Research on sensitive issues with specific reference to domestic violence

Research in domestic violence (DV) is gaining more attention in recent years. Having evolved into a growing public health problem, DV also poses a considerable threat to and violation of human rights. WHO (1) defines DV as, “a range of sexually, psychologically and physically coercive acts used against adult and adolescent women by a current or former male intimate partner.” Women are vulnerable to many forms of violence, DV being the most common form (2). In the context of DV, the victim is almost always the woman or the wife and the perpetrator is the man, very often the husband or intimate partner (3, 4). The terms DV and intimate partner violence (IPV) are used interchangeably in the literature and this paper.

Although DV is a global phenomenon and cuts through socio-cultural barriers, the bulk of the literature comes from developed countries. The reasons for the paucity of research in developing countries are: concealing or under reporting of facts, a reluctance to disclose facts, fear of jeopardising family respect, and poor orientation to research. Research on sensitive issues involving vulnerable subjects poses several methodological and ethical challenges to the researcher. Participants in a study of DV are doubly vulnerable, being female in a patriarchal society and being a victim of violence. Therefore, any attempt to study sensitive issues such as abuse/ violence among vulnerable subjects such as women ought to be ethically informed.

Impact of DV on mental health and the need for mental health assessment

Violence in any form impacts one’s mental health. Women exposed to both hysical/psychological violence and psychological violence alone had a higher incidence and severity of depressive and anxiety symptoms, post traumatic stress disorder (PTSD), and thoughts of suicide, than women in the control group (5). Participants reporting DV had higher scores on depression, anxiety, somatisation, low self esteem and were likely to have attempted suicide. Spouses of these participants were found to abuse drugs and alcohol (6). A strong correlation exists between the spouse’s problem drinking and violence against his partner (7, 8). Severe forms of DV were significantly associated with mental health symptoms and health behaviours, including anxiety, depression, difficulty concentrating, memory loss, suicide attempts, weight gain, past history of smoking and illicit drug use (9). While depression, anxiety, PTSD and dissociation are common psychiatric syndromes associated with violence, in some cases, psychological or emotional disturbances may not amount to a syndrome; instead, sub-threshold symptoms may be apparent. Further, mental health disturbances can either be the cause or the effect of violence (10).

Roche (11) found that one-third of all female patients reported at least one form of IPV in their lifetimes. IPV was associated with a low self-rating of physical and mental health, frequent visits to the emergency room and problems with alcohol, drugs, and mental health. The findings emphasise the importance of screening all women for IPV using protocols that address mental health and substance abuse.

Abused women in various settings had increased health problems such as injury, chronic pain, gastrointestinal and gynecological signs including sexually transmitted diseases, depression and post traumatic stress disorder (12). A recent study from India reported mental health correlates of domestic spousal violence based on a household survey method in both urban and rural areas of seven cities (13). About 40% of the sample reported experiencing violence in their marriage. They in turn were at increased risk for poor mental health .

Women who experienced partner violence, especially, in the West, Hispanic women, were at increased risk of not receiving needed mental health care. These findings, therefore, highlight the need for culturally sensitive and specific outreach services
that are aware of the effects of partner violence on women’s mental health, and are informed on how to access these services (14). In the West it is recommended that health service personnel routinely question women about abuse by their intimate partner. This is suggested because of the serious health consequences of DV and the low identification rate of abuse by health services (15). Psychological distress may be present prior to, during, or following the study. Hence assessing mental
health parameters becomes essential and integral to research of this nature. Since violence has a strong psychological sequel, it becomes ethically imperative to integrate mental health assessments in a study of DV.

Methodological issues

Research in the area of abuse involves two components that are unique. First, it includes a possibility that some revelations are occurring for the first time and hence are likely to be emotionally charged. Second, research of this nature, by its very act of disclosure, may involve risk to the respondent. Therefore, the method of inquiry employed in research of this nature ought to be sensitive (16).

Studies on DV both in the West and in India have used household surveys, structured interviews and self report measures. Since psychological distress may be present prior to, during or following the research, several issues in methodology are also important in mitigating the level of distress. The methodological and ethical issues addressed below are interrelated and therefore there may be some overlap. This is primarily because in studying sensitive issues, ethical concerns very often inform or influence the choice of the method. Instances of ethically driven methodology include the following:

1. Research on sensitive issues should make use of culturally relevant definitions of the constructs being studied, e.g., emotional abuse.

2. Tools that are used to assess abuse should either be developed in the same culture or be adapted to suit the requirements of a particular culture.

3. The order of questions should gradually progress from neutral to more sensitive and personal domains and the researcher ought to periodically check the participants’ comfort level.

4. Language and choice of words used to assess abuse ought to be sensitive and not be derogatory, callous and judgmental.

