, "mothers" included foster or adoptive mothers as well as female family members raising children. Participants spoke either English or Spanish, and participation was voluntary. Six mother groups where held (2 African American, 2 white, and 2 Latina); each group was composed of 9 to 12 participants. Recruitment was accomplished via wall posters in the emergency department. Along with contact information and place and time details, the posters stated that the goal of the focus groups was to help the emergency department staff best plan for a family violence prevention program. Each participant was given $50 compensation for her time. Written informed consent was obtained, and the University of Missouri institutional review board approved the study.

Focus groups were conducted in May 2001 by a research team from Lisboa Inc, a professional qualitative research group based in Washington, DC. Groups were conducted according to a semistructured format with the moderator using a discussion guide composed of open-ended questions. Each focus group lasted approximately 90 minutes and was videotaped and audiotaped. All groups except one were conducted in English by the chief moderator. One of the mothers groups, composed of women whose primary language was Spanish, was conducted in Spanish by a Spanish-speaking moderator.

All focus groups were held in a nonclinical meeting room on the hospital property and followed a similar format. The moderator introduced himself or herself to the participants, described the purpose of the group, and alerted those in attendance to the presence of the videocamera. Through the use of the discussion guide, participants were directed to discuss specific topic areas related to the subject of IPV. For mothers, these topic areas included (1) knowledge, beliefs, and attitudes toward IPV in general; (2) reactions to IPV screening in the PED; (3) concerns about IPV screening in the PED; and (4) ideas on methods and logistics of screening. For providers, a similar discussion guide was followed and explored, including (1) knowledge, beliefs, and attitudes toward IPV in general; (2) clinical experiences with IPV; (3) reactions to IPV screening in the PED, including concerns and logistics; and (4) current comfort level with knowledge about IPV and attitudes toward learning more about IPV.

The focus group moderator notes, as well as the audiotapes and videotapes, were the major sources of information for the study. The tapes were reviewed first by the chief moderator, who developed structured, topline summaries of the discussions for the following purposes: (1) to understand participants' attitudes and feelings about IPV and screening in a PED; (2) to identify common themes among participants and groups; and (3) to identify areas of disagreement among the study participants. The primary investigator (M.D.D.) and the study coordinator (J.S.-R.) reviewed the summaries to uncover possible areas of interpretative disagreement. No major areas of disagreement between the chief moderator and the investigators were discovered. In this article, primary recurring themes with direct illustrative participant quotes are presented.

A total of 59 women participated in the focus groups: 20 African American, 19 white, and 20 Latina. Results of the focus group discussion are grouped into 4 topical areas:

1. Recognition of the IPV problem and its effect on children. Women identified IPV as a significant problem in their community, and many spontaneously communicated personal experiences with IPV. Participants identified that exposure to IPV has numerous effects on children, and many added that they consider exposure to IPV as a form of child abuse or neglect. The identified effects on children included withdrawn behavior, difficulty expressing feelings, anger, depression, low self-esteem, and learning to continue the cycle of violence. "A child grows up in this and doesn't know what to do if the tension's not there." "They'll repeat what they've seen. They don't know how to handle it any other way."

2. Support for IPV screening in the PED. Mothers remarked that they thought the PED was a good place to ask about IPV. "Battered women want to feel cared for. So if they know that they are getting excellent care for their kids . . . why not come here and feel safe for your care for yourself." "You are bringing your child in and at the same time you have the opportunity to get some help."

3. Concerns identified. Many women felt that questions about IPV in the PED might be interpreted as a search for child abuse, and they feared that disclosure might lead to a report to child protective services (CPS) (Table 1). "Everything is not abuse, but sometimes the hospital doesn't look at it that way." "When I was in an abusive relationship it would have been a problem because I was always trying to hide it. If someone would have asked me I would have said `yes, I'm doing fine,' because I would have been afraid that my kids would have been taken away from me. You believe you're the only person in world who can protect your kids."

Women stated that understanding why the hospital staff was asking about IPV was very important, and that the reason for asking be communicated first. "Tell the truth. Domestic violence is on the rise and the hospital has taken an interest in this and we want to do whatever is possible to help victims of this." "Let us know you want to help the situation." "When the conversation begins, it should start with mention of a service; offering a chance to talk about and look over the program."

