Qualitative analysis of domestic violence detection and response in a tertiary hospital

KEYwords: Domestic and family violence; domestic violence; family violence; health-care services; research-to-practice.

Domestic and family violence in this article refers to any act of violence or control that is used to dominate another member of the household, such as an intimate partner, parent, child or other family member. Domestic violence contributes to more death, disability and illness in women aged 15–44 years than any other preventable risk factor.¹ The adverse outcomes of domestic violence range from the acute consequences (fatality, injury, homelessness) to long-term health consequences, chronic complaints, risk-taking behaviours and mental health issues.² The effects on children who witness domestic violence can also be severe and long-standing, such as increased risk of anxiety, depression, mood problems and other health problems, low self-esteem, aggression, peer conflict, lower social competence and antisocial behaviour.^2,3

Women who are living with domestic violence are likely to be high users of health-care services and have an estimated 30–50% higher rate of emergency department (ED) use than other women.^4,5 Despite this, health-care responses are often inadequate, with family and intimate partner violence not regularly screened for and, consequently, often undetected. Up to 88% of victims of murder or attempted murder may present to EDs in the previous 12 months.⁶ These figures highlight the missed opportunities for health-care providers to interpose in such cases, provide information, referrals and safety plans, or improved alert systems and documentation for follow-up care. Much of the domestic violence research is focused on women and children because they make up the majority of victims. However, the same principles apply to male victims.

Routine universal screening for domestic violence is now recommended by many professional bodies and organisations, but the execution of screening, and the particular parts of the health system in which it is appropriate, is widely contested.^4,7 Although around 80% of victims say they would disclose about a history of abuse if asked but are unlikely to disclose without being asked, only 12–20% report being asked.⁸ EDs in large regional hospitals are most commonly associated with presentations related to domestic violence, and are therefore a clinical area commonly targeted for screening. However, in light of evidence suggesting that pre- and postnatal women may be at high risk,⁹ there is an argument that the net of screening and detection should be wider and include maternity services.⁷ In addition, the evidence for an increase in, and close association of, drug and alcohol use and mental health issues in both victims and perpetrators of domestic violence suggests that screening should also be undertaken for people presenting to mental health clinicians and to drug and alcohol departments for treatment.¹⁰

In this study, a large tertiary hospital in south-east Queensland aimed to identify current responses and processes used across the hospital for managing the care of people who are experiencing domestic violence, explore the challenges in service provision for such people and identify opportunities for care improvement. The study was developed in response to the Not Now, Not Ever – Putting an End to Domestic and Family Violence in Queensland, 2014 Taskforce Report.⁶ Following the release of that report, the Hospital Board and Executive committed to implementing its recommendations. The present study was undertaken in May 2016 and provided a baseline assessment of current health service responses to people experiencing domestic violence and entering the hospital. It also identified some gaps in the delivery of services and highlighted potential areas for further development.

The project used semistructured, face-to-face or telephone interviews with key informants. Thematic analysis of interviews was used to define the key issues and areas of interest that were identified by participants.

Sixteen areas across the hospital were considered possible entry points for people who may have experienced domestic violence. These included EDs, Maternity Services, pre-admissions and out-patient departments, Mental Health Acute Care, Mental Health Inpatient Unit, Mental Health Continuing Care, Mental Health Homeless Health Outreach Team and Alcohol and Drugs Service, and Community Health – Adult and Child. Other relevant specialist services included the Homeless ED Liaison team, Sexual Health Service, Indigenous Health Liaison Service and the Child Safety Unit.

Managers from the 16 key services were invited to take part in the study because they would be able to recognise the strengths and challenges for their staff and teams when responding to and working with people experiencing domestic violence.

Ethics approval for the study was granted by the Gold Coast University hospital ethics committee (Reference HREC/16/QGC/327).

The semistructured interviews included the following topics: current domestic violence identification; screening; privacy and confidentiality; staff training; and referral pathways. Interview questions were developed by the researchers based on existing literature and after repeated consultations between members of the research team with expertise in the areas of qualitative research and domestic violence.