5. The conduct of the interview, manner of probing and termination of the interview often make a difference to the psychological impact. Preludes may be introduced to facilitate transition between different sections of an interview schedule and to provide a rationale for further enquiry.

6. Qualitative methods such as focus group discussions and semi-structured interview schedules are most suited, although questionnaires with a less structured and rigid approach may also be used. This method not only yields rich data, it also offers the researcher adequate flexibility to evaluate and address concerns.

7. Reports of adverse events are likely in violence research hence such studies cannot be bereft of mental health intervention, ongoing follow up, documentation and appropriate referral services (17).

8. Since the researcher and the subject are both affected in a study of this nature, adequate sensitisation, ongoing training, supervision and debriefing of research staff is equally essential (18, 19).

9. Finally, some women may also report positive mental health consequences. Participants often report a sense of relief at discussing abuse in the presence of empathic and nonjudgmental researchers, which should also be documented.

Ethical issues

The Council for International Organisation of Medical Sciences outlined basic ethical principles for human participants enrolled in epidemiological research. These are: respect for individuals, non-malfeasance (minimising harm), beneficence (maximising benefits) and justice (20).

WHO (21) subsequently published specific ethical guidelines for DV research based on the experiences of the International Research Network on Violence Against Women.These are:

1. The safety of respondents and the research team is paramount and should infuse all project decisions.
2. Protecting confidentiality is essential to ensure both women’s safety and data quality.
3. All research team members should be carefully selected and receive specialised training and ongoing support.
4. The study design must include a number of actions aimed at reducing any possible distress caused to the participants by the research.
5. Prevalence studies need to be methodologically sound and should build on current research experience about how to minimise the under reporting of abuse.
6. Field workers should be trained to refer women requesting assistance to available sources of support. Where few resources exist, it may be necessary for the study to create short-term support mechanisms.
7. Researchers and donors have an ethical obligation to help ensure that their findings are properly interpreted and used to advance policy and intervention development.
8. Violence questions should be incorporated into surveys designed for other purposes only when ethical and methodological requirements can be met.

Obtaining informed consent in violence research should be a process rather than a one-time formality. This is because the respondent is often unclear about the questions till they are
actually asked. One of the recommendations is to take informed consent at every level of sensitive information gathering, that is, while moving from questions related to emotional abuse to the section on physical abuse and so on. The informed consent form should clearly state the risks and benefits involved in volunteering for a study of this nature. Refusal rates also need
to be documented, not just for methodological reasons and interpretation, but also to indicate reticence on the part of the respondent or guardian to participate in such research (16).

The physical safety of respondents and interviewer from potential retaliatory violence by the abuser is of prime importance. At the same time women who are both the subjects and beneficiaries of research need to give full informed consent and ensure that the interview process is affirming and does not cause distress (22). The World Medical Association Declaration of Helsinki says, “Health and wellbeing of patients must have priority over any scientific result or benefit”, thus clearly emphasising ethical responsibilities of researchers towards the participants of their study.

Guidelines for integrating mental health assessments in DVresearch

1. Since symptoms of depression, anxiety and PTSD are most prevalent in victims of DV, scales such as Beck Depression Inventory, Hamilton’s Depression Rating Scale, Beck Anxiety Inventory, Hamilton’s Anxiety Rating Scale and PTSD Checklist may be used.
2. If detailed assessment using the above scales is not possible, a brief screening tool such as the GHQ-12 may be administered. Participants who are positive on this can be interviewed in detail to determine the severity of distress and to rule out false positives.
3. Studies using qualitative methodologies can integrate open ended questions on some common symptoms of depression, anxiety, and PTSD in the interview.
4. Mental health assessments may be incorporated after all other assessments, to ensure that participants responses are not influenced by their mood state. However, it is important to terminate the interview on a positive note.
5. If researchers are not trained in the area of mental health, they may have to take inputs from a mental health professional.
6. It also helps to ask the client directly if she has ever felt the need for professional psychological help.
7. Services of a mental health professional may be used to obtain inputs regarding assessment of mental health, training the research staff in the assessment of mental health and identifying victims who may require professional help, supervision and debriefing for the research staff, and to treat/ refer distressed participants.
8. Participants who are significantly distressed need to be referred to mental health professionals. Those who cannot or do not wish to avail of the above, and participants with subclinical distress, ought to be seen by the research staff on a follow up basis.
9. Follow up contact can be made at weekly or fortnightly intervals to assess for any negative outcomes which may not be obvious immediately and to also check for any worsening of violence.
10. It is important to have a liaison with the community and other organisations working with violence around that geographical area, so that they can provide ongoing support.
11. All participants can be given a pamphlet with official contact information of the research staff and available psychological services, which they can access if and when necessary.

Conclusion

Domestic violence affects different domains of a person’s life, and its impact on mental health is the most significant. Therefore, from an ethical standpoint, mental health assessments ought to be integral to research of this nature.