4. Considerations on screening methods/logistics. Participants mentioned that those doing the screening should show warmth, understanding, and empathy during the questioning. "I don't mind being asked sensitive questions if they show concern when I first come in." "Use a nonaccusatory tone of voice." "Be supportive, caring, and genuine." "Focus on helping."

Mothers discussed the importance of addressing the child's initial medical problems prior to asking about IPV. They preferred that the screening questions be asked after their child had received treatment. "You'd have to wait until the mother's focus is no longer on the child." "Ask after the child has been examined. He's going to be okay and you're at your comfort level."

Opinions on whether the screening should be done via a written questionnaire or face-to-face were mixed. "If you're going to ask questions it should probably be on paper." "I want to talk to you, not a clipboard." "I don't know who would later read the questionnaire."

Lastly, mothers emphasized the importance of having services, including shelter placement, immediately available to victims. One participant, a victim of IPV, related her experience, underscoring this theme. "The day I opened my mouth, I had to have a place to go. There's a good chance you'll die if you talk. You are more scared than ever because you don't know what's going to happen from then on."

Ethnic subgroup similarities far outweighed differences. The major themes discussed above were common to all focus groups. One subgroup difference is worth noting. African American women, in contrast to other subgroups, emphasized that much of the problem with IPV, as well as other violence, stems from a decline in spirituality in the family and community.

A total of 21 registered nurses (19 female and 2 male) and 17 physicians (10 female and 7 male) took part in the focus groups. This represented 42% of the department's total registered nursing staff (38% of the department's female nurses and 100% of the department's male nurses) and 58% of the total physician staff (53% of the department's female physicians and 70% of the department's male physicians). Results of the focus group discussions are grouped into 5 topical areas:

1. Recognition of the problem and its effect on children. Nurses and female physicians felt that IPV is prevalent in the community, noting that behaviors they witnessed in the PED seemed suggestive of IPV at home. Male physicians were less certain about the incidence and were less aware of in-hospital episodes suggestive of IPV. A female nurse stated, "I've seen situations where I've agonized because I know something is terribly amiss," and "You see it in the waiting room, so what do they do at home?" Providers cited numerous behavioral and emotional correlates of childhood exposure to IPV. Another quote from a female nurse: "Either they are very submissive because they're afraid or very aggressive; there's no middle ground." A female physician said, "It affects their perception of what's real and what should be. Younger kids think that's the way it is."

2. IPV as a form of child abuse or neglect. Most female physicians and nurses viewed exposure to IPV as a form of child abuse, with some indicating that they would feel obligated to report a child's exposure to IPV to CPS. Some of the male physicians were reluctant to view IPV exposure as a form of child abuse, suggesting that it be characterized as neglect. A female physician stated: "I would report it because there's probably more going on than you'll pick up in the ED [emergency department]," and "if I feel there is potential for harm [to the child] I have to let someone know." A male physican stated: "I don't think witnessing IPV falls under the classic definition of child abuse that we are required to report." A female nurse stated: "I think you should report it, but I don't think it would do any good."

3. Attitudes toward IPV screening and concerns identified. Time constraints were identified as a major obstacle for all providers. Nurses were sensitive to the problem of IPV, but said they already had too much to do and too little time to do it. Physicians more openly questioned the use of emergency department time and resources for IPV screening. A female nurse stated: "I would need more time to deal with it. We have huge numbers of people in the emergency department at any given time." A female physician stated: "There are bigger emergency issues that need to be taken care of." A male physician stated: "How many man-hours are wasted in the department every year? Will the cost of doing the screening outweigh the benefit?" A female physician stated: "You could really open up a Pandora's box."

