Recognising domestic violence in clinical practice using the diagnoses of posttraumatic stress disorder, depression and low self-esteem

Abstract

This discussion paper reviews the health impacts, physical and mental, of domestic violence and explores the link between domestic violence and psychological symptoms. This paper focuses more on posttraumatic stress disorder (PTSD) than depression and low self-esteem because doctors are less familiar with PTSD. The barriers preventing health workers from detecting domestic violence are reviewed and the fear of health professionals that asking about trauma can harm patients is explored. The article then outlines practical strategies to improve detection of domestic violence using patients' presenting psychological symptoms and the diagnoses frequently associated with domestic violence namely, PTSD, depression and low self-esteem. It is argued that it is inadvisable to try to implement a policy of screening for domestic violence in general practice when the public health model is currently inappropriate. The paper discusses why the diagnostic frameworks of depression and PTSD are helpful in general practice, not only in detecting domestic violence but in working with the patient to establish trust and ways forward that can be tailored to meet the needs of the patient and their children. Patients' and professionals' dilemmas about what to do once domestic violence is detected are briefly explored.

INTRODUCTION

The health impact of domestic violence

Surveys confirm that domestic violence is common. More specifically, intimate partner violence occurs at some point in the lives of one fifth^1,2 to two fifths³ of women attending general practice.

In addition to physical injuries, women in violent relationships have higher than average rates of irritable bowel syndrome, chronic pelvic pain, gynaecological problems and psychosomatic problems (such as chronic pain).⁴ In pregnancy, violence often escalates with higher than average rates of miscarriage, fetal damage and low birth-weight babies.⁴

Bradley reports that women who reported domestic violence were 32 times more likely to be afraid of their partner than women who did not report this;³ they experienced anxiety more than depression,³ and both states more commonly than the general population.³ Campbell⁴ confirms that the type of anxiety disorder most commonly associated with domestic violence is posttraumatic stress disorder (PTSD) (see Box 1).⁵ In a general practice study, PTSD was present in 35% of those who had experienced domestic violence, was often comorbid with major depression and was indicative of experiencing the severe end of the spectrum of domestic violence.¹

Box 1. PTSD summary derived from DSM1V.⁵.

PTSD is diagnosed when:

• ▸The causative trauma results in injury or feels life-threatening.
• ▸Feelings of terror or helplessness are engendered.
• ▸Symptoms diagnostic of PTSD are present at least 1 month after the trauma.
• ▸Symptoms include at least one of five ‘intrusive symptoms/re-experiencing’ symptoms, plus at least three of seven ‘avoidance’ symptoms plus at least two of five possible ‘hyper-arousal symptoms’.
• ▸Symptoms interfere with life.
• ▸PTSD is described as ‘chronic’ if it endures for longer than 3 months.
• ▸PTSD is of ‘delayed onset’ if it starts over 6 months after the trauma.

Patients experiencing domestic violence are likely to attend general practice or other NHS settings frequently, often with medically unexplained symptoms.⁴ A meta-analysis confirms that children witnessing domestic violence develop more psycho–social problems, particularly PTSD.⁶

Understanding PTSD and its connection to domestic violence

Lifetime PTSD affects about 10% of women and 5% of men in the general population.⁷ The sex difference may be due to genetics and/or to the fact that women experience the civilian traumas of childhood sexual abuse, rape and domestic violence more often than men.⁷ Intimate abuse is more common than combat trauma.⁷ Vulnerabilities to developing PTSD include: prior childhood abuse; a sense of loss of personal power and control (‘mental defeat’) during the trauma and subsequent sense of alienation; on-going injuries; anger; litigation and financial worries after the trauma.^8,9 Those who knew their assailant were more likely to experience mental defeat.⁸ This may be because power and control of the victim by the perpetrator is at the heart of abusive relationships (see the feminist, sex-biased wheel diagrams ‘Features of abusive relationships’ and ‘Features of healthy relationships’ from Domestic Abuse Intervention Project, 206 West Fourth Street, Duluth Minnesota 55802, US at http://www.duluth-model.org/wheels.html). Life-threatening traumatic events are common but any resultant PTSD is usually short-term, ‘acute’ PTSD. It is the chronic form of PTSD that is likely to result in large social and health costs.⁷ The types of trauma likely to cause chronic PTSD are abuse by intimates^7,10–12 and war experiences.

