Primary Care

General practitioner management of intimate partner abuse and the whole family: qualitative study

BMJ 2004; 328 doi: (Published 11 March 2004) Cite this as: BMJ 2004;328:618
  1. Angela Taft, research fellow (a.taft{at},
  2. Dorothy H Broom, senior fellow2,
  3. David Legge, associate professor3
  1. 1 Centre for the Study of Mothers' and Children's Health, La Trobe University, 251 Faraday Street, Carlton, Vic 3053, Australia
  2. 2 National Centre for Epidemiology and Population Health, Australian National University, Canberra, ACT 0200, Australia
  3. 3 School of Public Health, Faculty of Health Sciences, La Trobe University, Bundoora, Vic 3086, Australia
  1. Correspondence to: A Taft
  • Accepted 22 December 2003


Objective To explore management by general practitioners of victimised female patients, male partners who abuse, and children in the family.

Design Triangulated qualitative study comparing doctors' reported management with current recommendations in the literature.

Participants 28 general practitioners attending continuing medical education about management of domestic violence.

Results Doctors perceived partner abuse in diverse ways. Their gender, perceptions, and attitudes could all affect identification and management of this difficult problem. A few doctors practised in recommended ways, but many showed stress and aversion, difficulties in resolving the tensions involved in managing all family members, and neglect of the risks to children. Some doctors used contraindicated practices, such as breaking confidentiality and undertaking or referring for couple counselling. Doctors who were not familiar with community based agencies were reluctant to use them. A lack of expertise and support could have a negative impact on doctors themselves.

Conclusions General practitioners managing partner abuse need to be more familiar with and apply the central principles of confidentiality and safety of women and children. Recommended guidelines for managing the whole family should be developed. Doctors should consider referring one partner elsewhere and avoid couple counselling; always ask about and act on the children's welfare; refer to specialist family violence agencies; and seek training, supervision, and support for the inherent stress. Medical education and administration should ensure comprehensive training and support for doctors undertaking this difficult work.


Domestic violence, now commonly called intimate partner abuse, is generally defined to emphasise intimidation, coercion, and control.1 Twice as many patients of general practitioners (5-8%) report partner abuse within the previous 12 months, compared with women in the community (2.4%).2 General practitioners also see men who abuse3 and the children of such couples, whose health is also at risk.4 Most clinical guidance and training for general practitioners focuses on the management of the adult female victim, although one study highlighted that doctors' management decisions may be more influenced by the “dual relationship” with both partners in a couple than by the severity of the violence.5

Health professionals, including doctors, are increasingly being encouraged to screen for partner abuse,6 which is a serious health problem. Primary care is an important early intervention site,7 because general practitioners often have an ongoing therapeutic relationship with the whole family. The limited research on general practitioners' management of partner abuse focuses on perceptions of barriers to screening victims; intervention studies on partner abuse are rare.8

To be effective, guidelines for best practice in management of partner abuse should be informed by how family doctors manage the complexity of seeing all family members. This study explored how general practitioners managed all members of a family in which partner abuse occurred, the impact on doctors themselves, and what further training and support are needed. Our broad objectives were to explore the strength and limitations of continuing medical education on domestic violence, strategies to improve doctors' responses to women from ethnic minorities, the influence of the doctor's gender, the principles and strategies of doctors working with men who use violence, and the need for an integrated continuing medical education curriculum incorporating all family members.


In 1997 we undertook case studies of two general practice training projects on domestic violence in Victoria (projects 1 and 2) with nested case studies of doctors and patients.9 The continuing medical education projects covered management by general practitioners of both female victims and male perpetrators. However, the two projects varied in location, scope, method, and duration.

The project administrators contracted the lead investigator (AT), a social scientist, to do pre-training and post-training surveys. With their help, AT approached all doctors who registered for training. The study sample was purposive (that is, restricted to those doctors who sought training) and stratified in order to approximate the urban-rural and female-male composition of doctors in the projects. All doctors approached consented to a semistructured interview (consistent thematic questions about training in and management of domestic violence) before and after training. AT interviewed 15 doctors (seven women and eight men) from project 1 and 13 doctors (eight women and five men) from project 2 about their identification and management strategies and the usefulness of the training. Consistent with an Australian trend, twice as many male doctors (12) as female doctors (6) were over 40, and most (14) of the women were part time. All but one of the doctors were vocationally registered.

