Screening Mothers for Intimate Partner Abuse at Well-Baby Care Visits: Title and subTitle BreakThe Right Thing to Do

In this issue of THE JOURNAL, Martin et al¹ report on the prevalence of physical abuse before, during, and after pregnancy as determined from a random sample of North Carolina women (1997-1998). Women were surveyed by mail and telephone approximately 3.6 months after they delivered live infants. Reported physical abuse before and during pregnancy exceeded 6% and was 3.2% in the postpartum period. Abuse in an earlier period was strongly associated with further abuse in subsequent periods. Mothers reported a mean of 3 well-baby care visits during the first 3.6 postnatal months.

The findings from this study suggest that intimate partner abuse (also termed intimate partner violence or domestic violence) is occurring in the immediate postnatal period, and an opportunity exists for pediatric health care practitioners to identify this abuse. The American Academy of Pediatrics (AAP) issued a policy statement in June 1998 stating, "Pediatricians are in a position to recognize abused women in pediatric settings. Intervening on behalf of battered women is an active form of child abuse prevention. Knowledge of local resources and state laws for reporting abuse are emphasized."² This recommendation could apply to all who provide postnatal child care.

Martin et al suggest that abused mothers may have as many as 3 opportunities to be identified at well-baby care visits in the first 3½ months of their new infants' lives. The AAP recommends 3 visits in the first 3 months and 6 visits in the first 12 months of infancy. Utilization data from Group Health Cooperative (Seattle, Wash) indicate that the mean number of encounters for well-baby care in the first 3½ months is 2.5 and that in the first 12 months of life, there are 4.2 such visits (Virginia Immanuel, MPH, written communication, February 21, 2001). Using data provided by Martin et al,¹ the number of women who would have to be screened to detect 1 abused mother would be 14.5 before pregnancy, 16.3 during pregnancy, and 31 during the first 3½ months postpartum.

It is possible to implement this screening opportunity by building some questions into the standard forms used as a routine part of well-baby care. For instance, Group Health Cooperative includes a question about "family history of physical abuse or sexual abuse" on routine questionnaires for neonates or new child family members. Clearly, more in-depth questions are needed, but this example is a start, and previous work in family violence demonstrated that written questionnaires increased abuse screening by 14.3%.³

The second part of the AAP recommendation states that "intervening on behalf of battered women is an active form of child abuse prevention." The literature contains solid links between domestic violence and child abuse, and suggests that the link is approximately 50% in either direction.^{4 - 6} Information on the long-term adverse effects of child abuse is also well documented.^{7 - 9} Despite these links, the US Preventive Services Task Force concluded in 1996 that there was "insufficient evidence to recommend for or against the use of specific screening instruments"¹⁰ for detecting intimate partner abuse. Insufficiency issues boil down to instruments for detection of intimate partner abuse, modes for administering the instruments, and evidence for effectiveness of interventions using the best instruments.

Since the 1996 report, some progress has been made. Feldhaus et al¹¹ validated the Partner Violence Screen (PVS) scale for past physical violence and perceived personal safety compared with 2 other screening questionnaires (the Index of Spouse Abuse [ISA] and the Conflict Tactics Scale [CTS], administered orally). This research should be expanded to address not only physical abuse but also psychological and sexual abuse. Additionally, none of these scales (PVS, ISA, or CTS) directly assesses whether a person is currently seeking medical attention for an injury or illness related to intimate partner abuse. Coker et al¹² reported development of the Women's Experience With Battering scale, which encompasses physical, psychological, and sexual abuse domains. However, the issue of the most valid, reliable, and feasible screening questions remains unresolved.

The optimal mode of administration of screening to detect partner abuse also remains uncertain. Some suggest that direct oral assessment is superior to written questionnaire assessment for case finding.^{13 - 14} Conversely, others suggest that written questionnaires are better than face-to-face questioning.^{3 ,15} In a recent randomized controlled trial, Rhodes et al¹⁶ reported that touch-screen computer screening in an emergency department setting was acceptable to patients and resulted in significantly increased detection of intimate partner abuse. An emerging literature suggests that when data about sexual behaviors or other sensitive topics are to be collected, computer-based interviewing has the potential to reduce biases, improve measurement validity, and generally provide improved responses compared with written or face-to-face questionnaires.^{17 - 19}