Fear of offending caregivers was also a predominant theme among both nurse and physician groups. Concerns were expressed that caregivers might be reluctant to report true information because of fear that they would be reported to CPS. A female physician stated: "These are personally sensitive issues that people may not want you to ask them while they are here." A male physician stated: "If you're not careful you're going to offend the hell out of people." A male physician stated: "I could see a lot of people complain and say, ‘What does this have to do with my child's illness; why are you asking me this?'" A female physician stated: "They're afraid you might report it and take their kids away." A female nurse stated: "The clientele we have has had bad experiences with social workers and the police. The system is punitive. We have to approach [screening] with, ‘We're not here to give you one more bad experience with the Department of Family Services.'" These concerns are presented in Table 1.

4. Considerations on screening methods and logistics. Provider groups expressed concern that needed resources for intervention in IPV, such as shelter placement, might not always be available, and that they might be doing more harm than good if they uncovered the issue but did not offer effective services for the caregiver. A female physician stated: "It's irresponsible for us to initiate screening if we don't have the staff and resources. Can we appropriately direct them and meet their needs?" A male nurse stated: "If you ask me to do this, you've got to have something to give me. When a woman is trying to leave her relationship, that's when she is most vulnerable. We have to make sure that whatever we give her is open to her needs." A female nurse stated: "Once you know there's a problem, you want to have an intervention. We need to have resources."

Timing for screening was important; most thought it best that the child's medical condition be attended to first. Screening during routine medical intake, such as during the triage process, was not seen as appropriate. A female nurse stated: "If you have a kid who is having trouble breathing and you ask a social question during triage, I don't think that's good." Many agreed that there should be a dedicated professional or paraprofessional assigned to do the screening. Other providers mentioned that it might be more cost-effective to educate parents and guardians via public information materials such as posters and videos.

5. Comfort level with IPV and openness to education. Most providers said they are not comfortable with their current IPV knowledge level, but that they would be open to learning more. Participants mentioned that credible teachers on the subject would be victims of IPV, shelter staff, and law enforcement officers. A female nurse stated: "We need more education. Women from battered shelters or shelter directors could enlighten us on what we don't know."

We found more similarities in opinions expressed between the providers and the mothers than was expected. Mothers viewed IPV as a problem prevalent in the community, and were uniformly supportive of the idea of IPV screening in a PED; however, they mentioned several concerns. A dominant concern was that screening might be perceived as a search for child abuse, especially if it wasn't clear why the question was being asked. Additionally, screening would be seen as inappropriate if the child's medical needs were not addressed first and if there was not assistance (eg, shelter placement, orders of protection) readily available for the IPV victim. Physicians and nurses expressed opinions that were consistent with the mothers' views. They felt that those being screened could be offended if the question asked was interpreted as a screen for child abuse. Similar to the mothers, they felt that the child's medical needs should come first, and that readily available resources for victims must be in place. Additionally, providers identified time constraints and lack of education about IPV as barriers to screening.

This is one of the only studies examining attitudes of mothers and providers on the idea of screening for IPV in a PED or urgent care setting, so it is difficult to make comparisons with other studies. However, our findings are consistent with similar work on IPV screening in other clinical settings. Time constraints and fear of offending have been previously identified as physician-related barriers to screening.¹³ The importance of HCP empathy has been identified in previous focus group studies with female victims of IPV.¹⁴

This study identifies new findings of concern from both mothers and HCPs regarding the relationship between IPV and child abuse, as well as the legal and moral mandate to report a positive screen to CPS. Disclosure of a child witnessing IPV as the sole reason to report to CPS is controversial and the source of much medical and legal debate. Clearly, there is strong evidence that IPV committed in the presence of a child may be harmful to that child, and there is for many, a moral obligation to act on such information. Legally, there may also be an obligation for a mandated reporter to notify CPS. Statutory definitions of child abuse and neglect vary widely from state to state, and several now include the witnessing of acts of IPV in their definitions of child endangerment or abuse.¹⁵ (The current study took place in an institution that treats children from Kansas and Missouri—states that do not specifically include witnessing of IPV in their definitions of child maltreatment.) However, defining child exposure to IPV as neglect or maltreatment may imply that the victims are neglectful parents, or that they could have stopped the abuse. Child protection systems do not always have the capacity to intervene appropriately to ensure the safety of both the child and the mother in situations involving IPV. Lastly, the practice of routinely reporting IPV incidents occurring in the presence of children may discourage victims from seeking help.