Seven doctors (three men and four women) agreed to further in-depth interviews every two months on average for up to a year. These case study doctors described their perceptions of patients (patients were given a code name to ensure confidentiality), their management, their training experience, and any management stress. We abstracted more than 50 patient or family narratives from these interviews, of which 35 involved children in the home. We sent summaries of three of their most extensive narratives to each doctor for confirmation. This article draws on themes from the pre-training interviews, amplified in the case studies.

We recorded and transcribed all interviews with participants' consent, and we sent verbatim transcripts for checking to those who wished to see them. We entered all qualitative data into the NUD*IST4 software program (Qualitative Solutions and Research, Melbourne, Australia) and coded the data iteratively by using grounded theory coding methods up to nine times.10 We evaluated doctors' management practices by comparing them with recommended practice from the relevant literature concerning victims,7 11 men who abuse,12 13 and the management of couples.14 Although no published guidelines for doctors about managing children existed, several good studies have addressed children's needs.15 We discussed preliminary and final analyses with a research advisory group (project representatives, two general practice educators, and a domestic violence worker) to check that the analyses accorded with their experience.


The samples represented a wide range of forms of partner abuse. Doctors described patterns of partner abuse among their patients that varied from chronic and severe multiple abuses to mutual couple conflict; from what Johnson calls “intimate terrorism” to “common couple violence.”16 Some doctors expressed confusion about the boundaries beyond which normal conflict becomes abuse. Legal concepts of physical and sexual assault guided some respondents' awareness of domestic violence (suggested in the quote below), while other doctors, more often from rural areas, included forms of emotional abuse. “In a year, [I see] about three or four… There might be a lot more, but it was difficult; at least with the Sri Lankans… and the Vietnamese people, you can see their body, but with the Turkish and Arabic people you can't see their body.” (Urban male doctor 1.)

Doctors' perceptions could influence their estimation of the prevalence of partner abuse in their practice, and their identification and management of patients, which suggested that in some practices many victims were being overlooked. All these doctors had sought training. They were attempting to help their patients to manage the violence, yet they experienced many challenges. Limited time and the fee for service system are common restrictions affecting the ability to provide quality care for many Australian general practitioners, but other more specific difficulties existed.

Stress and aversion

Many problems contribute to doctors' aversion to working with partner abuse.17 In this study even those who expressed empathy for patients and wanted to help could find the work unrewarding and financially draining because of long consultations. Doctors could be frustrated because of patients who were “non-compliant” with their advice or who did not return.7 Some were discouraged because they got little positive feedback: “You often don't want to be too good at it because you get too many of them… you might find people start referring them to you.” (Rural female doctor 1.) Consequently, these doctors spoke of their own occasional reluctance to acknowledge the problem, even when they had grounds for suspicion.

Responding to victimised women

The doctors described very diverse abused women. Women presented with depression and anxiety, drug or alcohol problems, eating and sleeping disorders, and migraines and injuries as well as with children's ailments.1 Doctors' reactions to their victimised patients ranged from understanding, close identification, and distress to frustration with their inability to engender change. Several doctors were unaware of the barriers inhibiting women's disclosure. They were especially frustrated by women who would not disclose, even when doctors acted on their suspicions and asked directly.7 Most of the urban doctors and a third of the rural doctors believed that the best advice would always be to leave, despite the difficulties women experience.18

Impact of doctors' gender

For some doctors, a lack of professional effectiveness provoked feelings of despair or helplessness. Overall, female doctors believed their gender was advantageous because women would trust them more, they understood women's suffering, and they could identify with women's experiences. They also believed that male patients may find it easier to speak to them about emotional issues. Most male doctors also thought that male patients prefer to discuss their emotional problems with female doctors. However, female doctors said that seeing many patients with psychosocial problems came at some cost.19 Because of their empathy with victims' suffering they expressed more sadness, feelings of frustration, and distress that they had no “magic” remedy. Some consequently felt powerless or demoralised.

Responding to male patients who abuse

Most urban respondents and some rural respondents reported seeing male patients who abused female partners. Twenty men presented to case study doctors with depression, pain, and drug and alcohol abuse, and two presented with mental illness. These seven doctors saw men who attended both in couples and alone, for medical problems, “anger management,” or “wife-mandated” behavioural change. The doctors' responses to their male patients could be uncomprehending, hostile, and distancing: “On a dark night I'd run over them,” “Some Maltese from St Albans who's beaten the s**t out of his wife.”