Two researchers (KG and KB) analysed the data using a process of inductive thematic analysis. Each interview transcript was read several times to develop familiarity with the data. Initial codes were generated within the transcripts and recorded as individual notes by the researchers. Coding across the transcripts was continued until all the data extracts were coded.¹¹ Potential patterns in these coded data were examined by exploring any similarity or overlaps in the codes and the relevance to the entire dataset. Provisional themes were developed and refined as the data analysis continued over time. As themes and subthemes were formed from the data, a thematic map was developed to explore and refine the connections between the developing candidate themes.¹¹

All 16 managers invited to contribute to the study agreed to participate in an interview. Analysis of the data identified several strengths in hospital responses to domestic violence. Barriers to identification of domestic violence incidents were also identified in the processes and system responses of the hospital. These are discussed below.

The following main themes emerged from the data: limited opportunity for staff training, appointment of a specialist domestic violence coordinator, standardised referral pathways, improved communication processes, safety concerns for both patients and staff and multicultural support for Indigenous and culturally and linguistically diverse (CALD) victims of domestic violence.

Most clinicians who participated in the study disclosed that although they worked in key areas of the hospital for patient presentations related to domestic violence, they had not had any relevant training. They felt unable to offer support or provide informed and evidence-based information to patients experiencing domestic violence who accessed the hospital for health services. In some of these parts of the hospital, due to a lack of training and education, some staff members failed to respond effectively when domestic violence was disclosed. Some frontline team members who may have received a limited amount of training openly acknowledged that they may not be providing evidence-based and practice and assistance to victims due to a lack of confidence and training. At the time of the study, only three clinical areas across the health service had access to regular domestic violence training: Maternity Services, the ED and Child Protection Services. With the exception of the Maternity Services within Women, Newborn and Children’s Services, domestic violence training tended to be ad hoc, primarily due to time pressures and availability of clinical staff to attend training and education.

All respondents expressed the necessity for an established domestic violence specialist coordinator for consultation, liaison and guidance, stating that they would use this service if it were available. They also believed it was a necessary resource to provide safe and effective support to women, families and staff in complex situations. Four of the 16 participants identified Social Work (ED and Community Health) as the designated ‘go to’ domestic violence specialists. Maternity Services also liaise closely with the maternity social work team and have a designated domestic violence midwifery champion that midwives can approach for support and advice. The Homeless Health Outreach Team currently have two designated domestic violence specialists. The Sexual Health Service uses the sexual assault community service as their specialist resource. However, this still left a resource gap with at least nine clinical areas across the hospital and health services with no access to direct support when a person presents disclosing a history of domestic violence or requesting help and support. There was an imbalance in resources of the different departments. Although this may be somewhat warranted, practitioners from all departments encountered patients experiencing domestic violence, yet many feel poorly equipped to provide adequate care and support.

Respondents expressed the need for workplace instructions, clinical guidelines and referral pathways. The only referral pathways identified were the ED Social Work domestic violence pathway and Women’s and Newborn Services domestic violence clinical guideline. A lack of access to workplace guidelines can lead to an increased risk of unsafe practices. At the time of the study there was no standardised clinical handover or referral system to identify domestic violence across the service. In addition, aside from the Maternity Services, currently there are no standardised recording mechanisms for the statistics of domestic and family violence across the rest of the services.

Because there are several documentation systems currently in use in different clinical areas across the hospital service, this presents a barrier to effective communication. Currently, there is no standardised patient documentation recording system across the hospital, which can lead to fragmented care, which, in turn, can result in increased risk for domestic violence victims. Robust and safe documentation is important because the health service may be the only service that people may access where they feel able and safe to disclose domestic violence.

Sharing patient information across services is challenging because there is currently no way to automatically ensure that patient details recorded on one system transfer safely to another. This allows for human error and omission, placing victims of domestic violence at further risk, especially if a perpetrator calls the hospital looking for them. Despite attempts to make patient details anonymous on electronic systems, staff perpetrators who have access to electronic medical records or other databases may access patient records and see confidential information. There are also no secure in-patient areas that can provide safe spaces for people experiencing domestic violence, away from perpetrators. Some areas do not have access to security staff or emergency response alarms, placing both staff and patients at risk.

Most areas across the hospital do not display or provide domestic violence brochures or information, and many staff are unaware of how to access domestic violence resources, or know which resources are available. It is important for all areas to openly display information, because this shows that the services recognise that domestic violence exists and that they do not tolerate such behaviour. Every service within the hospital is responsible for becoming domestic violence aware and all staff should have, as a minimum, education to identify and refer people experiencing domestic violence to community services. Lack of available of resources minimises the opportunity for education and awareness raising for people who visit the hospital and are living in or surviving a violent relationship.

