What You Don’t Know Can Hurt You: The importance of family violence screening tools for family law practitioners
Analysis of Screening Tools
Summary of Screening Tools
Eighty-six (86) FVSTs were examined. Given the large number of tools reviewed, the tool statistics and content are best summarized by the research tool that was created to track and analyze tools. See Supplementary Document.
Use of FVSTs across Sectors and Settings
More than half of the FVSTs included in this review were developed for use in the health care sector (Figure 2). Approximately one-quarter of the tools were developed for use in family law and mediation; the remaining tools were developed for use in other sectors and settings.
Fifty-three (62%) of the tools are intended to be administered by the practitioner, 26 (30%) are intended to be self-administered, and four (5%) can be either practitioner- or self-administered.
There is an average of 11 questions per tool, with some containing 70 or more and some as few as one. Frequency counts for the number of questions in each tool were limited to primary questions that would be asked of everyone being screened, regardless of FV history or disclosure. Follow-up questions (i.e., those asked only if the client responded affirmatively to a previous question and therefore not necessarily asked of everyone being screened) were not included in the final count and therefore the total number of questions refers to the minimum number of questions that a client/patient could be asked specific to FV.
Question Type
Across all tools, the most common format for asking screening questions was to use yes/no questions (67 tools, 78%). The next most common format was to use open-ended questions (31 tools, 36%) and Likert-type rating scales (e.g., strongly disagree, disagree, neither agree nor disagree, agree, strongly agree) (25 tools, 29%). Most tools included multiple question types, ranging from one to five different question types. Tools used by mediators and, less commonly, lawyers most often used a combination of yes/no and open-ended questions.
Question Content and Phrasing
Based on all of the tools reviewed, 49 question categories were identified across the 86 tools. The five most common question themes (out of 49) related to physical abuse, fear of the partner, threats of harm, sexual abuse, and emotional abuse. Figure 4 illustrates these five common themes. The most common question phrasing for each theme is listed below.
1. Physical Abuse by Partner (80% of tools)
The majority of tools inquire about physical abuse by giving approximately three examples of physical abuse and asking whether or not they have occurred. The most commonly used examples are: “hit, kick, hurt, push, or slap,” with “hit” being the most commonly used example.
Of the 69 tools that contained questions relating directly to physical abuse, 61 (88%) of those tools asked behaviourally-specific questions. Fifteen tools specifically referenced acts of strangulation/choking (17% of all tools, 22% of tools referencing physical abuse).Non-fatal strangulation or choking is a form of physical violence and a risk factor that can predict serious future violence or homicide (Douglas & Fitzgerald, 2014). The Nursing Research Consortium on Violence Against Women altered the Abuse Assessment Screen (AAS) in 2007 to screen for this form of violence (Laughon, Renker, Glass, & Parker, 2008). While research does not reveal any studies on how this change improved or otherwise affected the tool’s sensitivity, Laughton and colleagues (2008) note that the addition of the word “choke” to a screening tool does not increase the length of the tool by any significant amount, nor should it decrease a tool’s usefulness.
Common Question Phrasing: Has [someone/your partner/the other parent etc.] hit, kicked, hurt, pushed or slapped you?
2. Fear of Partner/Feeling Unsafe at Home (65% of tools)
Fifty-six of the analyzed tools ask whether the interviewee is afraid of their partner. The majority ask this question as: “Do you ever feel afraid of [your partner/someone in your life/insert name etc.]?” or in a slightly varied form. Another common format is: “Do you feel safe at home/in your relationship?” A few of these tools (nine tools) expanded the question to ask about fears for the safety of children, typically phrasing this question as: “Do you have any concerns for your children’s safety?”
A third common format asks whether or not the other person causes the interviewee to worry about their safety or has done something or said something which had this effect. The DV Screening Tool for Consumer Lawyers, for example, asks, “Has your partner ever acted in ways that scare you?” (Sussman & Carter, 2007). These questions ask the interviewee for a reference point for their fear, when perhaps the cause for their fear cannot be reduced to particular statements and/or actions.
Stark (2007) points out that “violence in abusive relationships is ongoing rather than episodic, that its effects are cumulative rather than incident specific” (p. 12). Stark (2007) references a national survey conducted in Finland which found that fear was most prominent with women whose partners had not been physically violent for many years, but who experienced abuse in the form of coercive control and mental torment.