Men's violence could be variously ascribed to their class, ethnicity, or genetic predisposition. Other doctors spoke of the difficulty accepting that a charming male patient for whom they had long cared was abusing his partner: “It's hard to think of some of the men as abusers if you've been caring for them in other ways and really had no suspicion. I also think there's a tendency to minimise the violence and reassure yourself and the woman that, oh you know, its just bad temper.” (Rural female doctor 2.)

Hostile or ambivalent responses to abusive male patients made it difficult for doctors to respond appropriately to the men's violence. When doctors distance themselves from abusive male patients, they place greater responsibility on women to change their situation, irrespective of the women's ability to do so.

Managing couples

Like Ferris et al's (1999) study of Canadian family practitioners, we found that some doctors have difficulty in managing couples, and the severity of the violence did not necessarily guide decisions about management.5 With neither expertise nor practice evidence about managing the “dual relationship,” well meaning doctors could violate confidentiality, placing the woman at increased risk.14 One doctor described a Turkish couple, of which the woman eventually disclosed violence. The doctor discussed the husband's anger with him (without his wife's consent) when he next presented. The man realised that his wife had disclosed, and neither ever returned. The doctor sought counselling for her anxiety that she may have caused more abuse. As in overseas studies,20 two other urban doctors revealed that they had done this.

Some case study doctors constructed victims as “deserving,” which could affect their management: “[Andrea's] not terribly insightful… got a motor mouth, which is thrown into gear before you open the mouth… When Jack is not on drugs or booze he's a very clever man to talk to… I think the quieter Andrea gets [on tranquillisers], the better the system runs. When she's under control, everything else seems to fit in and go under control.” (Urban male doctor 2.)

Some doctors, while meaning well, offered couple counselling: “I usually get them in together first off and just play round with the words and just see whether she will accuse him in front of me… but I usually don't want to accuse a man. Was he guilty of it?” (Urban male doctor 3.) However, this is contraindicated because of the associated risks.7

A few case study doctors did not want to intervene in a couple, and so overlooked the violence. This might occur because the husband's illness required his wife's care (as in the case below), because the doctor-patient relationship was vested in the couple or family as a unit, or because the doctor did not want to lose them as patients. “It doesn't seem right to do it [intervene], because he's behaving himself at the moment. And I don't think, why bring up a new critical aspect to his life while he's staying off the grog and looking after his health.” (Urban male doctor 2, elderly couple.)

Invisible children

“I had one, just recently, how old was she?… seven. And this mother was saying very clearly, it's not affecting the kid. And yet she was a bed wetter, and she had lots of the classic symptoms… She said, well, I hate it when mummy and daddy fight. And then I said what do you do? I hide in the cupboard, and I take Jack with me. She had this whole behaviour, a way of protecting herself and her brother, who was about two, of getting into cupboards. (Rural female doctor 3.)

With attention firmly on the adult relationships, most doctors (with the exception of the doctor above) overlooked the impact on children. One concern was the potential jeopardising of the doctors' professional relationship with the child's parents: “You get that close to them and how difficult it is then to turn around and accuse them or suggest to them that they're damaging their child.” (Rural female doctor 2.)

Some doctors felt that they had no management skills for children's psychosocial issues. Many did not trust child protection services and were unaware of therapeutic services to assist children.

Disclosure, counselling, and referral

When domestic violence is suspected, doctors are advised to ask directly, inform the women of her options, support her, and refer her to specialist agencies.7 Particularly in rural areas, few agencies exist to which general practitioners can refer patients. Furthermore, doctors are often unfamiliar with those agencies relevant to victimised women, men who abuse, and least of all children. The absence of or ignorance about such agencies could sometimes be compounded by distrust of specific services. Some doctors expressed reservations about the benefits of referral, and little published evidence exists about whether intervention is beneficial or harmful to women or children in the longer term.8 Supportive counselling was the most common strategy doctors used. Insufficient time; absence of appropriate training, debriefing, or supervision; and unfamiliarity with or lack of trust in community based agencies all raised doctors' stress levels, which could lead to a reluctance to identify patients: “Sometimes myself I get depressed and frustrated, I don't know what to do… sometimes you ask yourself, did I do the right thing or not? Did I help or did I make it worse?” (Urban female doctor 1.)


