Domestic violence is a major social and medical problem. It occurs in all countries irrespective of social, economic, cultural or religious values. The World Health Organization (WHO) has reported that population studies around the world have found 10 to 69% of women reported being physically assaulted by an intimate partner at some point in their lives 1. In Malaysia, 39% of women above 15 years of age were estimated to have been physically beaten by their partner 2. A study in the outpatient clinic of the University Malaya Medical Centre (UMMC) revealed that one in seven female patients attending the clinic had a background of domestic violence 3.

The impact of domestic violence is alarming. Fatalities are related to partner homicides or women committing suicide 145. Morbidity as the consequences of domestic violence comes in the form of poor health status, poor quality of life and high use of health services 67. Many abused women suffer acute physical injuries and many other chronic health problems that present as ambiguous symptoms and physical findings 89. Psychosomatic complains and non-specific chronic pains are common 9. These presentations may be treated in health care facilities without identification of the underlying cause, leaving the patient at risk for subsequent episodes of abuse 1011121314.

Abuse usually escalates during pregnancy and represents a significant risk to the health of both mother and infant 15. Abused pregnant women have been reported to have late and poor antenatal check-up when compared to pregnant women without an abusive background 16. The impacts of domestic violence also extend to children of the abused women. A survey by Women Safety Australia (1996) found that 60% of the abused women have children under their care during the abuse and 38% of them reported that their children witness the violence episodes 17. These children suffered injury when they were caught between their fighting parents 18. Children brought up in a home where domestic violence occurs have the tendency to develop behavioral or psychological problems, with risk of poor health in later life 19. There is a close association between domestic violence and child abuse and it is estimated that child abuse occurs in 50% of families with domestic violence 17.

Studies have shown that most of the abused women will keep their experiences to themselves 202122. Those who sought help were most likely to disclose their experience to their close relatives or friends for help 17. In view of their poor health, these abused women were noted to be frequent attendees to health care facilities 15. However, only a small proportion of these victims were successfully identified. It was noted that the lifetime disclosure rate for abuse women to be around 30% and a low general practitioners' inquiry rate of 13% despite high levels of prevalence among those attending health care facilities 2324.

Factors such as shame, embarrassment, fear of partner's retaliation and perception that it is not the doctor's role to intervene are factors that prevent abused women from seeking help from health care providers 1225. Conversely, primary issues like lack of time, inadequate training, uncertainty about how to respond and perceptions of patient non-compliance affect professional response to domestic violence from these doctors 262728.

The medical practitioner's personal value system and beliefs about domestic violence also play an important role. A study in an emergency department in Hong Kong reported that the doctors found it difficult to optimally manage victims of domestic violence because of the belief in the importance of maintaining family unity and that domestic violence is a private issue 29. Fewer doctors were found to screen for domestic violence compared with other behavioral risks, such as alcohol and drug consumption, and risk of HIV/AIDS 30. More doctors also believed that domestic violence intervention was less successful than intervention for tobacco and HIV/AIDS risks 30.

The aim of this study was to assess the knowledge, attitude and practices of primary health care (PHC) provider teams (clinicians and nursing staff) related to the identification and management of domestic violence. The information gathered hopefully may assist in the development of appropriate intervention strategies that lead to improvements in the management of domestic violence.

Out of 188 staff available at the time of the study, only 108 responded to the questionnaires. The clinicians were made up of 61 respondents with response rate of 72.6%. They consisted of 18 consultants/physicians and 43 MOs. The rest of the respondents were from the nursing staff and consisted of 42 nurses and five medical assistants. The low response rate (45%) of the nursing staff has been contributed by difficulty in making contact with the participants in view of shift working time and perhaps minimal research culture exposure among them. Sixty-eight percent of the participants were female and nearly all (98.7%) of the participants had been in service for five years or more. Three of the clinicians (5%) and two of the nursing staff (4%) had personal experience of domestic violence at some point in their lives.

This study yielded important information about the current perceptions and approaches of the health care providers in Malaysia towards the identification and management of domestic violence in the primary care setting. The findings will be useful in guiding the development of appropriate clinical interventions to improve care in primary care settings in general as well as specifically in the UMMC.

However, caution should be taken when interpreting the results of this study in view of several limitations of this study. First, this study was a single study which was carried out in a single primary care location. Thus, the findings may not be applicable to other medical settings, such as in the inpatient or mental health settings. Second, the UMMC is a large public tertiary teaching hospital with its location in a suburban setting and because of this, the findings may not be representative of smaller primary care settings such as the general practice clinics or other medical facilities from non-urban locations. Third, since the data collection was based on self-reporting by the participants, respond and recall bias may result in desirable answer despite the confidential manner of the data collection.