Experiencing intimate abuse may feel similar for the individual to experiencing war incidents, as a journalist's account of her childhood witnessing of domestic violence demonstrates.

‘… I can't remember a day when I wasn't afraid. My mother and I lived like wary animals, ever watchful, tense and waiting, … for the next beating. … It's the noise that people don't understand, the shouting, … the sound of punching fists making contact with flesh and bone. Even blood makes a sound when it sprays onto the wall.’¹³

Barriers to detecting domestic violence and suggestions for overcoming them

Despite its high prevalence in general practice, GPs detect less than 20% of those experiencing domestic violence, even when there is a physical injury present.²

How this fits in

The government is encouraging health professionals to engage in ‘routine enquiry’ for domestic violence. This paper highlights concerns that this is an inappropriate policy for current general practice. This paper advocates instead a diagnostic strategy based on an improved understanding of the psychological consequences of domestic violence for the patient in order to increase detection and provide more appropriate support to the patient.

A summary of findings from survey based research that attempts to understand professionals' non-detection of domestic violence follows.

Health professionals' attitudes, beliefs and fears

Richardson et al¹⁴ surveyed the attitudes of 700 primary care staff. The majority recognised that domestic violence damaged health, and many also believed that they do have enough time to ask about domestic violence. However, only 10% of GPs thought they should routinely ask, and only 1% claimed to actually do this. Forty-five per cent of GPs and 85% of health visitors handed out information leaflets to patients if they discovered domestic violence.¹⁴

In another study, 90% of health workers in mental health, obstetrics and gynaecology and primary care felt domestic violence was an important healthcare issue and less than 20% thought that they did not have enough time to ask.¹⁵ More obstetric and gynaecology workers felt uncomfortable asking about domestic violence compared to mental health workers.¹⁵ Those clinicians with a greater knowledge about the likely mental and physical sequelae of domestic violence were less hesitant to enquire about domestic violence.¹⁵ Professionals with more knowledge about domestic violence displayed more helpful attitudes towards those experiencing domestic violence.¹⁵

What women thought about being asked about domestic violence

Bradley's survey showed that 78% of women attending general practice for any reason thought it would be ‘all right’ for their doctor to ask about violence in relationships, while 13% were uncertain, and 7% would not welcome direct enquiry when they were attending for some other problem.³ Another study suggests that traumatised individuals are actually desperate for health workers to ask them why they were ill, anxious or depressed, and not simply to diagnose and manage the condition.¹⁶ Results from a meta-analysis of qualitative studies found that what women who had experienced domestic violence wanted from healthcare professionals were non-judgmental, non-directive, individually tailored responses that demonstrated an appreciation of the complexity of partner violence.¹⁷ Asking women again about partner violence was seen as appropriate when the women were at a stage in their relationship with an abusive partner when they were thinking about making changes.¹⁷ This tends to confirm that asking about relationships in the context of presented distress is unlikely to cause offence, but that inquiring about relationships at home may not be welcomed when the patient has come about something else.

GPs and other health professionals often do not ask about traumatic experiences, including domestic violence, during their consultations with patients.^2–4,14 A possible explanation for this may be because of concerns that to do so may harm patients. The English language research on whether we harm patients by asking about trauma is limited. Electronic databases were searched for any type of study published from 1985. The databases were: MEDLINE, EMBASE, COCHRANE, PsycINFO, AMED, PAIS, CINAHL, RCN Journals Database, British Nursing Index, ASSIA, SIGLE and PILOTS database on PTSD (http://www.dartmouth.edu). Reference lists from retrieved research papers were also searched. The few papers identified^18–23 suggest that uncovering trauma is not harmful per se, but that rehearsing details of specific traumas in the wrong context, may be harmful. This is almost certainly true of the technique of ‘debriefing’ after a disaster.²³ In debriefing, the victim of a trauma receives counselling within the month following the trauma with the aim of preventing the development of PTSD. (See Box 1 for diagnostic criteria for PTSD). Unfortunately, this type of counselling paradoxically may make it more likely that the victim develops PTSD.²³

Ways forward

So what is the right approach to asking about trauma? We must remember that the traumatised and those with established PTSD are often experiencing intrusive reminders of the trauma such as flashbacks and nightmares. Thus, the trauma is frequently at the forefront of the patient's mind, although unbidden. The intrusive experiences of PTSD are not under conscious control but usually triggered by environmental cues. That is why such intrusive, unbidden memories are so disconcerting for the patient. This sense of being at the mercy of the traumatic memories and flashback re-experiences and in a state of heightened, anxious, somatic vigilance, which may include experiencing panic attacks and being startled easily, encourages the patient to consciously try to avoid all triggering reminders of the original trauma. These three combined behaviours and symptoms make up the cluster that is PTSD (Box 1.)