Because of the number of doctors in this study and the circumstances of their recruitment, our findings cannot be generalised. Nevertheless, they supply constructive directions for future research and policy. Some doctors in this study practised in ways that closely conformed to recommended practice. Others experienced serious difficulties, however, despite being motivated and concerned. Before training, doctors in this study were uncertain what to look for, they lacked confidence in knowing what to do if they found partner abuse, and most believed that they were underidentifying the problem. When they encountered a patient who was being abused, they sometimes used management practices that were potentially detrimental to victims and that differed from recommended practice.7 11

Doctors' management could be affected by several factors. Female victims could be perceived judgmentally. Doctors could unwittingly break confidentiality. Doctors could feel hostility towards or strong connections with male patients who abuse. Doctors could prioritise the preservation of the couple, preferring not to intervene in an intimate relationship, and hence ignore the abuse. Doctors could be unaware of, unfamiliar with, or lacking trust in appropriate referral services. Finally, doctors could overlook the children in a family.

Doctors' management strategies consisted mainly of counselling, occasionally marital counselling. Most doctors lacked appropriate expertise, debriefing facilities, support, or supervision to provide counselling. When doctors have no knowledge of recommended best practice or referral agencies and no support, trying to counsel patients in violent relationships could be stressful and disempowering, leading to a growing reluctance to “see” the problem.

Doctors were often not aware of the impact their gender, attitudes, and beliefs had on their practice. Most valued their long term relationships with all family members but had not been trained to reconceptualise “family practice” when violence is occurring. Such reframing would allow an acknowledgment of imbalances in power between men and women, the coercive nature of partner abuse, and the importance of safety.7 These data illustrate that family doctors see the sequelae of partner abuse among all family members. Managing these families can cause considerable tensions in family practice. Acknowledging these difficulties and strengthening support for doctors facing this common problem among their patients are important.


Intimate partner abuse is a damaging social problem affecting the health of many general practice patients. It cannot be solved by general practice alone; however, the doctor's role in identification and referral plays a critical part in society's coordinated response. In order for such a response to be effective, general practitioners need greater exposure to and familiarity with recommended good practice. To date, no guidelines exist about managing all family members, and this needs correcting. Several good medical curriculums cover partner abuse but have not integrated “whole of family” practice.21 22

What is already known on this topic

Primary care research and recommended practice on domestic violence or intimate partner abuse has largely focused on female victims

What this study adds

This study sheds light on doctors' dilemmas in managing all family members, especially couples and children

Future medical education and infrastructure planning should take account of these dilemmas to provide more expertise and effective support to general practitioners

A greater familiarity with recommended practice will increase doctors' ability to manage disclosure, so that if a woman is not ready to disclose, doctors neither feel ineffective nor blame her for her decision but maintain their vigilance and inquiry.23 They will use effective strategies with male patients who abuse,12 and they will inquire of both partners about the impact of violence on children, which can be a powerful catalyst for beneficial change. Doctors need heightened reflexivity, strict standards governing confidentiality and safety, judgment about the inappropriateness of undertaking or recommending couple counselling, and better links with specialist agencies.

Medical administration services should ensure that doctors have access to debriefing; supervision; and legal, police, and welfare agencies. When such assistance is in place doctors will feel better supported and more confident to engage with this critical underlying issue affecting their patients' health and wellbeing.

Key messages for managing partner abuse

  • Use recommended practice guidelines

  • Prioritise safety and confidentiality

  • Do not use couple counselling when violence is occurring

  • Always ask about children

  • Refer to specialist agencies


  • We thank the North West Melbourne and Central Highlands divisions of general practice for their support and involvement and all the general practitioners who generously gave their time to contribute to the study.

  • Contributors AT developed the idea for the study, in collaboration with DHB and DL. AT conducted all the fieldwork and analysis, in consultation with DHB and DL. AT drafted the initial manuscript, and all authors were involved in revising the manuscript for final submission. AT is guarantor for the paper.

  • Funding This research was carried out as part of the doctoral scholarship programme at the National Centre for Epidemiology and Population Health, Australian National University, in collaboration with La Trobe University. This paper was written up by AT as a staff member at CSMCH, La Trobe University.

  • Competing interests None declared.

  • Ethical approval Human ethics committee at Australian National University.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
View Abstract