As mandated reporters, pediatric HCPs should be familiar with and follow their local child abuse reporting laws. Most importantly, regardless of legal obligation, a positive IPV screen should prompt a thorough assessment of risk to the child. This assessment should include inquires about injury or abuse, or threats to the child; the current safety of the home; and whether the mother/victim can adequately protect the child. Answers to these questions will help the HCP make a decision about imminent risk of harm to the children. If the risk is not determined to be currently dangerous, the provider can refer the victim and her children to battered women's services, counseling, or child-focused services. In the event of a legal mandate or imminent risk, a report should be made to CPS, and the provider should inform the nonoffending parent of the obligation to report.

The mixed reaction of mother participants to face-to-face vs written screening brings up another area of controversy—that of documentation of IPV in the pediatric setting. Opposition to a notation of a positive IPV screen in the child's medical record may relate to the batterer potentially having access to the record. On the other hand, such information may be important for other providers who work with the family in the future. Evidence is currently lacking on the consequences or benefits of such written documentation in the child's medical record. Alternatives include creating a separate record for the mother/victim, maintaining a section of the child's record that is not released with the rest of the medical record, and using nonspecific terms to document IPV.

Because this is a qualitative study, it cannot be generalized to other settings. Also limiting generalizability, is the fact that participants all worked in or brought their children to a large inner-city PED. It is likely that many of the participants in both the mother and the provider groups had some form of direct experience with the public welfare system or CPS, which may have significantly shaped their attitudes. Additionally, since the poster announcing the study mentioned family violence, participants volunteering for the study might have had experience with, or specific opinions about, family violence; so a potential for selection bias exists. To maintain privacy, we did not specifically ask participants about their personal experiences. Because of the small number of male nurses, we were not able to hold a male-only nurse group, which did not allow us to compare gender differences across groups. Another study limitation is that the focus groups did not explore whether or not it is appropriate to screen in front of children, and if so, at what age? Previous authors have identified this as a major dilemma for which there are numerous opinions.¹⁶

Despite the limitations of this study, those developing an IPV screening protocol in a PED, particularly an urban one, should consider the concerns and themes elicited here. It seems that in this urban PED, there are 4 main conditions to enhancing the acceptability of a screening program. (1) Screeners should demonstrate empathy, warmth, and a helping attitude. Critical to the approach is framing the screening question in a manner that indicates why it is being asked. (2) Screening methods should recognize the priority of the child's medical care and be minimally disruptive to the emergency department. (3) There should be a defined, organized approach to assessing danger to the child, and how and when it is appropriate to notify CPS when a caregiver screens positive. (4) Resources must be immediately available to a victim who requests them.

Future studies that quantify some of the topics raised in the focus groups would be helpful by testing hypotheses and determining whether the concerns are generalizable. Research in other demographic settings would also be useful to understand whether attitudes are typical.

Accepted for publication October 8, 2001.

This study was funded in part by Public Health Service grant 1 H34 MC 0010601 from the Maternal and Child Health Bureau (Health Resources and Services Administration), US Department of Health and Human Services, Washington, DC.

Understanding the concerns and fears of both those being screened and those doing the screening will enhance the acceptability and effectiveness of an IPV screening protocol in a PED. To our knowledge, this study is among the first to qualitatively explore attitudes, beliefs, and feelings of both mothers and HCPs on the topic of IPV screening in a PED. Those developing screening protocols in pediatric emergency departments should consider the following: (1) that the screener must demonstrate empathy, warmth, and a helping attitude; (2) the importance of addressing the child's medical needs first and a screening process that is minimally disruptive to the emergency department; (3) that the disclosure of IPV should not routinely be viewed as an automatic report to CPS; a thorough assessment of danger to the child should be performed; and (4) that resources must be available immediately to a victim who requests them.

Corresponding author: M. Denise Dowd, MD, MPH, Division of Emergency Medicine, Children's Mercy Hospital, 2401 Gillham Rd, Kansas City, MO 64108 (e-mail: ddowd@cmh.edu).