So I would suggest that someone asking about the type of trauma the patient has experienced is unlikely to add substantially to their already substantial distress. In my experience as a GP, discussing the type of trauma, but not the content of the trauma, is safe in general practice and indeed is usually welcomed.

One third of those experiencing domestic violence develop PTSD¹ and for those patients making the link between life-threatening trauma and current PTSD symptoms such as flashbacks, nightmares and panic attacks can bring a sense of great relief. If the doctor clarifies the link between traumas suffered and symptoms of depression and/or PTSD symptoms, the patient realises that the symptoms they are experiencing are explicable as a recognised human response to life-threatening trauma. The relevance of the patient's domestic violence experiences to such diagnostic questioning is then clear to both professional and patient. PTSD is a psychiatric diagnostic label that patients find non-stigmatising.¹⁶ Additionally, a diagnosis of PTSD, by intrinsically acknowledging the role of trauma in patients' subsequent symptoms, encourages a non-blaming attitude in medical staff,²⁴ which is a welcome respite for patients whose traumatic experiences tend to generate feelings of shame and self-blame. Similarly, the role of domestic violence as a potent factor in the patient's declining self-esteem and development of depression becomes clear to doctor and patient once the doctor's enquiry about types of trauma the patient may have experienced has elicited a confirmation of domestic violence from the patient. Again, making the connection with the patient between symptoms and causative trauma, may help the patient to make choices about what they do next.

So, while professionals may be right to be cautious about when and how they ask about major traumas such as domestic violence, the non-judgmental acknowledgement by the doctor that a patient has experienced a significant trauma is usually helpful for the traumatised consulting patient.

How to detect domestic violence

To return to the problem of low detection rates of domestic violence, two possible approaches are outlined, the diagnostic approach and the screening approach:

Improving diagnostic skills

The possibility of intimate abuse (in childhood or adulthood) should be considered by clinicians when patients have multiple, ill-explained complaints. Multiple injuries and bruising, (especially to face, arms, breasts and abdomen), loss of consciousness, and drunkenness are significant, but non-specific markers of domestic violence.⁴ Given that physical signs of domestic violence have often faded before the victim presents to primary care, health professionals need to be able to detect the longer-lasting psychological scars, such as depression, low self-esteem and PTSD.⁵ There is most room for improvement in our recognition of PTSD (Box 1).⁵ Research shows that doctors frequently miss PTSD.^25–27 The condition is ‘misunderstood and incognito’.²⁸ In a recent study, just under 30% of GPs had the knowledge to recognise PTSD compared to 90% for depression.²⁹ Hence, this article has spent some time describing PTSD to make it less misunderstood, and focuses less on depression, which GPs are already more aware of.

PTSD is a new psychiatric name for human anxiety symptoms that are as old as history itself; but increasing familiarity with the diagnosis, and the judicious use of validated, self-report diagnostic questionnaires in the waiting room (such as the Posttraumatic Diagnostic Schedule),³⁰ should help us improve our detection rates of both PTSD and the domestic violence that may have caused it. For clinical use, copyright copies can be bought from http://www.pearsonassessments.com/tests/pds.htm.