Questions that ask if there were specific actions that inspired fear may not capture fear that is grounded in a pattern of behaviours including multiple tactics of abuse over time. First, there may not be a pinpoint action that contributes to the sense of fear. Second, there may be concern that the fear would seem disproportionate to a third party in relation to the specific actions which can be referenced without an understanding of these other patterns. It is possible that the more general phrasing of simply asking, “Are you afraid of [insert name]?”, will better capture situations where the fear is inspired by nonviolent patterns of coercive and controlling behaviours.
Common Question Phrasing: Do you ever feel afraid of [your partner/someone in your life/insert name etc.]?
3. Threats of Harm to Client/Patient or Others (59% of tools)
Threat-based questions focus primarily on threats of harm directed towards the interviewee and/or their children. Fifty-one of the analyzed tools contain questions touching on this theme. Threat-based questions that did not fall within this specific theme captured threats to take the children and/or threats relating to the interviewee’s immigration status. These types of threat-based questions were significantly less common in the analyzed tools. Approximately 17% of all tools contain questions relating to threats to take the childrenFootnote 11 and only 1% of all tools referenced immigration-related threats.Footnote 12 A history of threats to take the children (Araji, 2012; Bemiller, 2008; Coy et al., 2015) is more relevant in the family law context, however, given the need to arrange safe access schedules.
The majority of questions related to threats of harm focus on threats against the interviewee rather than against the children. Once again, it may be more important to inquire about threats against the children in the context of a separation, as an abuser may use children as substitutes for control after separation (Araji, 2012; Bemiller, 2008; Coy et al., 2015; Hayes, 2017). A history of using the children in this way could indicate that a former partner will engage in future tactics of control involving the children throughout the family court proceedings and in the resulting custody and access arrangement.
While most questions contain the generic phrasing, “threatened you”, or “threatened to harm/hurt you”, some contain specific references to a weapon (i.e., “threatened you with a weapon”) and to homicidal threats.
Common Question Phrasing: Has [insert name/your partner/the other party etc.] ever threatened [you/to harm you/to hurt you/to kill you]?
4. Sexual Abuse/Forced Sexual Activity (49% of tools)
Forty-two of the tools directly reference some form of forced sexual activity. This focus on force is not consistent with Canada’s sexual assault laws, which are focused on consent.Footnote 13 It is of note that 74% of the tools were developed in the United States, while 19% were developed in Canada (see Figure 5).
Few of the screening tools reference the use of threats to coerce sexual activity.Footnote 14 The Conflict Tactics Scale-Revised (Straus et al., 1999) is one tool which does ask whether threats were used to coerce sex and whether sex was insisted upon without the use of physical force. Other tools, such as the Initial Domestic Abuse Screening Guide from the Battered Women’s Project (Davis et al., 2015) and the Abusive Behavior Inventory (Shepard & Campbell, 1992), use the word “pressured” rather than “forced”. This wording may better capture instances of sexual abuse where physical force was not applied but the interviewee was not subjectively consenting to the sexual contact. While the word “forced” as it is used in the tools may well be read to capture instances of non-physical coercion, its typical affiliation with physical force may cause the interviewee to respond negatively even where sexual abuse has occurred.
Common Question Phrasing: Has [insert name/your partner] ever forced you to have [sex/sexual activities] that [you did not want/you were not comfortable with]?
5. Emotional Abuse/Insults (45% of tools)
Questions were included in this category when they either referenced emotional abuse directly, i.e., “Has your partner ever abused you emotionally?” or asked the interviewee if their partner engaged in activities that “put them down” in some way. Most tools inquired about emotional abuse by asking whether the interviewee’s partner ever put them down or called them names. For example, the Partner Abuse Assessment Questionnaire (New Brunswick Dept of Justice, 2011) includes a question that asks: “Did you or your partner call each other names or swear at each other?” If the client responds affirmatively, the tool recommends additional follow-up questions. To illustrate a question that asks about emotional abuse more broadly, the OAS (Weiss et al., 2003) asks, “Are you presently emotionally or physically abused by your partner or someone important to you?”