As evident in this study, more than sixty percent of health care providers in UMMC believed that the prevalence of domestic violence among patients attending their clinic to be low. This may be one of the many causes of low screening for domestic violence cases. Seventy percents of clinicians have reported screening for their patients but at times only. There was only one doctor among the 61 clinicians who screened all his/her patients for domestic violence.

Despite perception of low prevalence in domestic violence cases, 65% of the clinicians identified an abused victim within the past year. This is contrary to physicians' perception of low prevalence of domestic violence when the actual prevalence of domestic violence was found to be higher in those attending the health care facilities when compared to population 32. The prevalence of domestic violence cases among patients in various primary care settings varies from 8.5% to 41% 63233343536.

Physical injuries related to abuse may be one of the most obvious symptoms presented to medical facilities. However, only half of the clinicians reported 'always or almost always' asking their patients for any underlying abuse when treating cases of injury. This reported practice of inquiring about abuse is higher when compared to the study finding conducted by Sugg in 1999 31. Other presentations of domestic violence related to psychological, psychosomatic or antenatal problems which are more subtle to its relation to domestic violence were asked regarding abuse by the clinicians in a much lesser frequency. This finding is similar the study by Sugg 31. Clinicians failing to identify and to offer abused women help despite repeated presentations to health care facilities may cause them further abuse when they are send home to the same abusive environment.

Most clinicians in this study reported lack of time as a barrier to ask for domestic violence. This finding is similar to that reported in Sugg et al 27. Nearly a third of the clinicians were still unsure on how to ask regarding domestic violence among their patients which should raise major concern. It would seem reasonable to suggest that the clinicians have not been fully equipped on how to deal with domestic violence cases during their undergraduate or postgraduate training. The results also suggest the clinicians and the nursing staff received minimal or no training on violence management during their service.

Health care providers possess certain opinions and prejudices based on their own upbringing, culture and religious beliefs. These biases can affect their professional behavior including their intention to ask about abuse and create errors in clinical judgment in domestic violence cases. More than half of the clinicians and a third of the nursing staff reported a fear of offending patients in asking about domestic violence. This may be related to the underlying belief that domestic violence is a 'private matter' and not within the scope of medical treatment 29. Nearly a third of clinicians and half of nurses endorsed the view that the abused person must have done something to trigger the abuse. This 'blaming the victim' attitude is a very negative way to address the person who has been victimized when the abuser should be the one to be blamed for using violence to resolve conflict 37.

Traditional beliefs regarding the family privacy, family unity and gender role was found to have posed difficulties to health care providers in their management of domestic violence 29. However, many abused women do not mind being asked about violence and would like the health care providers to be more pro-active in asking questions on abuse 12252838. Furthermore, health care providers need to be aware that domestic violence is indeed a major medical problem and they have important roles to play in its detection and management 39.

Having a safe environment will also enable the health care providers to identify domestic violence. There should be a place for the health care providers to have a private consultation with the victim without the presence of the abuser. In this study, a very small proportion of the participants expressed concern on their safety but a large proportion was concerned about the safety of their patients with in a violent environment.

Perceived self-efficacy plays an important role in the management of various medical conditions. Most clinicians in this study were more confident in asking about smoking and alcohol intake rather than asking about different kinds of abuse. Most of them perceived lack of self-efficacy in the overall management of domestic violence, including the use of strategies to help the abused person and a lack of access to different management information. All these negatively impact on the health care provider's ability to adequately care for abused person or abusers. Factors, such as inadequate training or the perception of poor success in management of these cases are relevant 30.

There is no mandatory reporting for domestic violence in Malaysia. However, in this study, a proportion of health care providers have indicated that they would still report cases of abuse to police despite abused women's refusal to give consent. Not respecting the patient's autonomy can be considered as unethical and may represent institutional victimization. There should be support for the abused patients no matter what their decision is at that point of time. This will increase their self-esteem and confidence level, aspects of their self-image that may have been severely undermined by repeated abused by their partner 22.

Within institutional settings, having enabling factors for the management of domestic violence will make the health care providers more inclined to manage these cases. Less than three-quarters of the participants in this study had access to a social worker, while 10% of the participants were unsure of the availability of social workers to help them manage domestic violence cases. Less than a third of the participants knew of any written protocol for the management of domestic violence. Not even half of the participants knew of the existence of non-governmental organization that can support management of domestic violence. The health care providers may feel inadequate in helping the abused victims with the lack of knowledge on the availability of various domestic violence resources.

Primary health care providers in this study are more inclined to perceive domestic violence negatively and to manage it inappropriately with their limited skill and training in detection and management resulted from inadequate knowledge in domestic violence in general. Negative personal values also impact adversely on health care providers' detection and management.

The authors declare that they have no competing interests.

SO provided the initial concept of the study, conducted data collection, performed the statistical analysis and drafted the manuscript. NAMA participated in the design of the study and data collection. Both authors read and approved the final manuscript.