People from vulnerable populations who experience domestic violence frequently use hospital services. Indigenous women are 35-fold more likely to be hospitalised due to domestic violence than other Australian women,¹ yet currently there are no Indigenous interpreters employed in the study’s hospital service. The current practice is for Community Elders or friends to translate for Indigenous women.

We found from this study that the hospital already has some established foundations and processes to build upon, including a commitment from the hospital board and executive to prioritise domestic violence as a health response. Another strength of the service response was that the ED of the hospital already has a trained and experienced senior social worker who knows how to manage the care of people experiencing domestic violence. The hospital’s Maternity Services also have a policy of routine antenatal inquiry about domestic violence for all pregnant women, and the Homeless Health Outreach Team has domestic violence-trained staff based within their service. The hospital is also part of an established network of government and non-government agencies that provide a coordinated response to domestic violence, with several members of the hospital staff holding key positions on several local and state domestic violence committees.

There were also many identified barriers to an effective response to domestic violence, most notably the lack of sufficient training and education for staff, as well as limited available resources. Previous studies indicate that a domestic violence specialist can provide valuable ongoing training and support to staff. A specialist adviser can also give staff the confidence to discuss domestic violence with patients, where no discussion may have previously occurred.¹²

A prior longitudinal study found that the introduction of standardised domestic violence instruments improves data collection and documentation by staff, obtaining valuable data that may be otherwise unrecorded due to staff’s lack of knowledge or confidence.¹³ Implementation of standard documentation was more effective than training when it came to history taking and identifying risk of harm. When domestic violence is not identified or notified to other relevant staff across the health service, it can result in lack of continuity of care and potentially increase the risk for affected patients. Lapses in continuity of care mean patients may have to retell their stories many times to different people, causing unnecessary distress. Similarly, failure to collect relevant and accurate statistics may lead to underestimation of both the magnitude and effect of dealing with the consequences of domestic violence. This will lead to under-resourcing and an inability to provide adequate services to domestic violence victims.

Although a consistent protocol for screening and responding to domestic violence has been endorsed by several studies, it is also recognised that not all departments or patients can be approached or cared for in a uniform manner.¹⁴ It is evident that more clearly defined guidelines need to be created and made more readily available to staff.

For Aboriginal and Torres Strait Islander women, the incidence of violence is disproportionately high compared with the same types of violence within the Australian community as a whole.¹⁵ Intergenerational trauma, dispossession and colonisation play an important role in the experience and perpetration of violence, expressed in continued distrust of non-Aboriginal people, fear of retaliation or alienation from kinship community, lack of accurate information and awareness of services, lack of local service with capacity to assist and lack of culturally competent service providers.¹⁶ Women from CALD populations, and whose first language is not English, are also at an increased risk of not receiving appropriate or culturally acceptable support for domestic violence.

This study is the first in a series of research projects designed to identify the current hospital position of domestic violence identification and response, barriers and enablers, and necessary amendments and updates to existing policy and practice. The present study was of limited scope, interviewing only staff members, primarily of managerial ranking, within key hot spots of the hospital where it was considered that victims and survivors of domestic violence may likely be seen. Further research will be conducted with frontline staff, external organisations working closely with staff to provide services for people affected by domestic violence and women who have been directly affected by domestic violence and able to discuss their experiences and requirements of health-care services.

Results from the study demonstrate that, apart from Maternity Services, who routinely carry out routine antenatal enquiry for all pregnant women, no one discipline or profession is more likely to identify domestic violence. All clinicians in the key areas require training and support in identifying and responding to domestic violence. Although some services would explore domestic violence further once identified, at the time of the study no service was using a risk assessment tool.

The overall rates of family violence in Australia continue to be high and, as a result, many victims of domestic violence require hospitalisation for various forms for treatment. This can range from treatment for physical injuries to engaging with mental health services or drug and alcohol services. Health services have an important role to play in identifying and responding to domestic violence, but this requires health services to develop processes to support staff in responding effectively and safely, and to include workforce development and capacity-building policy development and implementation, and the development of evidence-based guidelines and policies.

The authors declare no competing interests.