Emotional abuse of this form is often tied to the presence of fear, which enhances the abuser’s power and control over their partner. Emotional abuse itself refers to a very broad category, including various activities beyond the commonly listed: “put them down or called them names.” Faver and Strand (2007) define emotional abuse as including a number of “verbal and behavioral tactics directed toward the aim of achieving and maintaining dominance, or power and control, over the victim”. Affirmative answers to the questions of, “Are you afraid of your partner”, and even, “Has your partner ever harmed/threatened to harm your pets?” are also, then, indicative of emotional abuse. For this reason, Faver and Strand (2007) recommend including questions relating to pet abuse as a part of the screening process to detect emotional abuse. Of the tools that subdivided questions into abuse categories, the MASIC (Holtzworth-Munroe et al., 2010) listed pet-related questions as a form of coercive control, the Duluth Power and Control Wheel (Domestic Abuse Intervention Programs, n.d.) captured harm to pets under “intimidation”, and the Interview Guide from the Battered Women’s Justice Project (Davis et al., 2015) listed “harm to pets” as a form of emotional abuse. No tool listed “harm to animals” as a tactic of emotional abuse, likely because of a focus on companion animals.
While the tools typically used the word “pets,” “harm to animals” is a broader category than “harm to pets,” and questions inquiring about harm to animals may capture patterns of abuse that, for instance, are economic in nature. Asking about “harm to animals” can include livestock animals that are used to generate income and service animals trained to provide assistance in addition to companion animals that are more associated with the term “pets”. Harm to service animals may have an additional physical and/or isolating impact on the interviewee compared to companion pets.
Including questions broad enough to capture harm to livestock (phrasing it as “harm to animals”) may be important in the family law context because these animals are assets whose value must be divided at the time of separation. If the abuser continues to harm the animals after separation or if the animal abuse escalates, this could affect the income the interviewee is able to receive from the sale of livestock if/when they are sold (Saskatchewan SPCA & STOPS to Violence, 2006). Also, broadening the question to include threats to animals captures instances where the animals were not actually harmed, but were used by the abuser to coerce and exert control over the interviewee (Newberry, 2017).
Common Question Phrasing:
- Has [insert name/your partner] ever put you down or called you names?
- Has [insert name/your partner] ever damaged/destroyed or harmed/hurt things that you cared about, including your pets?
Summary of Most Common Question Phrasing across 5 Most Common Themes
- Has [someone/your partner/the other parent etc.] hit, kicked, hurt, pushed or slapped you? (Physical)
- Do you ever feel afraid of [your partner/insert name etc.]? (Fear)
- Has [insert name/the person who hurt you/your partner/the other party etc.] ever threatened [you/to harm you/to hurt you/to kill you]? (Threats of Harm)
- Has [insert name/your partner] ever forced you to have [sex/sexual activities] that [you did not want/you were not comfortable with]? (Sexual)
- Has [insert name/your partner] ever put you down or called you names? (Emotional)
- Has [insert name/your partner] ever damaged/destroyed or harmed/hurt things that you cared about, including your pets? (Emotional - Pets)
FVST Comprehensiveness
The FVSTs vary with respect to their degree of comprehensiveness. “Comprehensiveness” refers to the number of question themes covered by each tool. For example, a tool that included questions about physical violence, emotional violence, and sexual violence is considered more comprehensive than a tool that includes questions about physical violence only. Table 2 lists the most and least comprehensive FVSTs included in the analysis.
Tool Name | No. of Question Themes (out of a possible 49) |
---|---|
Most Comprehensive | |
Mediator Assessment of Safety Issues and Concerns (MASIC, Holtzworth-Munroe et al., 2010) |
32 (65%) |
Partner Abuse Assessment Questionnaire (New Brunswick Dept of Justice, 2011) |
28 (57%) |
North Dakota Mediator Domestic Violence Screening Tool and Safety Planning (North Dakota Supreme Court, 2017) |
22 (45%) |
Minnesota State Bar Client Screening to Identify Domestic Violence Victimization (Minnesota State Bar, 2013) |
21 (43%) |
Battered Women’s Justice Project (Davis et al., 2015) |
20 (41%) |
Least Comprehensive | |
Simple Domestic Violence Screen (Siemieniuk, Krentz, Gish, & Gill, 2010) |
1 (2%) |
Partner Violence Screen (Feldhaus et al., 1997) |
2 (4%) |
Jellinek Inventory (Kraanen et al., 2013) |
3 (6%) |
Note: Percentages were calculated based on a possible total of 49 categories and refer to the percentage of categories represented in the tool.