Most important, assessment is rudimentary of the efficacy and effectiveness of interventions to identify and manage intimate partner abuse in the short and long term and the health outcomes of interventions. A systematic literature search for interventions in hospital outpatient, emergency department, community health center, health maintenance organization, and public health prenatal clinic settings revealed only 8 studies.^{3 ,20 - 26} While several studies demonstrated favorable changes in knowledge, attitudes, and beliefs, the positive effects on processes of care such as asking about intimate partner abuse and quality of record documentation are less impressive. Moreover, definitive case-finding methods that are effective over the long term have not yet been described. In most studies, the follow-up intervals for case-finding rates are short (≤6 months) and little information exists on the long-term outcomes for the cases found. What are the effects on health status? What are the effects on social isolation? What happens to health care utilization? Presently, much about the long-term history of intimate partner abuse is unknown.

Research is needed to develop valid, reliable, and feasible screening questionnaires, preferred modes of administration, and guidelines or programs for intervention over time to gauge long-term effects on processes of care, health outcomes from the patient perspective (eg, health status, social isolation, depression), and patterns and costs of health care utilization. This type of information is needed to develop and support sound training programs for education of health care professionals. To help fill this information void, use of systematic planning models^{27 - 28} to develop interventions is recommended. These models help specify the domains to be addressed by the intervention, such as the predisposition of practitioners to undertake the task, environmental enabling factors, and reinforcing factors.²⁷ Once the domains are clear, they are populated by evidence-based tools (eg, measurement and feedback, use of opinion leaders) addressing the specifics for a given domain, such as those in the Cochrane Reviews on effective clinical practice and in other articles.^{29 - 33} This approach is part of the emerging broader view of continuing medical education^{32 - 33} that is beginning to be espoused³⁴ and applied to the issue of family violence training for health care professionals in identification and management of intimate partner abuse.^{3 ,20}

To generate the information needed on all types of family violence, funding for concerted and ongoing research on intervention strategies and their effectiveness must be provided. In 1998, the Committee on the Assessment of Family Violence Interventions of the Institute of Medicine concluded that "to improve evaluations of family violence interventions and provide a research base that can inform policy and practice, major challenges must be addressed."³⁵ The committee "identified a set of interventions that are the focus of current policy attention and service innovation efforts, but have not received significant attention from research" and require evaluation.³⁵ The US Advisory Board on Child Abuse and Neglect (ABCAN) concluded that "although progress has been made, child maltreatment may still be the most under-researched social problem."³⁶ Several federal and independent reports^{35 - 38} over the last 15 years have noted the lack of an adequate scientific base for prevention of child abuse and intimate partner abuse.

In 1993, a National Academy of Sciences expert panel developed a research agenda for future studies of child maltreatment. This report³⁸ included a series of wide-ranging recommendations, many of which expanded on the ABCAN recommendations. The panel stated in its research priority 14:

When a sufficient research budget is available to support an expanded corps of research investigators from multiple disciplines, multidisciplinary research centers should be established to foster collaboration in research on child maltreatment. The proposed centers should have a regional distribution, be associated with major academic centers, have the capacity to educate professionals of various disciplines, and launch major research efforts. Examples of the cancer and diabetes centers funded by the National Institutes of Health could serve as models, as could the Prevention Intervention Research Centers of the National Institute of Mental Health.

However, in the 8 years since this report and the decade since the ABCAN reports, the 2 groups have ceased their work and their recommendations have not been acted on. An adequately funded major national focus to build the scientific base for family violence is still much needed. Anything short of this will not get the job done.

In conclusion, the study by Martin et al provides another useful piece to the overall puzzle of intimate partner abuse. There is a clear-cut gap in screening women for intimate partner abuse in the immediate postnatal period. An opportunity to fill this gap exists in health care practitioners' offices. At present, pediatricians and others involved in caring for infants should focus on addressing this need because "it is the right thing to do." In the future, with an increased focus on family violence research and the generation of evidence, the metric will become, "It is the right thing to do, and the evidence substantiates it." A long-term federal commitment to research on this issue is necessary and needed to improve health care for the 20% to 25% of US adults who have experienced this problem in their lifetime and the 1% to 3% who are experiencing it currently.^{39 - 40} The time to act is now. If intimate partner abuse were a new cancer affecting one quarter of adults, the money would be found. It should be found now for family violence.