Asking the questions required to make a diagnosis of PTSD in patients with symptoms suggesting anxiety and/or depression enables clinicians to identify domestic violence because establishing the type and severity of the causative trauma is intrinsically and specifically part of making the diagnosis. All the other markers for domestic violence (above) are non-specific. The importance of recognising PTSD, particularly in psychiatric and general practice, was recently emphasised to ensure more appropriate help is given to patients,²⁸ particularly when just under 40% of patients diagnosed with anxiety/depression by their American primary care physicians were found to meet diagnostic criteria for current PTSD.²⁵ Of the patients with PTSD in this latter study, 47.6% reported experiences of child abuse and 39.4% reported domestic violence.²⁵

Screening

Domestic violence does not meet many accepted criteria for screening.³¹ Richardson¹⁴ confirms the inadvisability of routine enquiry in healthcare settings until health professionals have gained confidence in how to ask about domestic violence. Without a reliable diagnostic test with good sensitivity and specificity, a screening programme cannot be implemented. Also the outcomes of the condition and benefit of early intervention needs to be clear. While the benefit of professional intervention in the more severe end of the domestic violence spectrum is clear, in terms of preventing deaths and improving health,³¹ when and how to benefit the rest of the spectrum requires further research.

The conundrum of how to achieve better detection rates remains. I suggest that a medical diagnostic approach continues to be useful but only if doctors' ability to diagnose depression, and particularly PTSD, is improved. Without such diagnostic skills we will fail to improve detection of domestic violence in primary care. These diagnostic methods have the advantage of concurring with current medical practice and acknowledge the harm that the domestic violence is doing to the recipient during the diagnostic process. But this approach is only as good as the clinicians' diagnostic skills.

Improving diagnostic skills is consistent with clinicians' current practice, whereas routinely asking people about private traumas feels uncomfortable. Improving diagnostic skills, particularly the professional awareness of how to make a diagnosis of PTSD, may enable professionals to ask about domestic violence more often, although this too is unresearched.

Diagnostic relevance may enable the clinician to ask about private intimate traumas, whereas an exhortation to screen all patients for domestic violence will only be done by the highly motivated few.

Patients' coping strategies and the health professionals' dilemmas if domestic violence is identified

Aside from fears that even asking questions about traumatic experiences may harm the consulting patient, health professionals also have concerns about what to do if domestic violence is identified¹⁵ and how best to intervene for optimum outcomes for the family.³² These fears of the professional may also inhibit them from asking their patients about possible domestic violence.

For their part, patients experiencing domestic violence have anxieties about what will happen to them and their children if they disclose abuse at home. They may prefer not to disclose domestic violence and to suffer in silence because to do otherwise may set in train mandatory professional responses that spin out of the patient's control and that they perceive as undesirable, such as the involvement of the social and healthcare services in their children's welfare.^33,34 For example, there is evidence to show that mandatory reporting of domestic violence to the police (such as in some American states) deterred abused women from disclosing domestic violence in the first place.³³ There is also evidence to suggest that the abused fear that their experience of domestic violence may escalate if the abuser becomes aware of public disclosure.³³ It is known that the victims of domestic violence are most at risk as they try to leave the relationship.³⁵

One dilemma for professionals is whether professional interventions for domestic violence are effective.³² Extensive searching of a number of electronic databases confirms that outcomes research in this area is sparse. A protocol for collecting research on outcomes has been established in the Cochrane database however.³⁶

It can be predicted that families suffer emotional and financial loss when they flee from their home and community. In many cultures, social stigma and shame accompanies divorce, whatever the grounds. This raises the question as to whether victims' losses are compensated by gains after professional intervention.

Most primary care health workers have information in the surgery about women's refuges, police, legal and housing services that are geared to help those escaping domestic violence. However, while many of these agencies may be knowledgeable and willing, they are also often under-resourced and may not function well as a result.³⁷ Lack of places in a refuge, over-stretched police departments, inadequate numbers of social workers with excessive caseloads, and the precarious funding of voluntary services supporting women and families, all place victims at risk once again. Under these circumstances, to do nothing may seem safer than referring women on. When the inter-agency journey works well, it may be life-enhancing for the patient. When it fails, it may even contribute to the death of the victim.³⁷

I would suggest that the detection of domestic violence in a distressed patient is always acceptable, provided that the decision about what to do remains with the patient as much as is legally possible. This does, however, raise a further professional dilemma for the health professional caught between professional responsibilities towards the nominal patient, including respecting confidentiality and autonomy, and wider responsibilities, now enshrined in law, to protect both mother and children. While the professional would like to support the patient in whatever they think appropriate, they will also need to assess the likely risk of damage to the patient's children that staying in the relationship poses. The professional is obliged by law (Children Act 1989) to put children's needs first.

In Britain, the Laming report (Laming rec 16, para 17.116),³⁸ the Human Rights Act 1998, and The General Medical Council (GMC) (page 16, para 29)³⁹ discuss when patient confidentiality should be breached and exhort professionals to share such information across agencies, irrespective of the mother's wishes if her consent cannot be obtained. The GMC is clear that if you choose not to share information, for example with a social worker, when children may be being harmed, then you must clearly record your reasons for not doing so, discuss the issues with an experienced colleague and be prepared to justify your decision.

Despite exhortations and protocols for action from government and professional bodies, when the longer-term outcomes of professionals' interventions are clearly complex and inadequately researched, it is unsurprising that many health professionals cite concern about the outcomes for families that might result from professional interventions as being a frequent reason for their failure to implement screening for domestic violence.³²

It is likely that even with much-needed, well-designed research into outcomes from interventions by professionals, the complexity of multiple social and psychological variables will still require professional responses to be tailored to individual patients or clients. Reports and research papers^16,17,33,34 suggest that victims of domestic violence want help but do not want to lose further control of their lives in the processes that can follow a disclosure of domestic violence.

What to do when domestic violence is found

For those experiencing domestic violence, the first priority is to discuss practical issues of safety and provide information leaflets with contact numbers for police, refuges, and solicitors. Supporting the patient to solve problems is likely to diminish depression and boost self-esteem. For those experiencing the symptoms of PTSD, it is less likely to become chronic if safety and adequate social supports are in place.⁴⁰ How much this can be achieved in reality must be explored with the patient. PTSD treatment⁴¹ cannot succeed unless the patient feels safe. A practice library of deposit-loaned self-help literature^42,43 may be helpful in supporting patients awaiting cognitive behavioural therapy for PTSD or depression. Selective serotonin reuptake inhibitors are helpful for both PTSD and depression.⁴¹

A key dynamic played out in abusive relationships revolves around power and control. Given that a sense of mental defeat and loss of personal power predicts the development and severity of PTSD,⁸ and the loss of self-esteem, another aid I find useful in supporting GP–patient communication, in what are often very complex decisions about what to do next, are the wheel diagrams mentioned previously that can be found at http://www.duluth-model.org/wheels.html. These diagrams outline respectful and disrespectful dynamics in relationships. The diagrams are sex-biased and culturally specific to the feminist, American Duluth Project. The rules of copyright do not allow modification of the wheel diagrams if they are reproduced. Some of the behaviours described in the wheels are as follows:

In the ‘power and control’ wheel:

• using coercion and threats of harm to her or the children;

• economic control;

• treating her like a servant;

• preventing her seeing others; and

• possessiveness.

In the ‘equality’ wheel:

• negotiation to resolve conflicts;

• economic partnership;

• shared responsibility for work distribution and children; and

• honesty, accountability, mutual support and trust.

Nevertheless, I have found them useful in clinical practice because patients can use the diagrams to assess their relationships and see the reality of their situation anew. I have found that this can allow them to decide to make changes. A summary of managing the consultation, or a number of serial consultations, and referral on as appropriate, is delineated in Box 2.

Box 2. Consultation summary points.

• ▸If health professionals detect depression, PTSD or low self-esteem in a patient, the professional should ask about domestic violence.
• ▸
  Questions to uncover domestic violence and establish the severity of trauma experienced include:

  • ‘How are things at home?’
  • ‘Do you ever feel controlled by or afraid of your partner?’
  • ‘Have you ever been hurt by your partner?’
  • ‘How long ago did this happen?’
  • ‘Did you think you might be killed?’
  • ‘Did you feel helpless or terrified?’
• ▸A structured approach to diagnosing PTSD and depression can help improve discussion with, and support of, the patient who is experiencing domestic violence.
• ▸Healthcare workers should be able to offer accurate information (written or verbal) regarding local support services for survivors of domestic violence.
• ▸Domestic violence survivors want to retain a sense of responsibility for their own care and future. This must be respected. The current legal framework and limits to confidentiality should be made explicit.

Given our current lack of research evidence about the outcomes of interventions in families' lives, I think it is reasonable to use the typically expanded medical paradigm usually employed by the GP where the social and psychological context of illness is always taken into account. By detecting domestic violence and identifying the needs of the patient and their children, the GP may be able to help prevent some deaths and certainly mental if not physical injuries. Outcome studies are needed to explore this further.

Competing interests

The author has stated that there are none