Generally, FVSTs developed specifically for mediation and family law contexts were more comprehensive (i.e., covered a broader range of question themes) than FVSTs developed for brief screening in health care settings. Rossi and colleagues (2015) compared the detection rates of the MASIC (the most comprehensive tool in our analysis) with the Multi-Door Screen (one of the less comprehensive tools in our analysis with only five question themes represented) and found that the MASIC, which assessed forms of abuse such as a coercive control and stalking not covered by the Multi-Door Screen, detected significantly more cases of abuse.
While Table 2 identifies the tools that were the least comprehensive, this does not mean these tools are ineffective. The Domestic Violence Evaluation (DOVE, Ellis & Stuckless, 2006)Footnote 15 closes the screening process by asking the open-ended question: “Is there anything you wish to add?”, which could lead to a number of disclosures even though this question does not fall within any of our established themes.
Less comprehensive tools may also be constructed anticipating that certain answers will result in further follow-up. Twenty-two (26%) of the tools examined either directly provided suggested follow-up questions or referenced a need for further follow-up if the main questions detected an abuse history. Follow-up questions (often separate from the main screening questions) were frequently meant to assess the level of risk. Some tools, such as the Jellinek Inventory for Assessing IPV (Kraanen et al., 2013) and the Mediator In-Person Screening Protocol (Michigan Supreme Court, 2006), that did not provide the risk assessment questions, still referenced separate risk assessment tools that could be used as a follow-up to the detection of violence. Many tools that did not reference a follow-up assessment still referenced procedural steps when a screening is positive.
Frequency of FVST Use Among Practitioners
A common finding from the literature reviewed was that FVSTs are underused by FLPs and in health care sectors (Davis & Harsh, 2001), whereas they are part of the mediation accreditation standards across the country.
For example, Part 3 of the Regulations to British Columbia’s Family Law Act (2012) sets out training that mediators must take to become accredited, which includes a component that covers identification and assessment of family violence and power dynamics. In Ontario, the standards set by the Ontario Association of Family Mediation require mediators to screen all clients for any occurrences of abuse and/or power imbalances (Ontario Association of Family Mediation, 2013). Elsewhere, Family Mediation Canada (2018) sets the standards for training and accreditation for family mediators.
In a study of family law practices in Canada conducted by the Department of Justice (Bertrand et al., 2016), almost 70% of lawyers and 47% of judges reported often or almost always screening clients for FV. Two percent of lawyers and almost 10% of judges reported that they never screened for FV. Of those lawyers who reported screening, over half (53%) reported never using a standardized screening tool, 25% reported that they rarely used a standardized tool, and only 13% reported often or almost always using a standardized tool. The Department of Justice will be conducting this survey again in 2018.
A recent literature review (Costa & Barros, 2016) reported the most frequently used FVSTs were the CTS-2 (Straus et al., 1999), Abuse Assessment Screen (Parker & McFarlane, 1991), and the WHO Violence Against Women Questionnaire (Ellsberg et al., 2008). Other commonly used tools in their review included in our analysis were the Sexual Experiences Survey (SES, Koss et al., 2007), Composite Abuse Scale (CAS, Hegarty et al., 2005), Partner Violence Screen (PVS, Feldhaus et al., 1997), Severity of Violence Against Women (Marshall, 1992), Women’s Experience with Battering (WEB, Smith et al., 1995), and the Hurt, Insult, Threat, Scream (HITS, Sherin et al., 1998).
The FVST analysis conducted for the current study did not identify information about the frequency of use of specific FVSTs. However, informational interview data corroborated the findings from the literature that FVSTs are largely underused by family law lawyers. Future research could further examine frequency of FVST use by surveying family law lawyers.
Assessment and Scoring Procedures for Determining the Presence of FV
An important component of FVSTs is the set of procedural steps practitioners use to make a determination as to the presence of FV. This determination will ultimately inform service provision, including the decision to terminate services (such as in the case of meditation) or the need to report the abuse to authorities (as is the case when child abuse is detected). The FVSTs analyzed revealed a range of assessment and scoring procedures (see Figure 6). These assessment/scoring procedures can be grouped into two broad categories:
- Determination of abuse based on practitioner judgement (52% of tools)
- Determination of abuse based on standardized scoring procedures (46% of tools)
Seven different assessment/scoring procedures for determining a positive screen for FV were identified:
- Practitioner judgement based on open-ended questions
- Practitioner judgement based on close-ended questions
- Practitioner judgement based on open- and close-ended questions
- Calculated total score, but no cut-off/threshold score provided (and thus the practitioner must determine its meaning)
- Calculated total and assessed against cut-off/threshold score
- Affirmative answer(s) to specific question(s)
- Unknown or other scoring procedures
The three most common assessment/scoring procedures are discussed in detail below.
1. Practitioner judgement based on open-ended questions (30%)
Across all FVSTs, the most common assessment/scoring procedure was for practitioners to make a determination of FV based on client responses to closed-ended questions, such as, “Have you been hit, kicked, punched, or otherwise hurt by someone within the past year?” or “Have the police ever been called to your home because of an argument?” Unlike the Danger Assessment instrument designed to assess the likelihood of lethality or attempted lethality in a case of intimate partner violence (Campbell, Webster, & Glass, 2009), there is no quantitative value or scoring associated with answering these FVST questions, nor are there guidelines for which questions are considered most important or should be given greater weight in making a determination of FV. The implied assumption is that practitioners will use the information from the screening tool to make an informed decision about the presence or absence of FV for any given client.
2. Client responds affirmatively to a specific question or set of questions (21%)
The second most common procedure for assessing the presence of abuse was to conclude that there is FV if the client responds affirmatively to a specific question or set of questions. For example, the Ongoing Abuse Screen (OAS, Weiss et al., 2003) is a brief 4-question screen that asks yes/no questions about physical abuse, sexual abuse, and fear of partner. If clients respond affirmatively to any of the 4 questions, the screen for FV is considered positive.
3. Client’s score reaches cut-off threshold (20%)
The third most common assessment/scoring procedure was to score the screening tool and make a positive determination of abuse if the client’s score reaches a specified cut-off threshold. For example, one of the most common FVSTs used in the health care sector, the HITS (Sherin et al., 1998), uses 4 questions measured on a 5-point scale from never to frequently. A positive screen is any score of six or higher (out of a possible 20).
The type of assessment/scoring procedure most commonly used differed by sector. FVSTs developed for family law contexts used practitioner judgement procedures. Ninety percent of the family law FVSTs used relied on practitioner judgement to make a determination of FV (see Figure 7).
Similarly, 60% of the FVSTs designed for mediation used practitioner judgement (see Figure 8). Family law and mediation FVSTs typically did not include scoring instructions or cut-off scores to make a determination of abuse. Rather, practitioners are expected to make a determination based on the client’s responses to a series of questions about abuse against the client, the children, history of police involvement, and so on.
In contrast, FVSTs in the health care sector included both practitioner-judgement and more standardized scoring procedures. Among the health care sector FVSTs analyzed, 41% relied on practitioner-judgement to make a determination of FV. Of these tools, 16 out of 19 were based on patient responses to close-ended questions. Over half of the health care sector tools (54%) used a more standardized scoring procedure. Of these tools, 15 out of 25 would result in a positive screen for violence if the patient responded in the affirmative to a specific question or set of questions.
Positive Screen Follow-up Procedures
Once a practitioner determines that FV is present, it is important that s/he use this information both to inform her/his practice (such as the decision of whether to modify or terminate mediation) and to provide appropriate support to women (and their children) experiencing abuse.
Approximately 60 of the analyzed tools (68%) mentioned the need for follow up post-screening or outlined some kind of follow-up procedure.Footnote 16 Often these procedures involved referrals to agencies specializing in domestic violence cases (e.g., domestic violence centres, shelters), which would themselves conduct further questioning or would engage in the risk assessment. These follow-up procedures may explain, to some extent, why some tools were not comprehensive according to our review process and why these tools did not themselves list any risk-assessment follow-up questions. In fact, the WEB (Smith et al., 1995) and STaT (Paranjape & Liebschutz, 2003) tools were the only ones identified as “least comprehensive” in Table 2 for which the researchers could not locate follow-up questions or procedure suggestions.Footnote 17
Some organizations have developed internal follow-up tools to supplement the screening tool. These are often designed for risk assessment and/or safety planning. There are also reporting/documentation tools and protocols. For example, the New York City Children’s Services Domestic Violence Screening for Parents,which includes a number of checklist questions, includes the instructions:
If domestic violence is present or suspected, conduct the Assessment for Identified Survivor of DV and follow with safety planning and/or other action steps. Once this is completed, review the Assessment and safety plan with your Supervisor and/or DV Specialist and after determining that it is safe to do so, conduct the Assessment for Identified Abusive Partner (n.d., p.3).
Other tools did not list particular procedures to put in place but referenced the need for each institution to establish the appropriate protocols for what should happen when a client screens positive for domestic violence (Furbee et al., 1998).
FVSTs developed for mediation typically included instructions for mediators and family law practitioners to terminate mediation or to recommend against mediation when FV is identified (e.g., Michigan Supreme Court, 2006; New Brunswick Department of Justice, 2011; North Dakota Supreme Court, 2018; see Rossi et al., 2015 for overview of research on mediation and FV). The Michigan Supreme Court’s (2006) Domestic Violence and Child Abuse/Neglect Screening for Domestic Relations Mediation protocol clearly outlines the concern behind these instructions:
Mediation presumes that participants can maintain a balance of power with the help of a mediator in order to reach a mutually satisfactory resolution of a dispute. When domestic violence is present among parties in a dispute, the abuser’s desire to maintain power and control over the victim is inconsistent with the method and objective of mediation. Fear of the abuser may prevent the victim from asserting needs, and the occasion of mediation may give abusers access to victims, which exposes the victim, the children, and the mediator to a risk of violence (Michigan Supreme Court, 2006, p. 1).
Recommended FV Screening Practices
As part of our analysis, we reviewed each FVST and accompanying article (if available) for FV screening practices recommended or endorsed by the FVST developers/authors (i.e., included in the instructions for practitioners using the tool). See Table 3 for recommended practices for practitioner expertise, FVST administration, and post-screening practices.
Category | Practice | No. of Organizations/Authors who Endorsed/Included Practice in Tool |
---|---|---|
Practitioner Expertise | Practitioners should receive training on specific tool and/or FV more broadly |
12 (14%) Examples:
|
Practitioners should recognize/understand the dynamics of abuse |
2 (2%)
|
|
Admin-istration | Screening should be administered in private, confidential setting and/or the practitioner should create safe setting |
33 (38%) Examples:
|
Screening should be administered in person/Screening should not be completed over the phone |
6 (7%) Examples:
|
|
Practitioner should inform client that screening is voluntary |
8 (9%) Examples:
|
|
Practitioner/client should not skip any questions |
2 (2%)
|
|
Practitioner should demonstrate cultural awareness (e.g., use appropriate language, be aware of cultural values, use interpreter) |
7 (8%) Examples:
|
|
Practitioner should administer screening as part of routine health history or initial interview/intake process |
16 (19%) Examples:
|
|
Practitioner should provide preamble to screening that informs client that FV screening is universal and routine |
20 (23%) Examples:
|
|
Practitioner/screening tool provides definition of FV/DV/IPV |
1 (1%)
|
|
Post-Screening Practices | Following positive screen, practitioners should affirm and validate the client’s experiences (e.g., acknowledge this is difficult, thank her for sharing, acknowledge this is not her fault) |
12 (14%) Examples:
|
Following positive screen, practitioner should assess client risk/safety (e.g., complete a safety plan) |
30 (35%) Examples:
|
|
Practitioners should offer resources and/or provide referrals |
42 (49%) Examples:
|
|
Practitioners should respect a client’s choice to refuse disclosure |
3 (3%)
|
Note: The results presented in the above table are intended to highlight broad themes for recommended practices rather than report the results of a systematic review/analysis. The totals for each category were calculated based on the literature reviewed and the information about the tool that was available to the researchers; as a result, there may be tools that were omitted from a particular category.
- Date modified: